DEVELOPMENT OF FORESTED WETLAND ECOLOGICAL FUNCTIONS
IN A HYDROLOGICALLY CONTROLLED FIELD EXPERIMENT IN
VIRGINIA, USA
____________________________________
A Dissertation
Presented to
The Faculty of the School of Marine Science
The College of William & Mary in Virginia
In Partial Fulfillment
Of the Requirements for the Degree of
Doctor of Philosophy
____________________________________
By
Herman W. Hudson III
2016
APPROVAL SHEET
This dissertation is submitted in partial fulfillment of
The requirements for the degree of
Doctor of Philosophy
_____________________________________
Herman W. Hudson III
_____________________________________
James E. Perry, Ph.D.
Committee Chairman/Advisor
_____________________________________
W. Michael Aust, Ph.D.
Virginia Polytechnic University
_____________________________________
Elizabeth A. Canuel, Ph.D.
_____________________________________
Randolph M. Chambers, Ph.D.
Department of Biology, College of William & Mary
_____________________________________
Frank P. Day, Jr. Ph.D.
Old Dominion University
DEDICATION
To all the citizen scientists who made this project possible.
iii
TABLE OF CONTENTS
ACKNOWLEDGMENTS ........................................................................................ vii
LIST OF TABLES................................................................................................... viii
LIST OF FIGURES ..................................................................................................xiv
ACRONYMS AND ABBREVIATIONS ..................................................................xix
ABSTRACT .............................................................................................................. xx
CHAPTER 1 : INTRODUCTION AND LITERATURE REVIEW..................................2
Forested Headwater Wetlands ......................................................................................3
Wetland Functions and Services ..................................................................................7
Downstream Connections ............................................................................................9
Wetland Loss ............................................................................................................. 11
Wetland Regulations and Protection .......................................................................... 12
Mitigation Sequence .............................................................................................. 12
Compensatory Mitigation Ecological Performance Standards................................. 14
Wetland Restoration .................................................................................................. 16
Assessment of Wetland Restoration ........................................................................... 16
Tree Establishment in Restored Wetlands .................................................................. 18
Natural Colonization .............................................................................................. 18
Planting ................................................................................................................. 20
Evaluating Planted Tree Survival and Growth............................................................ 27
Survival ................................................................................................................. 27
Morphology ........................................................................................................... 28
Biomass ................................................................................................................. 29
Growth................................................................................................................... 32
Factors Affecting Tree Survival and Growth .............................................................. 34
Prior to Outplanting ............................................................................................... 35
During Outplanting ................................................................................................ 39
Following Outplanting ........................................................................................... 42
Study Species ............................................................................................................ 50
Betula nigra L. ....................................................................................................... 50
Liquidambar styraciflua L...................................................................................... 52
Platanus occidentalis L. ......................................................................................... 53
Salix nigra Marshall ............................................................................................... 54
Quercus bicolor Willd............................................................................................ 55
Quercus palustris Münchh. .................................................................................... 55
Quercus phellos L. ................................................................................................. 56
Structure of Dissertation ............................................................................................ 58
Literature Cited.......................................................................................................... 61
CHAPTER 2 : EXPERIMENTAL FORESTED WETLAND PLANTED TREE
SURVIVAL AND MORPHOLOGY ............................................................................. 88
Abstract ..................................................................................................................... 89
Introduction ............................................................................................................... 90
Methods..................................................................................................................... 95
Study Site .............................................................................................................. 95
iv
Soil Sampling ........................................................................................................ 96
Planting Material.................................................................................................... 97
Species Grouping ................................................................................................... 99
Survival and Morphometric Measurements ............................................................ 99
Statistical Analysis ............................................................................................... 101
Results ..................................................................................................................... 103
Stocktype Comparison of Initial Morphology ....................................................... 103
Stocktype Comparison Over 5 Years .................................................................... 104
Species Group Comparison Over 5 Years............................................................. 109
Cell Comparison After 5 Years ............................................................................ 111
Discussion ............................................................................................................... 116
Stocktype Comparison of Initial Morphology ....................................................... 117
Stocktype Comparison Over 5 Years .................................................................... 118
Species Group Comparison Over 5 Years............................................................. 121
Cell Comparison After 5 Years ............................................................................ 122
Conclusions ............................................................................................................. 126
Literature Cited........................................................................................................ 128
Tables ...................................................................................................................... 143
Figures .................................................................................................................... 159
CHAPTER 3 : WOODY BIOMASS DEVELOPMENT OF SEVEN MID-ATLANTIC
SPECIES GROWN UNDER THREE HYDROLOGIC CONDITIONS OVER 6 YEARS
.................................................................................................................................... 177
Abstract ................................................................................................................... 178
Introduction ............................................................................................................. 179
Methods................................................................................................................... 181
Biomass Sampling ............................................................................................... 181
Biomass Estimation Model Development ............................................................. 183
Statistical Analysis ............................................................................................... 185
Results ..................................................................................................................... 186
Root-to-Shoot Ratio ............................................................................................. 186
Biomass Estimation Model Implementation ......................................................... 188
Discussion ............................................................................................................... 191
Root-to-Shoot Ratio ............................................................................................. 191
Elemental Concentration ...................................................................................... 192
Biomass Estimation Model Development ............................................................. 193
Accumulated Biomass .......................................................................................... 195
Conclusions ............................................................................................................. 198
Literature Cited........................................................................................................ 201
Tables ...................................................................................................................... 206
Figures .................................................................................................................... 215
CHAPTER 4 : EFFECT OF HYDROLOGIC CONDITIONS ON ABSOLUTE AND
RELATIVE GROWTH RATES OF SEVEN WETLAND TREE SPECIES ................ 219
Abstract ................................................................................................................... 220
Introduction ............................................................................................................. 221
Methods................................................................................................................... 223
Biomass Estimation Model .................................................................................. 223
v
Growth Rate Calculations .................................................................................... 224
Comparing Growth Rates ..................................................................................... 225
Results ..................................................................................................................... 226
Biomass Cumulative Growth ............................................................................... 226
Absolute Growth Rate (AGR) .............................................................................. 227
Relative Growth Rate (RGR) ............................................................................... 228
Discussion ............................................................................................................... 229
Biomass Cumulative Growth ............................................................................... 229
Absolute Growth Rate (AGR) .............................................................................. 230
Relative Growth Rate (RGR) ............................................................................... 232
Conclusions ............................................................................................................. 234
Literature Cited........................................................................................................ 235
Tables ...................................................................................................................... 240
Figures .................................................................................................................... 242
CHAPTER 5 : DEVELOPING AN ECOLOGICAL PERFORMANCE STANDARD
FOR WOODY VEGETATION IN COMPENSATORY MITIGATION WETLANDS OF
VIRGINIA .................................................................................................................. 246
Abstract ................................................................................................................... 247
Introduction ............................................................................................................. 248
Ecological Performance Standards ....................................................................... 249
Virginia EPS ........................................................................................................ 250
Methods................................................................................................................... 253
Existing EPS Relationship to Biomass Accumulation ........................................... 253
Development of Additional Woody EPS .............................................................. 253
Results ..................................................................................................................... 253
Sapling Survival EPS (>80% survival) and Stem Density EPS (>990 stems/ha) ... 253
Height Growth EPSs (>10%/year or > 1.5 m after 5 years) ................................... 254
Canopy Cover EPS (>30% Canopy Closure) ........................................................ 254
Development of Additional Woody EPS .............................................................. 255
Discussion ............................................................................................................... 256
Sapling Survival and Stem Density EPS ............................................................... 256
Height Growth EPSs ............................................................................................ 257
Canopy Cover EPS .............................................................................................. 258
Development of Additional Woody EPS .............................................................. 259
Literature Cited........................................................................................................ 263
Tables ...................................................................................................................... 267
Figures .................................................................................................................... 269
CHAPTER 6 : ECONOMIC ANALYSIS AND OVERALL CONCLUSIONS ............ 276
Abstract ................................................................................................................... 277
Economic Analysis .................................................................................................. 278
Initial Planting Density ............................................................................................ 281
Overall Conclusions ................................................................................................ 282
Literature Cited........................................................................................................ 284
Tables ...................................................................................................................... 286
VITA .......................................................................................................................... 289
vi
ACKNOWLEDGMENTS
I first and foremost would like to thank my advisor, Dr. Jim Perry, for the guidance, wit,
and wisdom he provided me during my time at VIMS. I am thankful for all of the
opportunities that you selflessly provided me and I enjoyed our times together in the field
identifying plants, teaching students, and learning about the natural world.
I am also deeply grateful to the 200+ citizen scientists who assisted with this project over
7 years. These friends, family members, Master Naturalists, Master Gardeners, Boy
Scouts, Girl Scouts, and students volunteered their free time to help monitor these trees.
Your herculean efforts in the field to measure all of the trees year after year yielded
fantastic data that is the foundation of this research. I am very glad that I was able to meet
and work with all of you and I hope that you can see your hard work reflected in this
research.
I would like to acknowledge Mike Aust, Liz Canuel, Randy Chambers, and Frank Day
for serving on my Ph.D. committee. Your support, guidance and encouragement greatly
enhanced this research and helped me become a better scientist.
I would like to thank the Virginia Department of Forestry for allowing us to establish our
field site at the New Kent Forestry Center. I would like to thank the staff at the New Kent
Forestry Center for their assistance with this project. I am particularly grateful to Jef
Stout and Paul Reier for the many times they helped fix the irrigation system and for their
help with the destructive harvesting. I would like to thank the Peterson Family
Foundation and Wetland Studies and Solutions, Inc. for providing the funding for this
project which allowed me to pursue this Ph.D. In particular, I would like to thank Mike
Rolband for personally supporting this research and providing insight into the wetland
regulatory framework.
I would like thank the former members of the Perry Nut House, Chris Hauser, Sean
Charles, and Lori Sutter, who helped get this project off the ground and running and
helped me along my way through many insightful conversations. The list of friends, here
at VIMS and elsewhere, who helped me in this endeavor, is too long to list, but I would
like to thank in particular, Eric Miller, Tim Russell, Jen Elliot, Doug DeBerry, Jackie
Roquemore, Eli Wright, and Jon Lefcheck. Your technical assistance and friendly
motivation kept this project and me going. Also, my constant mentor and friend, Rob
Atkinson, provided much needed and always helpful advice along the way.
I would like to thank my parents for supporting me whole heartily throughout my
academic career and most importantly, my wife Jeanna for her sacrifices, patience, and
enduring love. Last but not least, our daughter Violet, who is the light of my world,
I thank for coming into our lives and providing us with unending joy.
vii
LIST OF TABLES
CHAPTER 2
Table 2-1. Description of environmental parameters for the three experimental cells.
Numbers represent averages with associated standard deviations. Soil parameters
represent 44 samples taken in each cell. ....................................................................... 143
Table 2-2. Initial morphologies of planting stock and standard deviations. Same letters
denote no significant difference between stocktypes for each species (p>0.05). ........... 144
Table 2-3. Results of type 3 tests of fixed effects for species and stocktype for each
morphological variable over 5 years. ........................................................................... 145
Table 2-4. Results of type 3 tests of fixed effects for successional group and species for
each morphological variable over 5 years. ................................................................... 146
Table 2-5. Results of type 3 tests of fixed effects for cell for each morphological variable
after 5 years. ................................................................................................................ 147
Table 2-6. Number of species/measurement combinations that exhibited a particular
outcome when comparing stocktypes within each cell. The 28 species/measurement
combinations are 7 species paired with each of three morphological measurements
(CSAG, H, CD) and survival (e.g. B. nigra H, S. nigra survival, etc.) that were monitored
over 5 years. The outcomes (>,<,=) result from post-hoc comparisons of stocktypes (BR
vs GAL & TB vs GAL & BR vs TB) for each species/measurement combination. Percent
represents percentage of occurrence of each outcome for each group of stocktype posthoc comparisons (e.g. GAL>BR in 60.7% (17) of the 28 post-hoc BR vs. GAL
comparisons in the Ambient cell). Total represents sum of outcomes across all cells and
percent occurrence of outcomes for groups of post-hoc comparisons. See following tables
viii
representing comparisons of stocktypes for each individual species/measurements
combination. ................................................................................................................ 148
Table 2-7. Number of stocktype/measurement combinations that exhibited a particular
outcome when comparing successional groups within each cell. The 12
stocktype/measurement combinations are 3 stocktypes paired with each of three
morphological measurements (CSAG, H, CD) and survival (e.g. BR H, GAL survival,
etc.) that were monitored over 5 years. The outcomes (>,<,=) result from comparison of
successional groups (primary vs. secondary) for each stocktype/measurement
combination. Percent represents percentage of occurrence of each outcome (e.g. Pri>Sec
in 75% (9) of the 12 successional group comparisons in the Ambient cell). Total
represents sum of outcomes across all cells and percent occurrence of outcomes for
successional group comparisons. See following graphs representing comparisons of
successional groups for each individual stocktype/measurement combination. ............. 149
Table 2-8. Number of species/stocktype combinations that exhibited a particular outcome
when comparing cells for survival and morphological measurements (CSAG, CD, H)
after 5 years. The 21 species/stocktype combinations are 7 species paired with BR, GAL
and TB stocktypes (e.g. B. nigra BR, S. nigra TB, etc.). However, due to mortality all 21
combinations are not represented for all comparisons. The outcomes (>,<,=) result from
post-hoc comparisons of morphological measurements among cells (AMB vs. SAT,
AMB vs. FLD, SAT vs. FLD) for each species/stocktype combinations. Survival was not
compared statistically and represents absolute differences. Percent represents percentage
of occurrence of each outcome for each group of cell comparisons (e.g. AMB CSAG >
SAT CSAG in 42.9% (9) of the 21 post-hoc AMB vs. SAT comparisons). Total
ix
represents sum of outcomes across survival and morphological measures and percent
occurrence of outcomes for cell groups of post-hoc comparisons. See following tables
representing comparisons of cells for each individual species/stocktype combination. . 150
Table 2-9. Species exhibiting each outcome within each cell. < and > indicate significant
difference in percent survival (See Figure 2-11). Total represents a count of how many
times each outcome occurred across all cells. ............................................................... 151
Table 2-10. Species exhibiting each outcome for three stocktypes. < and > indicate
significant difference in percent survival (See Figure 2-11). Total represents a count of
how many times each outcome occurred across all stocktypes. .................................... 152
Table 2-11. Number of species exhibiting each outcome within each cell. < and >
indicate significant difference in stem cross-sectional area at groundline (CSAG) (See
Figure 2-12). Total represents a count of how many times each scenario occurred across
all cells. ....................................................................................................................... 153
Table 2-12. Number of species exhibiting each outcome for three stocktypes. < and >
indicate significant difference in stem cross-sectional area at groundline (CSAG) (See
Figure 2-12). Total represents a count of how many times each scenario occurred across
all stocktypes. .............................................................................................................. 154
Table 2-13. Number of species exhibiting each outcome within each cell. < and >
indicate significant difference in crown diameter (See Figure 2-13). Total represents a
count of how many times each outcome occurred across all cells. ................................ 155
Table 2-14. Number of species exhibiting each outcome for three stocktypes. < and >
indicate significant difference in crown diameter (See Figure 2-13). Total represents a
count of how many times each outcome occurred across all stocktypes. ....................... 156
x
Table 2-15. Number of species exhibiting each outcome within each cell. < and >
indicate significant difference in height (See Figure 2-14). Total represents a count of
how many times each outcome occurred across all cells............................................... 157
Table 2-16. Number of species exhibiting each outcome for three stocktypes. < and >
indicate significant difference in height (See Figure 2-14). Total represents a count of
how many times each outcome occurred across all stocktypes. .................................... 158
Chapter 3
Table 3-1. Morphological averages (standard deviation in parentheses) of trees
destructively harvested in 2011 and 2014. Species specific and successional group values
are averaged across three cells. .................................................................................... 206
Table 3-2. Average above (AGB) and belowground biomass (BGB), total biomass, and
root-to- shoot ratio (R:S) (standard deviation in parentheses) of trees destructively
harvested in 2011 and 2014. Species specific and successional group values are averaged
across three cells. ......................................................................................................... 207
Table 3-3. Average percentage carbon, nitrogen and C:N ratio of subsamples from trees
destructively harvested in 2014. Species specific and successional group values are
averaged across three cells. Values in parentheses represent standard deviation. .......... 208
Table 3-4. Results of fitting Eq. 1 (Y=aXb + ε) from saplings destructively harvested in
2011. Y=Total dry belowground biomass (kg). X=Total dry aboveground biomass
(excluding leaves) (kg). ............................................................................................... 209
Table 3-5. Results of fitting Eq. 1 (Y=aXb + ε) to saplings destructively harvested in 2011
and 2014. Y=Total dry biomass (excluding leaves) (kg). X= Equivalent stem crosssectional diameter at groundline (ESD) (cm)................................................................ 210
xi
Table 3-6. Average initial biomass estimates for each species/stocktype combination.
Same letters represent no significant difference among stocktype for each species
(p>0.05). n represents initial number of trees planted. .................................................. 211
Table 3-7. Average biomass accumulated 6 years following planted estimated for each
stocktype/species combination. Same letter represent no significance difference among
stocktype for each species within each cell (p>0.05). n represents count of live trees after
6 years. ........................................................................................................................ 212
Table 3-8. Average biomass accumulated 6 years following planted estimated for each
species (incorporating all stocktypes). Average biomass of species is compared within
cells and cells are compared for each species. Same letters represent no significant
difference (p<0.05). n represents count of live trees after 6 years. ................................ 213
Table 3-9. Average biomass accumulated 6 years following planted estimated for primary
and secondary groups (incorporating all species and stocktypes). Average biomass of
species groups are compared within cells and cells are compared for each species group.
Same letters represent no significant difference (p<0.05). n represents count of live trees
after 6 years. ................................................................................................................ 214
Chapter 4
Table 4-1. Initial biomass with standard deviation in parentheses. Same letters represent
no significant difference (p<0.05). N represents number of planted saplings. ............... 240
Table 4-2. Species AGR and RGR at standard masses. Values in parentheses represent
±95% confidence intervals. .......................................................................................... 241
xii
Chapter 5
Table 5-1. Total biomass (aboveground and belowground) accumulation (Mg/ha) of all
species/stocktype combinations across the 7-year study. .............................................. 267
Table 5-2. Total CSAG (m2/ha) across the 7-year study. .............................................. 268
Chapter 6
Table 6-1. Source of each species and stocktype and associated costs. Installation and
miscellaneous costs were obtained from Wetland Studies and Solutions, Inc and represent
prices in 2012. Material price is based on 2008 purchase price. Miscellaneous costs
include mulch, agriform fertilizer, shipping, and terrasorb. No mulch, fertilizer or
terrasorb were used in this study, but these products are commonly used in practice. ... 286
Table 6-2. Species/stocktype combinations return on investment ranked within each cell.
.................................................................................................................................... 287
Table 6-3. Initial densities required to meet 5.2 m2/ha CSAG EPS in FLD. ND represents
species/stocktype combinations for which initial density was not determined due to 1 or 0
surviving individuals. .................................................................................................. 288
xiii
LIST OF FIGURES
CHAPTER 2
Figure 2-1. A) The experimental site is located in the Mid-Atlantic Region of the United
States, in the Upper Coastal Plain Province of Virginia. B) The site is on an upland
terrace in the Virginia Department of Forestry, New Kent Forestry Center. C) The
hydrologically distinct cells (ambient (AMB), saturated (SAT) and flooded (FLD)) are 49
m x 95 m in size. ......................................................................................................... 159
Figure 2-2. Soil bulk density across experimental site. ................................................. 160
Figure 2-3. Percentage sand across experimental site ................................................... 161
Figure 2-4. Percentage silt across experimental site...................................................... 162
Figure 2-5. Percentage clay across experimental site .................................................... 163
Figure 2-6. Soil percentage carbon across experimental site. ........................................ 164
Figure 2-7. Soil percentage nitrogen across experimental site. ..................................... 165
Figure 2-8. Soil percentage phosphorus across experimental site.................................. 166
Figure 2-9. Average crown diameter of stocktypes for three cells. Line represents mean
of seven species and ribbons represent 95% confidence interval. ................................. 167
Figure 2-10. Average height of successional groups for all stocktypes across cells. Line
represents mean and ribbons represent 95% confidence interval. ................................. 168
Figure 2-11. Simple effects model results for survival of stocktypes (lines) among species
(columns) and cells (rows). X-axis represents time since planting. For stocktype
comparisons, no stars represent no significant difference (p-value > 0.05) while *
indicates p-value ≤ 0.05, ** indicates p-value ≤ 0.01 and *** indicates p-value ≤ 0.001.
.................................................................................................................................... 169
xiv
Figure 2-12. Simple effects model results for CSAG of stocktypes (lines) among species
(columns) and cells (rows). X-axis represents time since planting. Ribbons represent
95% confidence interval. Line represents mean. For stocktype comparisons, no stars
represent no significant difference (p-value > 0.05) while * indicates p-value ≤ 0.05, **
indicates p-value ≤ 0.01 and *** indicates p-value ≤ 0.001. ......................................... 170
Figure 2-13. Simple effects model results for CD of stocktypes (lines) among species
(columns) and cells (rows). X-axis represents time since planting. Ribbons represent 95%
confidence interval. Line represents mean. For stocktype comparisons, no stars represent
no significant difference (p-value > 0.05) while * indicates p-value ≤ 0.05, ** indicates pvalue ≤ 0.01 and *** indicates p-value ≤ 0.001. ........................................................... 171
Figure 2-14. Simple effects model results for H of stocktypes (lines) among species
(columns) and cells (rows). X-axis represents time since planting. Ribbons represent 95%
confidence interval. Line represents mean. For stocktype comparisons, no stars represent
no significant difference (p-value > 0.05) while * indicates p-value ≤ 0.05, ** indicates pvalue ≤ 0.01 and *** indicates p-value ≤ 0.001. ........................................................... 172
Figure 2-15. Simple effects model results for survival (time until death) of successional
group (lines) among cells (columns) and stocktype (rows). Ribbons represent 95%
confidence interval. Line represents mean. For successional group comparisons, no stars
represent no significant difference (p-value > 0.05) while * indicates p-value ≤ 0.05, **
indicates p-value ≤ 0.01 and *** indicates p-value ≤ 0.001. ......................................... 173
Figure 2-16. Simple effects model results for CSAG of successional group (lines) among
cells (columns) and stocktype (rows). Ribbons represent 95% confidence interval. Line
represents mean. For successional group comparisons, no stars represent no significant
xv
difference (p-value > 0.05) while * indicates p-value ≤ 0.05, ** indicates p-value ≤ 0.01
and *** indicates p-value ≤ 0.001. ............................................................................... 174
Figure 2-17. Simple effects model results for CD of successional group (lines) among
cells (columns) and stocktype (rows). Ribbons represent 95% confidence interval. Line
represents mean. For successional group comparisons, no stars represent no significant
difference (p-value > 0.05) while * indicates p-value ≤ 0.05, ** indicates p-value ≤ 0.01
and *** indicates p-value ≤ 0.001. ............................................................................... 175
Figure 2-18. Simple effects model results for H of successional group (lines) among cells
(columns) and stocktype (rows). Ribbons represent 95% confidence interval. Line
represents mean. For successional group comparisons, no stars represent no significant
difference (p-value > 0.05) while * indicates p-value ≤ 0.05, ** indicates p-value ≤ 0.01
and *** indicates p-value ≤ 0.001. ............................................................................... 176
Chapter 3
Figure 3-1. Results of fitting Eq. 1 (Y=aXb + ε) for primary successional species
destructively harvested in 2011. Y=Total dry belowground biomass (coarse roots) (kg)
and X= Total dry aboveground biomass (kg) (excluding leaves) (See Table 3-4 for results
of fitting model)........................................................................................................... 215
Figure 3-2. Results of fitting Eq. 1 (Y=aXb + ε) for secondary successional species
destructively harvested in 2011. Y=Total dry belowground biomass (coarse roots) (kg)
and X= Total dry aboveground biomass (excluding leaves) (kg) (See Table 3-4 for results
of fitting model)........................................................................................................... 216
Figure 3-3. Results of fitting Eq. 1 (Y=aXb + ε) for primary successional species
destructively harvested in 2011 and 2014. Y=Total biomass (kg) and X=Equivalent stem-
xvi
cross sectional diameter at groundline (cm) (See Table 3-5 for results of fitting model).
.................................................................................................................................... 217
Figure 3-4. Results of fitting Eq. 1 (Y=aXb + ε) for secondary successional species
destructively harvested in 2011 and 2014. Y=Total biomass (kg) and X=Equivalent stemcross sectional diameter at groundline (cm) (See Table 3-5 for results of fitting model).
.................................................................................................................................... 218
Chapter 4
Figure 4-1. Average predicted biomass of seven species (columns) in each cell (rows).
Shading represents 95% confidence intervals around mean. ......................................... 242
Figure 4-2. Natural log transformed biomass of seven species (columns) in each cell
(rows). Red line represents the results of fitting the monomolecular function. .............. 243
Figure 4-3. Absolute growth rate (AGR) of seven species (columns) in each cell (rows) as
a function of mass. Shading represents 95% confidence intervals around mean. ........... 244
Figure 4-4. Relative growth rate (RGR) of seven species (columns) in each cell (rows) as
a function of mass. Shading represents 95% confidence intervals around mean. ........... 245
Chapter 5
Figure 5-1. Relationship between percentage survival and average biomass accumulated
after 7 years for individual species/stocktype combinations. Points represent
species/stocktype combinations. Line represents simple linear regression and shading
represents 95% confidence interval. ............................................................................. 269
Figure 5-2. Relationship between 5-yr average percentage change in height and average
biomass after 5 years. Points represent species/stocktype combinations. ...................... 270
xvii
Figure 5-3. Relationship between canopy cover and average biomass after 7 years. Points
represent species/stocktype combinations. Line represents linear regression. ............... 271
Figure 5-4. The relationship between ESD and H after 7 years for primary species. Line
represents linear regression. ......................................................................................... 272
Figure 5-5. The relationship between ESD (cm) and H (cm) after 7 years for secondary
species. Line represents linear regression. .................................................................... 273
Figure 5-6. The relationship between ESD (cm) and CD (cm) after 7 years for primary
species. Line represents linear regression. .................................................................... 274
Figure 5-7. The relationship between ESD (cm) and CD (cm) after 7 years for secondary.
Line represents linear regression. ................................................................................. 275
xviii
ACRONYMS AND ABBREVIATIONS
AGB
Leafless woody aboveground biomass
AGR
Absolute growth rate
AMB
Ambient Cell
BEM
Biomass estimation model
BGB
Coarse root (>2mm) belowground biomass
BR
Bare root
CD
Crown diameter (cm)
CMS
Compensatory wetland mitigation sites
CSAG
Stem cross-sectional area at groundline (cm2)
CWA
Clean Water Act
dbh
Diameter at breast height
EPS
Ecological performance standards
ESD
Equivalent stem cross-sectional diameter at groundline (cm)
FHW
Forested headwater wetland
FLD
Flooded Cell
GAL
1-gallon container
H
Height (cm)
PRI
Primary successional species group
r:s
Root to shoot ratio
RGR
Relative growth rate
SAT
Saturated Cell
SEC
Secondary successional species group
TB
Tubeling
USACE
United States Army Corps of Engineers
USEPA
United States Environmental Protection Agency
VADEQ
Virginia Department of Environmental Quality
xix
ABSTRACT
Forested headwater wetlands provide numerous ecosystem functions and services
and are often disturbed, impacted, or destroyed due to their position in the landscape.
This has led to many successful and unsuccessful attempts to restore these important
ecosystems. Returning trees to these restored ecosystems has proven to be particularly
challenging. To increase successful forested headwater wetland restoration, this
dissertation recommends tree species and stocktypes that could be planted to return lost
ecosystem structure and ecological functions. This dissertation focuses on a created
experimental system that investigates the responses of seven species of native wetland
trees planted using three stocktypes in three distinct hydrologic conditions (normal
rainfall, constantly saturated root zone, and flooded above the root crown). This
dissertation has been divided into six chapters to facilitate investigation of several
research questions.
The first chapter provides an in depth literature review of forested wetland
restoration, highlighting wetland regulations, factors influencing planted trees and the
various responses that will be measured in the subsequent chapters. The second chapter
investigates survival and morphological development five years following planting and
provides recommendations for planting based on these responses.
The third chapter in this dissertation develops a biomass estimation model that
relates total biomass to stem morphology based on destructive harvests. The model is
then used to determine the biomass accumulated of all living trees after six years
allowing for a robust evaluation of each species and stocktypes’ response to the
hydrologic treatments. Building upon the model developed in chapter three, the fourth
chapter calculates biomass accumulation rates which provide greater insight to how these
species develop and change through time.
The fifth chapter investigates the regulatory context for wetland compensatory
mitigation standards in Virginia and provides an additional ecological performance
standard for evaluating compensatory mitigation site success. This performance standard
is based on stem morphology and provides a robust method for evaluating the ecological
functioning of compensatory mitigation sites. Finally the sixth chapter provides a brief
economic analysis that examines the return on investment for each species and stocktype
used in the study and provides overarching conclusions based on the preceding chapters.
The combination of these chapters provides an in-depth analysis of the responses
of these planted trees to the experimental hydrologic conditions from the time of planting
through six years of growth. These analyses yielded practical and regulatory
recommendations that can enhance forested wetland restoration.
xx
DEVELOPMENT OF FORESTED WETLAND ECOLOGICAL FUNCTIONS
IN A HYDROLOGICALLY CONTROLLED FIELD EXPERIMENT IN
VIRGINIA, USA
1
CHAPTER 1: INTRODUCTION AND LITERATURE REVIEW
2
There are a wide variety of scientific and legal definitions used to determine
which areas of the landscape qualify as a wetland; however, most encompass three
general components: 1) presence of water at the surface or within the root zone, 2) unique
soil conditions, 3) unique plants, animals, fungi, and bacteria that are adapted to the
presence of water (Mitsch and Gosselink 2007). Within Virginia, there are a large
number of ecosystems that fall under this general definition, from salt marshes in the
coastal plain to bogs and fens in the Blue Ridge Mountains. Each type of wetland has
unique physical, chemical, and biological properties often referred to as the ecosystem
structure. Additionally, each type of wetland performs unique physical, chemical and
biological processes. The processes that take place within an ecosystem are referred to as
ecosystem functions and those functions (or combinations of functions) that are of value
to humans are referred to as ecosystem services. An important group of wetlands in
Virginia are forested headwater wetlands that are valued for their unique landscape
position, large amount of area, and associated ecosystem structure, functions and
services.
Forested Headwater Wetlands
Forested headwater wetlands (FHW) are located in the upper reaches of non-tidal
freshwater streams where average annual stream flow is less than 5 ft3/second (33 CFR
Section 330.2 (d)). The primary hydrologic inputs are groundwater, overland flow, and
precipitation (Brinson et al. 1995, Noble et al. 2011). Several authors define headwater
wetlands as those surrounding 1st and 2nd order streams (smallest streams in the
landscape) using the Strahler method (Rheinhardt et al. 1998, Rheinhardt et al. 2009,
3
Noble et al. 2011, Rheinhardt et al. 2012). In the United States there are approximately
3,821,375 km of 1st and 2nd order streams representing approximately 73% of the total
length of all streams and rivers (Leopold et al. 1964, Brinson 1993) suggesting potential
for a large amount of FHW area. In Virginia, analysis of National Wetland Inventory
aerial maps revealed that there are approximately 156,795 ha of vegetated non-tidal
headwater wetlands surrounding 1st order streams with an addition 51,835 ha surrounding
2nd order streams (Hershner et al. 2003). These headwater wetlands represent
approximately 43% of the 485,623 ha of vegetated wetlands in Virginia (Hershner et al.
2003). Consequently, several authors have investigated various biotic and abiotic
characteristics of headwater wetlands in Virginia and the surrounding states (Glascock
and Ware 1979, Parsons and Ware 1982, Rheinhardt et al. 1998, Rheinhardt et al. 2000,
Brinson et al. 2006, Rheinhardt et al. 2009, Rheinhardt et al. 2012, 2012a).
As the name implies, FHW vegetative structure is dominated by trees and tree
biomass accounts for the majority (>96%) of living vegetative biomass (Rheinhardt et al.
2012). The species of trees found in FHW varies by physiographic province, successional
stage, hydrologic and soil conditions and site history (fire, logging, etc.). Glascock and
Ware (1979) investigated forest composition along small streams in the coastal plain of
Virginia that had high water tables and experienced temporary irregular flooding. They
concluded that the stands were not virgin stands but secondary growth based on the size
of stumps found. The species that ranked within the top three species in at least one stand
based on relative importance values (calculated using dominance and density) listed in
descending order were Acer rubrum (red maple), Fraxinus pennsylvanica (green ash),
Liquidambar styraciflua (sweetgum), Carpinus caroliniana (American hornbeam),
4
Ulmus americana (American elm), Quercus phellos (willow oak), Quercus michauxii
(swamp chestnut oak), Taxodium distichum (bald cypress), Quercus nigra (water oak),
Liriodendron tulipifera (tuliptree), Nyssa sylvatica (blackgum), Fagus grandifolia
(American beech), Ilex opaca (American holly), Betula nigra (river birch), Pinus taeda
(loblolly pine), and Quercus pagoda (cherrybark oak).
Parsons and Ware (1982) investigated soil and vegetation of coastal plain swamps
along small stream bottoms including 13 of the same sites investigated by Glascock and
Ware (1979). The purpose of their study was to measure additional environmental
parameters to determine what factors were influencing the distribution of tree species.
While they found similar species as Glascock and Ware (1979), Parsons and Ware (1982)
concluded that soil chemistry and soil moisture were the most important factors
influencing the distribution of tree species. Sites with higher soil moisture content were
dominated by F. pennsylvanica, A. rubrum and U. americana. Drier sites were dominated
by C. caroliniana, L. styraciflua, and Q. phellos.
Rheinhardt et al. (1998) investigated the geomorphic setting and vegetation of low
order streams within the inner coastal plain of North Carolina in order to provide
guidance for permit decisions, restoration efforts, and judging success of restoration
attempts. The authors of this paper defined 1st and 2nd order streams as headwater
systems. The most important species in the headwater systems of this study were Nyssa
biflora (swamp tupelo), A. rubrum, L. styraciflua, and L. tulipifera. They also found that
headwater systems had smaller drainage basins, narrower floodplains, steeper floodplain
gradients, steeper stream gradients, smaller channel cross-sectional areas, narrower
5
channel widths, higher graminoid cover and lower phosphorus concentrations than
midreach (stream order 3 and 4) systems.
Rheinhardt et al. (2000) described the vegetation of headwater wetlands in the
inner coastal plain of Maryland and Virginia with the purpose of developing the
hydrogeomorphic (HGM) functional assessment model for wetland on 1 st to 3rd order
streams. The most abundant canopy trees were A. rubrum, L. styraciflua, Fraxinus spp.,
N. biflora, L. tulipifera, Ulmus rubra (slippery elm), and U. americana. The density of
canopy trees ranged from 318 to 683 stems/ha. Quercus spp. had lower importance
values in this study than Parsons and Ware (1982) and the authors suggest that this may
have been due to preferential logging of stands or damming by beavers. The sapling
stratum (juveniles of potential canopy trees) was dominated by L. styraciflua and A.
rubrum. The density of saplings ranged from 0 to 3,111 stems/ha. The dominant
subcanopy (shrubs and trees restricted to the understory at maturity) species were C.
caroliniana, Lindera benzoin (northern spicebush) and I. opaca. The density of
subcanopy species ranged from 254 to 7,747 stems/ha.
Rheinhardt et al. (2009) investigated forest structure and composition in 219 sites
from three physiographic provinces (Coastal Plain, Piedmont, and Ridge & Valley) in the
Delaware River, Chesapeake Bay, and Albemarle/Pamlico Sound drainage basins. They
focused on low order riparian ecosystems (1st to 4th order streams) and defined headwater
as 1st to 2nd order streams. The goal was to outline strategies for restoring structure and
function to low-order riparian zone forests. To determine forest biomass, three-dimension
structure and successional status stand basal area (BA) was used. Early successional
systems were defined as BA ≤ 20 m2/ha and mid successional was 20 ≤ BA < 30 m2/ha
6
while late successional was defined as BA ≥ 30 m2/ha. The dominant species within the
late successional stands of the Piedmont region were A. rubrum, L. tulipifera, and
Quercus rubra (northern red oak). They conclude that Quercus spp. may have declined
because of fire suppression and that planting Quercus spp. and other heavy mast species
might facilitate restoration to a more historically accurate composition represented by late
successional stands that provide many habitat and biogeochemical functions.
Wetland Functions and Services
Wetland functions are generally defined as processes that occur in wetlands
(Hruby 1999) and vary based on wetland type, hydrologic and geomorphic conditions
(Brinson 1993), landscape position, season and many other factors. Wetland ecosystem
functions fall into three general categories: hydrological, biogeochemical, and habitat and
food web support (National Research Council (NRC) 1995). Hydrologic functions
include short-term and long-term surface water storage, storage of subsurface water,
moderation of groundwater flow or discharge, dissipation of energy, and maintenance of
high water tables (NRC 1995, Smith et al. 1995). Biogeochemical functions include
transformation, cycling, and storage of elements, retention and storage of dissolved
substances and particulates, accumulation of peat, export of organic carbon and
accumulation of inorganic sediments (NRC 1995, Smith et al. 1995). The functions
related to habitat and food web support include maintenance of characteristic plant and
animal communities (composition, abundance and age structure) and maintenance of
characteristic energy flow (NRC 1995, Smith et al. 1995).
7
Principal ecological functions provided by FHW are retention of sediments (Hupp
et al. 1993), transformation, cycling and retention of elements (carbon, nitrogen,
phosphorus, etc.) (Craft and Casey 2000, Noble et al. 2011), primary and secondary
productivity, water storage, groundwater recharge, as well as plant and animal habitat
(Morley 2008). Examples of services provided by FHW that are of value to humans
include but are not limited to flood mitigation (intercept storm runoff, store floodwater),
water quality enhancement (denitrification, nitrogen and phosphorus retention), timber
production, vegetation harvests, animal products (pelts, food), aesthetics (recreational
hunting), maintenance of biodiversity (rare and threatened species), aquifer recharge, and
air quality enhancement (carbon sequestration) (NRC 1995, Mitsch and Gosselink 2007).
The occurrence and rates of these processes within FHW are dependent upon
hydrologic and geomorphic conditions as well as nutrient and sediment inputs from
surrounding ecosystems (Brinson et al. 1981). Trees, being a major component of FHW,
also play an important role in many of the ecosystem functions and services provided by
FHW. Trees are often the dominant source of primary production (Rheinhardt et al. 2012)
and they heavily contribute to the cycling of elements. Trees in FWH also provide habitat
for plants, animals, fungi, and microbial communities. Living and shed bark, wood, roots,
flowers, fruits, seeds, leaves and sap are consumed by a number of different organisms
including, insects, mammals and birds. Leaf litter and fallen dead wood also provide
nutrients for fungus and other microorganisms in the detrital food web. Trees provide
shelter from weather and predators in the form of tree cavities, leaf litter and fallen dead
wood. Shade provided by trees surrounding streams, estuaries or rivers can reduce air and
water temperature enhancing habitat for aquatic organisms. Trees also provide space for
8
organisms to live, including insects living under and in the bark, birds and mammals
building nests in cavities and branches, caterpillars and other insects building nests in the
crown, and lichen, moss and fungi living on the bark.
Wetlands ecosystem services (those functions or combinations of functions that
are valued by humans) include storing flood waters (Brinson 1993a), enhancing water
quality (Sather and Smith 1984), retaining nutrients (Fisher and Acreman 2004),
providing wildlife corridors and habitat amenities (Balcombe et al. 2005, Shaw and
Fredine 1956), providing erosion control along streams (Silberhorn 1994), providing
intrinsic values such as recreation opportunities (Mitsch and Gosselink 2000, Office of
Technology Assessment (OTA) 1984), and trapping waterborne sediments that help
protect and restore sensitive aquatic ecosystems such as the Chesapeake Bay (USGS
1999). In addition to the value of these wetlands to society, they also play an important
role in the integrity of downstream ecosystems.
Downstream Connections
The functions and services provided by FHW influence the functioning and
integrity of ecosystems downstream mainly through alteration of water quality and export
of organic matter. In Virginia, these ecosystems include streams, rivers, and estuaries
including the Chesapeake Bay.
Headwater wetlands can enhance water quality through nutrient uptake and
retention, denitrification, and sediment trapping. Nutrient uptake and retention prevent
nutrients from entering the water column and are important in reducing eutrophication of
the Chesapeake Bay and surrounding estuaries (Lowrance et al. 1997). Through primary
9
production, nutrients are assimilated into vegetative biomass, preventing those nutrients
from being transported downstream (Dosskey et al. 2010). Trees are important to this
process because they provide long term storage of nutrients in woody tissue and short
term storage in leaves. Headwater wetlands play an important role in reducing
eutrophication because they are often located within agricultural landscapes that can have
increased nutrient loads. Freeman et al. (2007) found that alteration of headwater systems
for agricultural purposes can increase downstream eutrophication and coastal hypoxia.
Chemical processes, such as denitrification, that take place in wetland soil can
also enhance water quality. Denitrification can enhance water quality because
denitrifying bacteria convert nitrate to the nitrogen gases (NO, N 2O and N2), which are
released to the atmosphere (Seitzinger 1988, Paul 2007). Denitrifying bacteria utilize
NO3- rather than oxygen as the terminal electron acceptor; therefore, this process usually
takes place when oxygen concentrations are reduced, such as in wetland soils. Hefting et
al. (2005) found that rates of denitrification accounted for greater nitrogen removal than
plant uptake in wetter forested riparian sites. Trees can enhance denitrification because
their roots provide heterogeneous environments and energy (root exudates) for
denitrifying microbes (Groffman et al. 1996). In addition, tree biomass may contribute to
the organic carbon needed for denitrification. Through denitrification, water quality may
be enhanced because nitrogen, which drives eutrophication in estuaries and coastal water,
can be removed from the water prior to entering stream systems.
Water quality can also be enhanced through the trapping of sediment in FHW.
Excess sediment in stream systems can degrade the condition and functioning of
downstream systems by increasing turbidity, altering plant and animal habitats and
10
increasing sediment bound nutrients and pollutants. Trees help trap sediment because
their large stems and roots slow the flow of surface and ground water, which leads to
sediment trapping; they also stabilize existing soil (Dosskey et al. 2010). Additionally,
leaf litter from trees slows the overland flow of water increasing sediment retention.
Forested headwater wetlands also influence the functioning of downstream
systems through the export of organic matter. These systems can provide a substantial
proportion of the total organic matter found within a stream (Dosskey and Bertsch 1994)
and can contribute significantly higher concentrations of organic carbon when compared
to upland watersheds (Mulholland and Kuenzler 1979). This suggests that organic matter
produced in FHW influences the functioning of the detrital food webs found in adjacent
streams. Large tree debris increases channel bed roughness, which can slow stream
velocity, increase stability and provide habitat for microbes and animals (Harmon et al.
1986, Hilderbrand et al. 1997).
Overall, the ecosystem functions and services performed by FHW are justification
for the protection, preservation and restoration of these important ecosystems. However,
prior to changes in wetland regulations, a large amount of FHW were lost.
Wetland Loss
Between the 1780’s and 1980’s, prior to most wetland regulations, approximately
53% of the estimated 89,435,000 ha of wetlands was lost in the lower 48 states. During
the same time period, Virginia lost approximately 42% of the estimated 748,263 ha of
wetlands present in the 1780’s (Dahl 1990). Most of the wetlands lost in Virginia have
been forested wetlands, which are the most abundant wetland type in Virginia (Tiner and
11
Finn 1986, USGS 1999). Losses were mainly due to drainage activities associated with
agriculture and forestry practices, construction of reservoirs and urban/suburban
development (Dahl and Johnson 1991, Tiner and Finn 1986). Additionally, Brinson et al.
(1981a) estimated that over 70% of U.S. riparian forests have been lost since presettlement time.
Wetland Regulations and Protection
With the realization of the importance of wetlands, impacts to these ecosystems
are currently regulated by the Clean Water Act (CWA) (33 U.S.C. 1344). The Federal
Water Pollution Control Act (1948) was the first major law designed to reduce water
pollution and was later amended and renamed the CWA in 1977. Wetland impacts are
specifically regulated under Section 404 of the CWA which authorized the USACE under
the direction of the United States Environmental Protection Agency (USEPA) to issue
permits regulating the discharge of dredged or fill material into “waters of the United
States”, which includes wetlands (40 CFR Part 230.1). These permits are now used for
any activity that may impact wetlands.
Mitigation Sequence
In order to permit wetland impacts while maintaining the physical, biological and
chemical integrity of the waters of the United States (the overall goal of the CWA) and
fulfilling the 1988 policy goal of ‘No Net Loss’, these regulations now require that
impacts to wetlands follow a mitigation sequence. The mitigation sequence was first
defined in the 1978 Council on Environmental Quality (CEQ) National Environmental
12
Policy Act clarification and includes avoidance, minimization and compensation for
impacts to wetlands. After the impact has been found to be unavoidable and is
minimized, there are four methods of compensating for wetland impacts: 1) restoration of
converted wetlands, 2) creation of new wetlands, 3) enhancement of existing wetlands, 4)
preservation of existing wetlands (listed in descending priority (33 CFR PART 332,
USACE and USEPA 2008). On March 31, 2008, the USACE and USEPA released the
“Final Rule” for Compensatory Mitigation and announced the new priority of methods
for satisfying mitigation requirements (in order of decreasing priority) as 1) purchasing
credits from mitigation banks, 2) payment to an in-lieu-fee program, and 3) permitteeresponsible compensatory mitigation (USACE and USEPA 2008).
Development activities in wetlands in Virginia are regulated by the USACE
through Section 404 of the Clean Water Act, the Virginia Department of Environmental
Quality (VADEQ), through the Virginia Water Protection Permit program and Section
401 of the Clean Water Act; and the Virginia Marine Resources Commission and local
Wetland Boards through the Virginia Tidal Wetlands Act of 1972. All permits are
reviewed by the USEPA, U.S. Fish and Wildlife Service, Virginia Department of Game
and Inland Fisheries, the Virginia Department of Conservation and Recreation, Virginia
Department of Historic Resources, the Virginia Marine Resources Commission and any
other interested and affected agencies. There are exemptions for certain development
activities (e.g. normal farming, silvicultural and ranching activities) that are specified in
Section 404(f)(1).
13
Compensatory Mitigation Ecological Performance Standards
The overall goals of compensatory mitigation sites (CMS) are to replace the
structure and ecosystem functions and services that were lost during the permitted impact
(33 CFR PART 332). The current legislative-mandated method for determining if a CMS
is developing into the desired wetland type and is providing the expected ecological
functions is through meeting project specific ecological performance standards (EPS)
(aka: success criteria, success standards or release criteria) (33 CFR PART 332.5,
USACE and USEPA 2008). Performance standards for all CMSs are required to be clear,
objective, verifiable, based on the best available science and able to be assessed in a
practicable manner (33 CFR PART 332.5, USACE and USEPA 2008).
Ecological performance standards are developed on a project by project basis
(USACE and USEPA 2008). While EPSs vary across USACE districts, the majority are
based on soil, vegetation, and hydrologic indicators from the 1987 Federal Delineation
Manual (Environmental Laboratory 1987) and subsequent editions (Breaux and
Serefiddin 1999, Streever 1999, Matthews and Endress 2008, USACE 2008). Additional
EPSs may include survival of planted stock, specific density (or cover) of herbaceous or
woody plants, and limiting exotic and nuisance plants (Breaux and Serefiddin 1999,
Streever 1999, Matthews and Endress 2008, USACE 2008). Virginia has established
hydrologic, vegetation, and soil EPSs for use in CMSs (USACE Norfolk District and the
VADEQ 2004, VADEQ 2010). The majority of the Virginia’s EPSs are based on the
1987 Federal Delineation Manual (Environmental Laboratory 1987) and subsequent
editions.
14
Virginia has currently implemented a woody height growth EPS for mitigation
banks in particular; however, few projects have implemented this performance standard
(Mike Rolband, personal communication). Virginia is among few states to have
established a tree growth EPS (Streever 1999, Environmental Law Institute 2004). This
may be due to limited information on growth rates of planted trees in created or restored
wetlands (Denton 1990, Niswander and Mitsch 1995, Gamble and Mitsch 2006,
Pennington and Walters 2006, Henderson et al 2009). In order to ensure that EPSs are
being fulfilled for CMS, monitoring and compliance reports are required. The current
monitoring period in Virginia for CMS is to monitor for 6 years over a 10 year period
(typically years 1, 2, 3, 6, 7, and 10) (USACE Norfolk District and VADEQ 2004).
Most EPSs are structural measurements of the vegetative community and/or
physical environment (see above, Wilson and Mitsch 1996) and are not direct
measurements of wetland functions or services (Mitsch and Wilson 1996, Streever 1999,
NRC 2001). Additionally, many performance standards are not adequate indicators of
wetland functions or services (Kentula et al. 1992). Cole (2002) suggested that
measurement of herbaceous cover (a common performance standard) may not serve as an
accurate indicator of the replacement of wetland functions. Therefore, meeting site
specific performance standards does not guarantee that wetland functions and services are
being replaced (the overall goal of compensatory mitigation) (Matthews and Endress
2008). Streever (1999) stated comparisons between performance standards and the ability
of CMSs to replace lost functions are lacking. Several authors have attempted to develop
performance standards that are more closely linked with ecosystem functions and
services (Atkinson et al. 1993, Bedford 1996, Brinson 1996, Brinson et al. 1996, Breaux
15
and Serefiddin 1999, Environmental Law Institute 2004, Faber-Langendoen et al. 2006;
2008, DeBerry and Perry 2015). However, forested CMSs still lack woody EPSs that
quantify or are directly associated with ecosystem functions and services.
Wetland Restoration
Wetlands are restored for a variety of reasons in addition to restoration as part of
the mitigation sequence. These reasons include but are not limited to, failed agriculture
(or other failed land use), timber production, following timber harvest, recreation,
reclamation of disturbed habitat, re-establishment of bird habitat (Ducks Unlimited),
and/or agricultural easements (Agricultural Conservation Easement Program).
Additionally, there are several state and federal goals (Chesapeake 2000 Agreement) that
seek to increase the amount of wetland area through restoration.
Ecosystem restoration is an applied science that focuses on “the process of
assisting the recovery of an ecosystem that has been degraded, damaged or destroyed”
(Society for Ecological Restoration International Science & Policy Working Group
2004). The goals of wetland restoration are to return lost ecosystem structure, functions
and services to the landscape (Cairns Jr. 2000) and for the restored ecosystem to be selfsustaining, resilient, and have connections to other ecosystems.
Assessment of Wetland Restoration
There are many methods used to determine if restored wetlands are meeting these
general goals, and methods can differ based on the underlying reason for restoration. For
CMS, restoration success has been partitioned into regulatory and ecological success
16
(Wilson and Mitsch 1996, Kentula 2000). Successfully fulfilling regulatory obligations
does not guarantee successful functional wetland replacement and vice versa. For
wetlands restored for reasons other than compensatory mitigation, restoration success has
been evaluated using many different methods.
Several studies have found that CMSs are not meeting regulatory obligations. A
study completed by the United States Government Accountability Office (USGAO)
(2005) found that the USACE had failed to receive monitoring reports from 86% of the
surveyed permittee-responsible compensatory mitigation sites, 30% of the surveyed
mitigation bank sites, and 17% of the surveyed in-lieu-fee mitigation programs. A
national study completed by the National Research Council (NRC) (2001) found that
approximately 50% of surveyed CMSs failed to meet prescribed performance standards
and concluded that “the goal of no net loss of wetlands is not being met for wetland
functions by the mitigation program.” Several regional studies have also suggested that
forested CMS did not meet their prescribed vegetation performance standards (Brown
and Veneman 2001, Cole and Shafer 2002, Matthews and Endress 2008, Tischew et al.
2010).
Multiple regional and local studies have found that wetlands restored for a variety
of purposes are not ecologically successful and several studies have suggested that
wetland structure, functions and services are not developing in restored forested wetlands
(Atkinson and Cairns Jr. 2001, Brown and Veneman 2001, Cole et al. 2001, Sudol and
Ambrose 2002, Atkinson et al. 2005, Hoeltje and Cole 2007, Atkinson et al. 2010,
Moreno-Mateos et al. 2012, Stefanik and Mitsch 2012). Important to the present
investigation, numerous studies have found that tree density and tree growth were
17
significantly lower for restored sites as compared to conditions prior to conversion or
nearby mature forested wetlands (Brown and Veneman 2001, NRC 2001, Cole and
Shafer 2002, Morgan and Roberts 2003, Sharitz et al. 2006, Matthews and Endress 2008).
Tree Establishment in Restored Wetlands
Establishing trees in restored wetlands can be accomplished through natural
colonization and/or through tree planting. Tree establishment is often the most difficult
task in forested wetland restoration (Matthews and Endress 2008) and results from
inadequate natural colonization or through poor survival of planted trees (Robb 2002,
Morgan and Roberts 2003, Spieles 2005). At restoration sites where natural regeneration
is likely to be unsuccessful, trees are often planted.
Natural Colonization
Colonization of trees into a restoration site can be perceived as the process of tree
species immigration and local extinction of individuals that takes place over time
(MacArthur and Wilson 1967). Noon (1996) called changes in woody composition over
time in created wetlands the process of primary succession while van der Valk (1981)
called changes in restored wetlands secondary succession.
In Virginia, the most common early colonizing (pioneer) tree species into oldfields, timbered forested wetlands, and restored/created wetlands are L. styraciflua, A.
rubrum, F. pennsylvanica, Platanus occidentalis (American sycamore), Pinus taeda
(loblolly pine), Salix nigra (black willow), and Nyssa aquatica (water tupelo) (McQuilkin
1940, Monette and Ware 1983, Spencer et al. 2001, Atkinson et al. 2005, Hudson 2010).
18
These early colonizing species have characteristics that allow them to survive and grow
in the environmental conditions present in early successional ecosystems, including, high
acclimation potential, broad physiological responses, and increased growth rates (Bazzaz
1979). Additionally, the majority of these species seeds are wind dispersed (N. aquatica
is dispersed by animals and water) which increases their ability to colonize into early
successional ecosystems.
Tree species that arrive later in these ecosystems are typically Quercus sp., Carya
sp. (hickory), F. grandifolia, Pinus sp., and L. tulipifera, which are often the dominant
species in mesophytic and southern mixed forest regions of Virginia (Dyer 2006). The
composition and abundance of trees that will colonize these ecosystems are dependent
upon surrounding seed source composition and density as well as the biotic and abiotic
conditions that are present within a given ecosystem (De Steven 1991, Gill and Marks
1991, Myster and Pickett 1993, Hudson 2010, Casanova and Brock 2000, Keddy 1992).
The dominant factors that influence colonizing tree density are composition and distance
to seed source, size of the adjacent forest, hydrology, and the size (basal area, height) of
trees in the adjacent forest (Hudson 2010). Colonization by pioneer tree species from
adjacent forests should be considered during restoration design and preparation.
Inadequate natural colonization can result from a total lack of seed source,
insufficient amount of seeds of appropriate species in the existing seed bank, or lack of
seeds actively being supplied by nearby trees. Additionally, abiotic (hydrology, soil
conditions, etc.) and biotic (seed predation, competition, etc.) site conditions can limit
natural colonization.
19
Planting
In addition to colonization of trees from surrounding seed sources, trees can be
established at restoration sites through planting trees (seeds, seedlings, saplings) obtained
from nurseries or transplanted from other wetland areas (Havens 2004). Selecting the
appropriate species and planting stock and using correct planting techniques helps to
ensure survival and growth of the planted stock. The first decision when selecting
planting stock is whether to use seeds or seedlings/saplings. Whether using seeds or
seedlings the source or provenance of planting material can be important (Gardiner et al.
2002). Planting material should come from locations that are within the same cold
hardiness zone or ecoregion so that they have some tolerance to the local conditions (Dey
et al. 2008).
Direct Seeding
One method of establishing trees in restoration sites is through the planting of tree
seeds, often called direct seeding. Direct seeding is often less expensive than planting
nursery grown seedlings (Bullard 1992) and can be used to establish any tree species, but
efforts mainly focus on Pinus spp., Quercus spp. or Carya spp. (Harmer and Ker 1995,
Dey et al. 2008). Most methods of direct seedling come from restoration of bottomland
hardwoods for timber production. Direct seeding may be useful on sites that are large in
size or far away from seed sources (McKevlin 1992). Clewell and Lea (1989)
recommended planting acorns as opposed to broadcast seeding.
Williams et al. (1999) investigated the differences in survival of direct seeded
versus containerized and bare root Quercus texana (Nuttall oak) seedlings planted on old-
20
fields. Results from this study showed that containerized seedlings had greater survival
than bare root seedlings in areas with high water tables or flooding, suggesting that direct
seeding may not be an appropriate choice for most wetland restoration sites due to the
stressful hydrologic conditions.
Nursery Stocktypes and Production Techniques
There are numerous woody stocktypes available for afforestation or reforestation
projects, including wetland restoration, carbon sequestration projects, various
silvicultural and forestry projects, wildlife preserves or other conservation reserve
programs. The tree nursery industry in the United States and Canada utilizes many
different production techniques to produce a wide variety of woody stocktypes. The
overall effects of the various production methods are to alter the above and belowground
morphology and physiology of the nursery stock (Chavasse 1980). The goal of altering
these characteristics of seedlings is to produce a cost-effective nursery stock that will
successfully survive and grow following outplanting.
There are various descriptions used in the nursery industry to describe the size,
age, and how the stock was produced, that are not uniformly applied throughout the
nursery industry. This can lead to confusion when purchasing nursery stock and difficulty
in evaluating stocktypes from a scientific perspective. There have been several attempts
to standardize and describe the terminology of nursery stock including the American
Nursery & Landscape Association (2004), the USDA Forest Service (Landis 1990,
Landis et al. 2010) and the British Columbia Ministry of Forests (1998) Provincial
Seedling Stock Type Selection and Ordering Guidelines (Scagel et al. 1998). In 1984 the
21
Nursery Technology Cooperative at Oregon State University and the USDA Forest
Service produced the Forest Nursery Manual: Production of Bareroot Seedlings, which
described methods used for bare root seedling production (Duryea and Landis 1984).
Common descriptions of woody nursery stock available for outplanting include
cuttings (whips, live stakes), bare root seedlings, tubelings, plugs, containers (many
sizes), and balled and burlapped. These descriptive names can be used to describe the
age, size, and production technique used, all of which are not uniform throughout the
nursery industry. Therefore, when assessing the outplanting success of stocktype it is
important to focus on the individual morphology and physiology of the seedlings that
resulted from the unique production techniques (Pinto 2011b). In general bare root
seedlings range in age from one to three years old and are typically planted without soil
surrounding the roots. Tubelings are typically similar in age to bare root seedlings;
however, they are often planted with soil surrounding the roots and are grown in various
shaped (square, round) small containers. Many seedlings are grown in larger containers
ranging from 3.8 L to 94.6 L. Containerized seedlings can be grown to various ages and
sizes and are often planted with the soil intact around the root system, which allows for
planting later in the season. The Virginia Department of Forestry (2012) offers one- twoand three-year old bare root seedlings sold directly from the seedbed and specialty packs
(sets of different species that achieve a common goal, e.g. wildlife seedling pack).
Nursery production techniques and conditions that influence the morphology and
physiology of woody planting stock include; nursery location (local climate and weather),
container size, style and density (or lack thereof), fertilization timing, amount, type (or
lack thereof), lifting and storage times and conditions, root and/or shoot pruning, use of
22
pesticides (fungicides, herbicides etc.), irrigation techniques and amounts (flood pre
conditioning (Anderson and Pezeshki 2001), mycorrhizal inoculation (application time
and type), application of growth regulators, propagation methods (seeds, clones, tissue
culture, grafting), seed source, genetics, size, stratification time, germination period
(light, moisture), sowing season, potting soil type (nutrient/OM concentration), amount of
soil used, sterilization technique, light intensity and photoperiod, unintended stress,
grading/culling techniques, hardening off methods, seedbed conditions, growing time,
and overall input costs.
These nursery production techniques and conditions influence the morphological
characteristics of seedlings including; stem height, root collar diameter, seedling stem
volume, stem mass ratio, root to shoot ratio, aboveground biomass, number of branches,
branch diameter, leaf area, leaf mass ratio, foliar color, bud length and amount, disease
prevalence, root shape (length, area), root size (mass, volume), root mass ratio, root
fibrosity, and the amount of first order lateral roots.
These nursery techniques also influence the physiological characteristics of
seedlings including; water use efficiency, plant moisture stress, cold hardiness and heat
tolerance, nutrient concentrations and allocation, foliar nitrogen concentrations, vigor
(Oregon State University Vigor Test), CO2 assimilation rates, leaf respiration rates, stem
respiration rates, transpiration rates, chlorophyll concentrations, root growth potential
(RGP), electrolyte leakage, root carbohydrate concentrations, stem and root moisture
content and root respiration rates.
There have been many investigations into how nursery techniques and conditions
influence the morphology and physiology of woody seedlings. Duryea and Landis’
23
(1984) collection of papers on bare root production focused on how nursery design and
production techniques influence many characteristics of bare root seedlings. They
collected data from 21 nurseries in the northwest U.S. and Canada and interviewed 250
individuals to produce the manual. Brissette et al. (1991) investigated how container
nursery techniques such as seed quality, sowing techniques, growing medium moisture,
and seedling nutrition affect shoot and root morphology of southern Pinus spp.. They
concluded that containerized nursery stock provided a viable alternative when bare root
stock is not expected to perform well. Chavasse (1980) reviewed factors that affected the
seedling during production, including shelter (greenhouses), soil, irrigation, seed
treatments, weed control, insects, disease, spacing, conditioning, and lifting techniques.
He concluded that the ability of Pinus radiata (Monterey pine) to grow well after
planting depended more on nursery treatment than on size or any easily measurable
morphological characteristic. Duryea and McClain (1984) investigated how irrigation,
fertilization, and conditioning at the nursery influence the seedling physiology, including
frost hardiness, mineral nutrition, and carbohydrate reserves. Jackson et al. (2012)
investigated the effect of nursery fertilization and shelter (greenhouse vs. outdoor) on
several morphological and physiological responses of Pinus palustris (longleaf pine)
seedling and determined an optimal fertilization regiment. Alm and Schantz-Hansen
(1974) found that very small containers (14.3 mm diameter) restrict tree root growth of
Pinus banksiana (jack pine) and Pinus resinosa (red pine), while South and Mitchell
(2005) developed a “root bound index” to evaluate the effect of restricted root growth in
containers. Results suggest that containers produced restricted root growth that resulted in
decreased survival following outplanting.
24
Mycorrhizal Inoculations
The effect of mycorrhizal-root relationships on water absorption has not been
completely determined (Pallardy 2008). Dixon et al. (1984) investigated the effect of
mycorrhizal inoculation on containerized and bare root Quercus velutina (black oak)
seedlings. After two years the inoculated container grown seedlings had greater shoot
length, root collar diameter, and leaf area compared to the bare root stocktype. Burkett et
al. (2005) found that container grown Q. texana seedlings inoculated with vegetative
mycelia had greater survival than non-inoculated seedlings in flooded conditions.
However, additional studies are needed to determine if this nursery production technique
provides added benefit when planting trees into restored wetlands.
Alternative Stocktypes
In addition to the common stocktypes available, several tree nurseries have
developed specialized stocktypes by using particular production methods. One such
stocktype is the Root Production Method (RPM®) that was patented by Forrest Keeling
Nursery (Lovelace 1998). According to the Forrest Keeling website, RPM uses “superior
seed stock, air-root pruning, special nutrition and soil and proper timing to produce the
best plant stock on the market today.” Specifically, Root Production Methods collect seed
propagules within 161 km of the planned planting site, grade for quality, and then start
selected seeds in 4 cm of composted rice hull, Pinus spp. bark and sand. The compost is
prepared in a 4:4:2 ratio to create 35-40% porosity, and amended with slow release
fertilizers (Grossman et al. 2003). This combination promotes air pruning of the tap root,
25
and increased production of lateral roots (Shaw et al. 2003, Dey et al. 2004). One to two
months after emergence, seedlings are transplanted into individual, bottomless containers
10 cm deep to continue air root pruning of the tap root. Seedlings are then graded based
on height, stem diameter, and root development, with only individuals in the top 50%
selected (Grossman et al. 2003). As seedlings continue to develop, they undergo a series
of transplantings to larger containers and are acclimated to outdoor environments.
Containers continue to be shallow and bottomless to encourage air pruning of the tap
root.
Research showed after two growing seasons in the nursery, RPM produced
Quercus alba (white oak) seedlings had basal stem diameters greater than 2.0 cm and
heights exceeding 1.5 m, a significant improvement over conventionally produced
seedlings (Dey et al. 2004). Root dry weights and root volumes for the same RPM
seedlings averaged 101-117 g and 222-252 ml, compared to 18 g in weight and 26-33 ml
in volume for conventionally grown bare-root seedlings, also indicating a significant
improvement in growth (Dey et al. 2004). Research showed these trends are typical of
RPM grown trees (Grossman et al. 2003, Shaw et al. 2003). Because of the potential to
improve sapling development, naturally slow growing species, such as Quercus palustris
(pin oak), Quercus bicolor (swamp white oak), Quercus macrocarpa (bur oak), are often
the subject of RPM treatments (Dey et al. 2004).
Research comparing the field growth of RPM and conventionally produced Q.
alba seedlings indicated that early rapid shoot growth and acorn mass are significantly
greater and occur earlier in RPM produced trees (Grossman et al. 2003). Research also
showed that in forest restoration projects, first-year survival was significantly higher and
26
basal diameter was significantly greater in RPM produced seedlings than in bare-root
seedlings (Shaw et al. 2003). Tree height was also greater in RPM grown trees; however,
data analysis showed this difference to not be significant (Shaw et al. 2003).
Other studies have shown that survival rates can also be lower for RPM planting
stock compared to convention stock types (Henderson et al. 2009). Therefore, additional
research is needed to determine if this production technique provides added benefits
following outplanting into restored wetlands for a variety of species.
Evaluating Planted Tree Survival and Growth
There are many methods and measurements used to determine how individuals or
groups of planted trees respond following outplanting. Common measurements following
outplanting include (but are not limited to); survival rates, morphology (height, root
collar diameter, crown morphology, leaf area, number of branches, etc.), above and
belowground biomass (and allocation patterns), growth rates (morphological or biomass),
transplant shock duration, photosynthetic rates, respiration rates, transpiration rates,
stomatal conductance, and tissue nutrient concentrations. The sampling methods,
measurements, calculations, and statistics used should reflect the underlying research
question. The methods used to assess planted trees in this study focus on survival,
morphology, biomass, and growth rates.
Survival
One of the most common response variables measured following outplanting is
survival. Survival is relatively easy to determine; however, statistically it is a difficult
27
measure to analyze since it is a binomial response (dead vs. alive), except when analyzed
at the stand level (e.g. trees per acre). Survival analysis is a particular branch of statistics
that encompasses a variety of methods for survival or similar binomial events. Instead of
focusing on the binomial response, many of these methods focus on analyzing “time to
event data” (Allison 2010). For planted trees this represents the time between when the
tree was planted and when the tree died. This information and statistical methods can be
utilized to determine the effect of a given treatment (stocktype, nursery production
method, morphological or physiological characteristic) on the survival time of planted
trees. The Cox Proportional Hazards Model is one of the most highly cited survival
analyses (Cox 1972). The Cox model does not require choosing a particular probability
distribution to represent the survival times, can incorporate time-dependent covariates,
permits stratification, accommodates both discrete and continuous measurements, and
integrates censoring effectively (Allison 2010). These factors make it an appropriate
choice for modeling survival times of planted saplings (Burgman 1994, Clarke 2002,
Hastwell and Facelli 2003, Frey et al. 2007).
Morphology
There are numerous morphological measurements that can be used to evaluate
trees following planting, including but not limited to stem (trunk) diameter or area (at
groundline or breast height), overall height, number of branches, branch diameter, branch
length, crown diameter, area or cover, leaf area, leaf mass, root length, root density, total
volume, and total mass. Growth of each of these parameters can be used to quantify
development. While many of these morphological parameters are related, each gives
28
unique information about planted tree responses following outplanting. The
morphological factors that are of interest for this study include stem diameter at
groundline, height and crown diameter.
Biomass
Production and accumulation (storage) of plant biomass is an important
ecosystem function that will develop in successfully restored forested wetlands.
Therefore, it is an appropriate response for evaluating planted trees. In early successional
created and restored forested wetlands the majority of plant biomass is produced by
herbaceous vegetation (Atkinson et al. 2005, DeBerry and Perry 2012). However, as
restored forested wetlands develop, the production and accumulation of biomass shifts to
perennial woody vegetation (shrubs, trees, and woody vines) (DeBerry and Perry 2012),
where biomass can remain long-term in the boles, stems, and roots. Therefore, the
intended outcome of the restoration process of FHW is that the majority of biomass
accumulated will be produced by long lived species, particularly trees.
Quantifying standing biomass and production of planted trees is important in
understanding the development of ecosystem structure and functions in restored FHW.
Additionally, accurate quantification of tree biomass is an important step in
understanding carbon dynamics across a range of forested ecosystems (including
wetlands) as they play a large, but yet uncertain, role in the global carbon cycle
(Temesgen et al. 2015).
The most common method used to estimate the standing biomass of an individual
tree is the use of mathematical relationships between biomass and one or more
29
morphological woody vegetation characteristic, such as stem diameter and/or height (e.g.
Jenkins et al. 2003). This method is commonly referred to as dimensional analysis
(Whittaker and Woodwell 1968). Dimensional analysis relies on consistency in the
correlations between the changes in relative dimensions of parts of an organism with
changes in overall size, referred to as allometry (Gayon 2000, Stevens 2009). While
allometric relationships are “not confined to any form of mathematical expression”
(Gould 1966), the mathematical relationships describing the allometry of various
characteristics of many organisms (including trees) often conform to power laws (Niklas,
2004, Stevens 2009). One focus of this study is developing equations that relate total tree
biomass and morphological characteristics. These models will be referred to as biomass
estimation models (BEM), but have also been referred to as allometric equations by
others (Sileshi 2014).
When constructing BEMs for saplings and trees, biomass often only refers to
aboveground dry woody material but can refer to “green” (wet) weight and can include
leaves, buds, flowers, and roots depending on the goal of the research. The tree
dimension variable most often used as a predictor of biomass is diameter at breast height
(dbh). The most common height above the soil surface (breast height) where stem
diameter is measured is 1.37 m (in United States) but may vary for other countries. When
relating biomass to dbh, BEMs can take many forms (see Picard et al. 2015 for
examples); however, the most commonly used form is the following power law equation
and/or equivalent natural logarithmic transformation of this equation.
30
Y=aXb + ε
(Eq. 1)
Y = biomass (response or dependent variable)
a = model estimated parameter (normalization or proportionality constant or intercept or
allometric constant)
X = tree dimension variable, dbh (predictor or independent variable)
b = model estimated parameter ((allometric) scaling coefficient or exponent)
ε = error term (random normally distributed additive error term with constant variance)
This model form or equivalent natural logarithmic transformations have been
found to provide an accurate representation of the relationship between aboveground
biomass and dbh across a wide range of tree sizes from around the world (Whittaker and
Woodwell 1968, Tefler 1969, Tritton and Hornbeck 1982, Ter-Mikaelian and Korzukhin
1997, Jenkins et al. 2003, Xiao and Ceulemans 2004, Chojnacky et al. 2014). However,
for trees smaller than 1.37 m that lack dbh (seedlings or saplings), very few BEMs have
been constructed (Tefler 1969, Roussopoulos and Loomis 1979, Smith and Brand 1983,
Williams and McClenahem 1984, Wagner and Ter-Mikaelian 1999, Geudens et al. 2004,
Henry et al. 2011). These studies successfully related biomass to other stem
measurements, including root collar diameter (stem cross-sectional diameter at
groundline), stem diameter just above root collar swelling, and diameter at 15 cm above
the root collar.
Development of BEMs for seedlings and saplings is valuable because they can be
a major component of understory (Gemborys 1974) and canopy gaps (Ehrenfeld 1980) of
mature forests and can be significant components of the vegetative structure in early
31
successional stages of many ecosystems, including abandoned agricultural fields
(Monette and Ware 1983) and recently restored wetlands (Hudson 2010). Additionally,
BEMs for seedlings and saplings allow a more inclusive investigation of planted tree
development.
Growth
Growth is generally defined as an irreversible change in a measurable quantity
(Hunt 1990, Jørgensen et al. 2000). While this definition is simple, there are a multitude
of methods designed to describe growth. Common measurable quantities can include
size, form, or number (Hunt 1982). These measurable quantities can be applied to a vast
range of living and non-living systems from individual bacterium to human populations
and from crystals all the way to the entire universe. For the purpose of this paper, growth
will focus on groups of planted sapling’s increase in total biomass.
Plant growth analysis has produced many useful methods for describing growth
that were developed by researchers from many different backgrounds and schools of
thought (Evans 1972, Pommerening and Muszta 2016). These researchers developed
methods to address questions specific to their fields, but through time particular methods
proved more valuable than others and general trends became apparent. Increases in plant
mass through time (referred to as cumulative growth) tends to follow the same general
sigmoidal pattern that is found elsewhere in nature (Weiner and Thomas 2001,
Pommerening and Muszta 2016). Plants start with little to no detectable increases in
mass, then mass increases rapidly. As plants reach maturity, the increase in mass slows
(reaches an asymptote) and finally ceases (Hunt 1990). This pattern of growth remains
32
similar for many types of plants; however, the magnitude and symmetry of the curve can
change substantially. Additionally, recent studies have suggested that biomass (or carbon)
accumulation of trees may not reach an asymptote (Muller-Landau et al. 2006, Sillett et
al. 2010, Stephenson et al. 2014). This study focuses on sapling growth, when biomass is
rapidly increasing.
From this general sigmoidal pattern, two biomass growth rates can be determined.
The first is termed absolute growth rate (AGR) and is defined as the change in mass over
a given time interval (e.g. kg/day). Following the above sigmoidal pattern example, AGR
will increase rapidly, peak, rapidly decline, and then slowly taper off. While AGR yields
important information, it presents challenges when comparing growth of individuals of
different starting masses (Hunt 1982, Hunt 1990, Rees et al. 2010, Pommerening et al.
2016), which is the focus of this study.
Relative growth rate (RGR) can be used to compare individuals of different sizes
when calculated correctly (Rose et al. 2009). Relative growth rate is defined as the rate of
increase in mass per unit mass per unit of time (g/g/day) or growth per unit mass.
Relative growth rate is referred to as an ‘efficiency index’ because it measures the
efficiency of plant material to produce new material (Blackman 1919). It is often
compared to the rate of compound interest earned on capital in the financial world
(Blackman 1919, Hunt 1990). Relative growth rates of trees decline through time as trees
grow larger (ontogenetic drift) because of self-shading, increased allocation to structural
(non-photosynthetic) components, declines in leaf area ratio, and reduced nutrient
availability (Evans 1972, Hunt 1982, South 1995, Rees et al. 2010, Paine et al. 2012,
Philipson et al. 2012). Therefore, comparisons of RGR among species or experimental
33
treatments must account for differences in mass because high values of RGR can occur
because plants are either smaller is size or because they are growing faster (Turnbull et al.
2008, Rees et al. 2010, Turnbull et al. 2012).
Tree growth can be positively or negatively influenced by hydrologic conditions
depending on the species (Kozlowski 1984). However, there are very few studies that
measure total biomass (above and belowground) and calculate AGR and RGR for
individual seedlings or saplings grown across a hydrologic gradient (Mitsch and Ewel
1979, Farmer, Jr. 1980, Tang and Kozlowski 1982a,b, Megonigal and Day 1992). Several
studies have investigated morphological growth (basal area, stem diameter, tree rings,
height, leaf mass, etc.) of trees grown in different hydrologic regimes (Malecki et al.
1983, Mitsch and Rust 1984, Keeland et al. 1997, Kabrick et al. 2005, Anderson and
Mitsch 2008, McCurry et al. 2010, Rodríquez-González et al. 2010, Kabrick et al. 2012,
Smith et al. 2013). However, a large proportion of these studies fail to account for trees
that have a small stem dbh or are not tall enough to have a dbh entirely (Das 2012).
Overall, growth rates provide detailed information about how planted tree are responding.
Factors Affecting Tree Survival and Growth
The factors that influence planted tree survival and growth can be divided into
three groups based on when they occur relative to outplanting (before, during, after).
Outplanting is the act of planting trees or seeds into the ground. There are many factors
that will influence the survival and growth of planted trees, including (but not limited to)
planting technique, planting density, timing of planting (weather, air and soil
temperature), site characteristics, competition/facilitation, browsing/herbivory, soil bulk
34
density, soil nutrient content, hydrology, site preparation techniques (ripping, ripping,
tilling, burning etc.) and stochastic events (pests, diseases, hurricanes, floods, etc.). In
addition to these factors, the selection of species and stocktypes is important to the
establishment and growth of trees.
The survival and growth of planted trees and the factors that affect these
processes change through time. For example, during the first years after planting tree
seedlings are most sensitive to environmental factors and most subject to mortality
(McLeod and McPherson 1973, Alm and Schantz-Hansen 1974). Mortality becomes
important again later in time, following canopy closure, when self-thinning can occur.
Prior to Outplanting
Prior to outplanting several factors can greatly influence the survival and growth
of trees planted into restored wetland sites, including species selection, planting material
selection, and handling and transport of planting material.
Species selection
One of the most important decisions made prior to outplanting is which species
will be used. Tree species selected to be planted should be those that are found naturally
in nearby wetland ecosystems that have conditions (biotic, abiotic, climate, etc.) similar
to the restoration site (Nyland 2007). This is especially important for wetland restoration
sites having stressful hydrologic conditions. Tree species have unique traits and
adaptations that allow for survival and growth in particular environmental conditions. B.
nigra, P. occidentalis, S. nigra, and T. distichum have been shown to be well adapted to
35
the conditions in recently created wetlands (Spencer et al. 2001, DeBerry and Perry,
2012). An adaptation of these species is that they colonize during dry periods, but can
withstand prolonged saturation and inundation following establishment. Some hardwood
species (Quercus spp. and Carya spp.) are planted following wetland creation; however,
these species tend to naturally occur later in forest development (Spencer et al. 2001).
Several authors have investigated the survival and growth of various tree species
planted in environmental conditions similar to those present in recently restored wetlands.
McLeod (2000) began a series of experiments in 1990 that focused on 24 tree species for
restoration in bottomland hardwood forests in South Carolina. McLeod et al. (2006)
provided a ten year review of the series of experiments. The experimental trials focused
on fertilizer, initial morphology, stocktype, tree shelters, root pruning, competition
control and S. nigra control on the survival and growth of various tree species and
stocktypes. Results from this study suggested that woody species should be matched to
the hydrologic conditions (due to elevation differences). Taxodium distichum and N.
aquatica were found to be the most flood tolerant species. F. pennsylvanica, Carya
aquatica (water hickory) and Quercus lyrata (overcup oak) were found to be moderately
flood tolerant, while Q. michauxii, Q. texana, and Q. phellos were found to survive better
in higher elevations. Stanturf et al. (2004) similarly suggested that matching species to
site hydrology is a key factor for success in afforestation of bottomland hardwood forests.
Farmer (1980) compared first year growth of six deciduous species grown in nursery
conditions and found significant difference between the early successional species (L.
tulipifera and Pinus serotina (pond pine)) and late successional species (Q. rubra,
Quercus montana (chestnut oak)), Q. alba, and Quercus ilicifolia (bear oak)) with regard
36
to dry weight, leaf growth rate and net assimilation rate. The early successional species
had higher growth rates, net assimilation rates and high investment in leaf area.
The selection of species is of greater importance than stocktype selection. The
selection of species can influence the stocktype selection, as not all stocktypes are
commercially produced for all species. Both species and stocktype need to be matched to
site biotic and abiotic conditions as well as local climate conditions (Nyland 2007).
Selection of Planting Material
Following the selection of species, the selection of planting material is an
important factor that can greatly influence the survival and growth of planted trees
(Nyland 2007). Many studies have focused on the appropriate stocktype to ensure
survival and growth of planted trees for many different environmental conditions;
however, few have investigated how stocktype influences survival and growth in
conditions similar to restored wetlands.
Burdett et al. (1984) investigated differences in root growth capacity, height
growth, and needle length between container grown (336 mL container) and bare root (2
or 3 year olds) Picea glauca (white spruce) seedlings. Container grown seedlings had
greater root growth capacity, height growth and needle length compared to bare root
seedlings.
Grossnickle and Blake (1987) examined the water relations and root growth of P.
banksiana and Picea mariana (black spruce) container and bare root seedlings during the
first season following outplanting on a boreal cut-over site. Bare root seedlings of both
species had greater resistance to water-flow through the soil-plant-atmosphere continuum
37
early in the growing season. However, this trend did not persist through the growing
season. At the end of the growing season both stocktypes had similar new root growth.
Denton (1990) investigated the effect of stocktype on the growth of T. distichum
in a restored forested compensatory mitigation site in Florida. The results suggest that in
order to obtain 33% canopy closure the initial costs could be reduced by planting smaller
trees (1-gallon container) at ~2500 stems/ha as opposed to planting 7-gallon trees at
~1000 stems/ha.
A large scale long term study by McLeod (2000) found that bare root seedlings
had similar survival to more expensive containerized seedlings of F. pennsylvanica, N.
aquatica, and T. distichum when planted in a thermally impacted bottomland hardwood
forest.
South et al. (2005) investigated the effect of several containers and bare root
seedlings on the survival and growth of P. palustris outplanted on old-fields and cutover
sites. The results suggest that containerized seedlings had 20% better survival than bare
root seedlings having similar root-collar diameters. Mesh-covered plugs had lower field
performance compared to the three hardwall containers and styroblock containers. Low
root growth potential and high root bound index were correlated with low survival.
A meta-analysis of 122 trials comparing survival between bare root and
containerized stock planted across a variety of upland sites found that containerized stock
had greater survival than bare root stock in 60.7% of the trials (Grossnickle and ElKassaby 2015). Pinto et al. (2011a) found that increasing the size of containers for Pinus
ponderosa (ponderosa pine) increased total height and basal area following outplanting at
a mesic site.
38
Overall, the response of planted trees to stocktype selection is highly variable and
dependent upon the species and environmental conditions. How the stocktype is handled
prior to outplanting can also influence survival and growth.
Handling and Transport of Planting Material
The final factor to be considered prior to outplanting and once the planting stock
is selected is the handling and transport of seeds or seedlings to the restoration site. The
stocktype selection will determine the appropriate handing techniques used. For example,
bare root seedlings should be kept moist the entire time prior to outplanting and should
avoid high temperatures that can cause desiccation. Inappropriate handling of planting
stock can greatly reduce the potential for survival and growth following outplanting
(McKay 1997).
During Outplanting
During outplanting the methods used and their successful implementation can
influence the survival and growth of trees planted in restored wetlands. The factors that
can influence survival and growth include the timing of planting, density and
arrangement of planting, and the actual placement of trees in the ground. The optimal
time to plant trees and other woody vegetation in the Mid-Atlantic region of the US is fall
or early spring. By avoiding planting during winter or summer, trees are able to become
established prior to high or low (freezing) temperatures or fluctuations in water levels.
Deciduous trees are planted during dormancy to reduce water loss due to
evapotranspiration (Landis et al. 2010).
39
Planting Methods
The actual planting of seedlings into the ground is important for their survival and
growth. Incorrectly planting seedlings, too deep, too shallow or leaving a pocket of air
around the roots can negatively affect survival and growth (Landis et al. 2010).
Planting Arrangement
The density and arrangement of planting is important to ensure appropriate
density in the future and to avoid unnecessary woody competition. A number of studies
have focused on how initial planting of economically important trees influences the final
timbered product. Smith and Strub (1991) found that increasing initial planting density
(from 4.6 x 4.6m to 1.8 x 1.8 m spacing) of southern Pinus spp. (P. taeda and Pinus
elliottii (slash pine)) had a negative effect on size, value of the final product, as well as
the cost of harvesting and cultural treatments.
The USACE Norfolk District and VADEQ (2004) makes several
recommendations as to initial planting densities to establish forests. First they
recommend considering recruitment from surrounding areas when the site is narrow,
whether the site is exposed to flood waters bearing seeds, whether the original soils and
hydrology are not significantly altered, or if a seedbank is present. When direct seeding
Quercus spp., they recommend planting 2471-3707 acorns/ha. A 35% germination rate
would result in 300-500 seedlings/ha. They recommend planting bare root seedlings at a
density of 272-741 stems/ha when colonization from surrounding seed sources is
40
expected and suggest that lower planting densities of containerized saplings could be
used since containerized saplings have higher survival rates.
Few studies have investigated the influence of initial planting density on survival
and growth of planted trees. Bullock and Burkhart (2005) investigated the influence of
planting density (757 to 6,727 trees/ha) on P. taeda stem diameter and found that of the
23.2% of sampled plots exhibiting spatial dependencies, 73.1% showed negative
correlations. This suggests that increases in stem density lead to smaller diameter P.
taeda.
Several authors have suggested planting late successional species (Quercus spp.,
Carya spp.) under planted or naturally colonizing pioneer (early successional) tree
species to increase the survival and growth of the late successional species (Clewell and
Lea 1989). The underlying idea is that the establishment of early pioneer tree species,
often called nurse species, can create favorable conditions for the establishment of later
successional species. The favorable environmental conditions provided by the nurse
species are shade, moderate temperature, moderate moisture extremes, reduction of
competing herbaceous vegetation, organic matter added to the soil, decreased soil
compaction in the rooting zone, fixed atmospheric nitrogen, increased vertical structural
complexity, increased forest diversity, possible reduced costs (Clewell and Lea 1989,
Dulohery et al. 2000, Dey et al. 2010).
Dulohery et al. (2000) investigated the effects of a S. nigra overstory on planted
seedlings of F. pennsylvanica, T. distichum, N. aquatica, and Q. michauxii. During the
first 2 years the S. nigra overstory provided minor growth enhancements; however, by
age 5 there was no effect on seedling heights. McLeod et al. (2001) investigated the
41
response of container grown Q. lyrata, T. distichum, and C. aquatica seedlings to several
treatments involving S. nigra overstory. Results suggested that after three years survival
was not influenced by the S. nigra overstory. The heights of C. aquatica and Q. lyrata
were not affected by the S. nigra overstory. Twedt (2006) found that when Quercus spp.
plantings were supplemented with fast growing early successional trees the species
diversity, stem density, and maximum tree height were increased. Stanturf et al. (2009)
interplanted Q. texana with Populus deltoides (eastern cottonwood) and found that it did
not facilitate the development of the Q. texana seedlings.
Dey et al. (2010) provided a summary of many new techniques for afforestation
of old-field bottomland forests, including interplanting later successional species
(Quercus spp.) with early successional species (P. deltoides) to mimic natural succession.
Overall, the Q. texana were able to survive and grow successfully under the canopy of
the P. deltoides and should be able to replace the P. deltoides following harvest.
Following Outplanting
Following outplanting there are many factors that can influence survival and
growth of planted trees in restored wetlands, including site physical, chemical and
biological characteristics, herbaceous and woody competition, browsing/herbivory, and
stochastic events (storms, drought, etc.). During the first years after planting tree
seedlings are most sensitive to environmental factors and most subject to mortality
(McLeod and McPherson 1973, Alm and Schantz-Hansen 1974).
The environmental characteristics of restored wetlands are influenced by prior
land use and site preparation techniques (ripping, tilling, burning, fertilizing, etc.). Poor
42
survival and growth of planted trees can result from unfavorable site conditions,
including inappropriate hydrology, low soil organic material content, low soil nutrients,
high soil bulk density, and increased rock fragments (Stolt et al. 2000, Campbell et al.
2002, Bruland and Richardson 2004, Bergshneider 2005, Daniels et al. 2005, Bailey et al.
2007, Nyland 2007).
Hydrology
In natural and restored wetlands, low oxygen soil conditions associated with long
hydroperiods are among the greatest stressors on wetland vegetation (Mitsch and
Gosselink 2007), including planted trees. Saturated or flooded soil conditions lead to the
removal of plant available oxygen from the soil pore space (Brady and Weil 2002). This
leads to a lack of aerobic respiration in roots, which decreases the energy available for
trees to maintain functions of existing tissues (Hale and Orcutt 1987), and many trees
exhibit reduced growth when flooding exceeds a few weeks during the growing season
(Kozlowski 1984). Kozlowski (1984) provided an in depth review of the effects of
flooding on tree seed germination, seedling establishment, shoot growth, cambial growth,
root growth, biomass changes, morphological changes, and mortality. Several studies
have investigated the effect of hydrology on tree survival, growth, morphology, and
physiology following outplanting.
Megonigal and Day (1992) investigated the biomass production, carbon allocation
to roots and shoots, and root-system morphology of T. distichum seedlings in response to
continuous and periodic flooding in large watertight enclosures. After three growing
seasons, there was no significant difference in total biomass between the two treatments.
43
However, saplings in the continuously flooded treatment had lower root-to-shoot ratios,
shallower roots, lower belowground production, and higher aboveground production than
those in the periodically flooded treatment. The authors concluded that the saplings in the
continuously flooded treatment invested fewer resources belowground because of the
abundant water and dissolved nutrients available in the rooting zone.
Niswander and Mitsch (1995) planted 10 tree species in a created wetland along a
hydrologic gradient and found that trees located in the shallow water zone were either
dead or severely stressed, trees in the wet meadow were healthy, and the trees in the
extreme upland were the largest and had the densest foliage.
Perry et al. (1996) investigated survival and growth of planted trees in a created
forested wetland in Maryland and found that the canopy diameter increased over two
years but the height decreased and that mortality was lower than other created forested
wetlands. These results suggest that dieback and re-sprouting are common in created
wetlands and low mortality may have resulted from less hydrologic stress.
Burkett et al. (2005) investigated the effect of flooding (two elevations) on Q.
texana bare root seedlings, seedlings grown in cones, and seedlings inoculated with
vegetative mycelia planted into Sharkey clay soil. The overall results were that in the
lower elevation the mycorrhizal inoculated container seedlings had increased survival;
however, at both elevations the bare root seedlings had greater height after five years.
In created perched wetlands, Pennington and Walters (2006) found that the
survival and growth of planted trees was increased when trees were planted in transitional
zones between the created wetland and upland. These transitional zones were
44
characterized by lower soil saturation, higher soil redox potential, and distinct herbaceous
vegetation.
Dickinson (2007) investigated the influence of soil amendments and
microtopography in a created tidal freshwater swamp on various vegetation parameters,
including planted tree morphological responses. The results from this study suggested
that T. distichum tubelings planted in soil pits (increased water stress) had increased
height, root collar diameter, and buttressing compared to those grown at higher
elevations. Those grown at higher elevations had increased roots of greater length. T.
distichum is a unique tree because of its ability to survive and grow under very stressful
hydrologic conditions where other trees may not.
McCurry et al. (2010) investigated the effect of early season flooding on Q.
lyrata, Q. texana and Quercus phellos (willow oak) seedlings and found Q. lyrata to have
the highest flood tolerance followed by Q. texana and finally Q. phellos.
Soil Physical and Chemical Characteristics
Several studies have investigated the effect of soil chemical and physical
characteristics on the survival and growth of planted trees. Soil compaction leads to
increases in bulk density, runoff, and erosion, breakdown of soil aggregates, and
decreasing porosity, aeration and infiltration (Brady and Weil 2002). Soil compaction is
particularly important at CMS because construction methods typically increase soil
compaction (Galatowitsch and van der Valk 1996, Whittecar and Daniels 1999, Campbell
et al. 2002, Bruland and Richardson 2005, Daniels and Whittecar 2011). Many studies
have found that compacted soil reduces the survival and aboveground and belowground
45
growth of planted trees (Alberty et al. 1984, Kozlowski 1999, Siegel-Issem et al. 2005) In
particular, soil compaction reduces water and mineral absorption in woody plants and
limits the ability of roots to penetrate the soil (Kozlowski 1999). Several authors have
suggested that reducing soil compaction in restored wetlands will improve the survival
and growth of planted trees as well as the overall functioning of the restored wetland
(Daniels and Whittecar 2011). The Piedmont region of Virginia is characterized by
clayey soils that typically have higher bulk densities than other soil textures and lower
soil organic matter. These soils are often exposed during construction.
Soil organic matter (SOM) content and/or soil carbon content is often low in
restored wetlands when compared to natural wetlands (Bishel-Machung et al. 1996,
Galatowitsch and van der Valk 1996, Shaffer and Ernst 1999, Whittecar and Daniels
1999, Stolt et al. 2000, Campbell et al. 2002). Low SOM has been shown to contribute to
poor survival and growth of planted trees in restored wetlands (Bergshneider 2005,
Daniels et al. 2005, Bruland and Richardson 2006). However, several studies have found
that standing crop biomass does not increase when SOM is added (Cole et. 2001,
Anderson and Cowell 2004, Bailey et al. 2007).
Bailey et al. (2007) investigated the effect of organic matter loading rates and
elevation in a created wetland on several vegetation responses, including the growth of
planted B. nigra and Q. palustris. Results from this study suggest that the early growth of
planted trees responded to both SOM loading rates and hydrology related to elevation. In
the lower elevations (increased water stress) the tree growth rates were stunted.
Planted tree survival and growth is affected by presence and abundance of soil
nutrients. Stolt et al. (2000) found that plant growth may be limited in restored wetlands
46
because of low levels of nitrogen. In a greenhouse experiment investigating the effect of
flooding and soil nutrients on T. distichum and N. aquatica growth, Effler and Goyer
(2006) found that flooding in combination with low soil nutrients reduced growth, while
flooding in combination with fertilization lead to similar growth of trees grown without
flooding or fertilization.
Overall, poor soil conditions can reduce the survival and growth of planted trees
and can have compounding negative effects when other environmental stressors are
present.
Competition
The spatial location of herbaceous vegetation and other trees in relation to planted
trees can lead to competition for light, water, nutrients, CO 2, O2, and space. Competition
between planted trees and herbaceous vegetation has been found to negatively affect tree
survival, growth, morphology and physiology following outplanting. Species and
stocktypes may respond differently to herbaceous and woody competition.
There are several theories that can be used to describe competitive interactions
among plant species and how these processes lead to observed plant community structure
(MacArthur & Wilson 1967, Grime 2002, Tilman 1988, Taylor et al. 1990, Grace 1990).
These theories have been instrumental in determining the appropriate methods for
assessing competition among plants.
Gjerstad et al. (1984) reviewed the literature available at the time and found
several studies that suggested herbaceous competition control leads to increased survival
and growth of planted tree seedlings. Britt et al. (1991) investigated herbaceous control
47
around P. taeda and found that during the first two years following outplanting the trees
with low herbaceous competition had significantly greater root collar diameter relative
growth rate and net assimilation rate. Morris et al. (1993) found that competition for
water was the dominant factor in reducing the growth of P. taeda that was planted among
several species of herbaceous competitors. Davis et al. (1998) suggested that any factor
that reduces soil water content is likely to increase competition intensity. Davies et al.
(1999) investigated the effect of herbaceous competition along a water-light-nitrogen
gradient and found that seedling survival of Q. macrocarpa and Quercus ellipsoidalis
(northern pin oak) was significantly greater when herbaceous vegetation was removed in
the wetter shaded plots. Schweitzer et al. (1999) found that weed control (fabric mat and
chemical) increased survival of bare root Q. texana, F. pennsylvanica, and Diospyros
virginiana (common persimmon).
Twedt and Wilson (2002) found that physical weed barriers increased the survival
of P. deltoides and P. occidentalis, while chemical weed control adversely impacted P.
occidentalis survival. Groninger et al. (2004) found that the removal of herbaceous
competition using herbicide increased the height and diameter growth of planted F.
pennsylvanica in a bottomland hardwood site. In created wetlands, Pennington and
Walters (2006) suggested targeting transitional vegetation zones (between the created
wetland and upland) for tree planting to increase survival and growth. These zones were
identifiable by vegetation and had characteristics that increase tree performance. Gardiner
et al. (2007) found that herbaceous competitions reduced the survival of Q. texana by
8%, height growth by 69%, and root collar diameter by 61% after 2 years compared to
trees grown without competition. Pinto et al. (2012) investigated the effect of moisture
48
stress caused by vegetative competition on three stocktypes of P. ponderosa. The results
suggest that small stocktypes had low survival when exposed to low moisture conditions
caused by herbaceous competition, while larger stock had somewhat improved survival.
They concluded that appropriate moisture is critical for survival and that herbaceous
vegetation competes substantially for moisture.
Overall, herbaceous competition can lead to decreases in survival and growth of
planted trees, but certain methods of herbaceous vegetation control can also reduce
survival and growth.
Herbivory and Browsing
Survival and growth of planted trees can be significantly decreased by herbivory
(Myster and McCarthy 1989, Stange and Shea 1998, Conner et al. 2000, Sweeney et al.
2002). Many methods have been employed to reduce both herbivory and/or herbaceous
competition around planted trees including; tree tubes, tree shelters, weed mats,
herbicide, site fencing, trapping and hunting, chemical deterrents, attracting predators and
removing herbivore cover. These methods have been used in many forest reestablishment
projects with mixed levels of success (Myster and McCarthy 1989, Stange and Shea
1998, Schweitzer et al. 1999, Conner et al. 2000, McLeod 2000, Sweeney et al. 2002,
Taylor et al. 2004). Often these methods reduce herbaceous competition and/or
herbivory, but can lead to other deleterious effects.
49
Stochastic Events
While some of the factors that influence survival and growth of planted trees can
be selected for and controlled, stochastic events, by their nature, cannot. Storms, wind,
flooding (Day 1987), drought (Davis and Trettin 2006) and pathogens (insects, fungus,
bacteria, viruses) can severely impact planted trees. Restoration plans should always
include monitoring of planted material to ensure that planted trees are surviving and
growing.
Study Species
The tree species that are commonly planted in restored wetlands in the MidAtlantic region of the United States are; P. taeda, S. nigra, Morella cerifera (wax
myrtle), P. occidentalis, P. deltoides, A. rubrum, L. styraciflua, Quercus spp., Carya spp.,
F. grandifolia, F. pennsylvanica, Chamaecyparis thyoides (Atlantic white cedar), T.
distichum, N. aquatica, and Magnolia virginiana (sweetbay) (Clewell and Lea 1989,
USACE Norfolk District and VADEQ 2004).
The tree species planted in this study include B. nigra, L. styraciflua, P.
occidentalis, S. nigra, Q. bicolor, Q. palustris, and Q. phellos. These species and some of
their characteristics are introduced below. The Atlantic and Gulf Coastal Plain wetland
indicator status is also included for each species (Lichvar et al. 2012, Lichvar 2013).
Betula nigra L.
Betula nigra (river birch) (FACW) is a deciduous tree 25 m to 30 m in height,
with a stem diameter of 50 cm to 150 cm. B. nigra commonly features multiple trunks,
50
with highly variable, dark gray or brown, scaly bark which sometimes exfoliates in
papery sheets. Leaves are alternate, serrated ovals that range from 4 cm to 8 cm in length
and 3 cm to 6 cm in length. Betula nigra is native to the eastern United States and ranges
from New Hampshire south to northern Florida and west to southern Minnesota and
eastern Texas. Betula nigra is most often found growing in riparian zones, floodplains,
wetlands and other habitats featuring moist alluvial soils (Radford et al. 1968, Grimm
1983).
Betula nigra seeds are consumed by a number of bird species (Bonasa umbellus
(ruffed grouse) and Meleagris gallopavo (wild turkey)), leaves and twigs are consumed
by Odocoileus virginianus (white-tailed deer) and host the fungus Gloeosporium
betularum (Anthracnose leaf blight), and Phoradendron serotinum (Christmas mistletoe)
colonizes branches (Burns and Honkala 1990, Sullivan 1993, Van Dersal 1938, Horst
2001). European hornets (Vespa crabro) were observed removing bark and consuming
sap in this study consistent with findings from Santamour and Greene (1986). The
damage also attracted a variety of other insects (hornets, bees, flies and beetles). WitchHazel bud gall aphid (Hamamelistes spinosus) over winter on river birch and feed on
young leaves, bronze birch borer (Agrilus anxius) feed on vascular tissue, and many
caterpillars (including tent caterpillars) feed on and reproduce in branches and foliage
(Johnson and Lyon 1988, Adkins et al. 2012). Immature larvae of the birch leafminer
(Fenusa pusilla) feed between the leaf surface (Latimer and Close 2014). Acrobasis
betulivorella (species of snout moths) larvae feed on immature terminal leaves of B.
nigra (Johnson and Lyon 1988).
51
Liquidambar styraciflua L.
Liquidambar styraciflua (sweetgum) (FAC) is a deciduous tree ranging from 20
m to 35 m in height, with trunk diameters reaching 2 m. Bark is light brown and deeply
fissured with scaly ridges. Leaves range from 7 cm to 19 cm in length and resemble A.
rubrum leaves with five pointed lobes, except they are alternate rather than opposite.
Leaves are dark green and highly glossy, turning orange and red in autumn. In the United
States, L. styraciflua is a common, southern hardwood and ranges from southern New
York south to central Florida and west to Missouri. Liquidambar styraciflua is also
found in Mexico, Central America, and South America (Radford et al. 1968, Grimm
1983).
Liquidambar styraciflua seeds are eaten by birds (Colinus virginianus (northern
bobwhite), M. gallopavo (wild turkey) and several quail), Sciurus carolinensis (eastern
gray squirrel), and Tamias striatus (eastern chipmunk) (Martin et al. 1951, Burns and
Honkala 1990, Coladonato 1992, Van Dersal 1938). The small branches and buds are
consumed by O. virginianus (white-tailed deer) (Harlow et al. 1975, Coladonato 1992)
and the bark and cambium are eaten by Castor canadensis (North American Beaver)
(Martin et al. 1951, Silberhorn 1992). Dead sweetgum trunks (snags) are used by a
variety of birds as nesting, perching and foraging areas (Dickson et al. 1983). Many
fungi, bacteria, and other parasites use L. styraciflua as a host and many insects and
beetles feed on and live in leaves and bark of living and decaying trees (Johnson and
Lyon 1988, Burns and Honkala 1990). Additionally Meloidogyne sp. (nematode) feed on
the roots (Horst 2001). The treehopper (Stictocephala militaris) lives on sweetgum for its
entire life cycle (Ebel and Kormanik 1966).
52
Platanus occidentalis L.
Platanus occidentalis (American sycamore) (FACW) is a large tree 30 m to 40 m
in height and 1.5 m to 2 m in trunk diameter. The bark of the trunk and larger limbs is
highly rigid and flakes off as branches grow, leaving the surface a mottled mixture of
white, gray and brown. Leaves are alternate with three to five lobes, bright green on top
and paler on the underside. In the fall, they turn brown and wither. Platanus occidentalis
ranges from Maine, south to Texas, and as far west as Nebraska. The tree is typically
found growing in riparian and wetland soils, but also appears in fairly most, upland soils
(Radford et al. 1968, Grimm 1983).
Seeds of P. occidentalis are eaten by Haemorhous purpureus, (purple finch), and
Spinus tristis (American goldfinch), Pecille spp. (chickadees), Junco hyemalis (dark-eyed
junco), Anas platyrhynchos (mallard), Ondatra zibethicus (muskrat), Castro canadensis
(North American Beaver) and Sciurus carolinensis (eastern gray squirrel) (Van Dersal
1938, Martin et al. 1951, Sullivan 1994). P. occidentalis is of low value as food for O.
virginianus (white-tailed deer) and Meleagris gallopavo (wild turkey) (Sullivan 1994,
Allen and Kennedy 1989). P. occidentalis often develop hollow trunks as they grow,
which can provide shelter for waterfowl and the largest cavities can be used by Ursus
americanus (American black bear) (Allen and Kennedy 1989, Sullivan 1994). Cavities
can also be used by Strix varia (Barred owl), Megascops asio (Eastern screech-owl), and
Myiarchus crinitius (great crested flycatcher) (Hardin and Evans 1977, Allen 1987,
Sullivan 1994). Aix spnsa (wood duck) uses P. occidentalis for nesting (Dugger and
53
Fredrickson 1992). A large number of fungus and other pathogens live and feed on
Platanus spp. (Horst 2001).
Salix nigra Marshall
Salix nigra (black willow) (OBL) is the largest North American species in the
genus Salix, growing between 10 m to 30 m in height and 50 cm to 80 cm in trunk
diameter, with very dark, brown bark that becomes increasingly fissured with age. Young
shoots grow quickly and are light green or brown in color. The species’ leaves are simple
and alternate, growing 5 cm to 15 cm in length and 0.5 cm to 2 cm in width. Leaves are
lighter on the bottom with a serrated edge. Salix nigra is native to eastern North America,
ranging from southern Ontario, south to Florida, and west to Minnesota. Salix nigra is
typically found in wetlands, riparian areas, and other poorly drained sites with alluvial
soils (Radford et al. 1968, Grimm 1983).
Salix nigra provides habitat to a wide array of organisms. Sphyrapicus varius
(yellow-bellied sapsucker) pecks holes in bark to feed on the sap (Burns and Honkala
1990, Tesky 1992). The fungus Pollaccia saliciperda lives exclusively on members of
the Salicaceae family including S. nigra (Burns and Honkala 1990, Row and Geyer
2010). Mistletoes (Phoradendron spp.) colonize S. nigra branches (Burns and Honkala
1990). O. virginianus (white-tailed deer), Cercus canadensis (elk), Castro canadensis
(North American beaver) and other rabbits and rodents eat the twigs, leaves and buds
(Van Dersal 1938, Martin et al. 1951, Tesky 1992, Row and Geyer 2010). They flower
early in spring and are one of the first plants to provide nectar for bees and other insects
(Row and Geyer 2010). The larvae of Limenitis archippus (viceroy) and Limenitis
54
arthemis (red-spotted purple) among others live on S. nigra (Row and Geyer 2010).
Numerous caterpillars, moths, sawflys, nematodes, beetles, weevils and borers feed and
live on S. nigra (Johnson and Lyon 1988, Burns and Honkala 1990, Horst 2001, Row and
Geyer 2010).
Quercus bicolor Willd.
Quercus bicolor (swamp white oak) (FACW) is a small; 20 m to 25 m in height
and 50 cm to 70 cm in trunk diameter. The bark is dark gray, scaly and flat-ridged, often
resembling that of Q. alba. Leaves are lobed with five to seven lobes on each side,
ranging from 12 cm to 18 cm in length and 7 cm to 11 cm in width. Quercus bicolor is
found from southern Maine, south to South Carolina, and west to Minnesota. It typically
grows in forested wetlands, riparian zones, or other poorly drained soils (Radford et al.
1968, Grimm 1983).
Acorns of Q. bicolor are eaten by squirrels, mice, O. virginianus (white-tailed
deer), Castro canadensis (North American beaver), and Ursus americanus (American
black bear), other rodents and a variety of birds (Nixon et al. 1970, Burns and Honkala
1990, Snyder 1992, Nesom 2009).
Quercus palustris Münchh.
Quercus palustris (pin oak) (FACW) is a smaller member of the Lobatae (red
oak) section of the Quercus family. The species grows 18 m to 22 m in height and 50 cm
to 70 cm in trunk diameter with grayish-brown bark that features broad, shallow fissures.
The tree has a distinctive branching pattern highlighted by drooping lower branches,
55
horizontal middle branches, and ascending upper branches. Leaves are lobed with five to
seven lobes with bristles at the tip, and range from 5 cm to 16 cm in length and 5 cm to
12 cm in width. Young trees do not drop their leaves until new growth appears in the
spring. Quercus palustris is native to eastern North America, and is found from
Connecticut, south to Georgia, and as far west as eastern Kansas. The species is typically
found in poorly drained acidic, clay soils (Grimm 1983).
Acorns of Q. palustris are consumed by woodpeckers, Anas platyrhynchos
(mallard), Aix spnsa (wood duck), O. virginianus (white-tailed deer), squirrels, Meleagris
gallopavo (wild turkey), Cyanocitta cristata (blue jay) and other waterfowl (Burns and
Honkala 1990, Carey 1992a, Dickerson 2002).
Quercus phellos L.
Quercus phellos (willow oak) (FACW) is a medium sized tree belonging to the
Lobatae section of the Quercus family. Quercus phellos grows quickly and reaches 20 m
to 30 m in height and 1 m to 1.5 m in trunk diameter. In younger specimens, the bark is
dark gray and smooth, and then becomes even darker, and irregularly fissured, with
increasing age. Leaves are distinctive from other Quercus spp., being shaped like Salix
leaves. Leaves are green on top and paler and hairy on the bottom ranging from 5 cm to
12 cm in length, and 1 cm to 2.5 cm in width. Quercus phellos is native to eastern North
America and ranges from southern New York, south to northern Florida, and west to
eastern Kansas. The species typically grows on floodplains and riparian sites, or other
poorly drained soils that periodically become flooded (Radford et al. 1968, Grimm 1983).
56
Acorns of Q. phellos are eaten by ducks, Meleagris gallopavo (wild turkey),
Glaucomys Volans (southern flying squirrel) and other squirrels, O. virginianus (whitetailed deer), Urocyon cinereoargenteus (gray fox), Meleagris gallopavo (wild turkey),
Quiscalus quiscala (common grackle), Colaptes auratus (northern flicker), mice
(Peromyscus spp.), Cyanocitta cristata (blue jay), and Melanerpes erythrocephalus (redheaded woodpecker) (Van Dersal 1938, Cypert and Webster 1948, Burns and Honkala
1990, Carey 1992, Moore 2002).
These three Quercus spp. host a large number of caterpillars, moths, sawflys,
acorn weevils, beetles, leafminers, leafrollers, wood borers, gall-wasps, scale insects,
aphids, nematodes, midges, sap-suckers, mites, fungi, bacteria and viruses (Johnson and
Lyon 1988, Burns and Honkala 1990, Chong et al. 2012).
57
Structure of Dissertation
In Virginia, many forested headwater wetlands have been destroyed or altered.
Restoration of FHW has occurred following the recognition of their importance in the
landscape; however, these attempts have not been successful in returning lost ecological
structure, functions and services. In particular, successful establishment of trees has
proven difficult. There are limited recommendations regarding selection of species and
stocktypes with associated economic costs for planting trees in restored FHW.
Additionally, there is limited information regarding the biomass accumulation rates of
trees planted across hydrologic gradients that has precluded the development of
appropriate woody EPSs for CMS. The original research conducted to address these
limitations is presented in four chapters followed by a concluding chapter. The
overarching goal of this dissertation was to improve the probability of ecologically
successful FHW restoration.
CHAPTER 2: EXPERIMENTAL FORESTED WETLAND PLANTED TREE
SURVIVAL AND MORPHOLOGY
The objectives of this chapter were to investigate the differences in survival and
development of morphological structure (stem diameter, crown diameter, and height) of
planted trees over 5 years in response to; 1) species selection, 2) stocktype selection, and
3) soil physical, chemical and hydrologic conditions. The hypotheses of this chapter
were; 1) species’ responses (survival and morphology) would vary based on hydrologic
treatments, 2) stressful hydrologic treatments would reduce survival and morphological
growth, 3) the early successional species would have greater survival and growth than the
58
late successional (Quercus spp.) species, and 4) the larger stocktype (1-gallon container)
would have greater survival and growth than the smaller stocktypes (bare root and
tubeling).
CHAPTER 3: WOODY BIOMASS DEVELOPMENT OF SEVEN MIDATLANTIC SPECIES GROWN UNDER THREE HYDROLOGIC CONDITIONS
OVER 6 YEARS
The objective of this chapter was to develop biomass estimation models for the
seven study species using destructive harvests. In order to reach this objective the
following hypotheses were tested; 1) belowground biomass has a non-linear relationship
with aboveground woody biomass and 2) total aboveground and belowground biomass
has a non-linear relationship with stem cross-sectional diameter at groundline. The
resulting biomass estimation model was used to estimate the biomass of all living trees in
order to evaluate species and stocktype performance across hydrologic treatments after 6
years.
CHAPTER 4: EFFECT OF HYDROLOGIC CONDITIONS ON ABSOLUTE AND
RELATIVE GROWTH RATES OF SEVEN WETLAND TREE SPECIES
The objective of this chapter was to determine the absolute and relative biomass
growth rates of the seven study species over six years in response to soil physical,
chemical and hydrologic conditions. Absolute and relative growth rates were compared
among species and hydrologic treatments, with similar hypotheses to Chapter 2.
59
CHAPTER 5: DEVELOPING AN ECOLOGICAL PERFORMANCE STANDARD
FOR WOODY VEGETATION IN COMPENSATORY MITIGATION
WETLANDS OF VIRGINIA
The objective of this chapter was to develop a woody ecological performance
standard for use in forested wetland compensatory mitigation site monitoring. In order to
address this objective the following hypotheses were tested; 1) morphological
measurements are linearly related to each other and 2) existing woody ecological
performance standards for Virginia are statistically related to biomass accumulation.
Based on the results of testing these hypotheses, an additional ecological performance
standard was developed.
CHAPTER 6: ECONOMIC ANALYSIS AND OVERALL CONCLUSIONS
The objective of this chapter was to evaluate the economic cost associated with
planting trees in restored forested wetlands and to summarize the results from the
individual chapters to draw overarching conclusions.
60
Literature Cited
Adkins, C. R., Frank, S. D., Ward, N. A., & Fulcher, A. F. (2012). Birch - Betula spp. In
A. F. Fulcher, & S. A. White (Eds.), IPM for Select Deciduous Trees in
Southeastern US Nursery Production (2 ed.). Southern Nursery IPM working
group.
Alberty, C. A., Pellett, H. M., & Taylor, D. H. (1984). Characterization of soil
compaction at construction sites and woody plant responses. Journal of
Environmental Horticulture, 2, 48-53.
Allen, A. W. (1987). Habitat suitability index models: barred owl. Biological Report,
United States Fish and Wildlife Service.
Allen, J. A., & Kennedy, H. E. (1989). Bottomland hardwood reforestation in the Lower
Mississippi Valley. Tech. rep., United States Department of the Interior - Fish and
Wildlife Service, National Wetlands Research Center, Southern Forest
Experiment Station, Stoneville, MS.
Allison, P. (2010). Survival Analysis Using SAS®: A Practical Guide Second Edition.
Cary, NC: SAS Publishing.
Alm, A. A., & Schantz-Hansen, R. (1974). Tubeling research plantings in Minnesota. In
W. I. Richard W. Tinus (Ed.), Proceedings of the North American Containerized
Forest Tree Seedling Symposium (p. 458). Great Plains Agricultural Council.
American Nursery & Landscape Association (2004). American Standard for Nursery
Stock. Tech. rep., American Nursery & Landscape Association.
Anderson, C., & Cowell, B. (2004). Mulching effects on the seasonally flooded zone of
west-central Florida, USA. Wetlands, 24(4), 811-819.
Anderson, C., & Mitsch, W. (2008). Tree Basal Growth Response to Flooding in a
Bottomland Hardwood Forest in Central Ohio. JAWRA Journal of the American
Water Resources Association, 44(6), 1512-1520.
Anderson, P., & Pezeshki, R. S. (2001). Effects of flood pre-conditioning on responses of
three bottomland tree species to soil waterlogging. Journal of Plant Physiology,
158(2), 227-233.
61
Atkinson, R., & Cairns Jr, J. (2001). Plant decomposition and litter accumulation in
depressional wetlands: functional performance of two wetland age classes that
were created via excavation. Wetlands, 21(3), 354-362.
Atkinson, R., Perry, J., & Cairns Jr, J. (2005). Vegetation communities of 20-year-old
created depressional wetlands. Wetlands Ecology and Management, 13(4), 469478.
Atkinson, R., Perry, J., Noe, G., Daniels, W., & Cairns, J. J. (2010). Primary productivity
in 20-year old created wetlands in Southwestern Virginia. Wetlands, 30(2), 200210.
Atkinson, R., Perry, J., Smith, E., & Cairns Jr, J. (1993). Use of created wetland
delineation and weighted averages as a component of assessment. Wetlands,
13(3), 185-193.
Bailey, D., Perry, J., & Daniels, W. (2007). Vegetation dynamics in response to organic
matter loading rates in a created freshwater wetland in southeastern Virginia.
Wetlands, 27(4), 936-950.
Balcombe, C., Anderson, J., Fortney, R., & Kordek, W. (2005). Wildlife use of
mitigation and reference wetlands in West Virginia. Ecological Engineering,
25(1), 85-99.
Bazzaz, F. (1979). The physiological ecology of plant succession. Annual Review of
Ecology and Systematics, 10, 351-371.
Bedford, B. (1996). The need to define hydrologic equivalence at the landscape scale for
freshwater wetland mitigation. Ecological Applications, 6(1), 57-68.
Bergschneider, C. (2005). Determining an appropriate organic matter loading rate for a
created coastal plain forested wetland. Master's Thesis, Virginia Polytechnic
Institute and State University, Blacksburg, VA, USA.
Bishel-Machung, L., Brooks, R., Yates, S., & Hoover, K. (1996). Soil properties of
reference wetlands and wetland creation projects in Pennsylvania. Wetlands,
16(4), 532-541.
Blackman, V. (1919). The compound interest law and plant growth. Annals of Botany(3),
353.
62
Brady, N. C., & Weil, R. R. (2002). The Nature and Properties of Soil. Upper Saddle
River, NJ: Prentice Hall.
Breaux, A., & Serefiddin, F. (1999). Validity of performance criteria and a tentative
model for regulatory use in compensatory wetland mitigation permitting.
Environmental Management, 24(3), 327-336.
Brinson, M. (1993). A hydrogeomorphic classification for wetlands. Wetlands REsearch
Program Technical Report, United States Army Corps of Engineers Waterways
Experiment Station, Washington D.C.
Brinson, M. (1993a). Changes in the functioning of wetlands along environmental
gradients. Wetlands, 13(2), 65-74.
Brinson, M. (1996). Assessing wetland functions using HGM. National Wetlands
Newsletter, 18(1), 10-16.
Brinson, M., Lugo, A., & Brown, S. (1981). Primary productivity, decomposition and
consumer activity in freshwater wetlands. Annual Review of Ecology and
Systematics, 12, 123-161.
Brinson, M., Miller, K., Rheinhardt, R., Christian, R., Meyer, G., & O'Neal, J. (2006).
Developing reference data to identify and calibrate indicators of riparian
ecosystem condition in rural coastal plain landscapes in North Carolina. Report
to the Ecosystem Enhancement Program. North Carolina Department of
Environment and Natural Resources, East Carolina University.
Brinson, M., Rheinhardt, R. D., Hauer, R., Lee, L. C., Nutter, W. L., Smith, R. D., &
Whigham, D. (1995). A guidebook for application of hydrogeomorphic
assessments to riverine wetlands. United States Army Corps of Engineers and
Waterways Experiment Station.
Brinson, M., Swift, B., Plantico, R., & Barclay, J. (1981). Riparian Ecosystems:Their
Ecology and Status. Tech. rep., Eastern Energy and Land Use Team National
Water Resources Analysis Group U.S. Fish and Wildlife Service, Kearneyscille,
WV.
Brissette, J. C., Barnett, J. P., & Landis, T. D. (1991). Container seedlings. In M. L.
Duryea, & P. M. Dougherty (Eds.), Forest Regeneration Manual (pp. 117-141).
Kluwer Academic Publishers.
63
Britt, J. R., Mitchell, R. J., Zutter, B. R., South, D. B., Gjerstad, D. H., & Dickson, J. F.
(1991). The influence of herbaceous weed control and seedling diameter on six
years of loblolly pine growth--a classical growth analysis approach. Forest
Science, 37(2), 655-668.
Brown, S., & Veneman, P. (2001). Effectiveness of compensatory wetland mitigation in
Massachusetts, USA. Wetlands, 21(4), 508-518.
Bruland, G. L., & Richardson, C. J. (2005). Spatial variability of soil properties in
created, restored, and paired natural wetlands. Soil Science Society of America
Journal, 69(1), 273.
Bruland, G., & Richardson, C. (2004). Hydrologic gradients and topsoil additions affect
soil properties of Virginia created wetlands. Soil Science Society of America
Journal, 68(6), 2069-2077.
Bruland, G., & Richardson, C. (2006). Comparison of soil organic matter in created,
restored and paired natural wetlands in North Carolina. Wetlands Ecology and
Management, 14(3), 245-251.
Bullard, S., Hodges, J., Johnson, R., & Straka, T. (1992). Economics of direct seeding
and planting for establishing oak stands on old-field sites in the South. Southern
Journal of Applied Forestry, 16(1), 34-40.
Bullock, B., & Burkhart, H. (2005). An evaluation of spatial dependency in juvenile
loblolly pine stands using stem diameter. Forest Science, 51(2), 102-108.
Burdett, A. N., Herring, L. J., & Thompson, C. F. (1984). Early growth of planted spruce.
Canadian Journal of Forest Research, 14(5), 644-651.
Burgman, M., Incoll, W., Ades, P., Ferguson, I., Fletcher, T., & Wohlers, A. (1994).
Mortality models for mountain and alpine ash. Forest Ecology and Management,
67(1-3), 319-327.
Burkett, V. R., Draugelis-Dale, R. O., Williams, H. W., & Schoenholtz, S. H. (2005).
Effects of flooding regime and seedling treatment on early survival and growth of
nuttall oak. Restoration Ecology, 13(3), 471-479.
Burns, R., & Honkala, B. (1990). Silvics of North America, Volume 2, Hardwoods.
Agriculture Handbook.
64
Cairns Jr, J. (2000). Setting ecological restoration goals for technical feasibility and
scientific validity. Ecological Engineering, 15(3), 171-180.
Campbell, D., Cole, C., & Brooks, R. (2002). A comparison of created and natural
wetlands in Pennsylvania, USA. Wetlands Ecology and Management, 10(1), 4149.
Carey, J. H. (1992). Quercus phellos. In Rocky Mountain Research Station, Fire Sciences
Laboratory (Ed.), Fire Effects Information System. United States Department of
Agriculture, Forest Service.
Carey, J. H. (1992a). Quercus palustris. In Rocky Mountain Research Station, Fire
Sciences Laboratory (Ed.), Fire Effects Information System. United States
Department of Agriculture, Forest Service.
Casanova, M., & Brock, M. (2000). How do depth, duration and frequency of flooding
influence the establishment of wetland plant communities? Plant Ecology, 147(2),
237-250.
Chavasse, C. G. (1980). Planting stock quality: A review of factors affecting
performance. New Zealand Journal of Forestry Science, 25, 144-171.
Chojnacky, D. C., Heath, L. S., & Jenkins, J. C. (2014). Updated generalized biomass
equations for North American tree species. Forestry, 87(1), 129-151.
Chong, J.-H., Ward, N. A., Dunwell, W. C., & Chappell, M. R. (2012). Oak - Quercus
spp. In A. F. Fulcher, & S. A. White (Eds.), IPM for Select Deciduous Trees in
Southeastern US Nursery Production (2 ed.). Southern Nursery IPM working
group.
Clarke, P. (2002). Experiments on tree and shrub establishment in temperate grassy
woodlands: Seedling survival. Austral Ecology, 27(6), 606-615.
Clewell, A. F., & Lea, R. (1989). Creation and restoration of forested wetland vegetation
in the southeastern United States. In J. Kusler (Ed.), Wetland creation and
restoration: The status of the Science (pp. 199-237). United States Department of
Commerce, National Technical Information Service. Springfield, Virginia
Coladonato, M. (1992). Liquidambar styraciflua. In Rocky Mountain Research Station,
Fire Sciences Laboratory (Ed.), Fire Effects Information System. United States
Department of Agriculture, Forest Service.
65
Cole, C. (2002). The assessment of herbaceous plant cover in wetlands as an indicator of
function. Ecological Indicators, 2(3), 287-293.
Cole, C., & Shafer, D. (2002). Section 404 wetland mitigation and permit success criteria
in Pennsylvania, USA, 1986--1999. Environmental Management, 30(4), 508-515.
Cole, C., Brooks, R., & Wardrop, D. (2001). Assessing the relationship between biomass
and soil organic matter in created wetlands of central Pennsylvania, USA.
Ecological Engineering, 17(4), 423-428.
Conner, W. H., Inabinette, L. W., & Brantley, E. F. (2000). The use of tree shelters in
restoring forest species to a floodplain delta: 5-year results. Ecological
Engineering, 15, S47--S56.
Council on Environmental Quality (CEQ) (1978). National Environmental Policy Act –
Regulation: Implementation of the Procedural Provisions. Tech. rep., Council on
Environmental Quality.
Cox, D. R. (1972). Regression models and Life-Tables. Journal of the Royal Statistical
Society. Series B (Methodological), 34(2), 187-220.
Craft, C., & Casey, W. (2000). Sediment and nutrient accumulation in floodplain and
depressional freshwater wetlands of Georgia, USA. Wetlands, 20(2), 323-332.
Cypert, E., & Webster, B. S. (1948). Yield and use by wildlife of acorns of water and
willow oaks. The Journal of Wildlife Management, 12(3), 227-231.
Dahl, T. (1990). Wetlands losses in the United States, 1780's to 1980's. Report to the
Congress. Tech. rep., United States Department of the Interior, Fish and Wildlife
Service, Washington, D.C.
Dahl, T., & Johnson, C. (1991). Status and trends of wetlands in the conterminous United
States, mid-1970's to mid-1980's. Tech. rep., United States Department of the
Interior, Fish and Wildlife Service.
Daniels, W. L., & Whittecar, G. R. (2011). Assessing soil and hydrologic properties for
the successful creation of non-tidal wetlands. In L. M. Vasilas, & B. L. Vasilas
(Eds.), A Guide to Hydric Soils in the Mid-Atlantic Region (Vol. 2.0, pp. 120136). Mid-Atlantic Hydric Soils Committee.
Daniels, W. L., Perry, J., & Whittecar, R. G. (2005). Effects of soil amendments and other
practices upon the success of the Virginia Department of Transportation’s non-
66
tidal wetland mitigation efforts. Final Contract Report, Virginia Transportation
Research Council, Charlottesville, VA.
Das, A. (2012). The effect of size and competition on tree growth rate in old-growth
coniferous forests. Canadian Journal of Forest Research, 42(11), 1983-1995.
Davis, A. A., & Trettin, C. C. (2006). Sycamore and sweetgum plantation productivity on
former agricultural land in South Carolina. Biomass and Bioenergy, 30(8), 769777.
Davis, M. A., Wrage, K. J., Reich, P. B., Tjoelker, M. G., Schaeffer, T., & Muermann, C.
(1999). Survival, growth, and photosynthesis of tree seedlings competing with
herbaceous vegetation along a water-light-nitrogen gradient. Plant Ecology,
145(2), 341-350.
Davis, M., Wrage, K., & Reich, P. (1998). Competition between tree seedlings and
herbaceous vegetation: support for a theory of resource supply and demand.
Journal of Ecology, 86(4), 652-661.
Day, F. P. (1987). Effects of flooding and nutrient enrichment on biomass allocation in
Acer rubrum seedlings. American Journal of Botany, 74(10), 1541-1554.
De Steven, D. (1991). Experiments on mechanisms of tree establishment in old-field
succession: seedling emergence. Ecology, 1066-1075.
DeBerry, D. A., & Perry, J. (2012). Vegetation dynamics across a chronosequence of
created wetland sites in Virginia, USA. Wetlands Ecology 20(6), 521-537.
DeBerry, D. A., & Perry, J. (2015). Using the floristic quality concept to assess created
and natural wetlands: Ecological and management implications. Ecological
Indicators, 53, 247-257.
Denton, S. (1990). Growth rates, morphometrics and planting recommendations for
cypress trees at forested mitigation sites. In F. Webb (Ed.), Proceedings of the
17th Annual Conference on Wetlands Restoration and Creation, (pp. 24-32).
Tampa, FL.
Dey, D. C., Gardiner, E. S., Kabrick, J. M., Stanturf, J. A., & Jacobs, D. F. (2010).
Innovations in afforestation of agricultural bottomlands to restore native forests in
the eastern USA. Scandinavian Jornal of Forest Research, 25(Suppl 8), 31-42.
67
Dey, D. C., Jacobs, D., McNabb, K., Miller, G., Baldwin, V., & Foster, G. (2008).
Artificial regeneration of major oak (Quercus) species in the eastern United
States—a review of the literature. Forest Science, 54(1), 77-106.
Dey, D. C., Lovelace, W., Kabrick, J. M., & Gold, M. A. (2004). Production and early
field performance of RPM® seedlings in Missouri floodplains. In C. H. Michler,
P. M. Pijut, J. W. Van Sambeek, M. V. Coggeshall, J. Seifert, K. Woeste, et al.
(Ed.), Proceedings of the 6th Walnut Council Research Symposium, (pp. 59-65).
Dickerson, J. (2002). Pin Oak Quercus palustris Muenchh. Plant Fact Sheet. Tech. rep.,
United States Department of Agriculture, Natural Resources Conservation
Service.
Dickinson, S. B. (2007). Influences of soil amendments and microtopography on
vegetation at a created tidal freshwater swamp in southeastern Virginia. Master's
thesis, Virginia Polytechnic Institute and State University.
Dickson, J. G., Conner, R. N., & Williamson, J. H. (1983). Snag retention increases bird
use of a clear-cut. The Journal of Wildlife Management, 47(3), 799-804.
Dixon, R. K., Garrett, H. E., Cox, G. S., & Pallardy, S. G. (1984). Mycorrhizae and
reforestation success in the oak hickory region. In M. L. Duryea, & G. N. Brown
(Eds.), Seedling physiology and reforestation success. Martinus Nijhoff/Dr W.
Junk Publishers.
Dosskey, M., & Bertsch, P. (1994). Forest sources and pathways of organic matter
transport to a blackwater stream: a hydrologic approach. Biogeochemistry, 24(1),
1-19.
Dosskey, M., Vidon, P., Gurwick, N., Allan, C., Duval, T., & Lowrance, R. (2010). The
Role of Riparian Vegetation in Protecting and Improving Chemical Water Quality
in Streams. JAWRA Journal of the American Water Resources Association, 46(2),
261-277.
Dugger, K. M., & Fredrickson, L. H. (1992). Life history and habitat needs of the wood
duck. Fish and Wildlife Leaflet, United States Department of the Interior, Fish
and Wildlife Service, Washington, DC.
68
Dulohery, C., Kolka, R., & McKevlin, M. (2000). Effects of a willow overstory on
planted seedlings in a bottomland restoration. Ecological Engineering, 15, S57-S66.
Duryea, M. L., & Landis, T. D. (Eds.). (1984). Forest Nursery Manual: Production of
Bareroot Seedlings. Martinus Nijhoff/Dr W. Junk Publishers.
Duryea, M. L., & McClain, K. M. (1984). Altering seedling physiology to improve
reforestation. In M. L. Duryea, & G. N. Brown (Eds.), Seedling physiology and
reforestation success. Martinus Nijhoff/Dr W. Junk Publishers.
Dyer, J. (2006). Revisiting the deciduous forests of eastern North America. BioScience,
56(4), 341-352.
Ebel, B. H., & Kormanik, P. P. (1966). Stictocephala militaris, a mebracid Homoptera
associated with sweetgum, Liquidambar styraciflua. Annals of the Entomological
Society of America, 59(3), 600-601.
Effler, R., & Goyer, R. (2006). Baldcypress and water tupelo sapling response to multiple
stress agents and reforestation implications for Louisiana swamps. Forest Ecology
and Management, 226(1-3), 330-340.
Ehrenfeld, J. G. (1980). Understory response to canopy gaps of varying size in a mature
oak forest. Bulletin of the Torrey Botanical Club, 107, 29-41.
Environmental Law Institute. (2004). Measuring Mitigation: A review of the science for
compensatory mitigation performance standards. Tech. rep., Environmental Law
Institute.
Evans, G. C. (1972). The Quantitative Analysis of Plant Growth. Berkley and Los
Angeles: University of California Press.
Faber-Langendoen, D., & (Program), N. (2006). Ecological Integrity Assessment and
Performance Measures for Wetland Mitigation. NatureServe.
Faber-Langendoen, D., Kudray, G., Nordman, C., Sneddon, L., Vance, L., Byers, E., et
al. (2008). Ecological performance standards for wetland mitigation: An
approach based on ecological intergrity assessments. Arlington, VA:
NatureServe.
Farmer Jr., R. E. (1980). Comparative analysis of first-year growth in six deciduous tree
species. Canadian Journal of Forestry, 10, 35-41.
69
Fisher, J., & Acreman, M. (2004). Wetland nutrient removal: a review of the evidence.
Hydrology and Earth System Sciences, 8(4), 673-685.
Freeman, M., Pringle, C., & Jackson, C. (2007). Hydrologic connectivity and the
contribution of stream headwaters to ecological integrity at regional scales.
JAWRA Journal of the American Water Resources Association, 43(1), 5-14.
Frey, B. R., Ashton, M. S., McKenna, J. J., Ellum, D., & Finkral, A. (2007). Topographic
and temporal patterns in tree seedling establishment, growth, and survival among
masting species of southern New England mixed-deciduous forests. Forest
Ecology and Management, 245(1-3), 54-63.
Galatowitsch, S. M., & van der Valk, A. G. (1996). Vegetation and environmental
conditions in recently restored wetlands in the prairie pothole region of the USA.
Plant Ecology, 126(1), 89-99.
Gamble, D. L., & Mitsch, W. J. (2006). Tree growth and hydrologic patterns in urban
forested mitigation wetlands. Tech. rep., Schiermeier Olentangy River Wetland
Research Park, School of Natural Resources, The Ohio State University.
Gardiner, E. , Russell, D., Oliver, M, & Dorris JR, L. (2002). Bottomland hardwood
afforestation: State of the art. In Holland, M., Warren, M., & Stanturf, J. (Ed.)
Proceedings of a conference on sustainability of wetlands and water resources:
how well can riverine wetlands continue to support society into the 21st century?
Gen. Tech. Rep. SRS-50. United States Department of Agriculture Forest Service
Southern Research Station. Asheville, NC.
Gardiner, E., Salifu, K., Jacobs, D., Hernandez, G., & Overton, R. (2007). Field
performance of Nuttall Oak on former agricultural fields: Initial effects of nursery
source and competition control. In L. E. Riley, R. Dumroese, & T. Landis (Ed.),
National proceedings: Forest and Conservation Nursery Associations. Fort
Collins, CO.
Gayon, J. (2000). History of the concept of allometry. American Zoologist, 40(5), 748758.
Gemborys, S. (1974). The structure of hardwood forest ecosystems of Prince Edward
County, Virginia. Ecology, 55(3), 614-621.
70
Geudens, G., Staelens, J., Kint, V., Goris, R., & Lust, N. (2004). Allometric biomass
equations for Scots pine (Pinus sylvestris L.) seedlings during the first years of
establishment in dense natural regeneration. Annals of Forest Science, 61(7), 653659.
Gill, D., & Marks, P. (1991). Tree and shrub seedling colonization of old fields in central
New York. Ecological Monographs, 183-205.
Gjerstad, D. H., Nelson, L. R., Dukes Jr., J. H., & Retzlaff, W. A. (1984). Growth
response and physiology of tree seedlings as affected by weed control. In M. L.
Duryea, & G. N. Brown (Eds.), Seedling physiology and reforestation success.
Dordrecht, The Netherlands: Martinus Nijhoff/Dr W. Junk Publishers.
Glascock, S., & Ware, S. A. (1979). Forests of small stream bottoms in the peninsula of
Virginia. Virginia Journal of Science, 30, 17-21.
Grace, J., & Tilman, D. (Eds.). (1990). Perspectives on plant competition. Academic
Press, Inc.
Grime, J. (2002). Plant strategies, vegetation processes, and ecosystem properties. John
Wiley & Sons Inc.
Grimm, W. C. (1983). The Illustrated Book of Trees. Harrisonburg, PA: Stackpole
Books.
Groffman, P., Howard, G., Gold, A., & Nelson, W. (1996). Microbial nitrate processing
in shallow groundwater in a riparian forest. Journal of Environmental Quality,
25(6), 1309-1316.
Groninger, J. W., Baer, S. G., Babassana, D. A., & Allen, D. H. (2004). Planted green ash
(Fraxinus pennsylvanica Marsh.) and herbaceous vegetation responses to initial
competition control during the first 3 years of afforestation. Forest Ecology and
Management, 189(1-3), 161-170.
Grossman, B., Gold, M., & Dey, D. (2003). Restoration of hard mast species for wildlife
in Missouri using precocious flowering oak in the Missouri River floodplain,
USA. Agroforestry Systems, 59(1), 3-10.
Grossnickle, S. C., & Blake, T. J. (1987). Water relation patterns of bare-root and
container jack pine and black spruce seedling planted on boreal cut-over sites.
New Forests, 1, 101-116.
71
Grossnickle, S. C., & El-Kassaby, Y. A. (2015). Bareroot versus container stocktypes: a
performance comparison. New Forests 47(1), 1-51.
Hale, M. G., & Orcutt, D. M. (1987). The Physiology of Plants Under Stress. New York,
New York: John Wiley & Sons.
Hardin, K. I., & Evans, K. E. (1977). Cavity nesting bird habitat in the oak-hickory
forests -- a review. Gen. Teck. Rep., United States Department of Agriculture,
Forest Service, North Central Forest Experiment Station, St. Paul, MN.
Harlow, R. F., Shrauder, P. A., & Seehorn, M. E. (1975). Deer browse resources of the
Oconee National Forest. Research Paper, United States Department of
Agriculture, Forest Service, Southeastern Forest Experiment Station, Asheville,
NC.
Harmer, R., & Ker, R. G. (1995). Creating Woodlands: To Plant Trees or Not? In R.
Ferris-Kaan (Ed.), The Ecology of Woodland Creation (pp. 113-128). John Wiley
& Sons Ltd.
Harmon, M., Franklin, J., Swanson, F., Sollins, P., Gregory, S., Lattin, J., Anderson, N.,
Cline, S., Aumen, N., Sedell, J., Lienkaemper, G., Cromack Jr., K. & Cummins,
K. (1986). Ecology of coarse woody debris in temperate ecosystems. Advances in
Ecological Research, 15(133), 302.
Hastwell, G., & Facelli, J. (2003). Differing effects of shade-induced facilitation on
growth and survival during the establishment of a chenopod shrub. Journal of
Ecology, 91(6), 941-950.
Havens, K. (2004). A comparison of C. caroliniana, Q. michauxii, Q. pagoda, and T.
distichum seedlings of upland and wetland stock for use in created or restored
forested wetlands. Ecological Engineering 23, 341-349.
Hefting, M., Clement, J., Bienkowski, P., Dowrick, D., Guenat, C., Butturini, A., et al.
(2005). The role of vegetation and litter in the nitrogen dynamics of riparian
buffer zones in Europe. Ecological Engineering, 24(5), 465-482.
Henderson, D., Botch, P., Cussimanio, J., Ryan, D., Kabrick, J., & Dey, D. (2009).
Growth and mortality of pin oak and pecan reforestation in a constructed
wetland- Analysis with management implications. Science and Management
Technical Series, USDA Forest Service, Jefferson City, MO.
72
Hershner, C., Havens, K., Berman, M., Rudnicky, T., & Schatt, D. (2003). Wetlands of
Virginia: total, isolated and headwater. Special Report, Center for Coastal
Resources Management, Virginia Institute of Marine Science.
Hilderbrand, R. H., Lemly, A. D., Dolloff, C. A. & Harpster, K. L. (1997). Effects of
large woody debris placement on stream channels and benthic macroinvertebrates.
Canadian Journal of Fisheries and Aquatic Sciences, 54(4), 931-939.
Hoeltje, S., & Cole, C. (2007). Losing function through wetland mitigation in central
Pennsylvania, USA. Environmental Management, 39(3), 385-402.
Hruby, T. (1999). Assessments of wetland functions: What they are and what they are
not. Environmental Management, 23(1), 75-85.
Hudson, H.W. (2010). The Effect of Adjacent Forests on Colonizing Tree Density in
Restored Wetland Compensation Sites in Virginia. Master's Thesis, Christopher
Newport University. Newport News, VA.
Hunt, R. (1982). Plant Growth Curves - The functional approach to plant growth
analysis. Edward Arnold.
Hunt, R. (1990). Basic growth analysis - Plant growth analysis for beginners. Broadwick
Street, London: Unwin Hyman, Ltd.
Hupp, C., Woodside, M., & Yanosky, T. (1993). Sediment and trace element trapping in
a forested wetland, Chickahominy River, Virginia. Wetlands, 13(2), 95-104.
Jackson, D. P., Dumroese, R. K., & Barnett, J. P. (2012). Nursery response of container
Pinus palustris seedlings to nitrogen supply and subsequent effects on outplanting
performance. Forest Ecology and Management, 265, 1-12.
Jenkins, J., Chojnacky, D., Heath, L., & Birdsey, R. (2003). National-scale biomass
estimators for United States tree species. Forest Science, 49(1), 12-35.
Johnson, W. T., & Lyon, H. H. (1988). Insects that feed on trees and shrubs (Second ed.).
Comstock Publishing Associates.
Jørgensen, S. E., Patten, B. C., & Straškraba, M. (2000). Ecosystems emerging:: 4.
growth. Ecological Modelling, 126(2), 249-284.
Kabrick, J. M., Dey, D. C., Van Sambeek, J., Wallendorf, M., & Gold, M. A. (2005). Soil
properties and growth of swamp white oak and pin oak on bedded soils in the
73
lower Missouri River floodplain. Forest Ecology and Management, 204(2), 315327.
Kabrick, J., Dey, D., Van Sambeek, J., Coggeshall, M., & Jacobs, D. (2012). Quantifying
flooding effects on hardwood seedling survival and growth for bottomland
restoration. New Forests, 1-16.
Keddy, P. (1992). Assembly and response rules: two goals for predictive community
ecology. Journal of Vegetation Science, 3(2), 157-164.
Keeland, B., Conner, W., & Sharitz, R. (1997). A comparison of wetland tree growth
response to hydrologic regime in Louisiana and South Carolina. Forest Ecology
and Management, 90(2-3), 237-250.
Kentula, M. E. (2000). Perspectives on setting success criteria for wetland restoration.
Ecological Engineering, 15, 199-209.
Kentula, M. E., Brooks, R. P., Gwin, S. E., Holland, C. C., Sherman, A. D., & Sifneos, J.
C. (1992). Wetlands: an approach to improving decision making in wetland
restoration and creation. (A. J. Hairston, Ed.) Island Press.
Kozlowski, T. (1984). Responses of woody plants to flooding. In Kozlowski, T. (Ed.)
Flooding and Plant Growth. (pp. 129-163). Academic Press, Inc.
Kozlowski, T. T. (1999). Soil compaction and growth of woody plants. Scandinavian
Journal of Forest Research, 14, 596-619.
Landis, T. (1990). Containers: Types and Functions. In T. D. Landis, R. Tinus, S.
McDonald, & J. Barnett (Eds.), The Container Tree Nursery Manual (Vol. 2, pp.
1-39). Washington, D.C.: United States Department of Agriculture, Forest
Service.
Landis, T. Dumroese, R. K., & Haase, D.L. (2010). The Container Tree Nursery Manual
(Vol. 7). Agriculture Handbook 674. Washington, D.C.: United States
Department of Agriculture, Forest Service.
Latimer, J. G. & Close, D. (Eds.). (2014). Home Grounds and Animals: 2014 Pest
Management Guide. Virginia Cooperative Extension.
Leopold, L., Wolman, M. G., & Miller, J. (1964). Fluvial Processes in Geomorphology.
San Fancisco, CA: W.H. Freeman and Company.
74
Lichvar, R. W. (2013). The National Wetland Plant List: 2013 Wetland Ratings.
Phytoneuron, 49, 1-241.
Lichvar, R. W., Melvin, N. C., Butterwick, M. L., & Kirchner, W. N. (2012). National
wetland plant list indicator rating definitions. Tech. rep., United States Army
Corps of Engineers, Engineer Research and Development Center.
Lovelace, W. (1998). The Root Production Method (RPM) system for producing
container trees. Combined Proceedings International Plant Propagators' Society.
Lowrance, R., Altier, L., Newbold, J., Schnabel, R., Groffman, P., Denver, J., et al.
(1997). Water quality functions of riparian forest buffers in Chesapeake Bay
watersheds. Environmental Management, 21(5), 687-712.
MacArthur, R. H., & Wilson, E. O. (1967). The Theory of Island Biogeography.
Princeton University Press.
Martin, A. C., Zim, H. S., & Nelson, A. L. (1951). American Wildlife and Plants: A guide
to wildlife food habits: the use of trees, shshrub, wweed, and herbs by birds and
mammals of the United States (Second Edition ed.). McGraw-Hill BMcGraw-Hill,
Inc.
Matthews, J., & Endress, A. (2008). Performance criteria, compliance success, and
vegetation development in compensatory mitigation wetlands. Environmental
Management, 41(1), 130-141.
McCurry, J., Gray, M., & Mercker, D. (2010). Early Growing Season Flooding Influence
on Seedlings of Three Common Bottomland Hardwood Species in Western
Tennessee. Journal of Fish and Wildlife Management, 1(1), 11-18.
McKay, H.M. (1997). A review of the effect of stresses between lifting and planting on
nursery stock quality and performance. New Forests 13, 369-399.
McKevlin, M. (1992). Guide to regeneration of bottomland hardwoods. Gen. Tech. Rep.,
United States Department of Agriculture Forest Service, Asheville, N.C.
McLeod, K. (2000). Species selection trials and silvicultural techniques for the
restoration of bottomland hardwood forests. Ecological Engineering, 15, S35-S46.
McLeod, K., Reed, M. R., Moyer, B. P., & Ciravolo, T. G. (2006). Species selection trials
and silvicultural techniques for the restoration of bottomland hardwood forests -
75
A 10 year review. In K. F. Connor (Ed.), Proceedings of the 13th biennial
southern silvicultural research conference. Asheville, NC.
McLeod, K., & McPherson, J. (1973). Factors limiting the distribution of Salix nigra.
Bulletin of the Torrey Botanical Club, 100(2), 102-110.
McLeod, K., Reed, M., & Nelson, E. (2001). Influence of a willow canopy on tree
seedling establishment for wetland restoration. Wetlands, 21(3), 395-402.
McLeod, K., Reed, M., & Wike, L. (2000). Elevation, competition control, and species
affect bottomland forest restoration. Wetlands, 20(1), 162-168.
McQuilkin, W. (1940). The natural establishment of pine in abandoned fields in the
Piedmont Plateau region. Ecology, 21(2), 135-147.
Megonigal, J., & Day, F. (1992). Effects of flooding on root and shoot production of bald
cypress in large experimental enclosures. Ecology, 73(4), 1182-1193.
Mitsch, W., & Ewel, K. (1979). Comparative biomass and growth of cypress in Florida
wetlands. American Midland Naturalist, 101(2), 417-426.
Mitsch, W., & Gosselink, J. (2007). Wetlands (4 ed.). John Wiley & Sons, Inc.
Mitsch, W., & Gosselink, J. (2000). The value of wetlands: importance of scale and
landscape setting. Ecological Economics, 35(1), 25-33.
Mitsch, W., & Rust, W. (1984). Tree growth responses to flooding in a bottomland forest
in northeastern Illinois. Forest Science, 30(2), 499-510.
Mitsch, W., & Wilson, R. (1996). Improving the success of wetland creation and
restoration with know-how, time, and self-design. Ecological Applications, 6(1),
77-83.
Monette, R., & Ware, S. (1983). Early forest succession in the Virginia Coastal Plain.
Bulletin of the Torrey Botanical Club, 110(1), 80-86.
Moore, L. (2002). Willow Oak Quercus phellos L. Plant Fact Sheet. Tech. rep., United
States Department of Agriculture, Natural Resources Conservation Service.
Moreno-Mateos, D., Power, M. E., Comin, F. A., & Yockteng, R. (2012). Structural and
functional loss in restored wetland ecosystems. PLoS Biology, 10(1).
Morgan, K., & Roberts, T. (2003). Characterization of wetland mitigation projects in
Tennessee, USA. Wetlands, 23(1), 65-69.
76
Morley, T. R. (2008). The functions of forested headwater wetlands in a New England
Landscape. Ph.D. dissertation, The University of Maine.
Morris, L. A., Moss, S. A., & Garbett, W. S. (1993). Competitive interference between
selected herbaceous and woody plants and Pinus taeda L. during two growing
seasons following planting. Forest Science, 39(1), 166-187.
Mulholland, P., & Kuenzler, E. (1979). Organic carbon export from upland and forested
wetland watersheds. Limnology and Oceanography, 960-966.
Muller-Landau, H. C., Condit, R. S., Chave, J., Thomas, S. C., Bohlman, S. A.,
Bunyavejchewin, S., Davies, S., Foster, R., Gunatilleke, S., Gunatilleke, N., and
others (2006). Testing metabolic ecology theory for allometric scaling of tree size,
growth and mortality in tropical forests. Ecology Letters, 9(5), 575-588.
Myster, R., & McCarthy, B. (1989). Effects of herbivory and competition on survival of
Carya tomentosa (Juglandaceae) seedlings. Oikos, 56, 145-148.
Myster, R., & Pickett, S. (1993). Effects of litter, distance, density and vegetation patch
type on postdispersal tree seed predation in old fields. Oikos, 381-388.
National Research Council (NRC) (1995). Wetlands: Characteristics and Boundaries.
National Academy Press.
Nesom, G. (2009). Swamp white oak Quercus bicolor Willd. Plant Guide. Tech. rep.,
United States Department of Agriculture Natural Resources Conservation Service.
Niklas, K. J. (2004). Plant allometry: is there a grand unifying theory? Biological
reviews, 79(4), 871-889.
Niswander, S., & Mitsch, W. (1995). Functional analysis of a two-year-old created instream wetland: hydrology, phosphorus retention, and vegetation survival and
growth. Wetlands, 15(3), 212-225.
Nixon, C. M., McClain, M. W., & Russell, K. R. (1970). Deer food habits and range
characteristics in Ohio. The Journal of Wildlife Management, 34(4), 870-886.
Noble, C. V., Murray, E. O., Klimas, C. V., & Ainslie, W. (2011). Regional Guidebook
for Applying the Hydrogeomorphic Approach to Assessing the Functions of
Headwater Slope Wetlands on the South Carolina Coastal Plain. Wetlands
Regulatory Assistance Program, United States Army Corps of Engineers Engineer
Research and Development Center.
77
Noon, K. (1996). A model of created wetland primary succession. Landscape and Urban
planning, 34(2), 97-123.
Nyland, R. D. (2007). Silviculture: Concepts and Applications (Second Edition ed.).
Waveland Press, Inc.
Office of Technology Assessment (OTA) (1984). Wetlands: Their use and regulation.
Washington, D.C.: DIANE Publishing.
Paine, C. E., Marthews, T., Vogt, D., Purves, D., Rees, M., Hector, A., & Turnbull, L.
(2012). How to fit nonlinear plant growth models and calculate growth rates: an
update for ecologists. Methods in Ecology and Evolution, 3(2), 245-256.
Pallardy, S. G. (2008). Physiology of Woody Plants (Third ed.). Burlington, MA:
Academic Press, Inc.
Parsons, S., & Ware, S. (1982). Edaphic factors and vegetation in Virginia Coastal Plain
swamps. Bulletin of the Torrey Botanical Club, 365-370.
Paul, E. (Ed.). (2007). Soil microbiology, ecology, and biochemistry. Academic Press.
Pennington, M. R., & Walters, M. B. (2006). The response of planted trees to vegetation
zonation and soil redox potential in created wetlands. Forest Ecology and
Management, 233(1), 1-10.
Perry, M., Sibrel, C., & Gough, G. (1996). Wetlands mitigation: Partnership between an
electric power company and a federal wildlife refuge. Environmental
management, 20(6), 933-939.
Philipson, C. D., Saner, P., Marthews, T. R., Nilus, R., Reynolds, G., Turnbull, L. A., &
Hector, A. (2012). Light-based Regeneration Niches: Evidence from 21
Dipterocarp Species using Size-specific RGRs. Biotropica, 44(5), 627-636.
Picard, N., Rutishauser, E., Ploton, P., Ngomanda, A., & Henry, M. (2015). Should tree
biomass allometry be restricted to power models? Forest Ecology and
Management, 353, 156-163.
Pinto, J. R., Marshall, J. D., Dumroese, R. K., Davis, A. S., & Cobos, D. R. (2012).
Photosynthetic response, carbon isotopic composition, survival, and growth of
three stock types under water stress enhanced by vegetative competition.
Canadian Journal of Forest Research, 42, 333-344.
78
Pinto, J., Dumroese, R., Davis, A., & Landis, T. (2011b). Conducting seedling stocktype
trials: a new approach to an old question. Journal of Forestry, 109(5), 293-299.
Pinto, J., Marshall, J., Dumroese, R., Davis, A., & Cobos, D. (2011a). Establishment and
growth of container seedlings for reforestation: A function of stocktype and
edaphic conditions. Forest Ecology and Management, 261, 1876-1884.
Pommerening, A., & Muszta, A. (2016). Relative plant growth revisited: Towards a
mathematical standardisation of separate approaches. Ecological Modelling, 320,
383-392.
Radford, A. E., Ahles, H. E., & Bell, C. R. (1968). Manual of the vascular flora of the
Carolinas. Chapel Hill, NC: University of North Carolina Press.
Rees, M., Osborne, C. P., Woodward, F. I., Hulme, S. P., Turnbull, L. A., & Taylor, S. H.
(2010). Partitioning the components of relative growth rate: how important is
plant size variation? The American Naturalist, 176(6), E152--E161.
Rheinhardt, R., Brinson, M., Meyer, G., & Miller, K. (2012). Integrating forest biomass
and distance from channel to develop an indicator of riparian condition.
Ecological Indicators, 23, 46-55.
Rheinhardt, R. D., Brinson, M., Meyer, G., & Miller, K. (2012a). Carbon storage of
headwater riparian zones in an agricultural landscape. Carbon Balance and
Management, 7(4).
Rheinhardt, R., McKenney-Easterling, M., Brinson, M., Masina-Rubbo, J., Brooks, R.,
Whigham, D., O’Brien, D., Hite, J., & Armstrong, B. (2009). Canopy
Composition and Forest Structure Provide Restoration Targets for Low-Order
Riparian Ecosystems. Restoration Ecology, 17(1), 51-59.
Rheinhardt, R., Rheinhardt, M., Brinson, M., & Faser, K. (1998). Forested wetlands of
low order streams in the inner coastal plain of North Carolina, USA. Wetlands,
18(3), 365-378.
Rheinhardt, R., Whigham, D., Khan, H., & Brinson, M. (2000). Vegetation of headwater
wetlands in the inner coastal plain of Virginia and Maryland. Castanea, 21-35.
Robb, J. (2002). Assessing wetland compensatory mitigation sites to aid in establishing
mitigation ratios. Wetlands, 22(2), 435-440.
79
Rodriguez-Gonzalez, P., Stella, J., Campelo, F., Ferreira, M., & Albuquerque, A. (2010).
Subsidy or stress? Tree structure and growth in wetland forests along a
hydrological gradient in Southern Europe. Forest Ecology and Management,
259(10), 2015-2025.
Rose, K. E., Atkinson, R. L., Turnbull, L. A., & Rees, M. (2009). The costs and benefits
of fast living. Ecology Letters, 12(12), 1379-1384.
Row, J. M., & Geyer, W. A. (2010). Black willow Salix nigra Marsh. Plant Guide, United
States Department of Agriculture Natural Resources Conservation Service.
Santamour, J. F., & Greene, A. (1986). European hornet damage to ash and birch trees.
Journal of arboriculture (USA) 12(11), 273-279.
Sather, J. H., & Smith, R. D. (1984). An overview of major wetland functions and values.
Fort Collins, CO: United States Department of the Interior Fish and Wildlife
Service.
Scagel, R. (1993). Provincial seedling stock type selection and ordering guidelines.
British Columbia Ministry of Forests.
Schweitzer, C., Gardiner, E., Stanturf, J., & Ezell, A. (1999). Methods to improve
establishment and growth of bottomland hardwood artificial regeneration. In J. W.
Stringer, & D. L. Loftis (Ed.), Proceedings of the 12th Central Hardwood Forest
Conference, Gen. Tech. Rep. SRS-24. Asheville, NC.
Seitzinger, S. P. (1988). Denitrification in freshwater and coastal marine ecosystems:
ecological and geochemical significance. Limnology and Oceanography, 33(4),
702-724.
Shaffer, p. W., & Ernst, T. L. (1999). Distribution of SOM in freshwater wetlands in the
Protland, Oregon area. Wetlands, 19, 505-516.
Sharitz, R., Barton, C., & De Steven, D. (2006). Tree plantings in depression wetland
restorations show mixed success (South Carolina). Ecological Restoration, 24(2),
114-115.
Shaw, G. W., Dey, D., Kabrick, J., Grabner, J., & Muzika, R.-M. (2003). Comparison of
site preparation methods and stock types for artificial regeneration of oaks in
bottomlands. In J. W. Van Sambeek, J. O. Dawson, F. Ponder Jr., E. F.
80
Loewenstein, & J. S. Fralish (Ed.), Proceedings of the 13th Central Hardwood
Forest Conference, (pp. 186-198). St. Paul, MN.
Shaw, S. P., & Fredine, C. G. (1956). Wetlands of the United States – Their extent and
their value to waterfowl and other wildlife. Washington, D. C.: United States
Department of the Interior.
Siegel-Issem, C. M., Burger, J. A., Powers, R. F., Ponder, F., & Patterson, S. C. (2005).
Seedling root growth as a function of soil density and water content. Soil Science
Society of America Journal, 69, 215-226.
Silberhorn, G. (1992). Sweet Gum. Wetland Flora, Virginia Institute of Marine Science.
Silberhorn, G. M. (1994). A characterization of the plant community structure of
palustrine forested wetlands in the coastal plain of Virginia. Proceedings:
Saturated Forested Wetlands in the Mid-Atlantic Region: State of the Science. A
symposium sponsored by the U.S. Fish and Wildlife Service.
Sileshi, G. W. (2014). A critical review of forest biomass estimation models, common
mistakes and corrective measures. Forest Ecology and Management, 329, 237254.
Sillett, S., Van Pelt, R., Koch, G., Ambrose, A., Carroll, A., Antoine, M., & Mifsud, B.
(2010). Increasing wood production through old age in tall trees. Forest Ecology
and Management, 259(5), 976-994.
Smith, R. D., Ammann, A., Bartoldus, C., & Brinson, M. M. (1995). An approach for
assessing wetland functions using hydrogeomorphic classification, reference
wetlands, and functional indices. Technical Report, United States Army Engineer
Waterways Experiment Station, Vicksburg, MS.
Smith, M. C., Anthony Stallins, J., Maxwell, J. T., & Van Dyke, C. (2013). Hydrological
shifts and tree growth responses to river modification along the Apalachicola
River, Florida. Physical Geography, 34(6), 491-511.
Smith, W. B., & Brand, G. J. (1983). Allometric biomass equations for 98 species of
herbs, shrubs, and small trees. Research Note, United States Department of
Agriculture Forest Service, North Central Forest Experiment Station, St. Paul,
MN.
81
Smith, W. D., & Strub, M. R. (1991). Initial spacing: How many trees to plant. In M. L.
Duryea, & P. M. Dougherty (Eds.), Forest Regeneration Manual (pp. 281-289).
The Netherlands: Kluwer Academic Publishers.
Snyder, S. A. (1992). Quercus bicolor. In Rocky Mountain Research Station, Fire
Sciences Laboratory (Ed.), Fire Effects Information System. United States
Department of Agriculture, Forest Service.
Society for Ecological Restoration International Science & Policy Working Group.
(2004). The SER International Primer on Ecological Restoration. www.ser.org
Society for Ecological Restoration International.
South, D. B., & Mitchell, R. G. (2005). A Root-Bound Index for Container-Grown Pines.
In S. J. Colombo (Ed.), The Thin Green Line: A Symposium on the State-of-theArt in Reforestation. Ontario, Canada: Queen's Printer for Ontario.
South, D., Harris, S., Barnett, J., Hainds, M., & Gjerstad, D. (2005). Effect of container
type and seedling size on survival and early height growth of Pinus palustris
seedlings in Alabama, USA. Forest Ecology and Management, 204(2-3), 385398.
Spencer, D., Perry, J., & Silberhorn, G. (2001). Early secondary succession in bottomland
hardwood forests of southeastern Virginia. Environmental Management, 27(4),
559-570.
Spieles, D. (2005). Vegetation development in created, restored, and enhanced mitigation
wetland banks of the United States. Wetlands, 25(1), 51-63.
Stange, E. E., & Shea, K. L. (1998). Effects of deer browsing, fabric mats, and tree
shelters on Quercus rubra seedlings. Restoration Ecology, 6(1), 29-34.
Stanturf, J., Conner, W., Gardiner, E., Schweitzer, C., & Ezell, A. (2004). Recognizing
and overcoming difficult site conditions for afforestation of bottomland
hardwoods. Ecological Restoration, 22(3), 183.
Stanturf, J., Gardiner, E., Shepard, J., Schweitzer, C., Portwood, C., & Dorris Jr, L.
(2009). Restoration of bottomland hardwood forests across a treatment intensity
gradient. Forest Ecology and Management, 257(8), 1803-1814.
82
Stefanik, K., & Mitsch, W. (2012). Structural and functional vegetation development in
created and restored wetland mitigation banks of different ages. Ecological
Engineering, 39, 104-112.
Stephenson, N., Das, A., Condit, R., Russo, S., Baker, P., Beckman, N., . . . others.
(2014). Rate of tree carbon accumulation increases continuously with tree size.
Nature, 507, 90-93.
Stolt, M., Genthner, M., Daniels, W., Groover, V., Nagle, S., & Haering, K. (2000).
Comparison of soil and other environmental conditions in constructed and
adjacent palustrine reference wetlands. Wetlands, 20(4), 671-683.
Streever, B. (1999). Examples of performance standards for wetland creation and
restoration in Section 404 permits and an approach to developing performance
standards. Tech. rep., United States Army Corps of Engineers, Wetlands
Research Program.
Sudol, M., & Ambrose, R. (2002). The US Clean Water Act and habitat replacement:
evaluation of mitigation sites in Orange County, California, USA. Environmental
Management, 30(5), 727-734.
Sullivan, J. (1993). Betula nigra. In Rocky Mountain Research Station, Fire Sciences
Laboratory (Ed.), Fire Effects Information System. United States Department of
Agriculture, Forest Service.
Sullivan, J. (1994). Platanus occidentalis. In Rocky Mountain Research Station, Fire
Sciences Laboratory (Ed.), Fire Effects Information System. United States
Department of Agriculture, Forest Service.
Survey, U. S. (1999). National Water Summary - Wetland Resources: Virginia. Supply
Paper, United States Geological Survey, USGS National Center, Reston, VA.
Sweeney, B., Czapka, S., & Yerkes, T. (2002). Riparian forest restoration: increasing
success by reducing plant competition and herbivory. Restoration Ecology, 10(1),
392-400.
Tang, Z. C., & Kozlowski, T. T. (1982a). Physiological, morphological, and growth
responses of Platanus occidentalis seedlings to flooding. Plant and Soil, 66,
243:255.
83
Tang, Z. C., & Kozlowski, T. T. (1982b). Some physiological and growth responses of
Betula papyrifera seedlings to flooding. Physiologia Plantarum, 55(4), 415-420.
Taylor, D., Aarssen, L., & Loehle, C. (1990). On the relationship between r/K selection
and environmental carrying capacity: a new habitat templet for plant life history
strategies. Oikos, 239-250.
Taylor, T., Loewenstein, E., & Chappelka, A. (2004). Improving species composition in
mismanaged bottomland hardwood stands in western Alabama. In K. F. Connor
(Ed.), Proceedings of the 12th biennial southern silvicultural research
conference, (pp. 565-570). Asheville, NC.
Telfer, E. S. (1969). Weight-diameter relationships for 22 woody plant species. Canadian
Journal of Botany, 47(12), 1851-1855.
Temesgen, H., Affleck, D., Poudel, K., Gray, A., & Sessions, J. (2015). A review of the
challenges and opportunities in estimating above ground forest biomass using
tree-level models. Scandinavian Journal of Forest Research, 30(4), 326-335.
Ter-Mikaelian, M., & Korzukhin, M. (1997). Biomass equations for sixty-five North
American tree species. Forest Ecology and Management, 97(1), 1-24.
Tilman, D. (1988). Plant strategies and the dynamics and structure of plant communities
(Vol. 26). Princeton University Press.
Tiner, R. W., & Finn, J. T. (1986). Status and recent trends of wetlands in five MidAtlantic states: Delaware, Maryland, Pennsylvania, Virginia, and West Virginia.
Tech. rep., United States Department of the Interior, Fish and Wildlife Service,
Fish and Wildlife Enhancement, National Wetlands Inventory Project, Newton
Corner, MA.
Tischew, S., Baasch, A., Conrad, M., & Kirmer, A. (2010). Evaluating restoration
success of frequently implemented compensation measures: results and demands
for control procedures. Restoration Ecology, 18(4), 467-480.
Tritton, L. M., & Hornbeck, J. W. (1982). Biomass equations for major tree species of
the Northeast. General Technical Report, United Stated Department of
Agriculture, Forest Service Northeastn Forest Experiment Station, Broomall,
Pennsylvania.
84
Turnbull, L. A., Paul-Victor, C., Schmid, B., & Purves, D. W. (2008). Growth rates, seed
size, and physiology: do small-seeded species really grow faster. Ecology, 89(5),
1352-1363.
Turnbull, L. A., Philipson, C. D., Purves, D. W., Atkinson, R. L., Cunniff, J.,
Goodenough, A., . . . others. (2012). Plant growth rates and seed size: a reevaluation. Ecology, 93(6), 1283-1289.
Twedt, D. (2006). Small Clusters of Fast-Growing Trees Enhance Forest Structure on
Restored Bottomland Sites. Restoration Ecology, 14(2), 316-320.
Twedt, D., & Wilson, R. (2002). Supplemental planting of early successional tree species
during bottomland hardwood afforestation. In K. W. Outcalt (Ed.), Proceedings of
the eleventh biennial southern silvicultural research conference, (pp. 358-364).
Asheville, NC.
United States Army Corps of Engineers (USACE) (2008). Final Environmental
Assessment, Finding of No Significant Impact, and Regulatory Analysis for the
Compensatory Mitigation Regulation. Tech. rep., Directorate of Civil Works
Operations and Regulatory Community of Practice.
United States Army Corps of Engineers (USACE) & United States Environmental
Protection Agency (USEPA) (2008). Compensatory Mitigation for Losses of
Aquatic Resources; Final Rule. Tech. rep., United States Army Corps of
Engineers and United States Environmental Protection Agency, Washington D.C.
United States Army Corps of Engineers (USACE) Norfolk District & Virginia
Department of Environmental Quality (VADEQ) (2004). Recommendations for
wetland compensatory mitigation: Including site design, permit conditions,
performance and monitoring criteria. Recomendations, United States Army
Corps of Engineers Norfolk District and Virginia Department of Environmental
Quality.
United States Government Accountability Office (USGAO) (2005). Corps of Engineers
does not have an effective oversight approach to ensure that compensatory
mitigation is occurring. Tech. rep., United States Government Accountability
Office, Washington D.C.
85
Van der Valk, A. (1981). Succession in wetlands: a Gleasonian approach. Ecology, 62(3),
688-696.
Van Dersal, W. R. (1938). Native woody plants of the United States, their erosion-control
and wildlife values. United States Department of Agriculture.
Virginia Department of Environmental Quality (VADEQ) (2010). Mitigation Banking
Instrument Template. Template, Virginia Department of Environmental Quality,
Richmond, VA.
Virginia Department of Forestry (VADOF) (2012). Virginia Trees for Virginia
Landowners Seedling Catalog 2011-2012. Tech. rep., Virginia Department of
Forestry.
Wagner, R., & Ter-Mikaelian, M. (1999). Comparison of biomass component equations
for four species of northern coniferous tree seedlings. Annals of Forest Science,
56(3), 193-199.
Weiner, J., & Thomas, S. (2001). The nature of tree growth and the "age-related decline
in forest productivity". Oikos, 94(2), 374-376.
Westcott, C. (2001). Westcott's Plant Disease Handbook (6th Edition ed.). (R. K. Horst,
Ed.) Norwell, Massachusetts: Kluwer Academic Publishers.
Whittaker, R., & Woodwell, G. (1968). Dimension and production relations of trees and
shrubs in the Brookhaven Forest, New York. The Journal of Ecology, 56(1), 1-25.
Whittecar, G., & Daniels, W. (1999). Use of hydrogeomorphic concepts to design created
wetlands in southeastern Virginia. Geomorphology, 31(1-4), 355-371.
Williams, H. M., Burkett, V., & Craft, M. (1999). The Effects of Seedling Stock-Type
and Direct-Seeding on the Early Field Survival of Nuttall Oak Planted on
Agricultural Land. In T. D. Landis, & J. Barnett (Ed.), National Proceedings:
Forest Conservation Nursery Associations, (pp. 29-34).
Williams, R., & McClenahen, J. (1984). Biomass prediction equations for seedlings,
sprouts, and saplings of ten central hardwood species. Forest Science, 30(2), 523527.
Wilson, R., & Mitsch, W. (1996). Functional assessment of five wetlands constructed to
mitigate wetland loss in Ohio, USA. Wetlands, 16(4), 436-451.
86
Xiao, C., & Ceulemans, R. (2004). Allometric relationships for below-and aboveground
biomass of young Scots pines. Forest Ecology and Management, 203(1-3), 177186.
87
CHAPTER 2: EXPERIMENTAL FORESTED WETLAND PLANTED TREE
SURVIVAL AND MORPHOLOGY
88
Abstract
The destruction or conversion of forested wetland ecosystems has resulted in loss of
valuable ecosystem structures, functions and services from the landscape. Assisting the
recovery of these degraded ecosystems through restoration practices is often
unsuccessful. A major challenge associated with forested wetland restoration has been
establishment and growth of trees, driven by a lack of planting recommendations. The
purpose of this study was to investigate the differences in survival and development of
morphological structure (stem diameter, crown diameter, and height) of planted trees
over 5 years in response to: 1) species selection 2) stocktype selection and 3) soil
physical, chemical and hydrologic conditions. Seven native wetland trees (Betula nigra
L., Liquidambar styraciflua L., Platanus occidentalis L., Salix nigra Marshall, Quercus
bicolor Willd., Quercus palustris Münchh., and Quercus phellos L.) were planted using
three stocktypes (bare roots (BR), tubelings (TB), and 1-gallon containers (GAL)) in a
large scale, hydrologically manipulated field experiment. Hydrologic manipulations were
applied over three 0.7 ha cells and included: 1) ambient treatment received only
precipitation, 2) saturated treatment was saturated within the root-zone (10 cm) for a
minimum of 90% of the growing season, 3) flooded treatment was inundated above the
root crown for a minimum of 90% of each year and had increased amounts of clay,
higher bulk density and decreased nitrogen and phosphorus concentrations compared to
the ambient and saturated treatments. Results from the flooded treatment suggest that
planting primary successional species (especially S. nigra) with larger containerized
stocktypes may enhance the return of woody ecosystem structure and ecological
functions in stressful environmental conditions of recently restored forested wetlands. In
less stressful environmental conditions (ambient and saturated treatment), the bare root
stocktype grew similarly to the gallon stocktype, suggesting that the less expensive bare
root stocktype could be used successfully. These results apply to a variety of tree planting
situations including carbon sequestration projects, wildlife habitat creation and other
conservation projects across the landscape.
89
Introduction
Palustrine forested wetlands, the most abundant wetland type in Virginia, make
up a majority of wetland losses in Virginia over the past few decades (Tiner and Finn
1986, USGS 1999, Tiner et al. 2005). These wetlands are valued for the ecological
functions and services they provide (Sather and Smith 1984, NRC 1995) including but
not limited to, accumulation and retention of sediments (Hupp 1993, Craft and Casey
2000), accumulation, retention and cycling of elements (carbon, nitrogen, phosphorus,
etc.) (Craft and Casey 2000, Mitsch et al. 2012, Rheinhardt et al. 2012), maintenance of
plant communities (Walbridge 1993, Rheinhardt et al. 2000), provision of animal habitat
(Morley 2008), water quality enhancement (Whigham et al. 1988), and surface and
subsurface water storage (Brooks et al. 2013). The destruction or conversion of forested
wetland ecosystems removes the ecosystem functions and services they provided from
the landscape.
Realization of these lost functions and services has led to the restoration of
wetlands for a variety of reasons, including but not limited to, compensation for Clean
Water Act Section 404 permitted impacts to existing wetlands (wetland compensatory
mitigation), state and federal goals (Chesapeake 2000 Agreement), re-establishing bird
habitat (Ducks Unlimited) and/or agricultural easements (Agricultural Conservation
Easement Program). Regardless of the underlying reason, many attempts at forested
wetland restoration have not been ecologically successful (i.e. ecosystem structures,
functions and services have not been restored) (Atkinson and Cairns Jr. 2001, Brown and
Veneman 2001, Cole et al. 2001, Sudol and Ambrose 2002, Atkinson et al. 2005,
Atkinson et al. 2010, Moreno-Mateos et al. 2012, Stefanik and Mitsch 2012). One of the
90
major challenges associated with forested wetland restoration has been establishment and
growth of trees. Numerous studies have found that tree density and tree growth were
significantly lower for restored sites as compared to conditions prior to conversion or
nearby mature forested wetlands (Brown and Veneman 2001, NRC 2001, Cole and
Shafer 2002, Sharitz et al. 2006, Matthews and Endress 2008). This study was designed
to assist wetland managers in selecting appropriate species and stocktypes for site
conditions in order to increase the probability of successful forested wetland restoration
as well as providing insight about the survival and development of planted trees.
Successful establishment and growth of trees in restored forested wetlands is important
because trees contribute greatly to the ecological structure and multiple ecological
functions and services that forested wetland ecosystems provide.
For example, one ecosystem function that trees contribute to is providing and
enhancing habitat for plants, animals, fungi, and microbial communities both within the
restored wetland and in adjacent and downstream ecosystems. Living and shed bark,
wood, roots, flowers, fruits, seeds, leaves and sap are consumed by a number of different
organisms, including insects, mammals and birds. Leaf litter and fallen dead wood also
provide nutrients for fungus and other microorganisms in the detrital food web. Trees
provide shelter from weather and predators in the form of tree cavities, leaf litter and
fallen dead wood. Shade provided by trees surrounding streams, estuaries or rivers can
reduce air and water temperature, enhancing habitat for aquatic organisms. Trees also
provide space for organisms to live, including insects living under and in the bark, birds
and mammals building nests in cavities and branches, caterpillars and other insects
building nests in the crown, and lichen, moss and fungi living on the bark. Organic matter
91
(OM) produced by trees provides nutrients and shelter for organisms downstream.
Forested headwater wetlands in particular provide a substantial proportion of the total
OM found within streams (Dosskey and Bertsch 1994) and contribute significantly higher
concentrations of organic carbon when compared to upland watersheds (Mulholland and
Kuenzler 1979). Large woody debris increases channel bed roughness, which can slow
stream velocity, increase stability and provide habitat for microbial organisms and
animals (Harmon et al. 1986). All species used in this study (Betula nigra L. (river birch),
Liquidambar styraciflua L. (sweetgum), Platanus occidentalis L. (American sycamore),
Quercus bicolor Willd. (swamp white oak), Quercus palustris Münchh. (pin oak),
Quercus. phellos L. (willow oak) and Salix nigra Marshall (black willow)) are known to
provide habitat for a wide variety of organisms (Van Dersal 1938, Cypert and Webster
1948, Martin et al. 1951, Ebel and Kormanik 1966, Nixon et al. 1970, Hardin and Evans
1977, Dikson et al. 1983, Allen 1987, Allen and Kennedy 1989, Johnson and Lyon 1988,
Burns and Honkala 1990, Coladonato 1992, Carey 1992, 1992a, Dugger and Fredrickson
1992, Silberhorn 1992, Snyder 1992, Sullivan 1993, Tesky 1992, Santamour and Greene
1986, Sullivan 1994, Horst 2001, Moore 2002, Dickerson 2002, Nesom 2009, Row and
Geyer 2010, Adkins et al. 2012, Chong et al. 2012, Latimer and Close 2014).
Establishing trees in restored wetlands can be achieved through natural
colonization or planting seeds or saplings. Saplings are often planted when natural
colonization is anticipated to be limited due to insufficient amount of propagules of
appropriate species in the existing seed bank or being actively supplied by nearby trees.
Additionally, abiotic (hydrology, soil conditions, etc.) and biotic (seed predation,
competition, etc.) site conditions can limit natural colonization. If planting is deemed
92
necessary for restoration, selection of species and stocktype are two critical choices for
ensuring survival and subsequent growth following outplanting. These two choices are
influenced by the stressful biotic (herbaceous competition, herbivory, deer browsing,
pathogens) and abiotic conditions (low soil oxygen concentrations, low organic material,
low soil nutrient concentrations, high soil bulk density, and increased rock fragments)
typically present in areas undergoing wetland restoration.
Tree species selected for planting should be those that are found naturally in
nearby wetland ecosystems that have environmental conditions similar to the anticipated
conditions of the restoration site (Nyland 2007). In particular, matching species to
restoration site hydrology has been shown to be a key factor for successful establishment
of trees in bottomland hardwood forests (Stanturf et al. 2004, McLeod et al. 2006).
Following species selection, selection of stocktype (e.g. bare root seedlings of various
ages, tubelings or plugs, containerized, balled and burlapped, live stakes, etc.) becomes
relevant. These descriptive names can be used to describe the age, size, and production
techniques used, which are not uniform throughout the nursery industry. In general bare
root seedlings range in age from one to three years old and are typically planted during
dormancy without soil surrounding the roots. Tubelings are typically similar in age to
bare root seedlings; however, they are planted with tube soil surrounding the roots and
are grown in various shaped (square, round) small containers. Seedlings are also grown in
larger containers ranging from 1 gallon to >25 gallons. These containerized seedlings are
grown to various ages and sizes and are often planted with the soil intact around the root
system, which allows for planting later in the season. In general, bare root seedlings are
93
less expensive and are cheaper to plant, while containerized seedlings are more expensive
and require additional labor to plant.
Many studies have investigated how different stocktypes survive and grow
following outplanting on upland sites (See Grossnickle and El-Kassaby 2015 for review
of >400 containerized and bare-root stocktypes comparison studies). However, few
studies have investigated how stocktype selection influences survival and growth in
restored forested wetlands (See companion study - Roquemore et al. 2014). Of the studies
investigating stocktype selection in relation to wetland restoration the results suggests
that the stocktype responses are dependent upon the species used and the environmental
conditions. For example, McLeod (2000) found that bare root seedlings had similar
survival to more expensive containerized seedlings of Fraxinus pennsylvanica, Nyssa
aquatica, and Taxodium distichum when planted in a thermally impacted bottomland
hardwood forest. Whereas, a meta-analysis of 122 trials comparing survival between
bare-root and containerized stock planted across a variety of upland sites found that
containerized stock had greater survival than bare-root stock in 60.7% of the trials
(Grossnickle and El-Kassaby 2015). Due to the complexities of the species and
stocktypes available and uncertainty regarding the influence of species and stocktype
selection on the survival and growth of woody species under various hydrologic
conditions, practitioners have encountered difficulties establishing trees in restored
forested wetlands that will lead to the desired ecological structures and ecosystem
functions and services.
The purpose of this study was to investigate the differences in survival and
development of morphological structure (stem diameter, crown diameter, and height) of
94
planted trees in response to: 1) stocktype selection 2) species selection and 3) soil
physical, chemical and hydrologic conditions. Seven native wetland trees common to the
mid-Atlantic region of the United States were planted using three stocktypes (bare roots,
tubelings, and 1-gallon containers) in a large scale, hydrologically controlled, field
experiment.
Methods
Study Site
The experimental site (hereto referred to as “the site”) was established in New
Kent County, Virginia, USA at the Virginia Department of Forestry, New Kent Forestry
Center in 2008-2009 (Figure 2-1). The site was located in the Coastal Plain Region of
Virginia and the average annual temperature is 15o C and average annual precipitation is
116.2 cm/year (39 year average; WEST POINT 2 NW, Coop ID: 449025). The 1.4-ha
experimental site was located 8.8 m above sea level and ~1 km north of the
Chickahominy River (latitude 37° 25' 25.9026" N, longitude -77° 0' 53.3628" W). The
site was located on a terrace adjacent to a mature palustrine forested wetland to the west
and north with managed upland fields to the east and south. Soil series (and associated
taxonomic class) on the site included Catpoint fine sand (thermic, coated Lamellic
Quartzipsamments), State very fine sandy loam (fine-loamy, mixed, semiactive, thermic
Typic Hapludults) and Altavista fine sandy loam (fine-loamy, mixed, semiactive, thermic
Aquic Hapludults) (USDA NRCS 2015). These soils are classified as somewhat
excessively drained, well drained, and moderately well drained respectively. Based on
95
observations at the site, depth to the natural water table was estimated to be greater than 1
m prior to construction.
The site consisted of three hydrologically distinct cells (ambient (AMB), saturated
(SAT) and flooded (FLD)) each 49 m x 95 m in size (Figure 2-1). Each cell was equipped
with an on-site irrigation system capable of producing a minimum of 2.54 cm of
irrigation per hour. The three cells were hydrologically manipulated such that the ambient
treatment received only precipitation, the saturated treatment received irritation as needed
to exhibit saturated soil conditions within the root-zone (upper 10 cm) for a minimum of
90% of the growing season, and the flooded treatment received irrigation as needed to
maintain inundations above the root crown for a minimum of 90% of each year. The
AMB cell received irrigation when the Palmer Drought Severity Index weekly value was
-3 or less indicating a severe drought. Irrigation water was drawn from the non-tidal
portion of the Chickahominy River approximately 8 km upriver above Walkers Dam in
Walkers, VA. Soils in AMB and SAT treatments were tilled using a finger plow to a
depth of 20 cm in February 2009 prior to planting, whereas, the FLD treatment was
excavated using a 5-ton backhoe to a depth of 1 m to an existing clay layer.
Soil Sampling
Forty four soil samples were collected from each cell (Total=132) in 2013 and
analyzed for soil carbon, nitrogen and phosphorus concentrations and soil sand, silt and
clay percentages. Samples were evenly spaced within each cell and collected 70 cm
diagonally from the tree base. The top 15 cm of soil were collected using a soil probe
(tube sampler). Samples were dried in an oven at 60o C until constant mass was obtained.
96
Bulk density was determined by dividing mass of oven dry soil by volume collected
(Brady & Weil 2002). Particle size distribution (sand (>63µm), silt (<63µm-4µm), and
clay (<4µm)) was determined by the standard sieve-pipette method (Brady & Weil 2002).
Carbon and nitrogen concentrations from a subsample of homogenized oven dry
samples were measured using a PE2400 CHNS/O Elemental Analyzer (Perkin Elmer,
Massachusetts, USA). Phosphorus concentrations were measured using a
spectrophotometer following a modified ashing and extraction technique (Chambers &
Fourqurean 1991). Carbon, nitrogen and phosphorus concentration are presented as
percentages of total soil mass.
There were differences in soil physical and chemical characteristics among the
cells (Table 2-1). Within each cell there were also differences in the spatial distributions
of soil bulk density (Figure 2-2), percentage sand (Figure 2-3), percentage silt (Figure
2-4), percentage clay (Figure 2-5), percentage soil carbon (Figure 2-6), percentage soil
nitrogen (Figure 2-7), and percentage phosphorus (Figure 2-8). In general the soil bulk
density and percentage clay was higher in the FLD than in the AMB and SAT and the
soil elemental concentrations were lower in the FLD than in the AMB and SAT.
Planting Material
The seven planted species (and Atlantic and Gulf Coastal Plain wetland indicator
statuses (Lichvar et al. 2012; Lichvar 2013)) were Betula nigra L. (river birch, FACW),
Liquidambar styraciflua L. (sweetgum, FAC), Platanus occidentalis L. (American
sycamore, FACW), Quercus bicolor Willd. (swamp white oak, FACW), Quercus
97
palustris Münchh. (pin oak, FACW), Quercus. phellos L. (willow oak, FACW) and Salix
nigra Marshall (black willow, OBL).
Three stocktypes of each species were used: bare-root (BR), tubeling (TB), and 1gallon containers (GAL) (tubelings of P. occidentalis, Q. phellos, and S. nigra had their
soil removed by the nursery prior to shipment). Bare root seedlings range in age from one
to three years old and were planted during dormancy without soil surrounding the roots.
Tubelings are typically similar in age to bare root seedlings; however, they were planted
with potting soil surrounding the roots and were grown in small square containers. One
gallon containerized seedlings are larger and older than BR and TB and were planted
with potting soil intact around the root system. The GAL was most expensive, followed
by TB, with BR being the least expensive.
In spring 2009, all combinations of species and stocktypes were planted randomly
along rows within each cell. A total of 2,772 trees were planted; ~44 of each
species/stocktype combination, for a total of 924 trees per cell. Saplings were arranged in
22 rows per cell (42 saplings per row) that were staggered. Therefore, saplings were
spaced 2.29 m from saplings within the row and 2.56 m from saplings in adjacent rows.
This led to a density of 1969 stems/ha. In the spring of 2010, a total of 482 additional
saplings were planted to ensure adequate survival for biomass sampling, but these were
not considered in the present analysis.
Saplings were obtained from four nurseries located in Virginia, North Carolina,
New Jersey, and Tennessee. Replacement stock did not necessarily come from the same
nursery as the original stock. No fertilizers were applied prior to or following planting.
Herbaceous competition was controlled around plantings in AMB and SAT through
98
bimonthly grass cutting and application of glyphosate at the beginning and middle of the
growing season using commercial backpack sprayers.
Species Grouping
To facilitate analysis and interpretation of plant material selection, species were
assigned to early (primary) or late (secondary) successional categories based on common
Mid-Atlantic Coastal Plain regional trends in dominance during stand development as
expressed through differences in maturation and growth rates, dispersal mechanisms, and
disturbance tolerance. The primary species group (PRI) consisted of four species (B.
nigra, L. styraciflua, P. occidentalis and S. nigra) that are typically dominant during the
early stages of succession, have rapid growth and maturation rates, have wind dispersed
seeds and are moderately tolerant of disturbance (Bazzaz 1979). The secondary species
group (SEC) consisted of three species (Q. bicolor, Q. palustris, and Q. phellos) typically
dominant in the later stages of succession, have slower growth and maturation rates, have
large seeds that are dispersed mainly by animals and are generally less tolerant of
disturbance. (Guyette et al. 2004).
Survival and Morphometric Measurements
Providing habitat for other organisms requires that trees planted in restored
wetlands must survive transplanting and grow. Structural measurements of sapling
morphology used in this study (crown and stem diameter and height) and survival rates
can be used to provide inferences about the amount of habitat provided by individual
trees and stands of trees. These measurements are useful since specific amounts of habitat
resources provided by trees are difficult to quantify, they are commonly measured in
99
forest inventories, and they have direct and indirect relationships to the occurrence and
abundance of other organisms.
Survival and morphology were measured in mid-April, mid-August, and midOctober for seven years (from April 2009 to October 2015). Individual saplings were
considered “live” based on the presence of green leaves or a green layer in the cortex
under the bark. We often found it necessary to search for the latter in many saplings that
exhibited die-back and re-growth. To check for a live cortex, a small longitudinal incision
scratch was made at the highest point on the stem. If brown (i.e. not alive), a second
incision was made approximately one half way down the stem. If brown, a final incision
was made at the base. If any of the incisions showed a green cortex, the individual
sapling was considered alive. Percent survival calculations excluded live trees removed
for biomass measurements.
Methods for sampling morphology (stem cross-sectional diameter at groundline,
crown diameter (CD), and height of tallest stem (H)) were modified from Bailey et al.
(2007). The diameter of stems at ground level was measured using micro-calipers (6-inch
stainless steel digital caliper, General, Secaucus, New Jersey) or macro-calipers (127-cm
Mantax Precision Blue Calipers, Haglöf, Inc., Långsele Sweden). If root swelling was
present (defined as stem diameter > 10% larger than stem above swelling), stem crosssectional diameter was measured just above the visual top of swelling. For trees with
multiple stems originating from below the soil surface, the stem diameter of the five
largest stems was measured. A single cross-sectional area at groundline (CSAG) was
calculated by summing CSAG for each stem of an individual. In subsequent chapters, this
CSAG was then converted to a single stem cross-sectional diameter at groundline
100
(referred to as equivalent stem cross-sectional diameter at groundline (ESD). The sum of
the area of the measured stems equals an area equivalent to the corresponding diameter of
a single stemmed sapling as described in Paul et al. (2013).
Crown diameter was measured in three evenly distributed angles using a meter
stick, macro-calipers, 5-m stadia rod, or tape measure. The mean crown diameter was
determined by averaging three crown measurements. In subsequent chapters, canopy
cover was calculated by converting average crown diameter (cm) to area (cm2).
Total heights were measured with a standard meter stick, 5-m stadia rod,
clinometer (PM-5/360 PC, Suunto, Co., Vantaa, Finland ), or hypsometer (Vertex III,
Haglöf, Inc., Långsele Sweden).
Statistical Analysis
Initial morphological means of stocktypes were compared for each species.
Unequal sample size, unequal variance and multiple comparisons were accounted for by
modeling variance for each stocktype and using the Games-Howell adjustment (Westfall
et al. 2011).
To determine the differences in survival over five years among stocktypes,
species, and successional groups, the Cox proportional hazards model was applied to
each species or stocktype within each cell (Firth correction and Breslow method for ties).
To determine the differences in morphological variables over five years among stocktype,
species, and successional groups, repeated measures analysis of variance (rANOVA) was
used within each cell (Covariance structure: Autoregressive 1, Estimation method:
Residual maximum likelihood). If significant interactions were found for either survival
101
or morphological variables, a simple effects model was used to determine differences
among the stocktypes or successional group for each species within each cell. Least
squares post hoc test using a Bonferroni adjustment was used to determine differences
among each stocktype. All alpha values were set at 0.05.
Due to the cells unreplicated design the survival and morphology of the 21
unique combinations of species and stocktypes were compared among the cells. Survival
was not compared statistically but was compared using only percent survival among
species/stocktype combinations. Morphological means after 5 years for each
species/stocktype combination were compared across cells. Unequal sample size, unequal
variance and multiple comparisons were accounted for by modeling variance for each cell
and using the Games-Howell adjustment (Westfall et al. 2011).
Results from each post hoc comparison (stocktype, successional group, cells) for
each measured variable (survival, CSAG, H, and CD) were tallied by unique outcome.
For example, when comparing the survival among the stocktypes the number of times a
particular outcome occurred (BR>GAL, BR=GAL, BR<GAL, TB<GAL, TB=GAL,
TB>GAL, BR=TB, or BR<TB) was counted for each cell. The total number of times
each outcome occurred across all measured variables was obtained for each cell and in
total, to determine which outcome was most common. This analysis also allowed for
survival and morphology results to be combined and will be presented as such.
102
Results
Stocktype Comparison of Initial Morphology
The average initial stem cross-sectional area at groundline (CSAG) for the bare
root (BR), gallon (GAL), and tubeling (TB) stocktypes across all species was 0.54 cm2
(Standard Deviation (SD = 0.64), 0.56 cm2 (SD = 0.63) and 0.96 cm2 (SD = 0.98)
respectively. The average initial height (H) for BR, GAL, and TB across all species was
52.6 cm (SD = 31.0), 70.0 cm (SD = 42.1) and 53.3 cm (SD = 34.0) respectively. The
average crown diameter (CD) for BR, GAL, and TB across all species was 8.1 cm (SD =
10.5), 16.2 cm (SD = 17.0) and 8.4 cm (SD = 10.9) respectively. However, there were
significant interactions between species and stocktype for each morphological parameter
at the time of planting. This suggests that the initial size of the stocktype depended upon
the species and vice versa (e.g. GAL stocktype may not be the largest stocktype across all
species). As a result of significant interactions, subsequent analysis focused on
determining differences among stocktypes for each species separately (Table 2-2).
The GAL CSAG and CD were significantly greater than the BR and TUB
stocktypes for all species except Q. bicolor. The GAL H was significantly greater than
the BR and TUB stocktypes for all species except L. styraciflua and Q. bicolor.
The CSAG and CD were not significantly different between the BR and TB for all
species. Initial H was not significantly different between the BR and TB stocktype for all
species except B. nigra, where the TB was significantly taller than the BR at the time of
planting (Table 2-2).
103
Stocktype Comparison Over 5 Years
There were significant interactions between species and stocktype within the
AMB, SAT, and FLD for each parameter (survival, CSAG, H and CD) over the 5 years
of monitoring (Figure 2-10). This suggests that within each cell, response of stocktype
depended upon species and vice versa (e.g. GAL stocktype may not have greatest
survival or H across all species). As a result of significant interactions, subsequent
analysis focused on determining differences among stocktype in each cell for each
species separately.
Survival
After five years average percent survival across all cells, species and stocktypes
was 57%. Average percent survival for BR, GAL, and TB stocktype (across all cells and
species) was 48%, 81%, and 43%, respectively. In order to further investigate differences
among stocktypes, survival of stocktypes by species were compared within each cell.
In AMB, GAL survival was greater than BR survival for B. nigra, L. styraciflua,
P. occidentalis, Q. palustris, Q. phellos, and S. nigra. Similarly, GAL survival was
greater than TB survival for B. nigra, L. styraciflua Q. bicolor, Q. palustris, Q. phellos,
and S. nigra. BR survival was greater than TB survival for L. styraciflua, Q. bicolor
while Q. palustris had greater survival in BR than TB (Figure 2-11 and Table 2-9).
Overall, BR survival was similar to TB for B. nigra and Q. phellos.
In SAT, GAL survival was greater than BR survival for B. nigra, L. styraciflua, P.
occidentalis, Q. phellos, and S. nigra and had greater survival than TB for all seven
species. BR survival was not different than TB survival for B. nigra, P. occidentalis, Q.
104
phellos, and S. nigra. However, BR survival was greater than TB survival for L.
styraciflua Q. bicolor and Q. palustris (Figure 2-11 and Table 2-9).
In FLD, GAL survival was greater than BR survival for B. nigra, L. styraciflua, P.
occidentalis, Q. bicolor, Q. palustris, and Q. phellos. GAL survival was greater than TB
survival for L. styraciflua, P. occidentalis, Q. bicolor, Q. palustris, and Q. phellos. There
was no difference in survival between BR and TB for L. styraciflua, P. occidentalis, and
Q. palustris. S. nigra had no differences in survival among all stocktypes in FLD (Figure
2-11 and Table 2-9).
Morphology
Stem Cross-sectional Area at Groundline
Average stem cross-sectional area at groundline (CSAG) and standard deviation
(SD) for BR, GAL, and TB stocktype (across all cells and species) was 59.6 cm2 (n=440,
SD=114.6 cm2), 79.8 cm2 (n=678, SD=137.4 cm2), and 75.8 cm2 (n=346, SD=135.9 cm2),
respectively after five years. In order to further investigate differences among stocktypes,
CSAG of stocktypes by species were compared within each cell.
In AMB, GAL CSAG was greater than BR CSAG for B. nigra, Q. phellos and S.
nigra. There was no difference in CSAG between BR and GAL stocktypes for L.
styraciflua, P. occidentalis, Q. bicolor, or Q. palustris. GAL CSAG was greater than TB
CSAG for B. nigra, L. styraciflua, Q. bicolor, Q. palustris and Q. phellos. There was no
difference in CSAG between TB and BR stocktype for B. nigra, P. occidentalis, Q.
palustris, Q. phellos, or S. nigra (Figure 2-12 and Table 2-11).
105
In SAT, GAL CSAG was greater than BR CSAG for B. nigra, Q. palustris, and
Q. phellos. There was no difference in CSAG between BR and GAL stocktypes for L.
styraciflua, P. occidentalis, Q. bicolor and S. nigra. GAL CSAG was greater than TB
CSAG for B. nigra, L. styraciflua, Q. bicolor, Q. palustris and Q. phellos. There was no
difference in CSAG between BR and TB for B. nigra, L. styraciflua, P. occidentalis, Q.
palustris, Q. phellos, and S. nigra (Figure 2-12 and Table 2-11).
In FLD, GAL CSAG was greater than BR CSAG for B. nigra, L. styraciflua, P.
occidentalis, Q. palustris, Q. phellos, and S. nigra. GAL CSAG was greater than TB
CSAG for all seven species. BR CSAG was greater than TB CSAG for Q. bicolor only
(Figure 2-12 and Table 2-11).
Crown Diameter
BR, GAL, and TB average crown diameter (CD) after five years was 209.3 cm
(n=440, SD=158.9 cm), 241.2 cm (n=678, SD=172.6), and 220.2 cm (n=346, SD= 189.9
cm) respectively (across all cells and species). In order to further investigate differences
among stocktypes, CD of stocktypes by species were compared within each cell.
In AMB, GAL CD was greater than BR CD for B. nigra, Q. palustris, Q. phellos,
and S. nigra. There was no difference between BR and GAL CD for L. styraciflua, P.
occidentalis, and Q. bicolor. GAL CD was greater than TB CD for all seven species
except P. occidentalis (no difference). BR CD was not different than TB CD for B. nigra,
P. occidentalis, Q. bicolor, Q. phellos and S. nigra. BR CD was greater than TB CD for
L. styraciflua and Q. palustris (Figure 2-13 and Table 2-13).
106
In SAT GAL CD was greater than BR CD for B. nigra, Q. palustris, Q. phellos,
and S. nigra. There was no difference between BR and GAL CD for L. styraciflua, P.
occidentalis, and Q. bicolor. GAL CD was greater than TB CD for L. styraciflua, Q.
bicolor, Q. palustris, Q. phellos, and S. nigra. There was no difference in CD between
TB and GAL stocktype for B. nigra and P. occidentalis. There was no difference between
BR and TB CD for all species except Q. bicolor (Figure 2-13 and Table 2-13).
In FLD, GAL CD was greater than BR and TB CD for all seven species. BR CD
was greater than TB for all seven species except Q. bicolor (Figure 2-13 and Table 2-13).
Height
After five years, average height (H) for BR, GAL, and TB (across all cells and
species) was 314.6 cm (n=440, SD=249.1 cm), 345.6 cm (n=678, SD=241.7 cm), and
337.5 cm (n=346, SD=303.0 cm) respectively. In order to further investigate differences
among stocktypes, H of stocktypes by species were compared within each cell.
In AMB, GAL H was greater than BR H for B. nigra, Q. palustris, Q. phellos, and
S. nigra. There was no difference in H between BR and GAL for L. styraciflua, P.
occidentalis, and Q. bicolor. GAL H was greater than TB H for B. nigra, L. styraciflua,
Q. bicolor, Q. palustris, and Q. phellos. There was no difference between TB and GAL H
for P. occidentalis and S. nigra. There was no difference in H between BR and TB for B.
nigra, P. occidentalis, Q. bicolor, Q. phellos and S. nigra. BR H was greater than TB H
for L. styraciflua and Q. palustris (Figure 2-14 and Table 2-15).
In SAT, there was no difference in H between BR and GAL for L. styraciflua, P.
occidentalis, Q. bicolor, Q. palustris and S. nigra. GAL H was greater than BR for B.
nigra and Q. phellos. GAL H was greater than TB H for L. styraciflua, Q. bicolor, Q.
107
palustris, and Q. phellos. There was no difference between TB H and GAL H for B.
nigra, P. occidentalis, and S. nigra. There was no difference in H between BR and TB for
B. nigra, P. occidentalis, Q. bicolor, Q. phellos and S. nigra. BR H was greater than TB
H for L. styraciflua and Q. palustris (Figure 2-14 and Table 2-15).
In FLD, GAL H was greater than BR height for all species except Q. bicolor, for
which BR H was greater than GAL height. GAL H was greater than TB H for all species.
There was no difference in BR H and TB H for B. nigra, P. occidentalis, Q. phellos, and
S. nigra. BR H was greater than TB H for L. styraciflua, Q. bicolor, and Q. palustris
(Figure 2-14 and Table 2-15).
Combining survival and morphology
When combining survival and morphological comparisons for each species across
all three cells and counting the number of times an outcome occurred, GAL was greater
than BR and TB in 67% and 82% of all comparisons respectively and BR was not
different than TB in 69% of comparisons (Table 2-6). To further investigate differences
among stocktypes, outcomes were counted within each cell.
For all species in AMB, GAL was greater than BR and TB in 61% and 79% of all
survival and morphological comparisons respectively and BR was not different than TB
in 61% of all comparisons. BR was not different than GAL in 39% of all comparisons
and BR was greater than TB in 32% of all comparisons (Table 2-6).
For all species in SAT, GAL was greater than BR in 50% of combined
comparisons and was not different than BR in 50% of all comparisons. GAL was greater
108
than TB in 75% of the comparisons while BR was not different than TB in 75% of all
comparisons. BR was greater than TB in 25% of all comparisons (Table 2-6).
For all species in FLD, GAL was greater than BR in 89% of all survival and
morphological comparisons and was greater than TB in 93% of all comparisons. BR and
TB were not different in 71% of all comparisons (Table 2-6).
Species Group Comparison Over 5 Years
The seven species were divided into two groups (primary and secondary) based
on dominance during the traditional forest successional sequence, differences in
maturation and growth rates, dispersal mechanisms, and disturbance tolerance in order to
facilitate comparisons among species. When analyzing differences in survival, CSAG and
H among successional groups and stocktype there were significant interactions between
successional group and stocktype within AMB and FLD, SAT and FLD, and FLD
respectively. This suggests that the survival, CSAG and H response of stocktype
depended upon successional group and vice versa. There was no significant interaction
among successional group and stocktype when analyzing differences in CD (Table 2-4).
This suggests that CD response was similar among stocktypes for all successional groups
and vice versa. As a result of significant interactions, subsequent analysis of each
parameter focused on determining differences among successional groups in each cell for
each stocktype separately.
109
Survival
Survival of secondary species (Quercus spp.) was greater than survival of primary
species when planted as BR in AMB. When planted as GAL or TUB there was no
difference between the survival of the primary and secondary successional species
(Figure 2-15). In SAT, secondary species (Quercus spp.) had greater survival than
primary species when planted as BR. When planted as GAL or TUB there were no
differences between survival of primary and secondary successional species (Figure
2-15). For all stocktypes primary successional species had greater survival than
secondary species in FLD (Figure 2-15).
Morphology
Primary species had greater CSAG and H than secondary species for all
stocktypes across all cells (Figure 2-16 and Figure 2-18). Primary species had greater CD
than secondary species for all stocktypes in AMB and SAT. In FLD primary species had
greater CD than secondary species for GAL and TB stocktype. There was no difference
in CD between primary and secondary species for BR (Figure 2-17).
Primary successional species were greater than secondary successional species in
81% of all comparisons when merging survival and morphological comparisons for each
stocktype across all three cells (Table 2-7). In order to investigate these results further,
survival and morphology comparisons were combined for each cell and differences
between stocktype and successional stages were investigated.
110
Combining survival and morphology
In AMB, when combining survival and morphological measurements for each
stocktype, primary species were greater than secondary species in 75% of comparisons
(Table 2-7). Primary species were greater than secondary species in 75% of combined
survival and morphological comparisons in SAT (Table 2-7). In FLD, primary species
were greater than secondary species in 92% of all comparisons. There were no
differences between primary and secondary species in 8% of comparisons in FLD (Table
2-7).
Cell Comparison After 5 Years
Individual species/stocktype combination’s responses after five years were
compared among cells in order to make inferences about their responses to environment
conditions and to infer about differences among cells. Due to the unreplicated nature of
cells, survival was compared using absolute values, while CSAG, H and CD were
compared statistically (Table 2-5). The majority of species/stocktype combinations had
significantly different responses among the cells, except Q. palustris and Q. phellos TB
and P. occidentalis BR (Table 2-5). The results of the individual measurements
comparisons are presented below.
Survival
BR survival after five years was greater in AMB than SAT for P. occidentalis. BR
survival in SAT was greater than AMB for B. nigra, L. styraciflua, Q. bicolor, Q.
palustris, Q. phellos, and S. nigra. All seven species BR survival was greater in AMB
111
and SAT compared to FLD, except for S. nigra. S. nigra BR had greater survival in FLD
compared to both SAT and AMB (Table 2-10).
GAL survival after five years was greater in AMB than SAT for B. nigra, Q.
palustris, and S. nigra. GAL survival in SAT was greater than survival in AMB for L.
styraciflua, P. occidentalis, and Q. phellos. While, Q. bicolor GAL had no difference in
survival between the AMB and SAT. All seven species GAL survival was greater in the
AMB and SAT compared to FLD, except for S. nigra. S. nigra GAL had greater survival
in FLD compared to both SAT and AMB (Table 2-10).
TB survival after five years was greater in AMB than SAT for P. occidentalis and
S. nigra while the remaining 5 species TB survival was greater in SAT than AMB. P.
occidentalis, Q. bicolor, Q. palustris and Q. phellos TB survival in AMB and SAT was
greater than FLD. However, B. nigra, L. styraciflua, and S. nigra TB survival in FLD was
greater than AMB and SAT (Table 2-10).
Morphology
Stem Cross-sectional Area at Groundline
The average BR CSAG after five years in AMB, SAT and FLD was 90.7 cm2
(n=182, SD=147.0 cm2), 49.0 cm2 (n=183, SD=89.3 cm2), 10.6 cm2 (n=76, SD=21.3 cm2)
respectively. The average GAL CSAG after five years in AMB, SAT and FLD was 126.8
cm2 (n=261, SD=178.2 cm2), 73.3 cm2 (n=268, SD=108.9 cm2), 9.2 cm2 (n=149,
SD=12.1 cm2) respectively. Average TB CSAG after five years in AMB, SAT and FLD
was 148.5 cm2 (n=111, SD=196.1 cm2), 58.3 cm2 (n=157, SD=84.8 cm2), 7.4 cm2 (n=78,
112
SD=15.0 cm2) respectively. In order to further investigate differences among cells, CSAG
of species within cells were compared for each stocktype.
BR CSAG in AMB was greater than SAT after five years for L. styraciflua, P.
occidentalis, and Q. palustris. There was no difference in BR CSAG between AMB and
SAT for B. nigra, Q. bicolor, Q. phellos, and S. nigra. BR CSAG in AMB was greater
FLD for all species except S. nigra and BR CSAG in SAT was greater than FLD for all
seven species (Table 2-12).
GAL CSAG in AMB was greater than SAT for B. nigra, P. occidentalis, Q.
bicolor, and Q. palustris. There was no difference between GAL CSAG in AMB and
SAT for L. styraciflua, Q. phellos, and S. nigra. All seven species CSAG for GAL was
greater in the AMB compared to FLD. GAL CSAG was greater in SAT than FLD for all
species except Q. palustris (Table 2-12).
TB CSAG after five years was greater in AMB than SAT for P. occidentalis, and
Q. bicolor. There was no difference in TB CSAG between AMB and SAT for B. nigra, L.
styraciflua, Q. palustris, Q. phellos, and S. nigra. TB CSAG was greater in AMB than
FLD for B. nigra, L. styraciflua, Q. bicolor, and S. nigra. There was no difference in TB
CSAG between AMB and FLD for P. occidentalis and Q. palustris. TB CSAG was
greater in SAT than FLD for B. nigra, L. styraciflua, Q. bicolor, and S. nigra. There was
no difference in TB CSAG between SAT and FLD for P. occidentalis and Q. palustris
after five years (Table 2-12).
113
Crown diameter
The average BR CD after five years in AMB, SAT and FLD was 277.2 cm
(n=182, SD=168.9 cm), 205.5 cm (n=182, SD=131.8 cm), 56.1 cm (n=76, SD=45.0 cm)
respectively. Average GAL CD after five years in AMB, SAT and FLD was 326.8 cm
(n=261, SD=175.8 cm), 257.9 cm (n=268, SD=136.2 cm), 61.2 cm (n=149, SD=50.7 cm)
respectively. Average TB CD after five years in AMB, SAT and FLD was 329.7 cm
(n=111, SD=214.1 cm), 228.9 cm (n=157, SD=151.4 cm), 46.7 cm (n=78, SD=42.1 cm)
respectively. In order to further investigate differences among cells, CD of species within
cells were compared for each stocktype.
BR CD after five years was greater in AMB than SAT for L. styraciflua, P.
occidentalis, Q. bicolor, and Q. palustris. There was no difference in BR CD between
AMB and SAT for B. nigra, Q. phellos and S. nigra. All seven species had greater BR
CD in AMB than FLD except S. nigra (no difference). All seven species had greater BR
CD in SAT than FLD (Table 2-14).
GAL CD was greater in AMB than SAT for B. nigra, P. occidentalis, Q. bicolor,
and Q. palustris. There was no difference in GAL CD between AMB and SAT for L.
styraciflua, Q. phellos, and S. nigra. All seven species GAL CD in AMB and SAT was
greater than FLD (Table 2-14).
TB CD was greater in the AMB than SAT for Q. bicolor, while the remaining six
species had no difference in TB CD between AMB and SAT. TB CD was not different
between AMB and FLD for Q. palustris, while the remaining six species TB CD was
greater in the AMB than FLD. TB CD was greater in SAT than FLD for B. nigra, L.
114
styraciflua, Q. bicolor, and S. nigra. TB CD was not different between SAT and FLD for
P. occidentalis, and Q. palustris (Table 2-14).
Height
The average BR H after five years in AMB, SAT and FLD was 435.7 cm (n=182,
SD=278.2 cm), 286.7 cm (n=182, SD=188.8 cm), 91.2 cm (n=76, SD=47.7 cm)
respectively. Average GAL H after five years in the AMB, SAT and FLD was 485.8 cm
(n=261, SD=251.4 cm), 345.8 cm (n=268, SD=179.4 cm), 99.7 cm (n=149, SD=52.3 cm)
respectively. Average TB H after five years in the AMB, SAT and FLD was 545.3 cm
(n=111, SD=365.1 cm), 318.1 cm (n=157, SD=210.0 cm), 80.8 cm (n=78, SD=41.3 cm)
respectively. In order to further investigate differences among cells, H of species within
cells were compared for each stocktype.
BR H after five years was not different between AMB and SAT for P.
occidentalis and S. nigra, while the remaing 5 species BR H was greater in the AMB than
SAT. BR H was greater in AMB and SAT than FLD for B. nigra, L. styraciflua, Q.
bicolor, Q. palustris, Q. phellos, and S. nigra (Table 2-16).
After five years GAL H was greater in AMB than SAT for all species except L.
styraciflua. GAL H in AMB and SAT were greater than FLD for all 7 species (Table
2-16).
TB H was greater in AMB than SAT for B. nigra, P. occidentalis, Q. bicolor, and
S. nigra. TB H was not different between AMB and SAT for L. styraciflua, Q. palustris,
and Q. phellos. TB H was greater in AMB than FLD for B. nigra, L. styrcaiflua, P.
occidentalis, Q. bicolor, and S. nigra. There was no difference in H between AMB and
115
FLD for Q. palustris TB. TB H was greater in SAT than FLD for B. nigra, L.
styrcaciflua, Q. bicolor, and S. nigra. TB H was not different between SAT and FLD for
P. occidentalis and Q. palustris (Table 2-16).
Combining survival and morphology
To further analyze the differences among cells, the survival and morphology
measurements across all species and stocktypes were combined (Table 2-8). In total, the
outcomes of the comparisons of species/stocktype comparisons among cells show that
AMB was greater than SAT in 46.4% of comparisons, while there was no difference
among AMB and SAT in 36.9% of comparisons and SAT exceeded AMB in 16.7% of
comparisons. The SAT exceeded AMB only in percent survival comparisons. When
comparing AMB to FLD, AMB was greater than FLD in 85.9% of comparisons and
equal to in 7.7%. The SAT exceeded FLD in 84.6% of comparisons and was similar to
FLD in 9.0% of comparisons. Based on the responses of the species/stocktype
combinations, these results suggest that AMB was more similar to SAT than SAT was
similar to FLD, while AMB and FLD are most dissimilar.
Discussion
The goal of wetland restoration is to return lost ecological structure and functions
to the landscape, including plant, animal, and microbial habitat. Habitat in restored
wetlands is obtained primary through the successful establishment of vegetative structure,
which provides cover, food and space for a variety of organisms. Numerous studies have
found that tree density and tree growth were significantly lower for restored sites as
116
compared to conditions prior to conversion or nearby mature forested wetlands (Brown
and Veneman 2001, NRC 2001, Cole and Shafer 2002, Sharitz et al. 2006, Matthews and
Endress 2008). Poor establishment and growth may result from inadequate colonization
from surrounding seed sources or low survival of planted woody vegetation (Robb 2002,
Morgan and Roberts 2003). Poor survival and growth of planted trees results from
unfavorable site conditions (inappropriate hydrology, low organic content, high bulk
density, increased rock fragments), competition from non-desired species, improper
species or stocktype selection, and/or improper planting techniques (Stolt et al. 2000,
Campbell et al. 2002, Bruland and Richardson 2004, Bergshneider 2005, Daniels et al.
2005, Bailey et al. 2007). The effect of stocktype, species and hydrologic and soil
conditions on planted tree survival and morphology were the focus of this project.
Stocktype Comparison of Initial Morphology
In general the initial morphology of GAL stocktype was greater than the BR and
TB stocktypes (except Q. bicolor). This suggests that the GAL stocktype with nutrient
rich potting soil around the roots may have more initial above and belowground biomass
than BR and TB. There were no differences in initial morphology between the BR and
TB stocktypes (except H of B. nigra TB>BR). The lack of initial morphological
differences between the BR and TB suggests that the major difference between these two
stocktypes is the presence of potting soil around the roots and possibly other nursery
production techniques. However, P. occidentalis, Q. phellos, and S. nigra TB had soil
removed by the nursery prior to shipment.
117
Stocktype Comparison Over 5 Years
In the stressful hydrologic, soil and competitive herbaceous conditions of the
flooded cell (FLD), the larger stocktype (GAL) exhibited increased survival, CD (Figure
2-9), and CSAG compared to smaller stocktypes (BR and TB) for most species. The
characteristics of GAL (larger initial size, organic rich potting soil surrounding roots)
may have increased its ability to overcome transplant shock, competition from
herbaceous vegetation and low soil nutrient concentrations in the FLD treatment.
Transplant shock (also called planting check) is a temporary setback in growth
that occurs after outplanting, which if severe enough can result in tree mortality
(Kozlowski and Davies 1975, Acquaah 2005, Grossnickle 2005, South and Zwolinski
1996). Transplant shock is associated with decreased water absorption as a result of poor
root-soil contact, low permeability of suberized roots (older woody roots) and a low
amount of roots in relation to shoots (Beineke and Perry 1965, Carlson and Miller 1990,
South and Zwolinski 1996, Grossnickle 2005). In order to overcome transplant shock,
saplings must absorb enough water to satisfy evapotranspiration and
metabolic/physiologic processes. The stressors associated with transplant shock in
recently restored wetlands may be greater due to the low oxygen soil conditions present
(Kozlowski and Davies 1975).
The larger initial size of GAL suggests that it may have had greater initial
aboveground and belowground biomass than the BR and TB stocktypes. Increased
belowground biomass has been shown to increase the amount of water absorbed by roots
(Carlson 1986) and trees with increased initial aboveground biomass have been shown to
overcome herbaceous competition (Grossnickle and El-Kassaby 2015). Additionally, the
118
gallon stocktype was planted with organic rich potting soil surrounding the root mass,
which may have enhanced the probability for survival and overall growth because the
roots would have remained in contact with the potting soil and continued to take up
water. Furthermore, the potting soil may have provided additional nutrients not available
in the surrounding soil. Overall, the initial characteristics of GAL (larger initial size,
organic rich potting soil surrounding roots) may be reasons for the greater survival and
overall growth than the BR and/or TB stocktypes in FLD.
Previous research has also demonstrated that large containerized woody stock had
better survival and/or growth than smaller planting stocks. Burdett et al. (1984) showed
that container grown seedlings can have greater root growth during their first growing
season after outplanting compared to bare root seedlings. South et al. (2005) also showed
that containerized seedlings of Pinus palustris had 20% better survival than bare root
seedlings having similar root-collar diameters when outplanted on old-fields and cutover
sites. Pinto et al. (2011) found that larger containers of Pinus ponderosa planted at a
mesic site had increased total height and basal area. A meta-analysis of 122 trials
comparing survival between bare-root and containerized stock planted across a variety of
upland sites found that containerized stock had greater survival than bare-root stock in
60.7% of the trials (Grossnickle and El-Kassaby 2015).
In more aerobic soil conditions (AMB and SAT) the GAL stocktype has similar
morphology compared to the BR for most species, but was often larger than the TB
stocktype (Table 2-6). This suggests that the less expensive BR stocktype may return
ecosystem structure and possibly ecosystem habitat functions in a similar manner as a
larger stocktype if hydrologic stress and herbaceous vegetation competition is reduced
119
and there are better soil conditions (low bulk density, high soil nutrient concentrations).
However, the TB stocktype may not be appropriate for planting into restored wetlands
because for many species, the survival was less and the morphology was smaller than the
GAL and BR stocktypes, particularly in FLD (Table 2-6).
Several previous studies have similarly found that bare-root seedlings have
similar survival and growth compared to the more expensive containerized stocktype. A
large scale long term study by McLeod (2000) found that bare root seedlings had similar
survival to more expensive containerized seedlings of Fraxinus pennsylvanica, Nyssa
aquatica, and Taxodium distichum when planted in a thermally impacted bottomland
hardwood forest. Additionally, Denton (1990) investigated the effect of stocktype on the
growth of T. distichum in restored forested wetlands in Florida and their results suggest
that in order to obtain 33% canopy closure the initial costs could be reduced by planting
smaller trees (1 gallon container) at ~2500 stems/ha as opposed to planting 7 gallon trees
at ~1000 stems/ha. A large meta-analysis comparing container grown seedlings and bareroot seedlings found that on a variety of sites with low stress, the two stocktypes had
similar survival rates (Grossnickle and El-Kassaby 2015). While not focusing on
wetlands, these results are similar to the present study.
Recently restored wetlands often have stressful hydrologic conditions (persistent
high water tables) and vegetative competition (Cole and Brooks 2000, Campbell et al.
2002, Bruland and Richardson 2004, DeBerry and Perry 2004). The overall results from
the present study suggest that using a larger stocktype that has increased survival and
grows quickly is returning lost ecological structure and functions more than other smaller
stocktypes when planting in recently restored wetlands. However, less expensive
120
stocktype could be utilized in less stressful environmental conditions to obtain similar
amounts of ecological structure and functioning.
Species Group Comparison Over 5 Years
In stressed environmental conditions (FLD) primary successional species
(especially S. nigra) exhibited greater survival than secondary successional species, while
in less stressed conditions (AMB and SAT) the survival of secondary species equaled or
exceeded the survival of primary species (Figure 2-15).
Primary successional species have adaptations that allow for establishment
following planting in harsh environmental conditions while secondary species may lack
these adaptations. The physiological and morphological traits that may enhance
establishment of primary successional tree species are high photosynthetic and growth
rates, high acclimation potential, fast recovery from resource limitation, fast resource
acquisition rates, and high competitive ability in early successional stages (Bazzaz, 1979,
Brzeziecki and Kienast, 1994, Huston and Smith 1987).
Simmons et al. (2012) found that the survival, growth and vigor of early
successional species were greater than later successional species when planted in
different microtopographic treatments (ridges, flats, and mound-and-pool) after 2 years in
a riparian forest restoration. They suggested that some early successional species may be
more appropriate for restoration if they are adapted to disturbed environmental
conditions. Our results confirm that secondary species alone may not be appropriate for
returning ecological structure or restoring habitat functions in recently created or restored
wetlands because of the harsh environmental conditions often found during this time.
121
Therefore, we conclude that primary species will more quickly return more tree stem
structure and ecological functions than secondary successional species, especially in
harsh environmental conditions. For example, S. nigra, though short lived, has many
adaptations (shallow roots, rapid growth, adventitious rooting etc.) to harsh
environmental conditions and appears to be a good species for forested wetland
restoration in degraded habitats.
The H, CD and CSAG of primary successional species, regardless of stocktype,
were almost always larger than secondary successional species. Primary successional
species most often have faster growth rates than secondary species. Farmer (1980)
compared first-year growth of six deciduous species grown in nursery conditions and
found that early successional species (Liriodendron tulipifera and Prunus serotina) had
higher growth rates, net assimilation rates and higher investment in leaf area than late
successional species (Q. rubra, Q. prinus, Q. alba, and Q. ilicifolia). Results from Farmer
(1980) and the present study suggest that primary successional species are returning
ecological structure that provides ecosystem functions (such as habitat) at a faster rate
than secondary successional species across a variety of environmental conditions.
Cell Comparison After 5 Years
When survival and morphology measurements were combined across all
species/stocktype combinations, in general those trees grown in AMB had greater
survival and were structurally larger than those in FLD and were greater than or equal in
size to those grown in SAT, which generally had greater survival and increased size
compared to those grown in FLD. These results suggest that the environmental conditions
122
in FLD (flooded hydrologic conditions, uncontrolled herbaceous competition, higher clay
content, higher bulk density, reduced soil nutrient pools) caused stress to the trees planted
there in excess of their physiological tolerances. This also suggests that the initial soil,
hydrologic and competitive conditions present during restoration can affect the
development of ecological structure and functions provided by trees.
Reduction in tree survival and growth can be attributed to prolonged saturated or
flooded soil conditions that remove the plant available oxygen from the soil pore space
(Kozlowski and Davies 1975). The reduction in oxygen leads to a lack of aerobic
respiration in roots, which decreases the energy available for trees to maintain functions
of existing tissues (Hale and Orcutt 1987, Brady and Weil 2002). Many growth chamber,
greenhouse, mesocosm and field experiments have investigated the effect of hydrology
on a multitude of responses across many species of trees. While species specific
responses may vary (e.g. Taxodium distichum, mangroves) most species exhibit
decreased survival and growth when grown under prolonged inundation. Niswander and
Mitch (1995) planted ten tree species (three of which were used in this study, B. nigra, L.
styraciflua, Q. palustris) across a hydrologic gradient in a created wetland. Similar to the
results in this study, they found that trees planted in shallow water died or were severely
stressed, and that trees planted in the wet meadow portion were able to survive and grow,
while trees planted in the upland section were the largest and had the densest foliage.
Pennington and Walters (2006) investigated growth and survival four species (two of
which were used in this study, Q. palustris and Q. bicolor) planted in three hydrologic
zones (wetland, transition, upland) of created perched wetlands. Again, similar to the
present study, trees grown in the transitions zone (high soil water availability with
123
oxidized root zone) had greater height growth and survival after 5 years than those
planted in the wetland zone (reduced oxygen in the root zone). Bailey et al. (2007)
investigated the effect of organic matter loading rates and elevation in a created wetland
on several vegetation responses, including the growth of planted B. nigra. Results
suggested that the early growth of planted trees responded to both OM loading rates and
hydrology related to elevation. In the lower elevations (higher water table) the tree
growth rates were reduced compared to those in the higher elevations, consistent with
results of B. nigra from the present study. From the present study and previous studies,
stressful hydrologic conditions reduce the ecological structure and functions associated
with planted trees.
The spatial location of herbaceous vegetation and other trees in relation to planted
trees can lead to competition for resources including, light, water, nutrients, CO2, O2, and
space. Davis et al. (1999) investigated the effect of herbaceous competition along a
water-light-nitrogen gradient and found that seedling survival of Q. macrocarpa and Q.
ellipsoidalis was significantly greater when herbaceous vegetation was removed in the
wetter shaded plots. These results are consistent with the results from the present study
where secondary species had increased survival and growth in SAT where competition
was reduced and hydrologic stress was less than FLD. Pinto et al. (2012) investigated the
effect of moisture stress caused by vegetative competition on three stocktypes of P.
ponderosa. The results suggest that small stocktypes had low survival when exposed to
low moisture conditions caused by herbaceous competition, while larger stock had
somewhat improved survival. They concluded that appropriate moisture is critical for
survival and that herbaceous vegetation competes substantially for moisture. A related
124
finding from the present study was that more species/stocktype combinations had greater
survival in SAT than in AMB. This suggests that the hydrologic regime and/or reduction
in competition in SAT provided conditions that increased survival, which lead to the
restoration of ecological structure and functions.
The soil physical and chemical characteristics of FLD compared to AMB (higher
bulk density, higher clay content, lower soil nutrients) are characteristic of restored
wetlands in the Mid-Atlantic region (Bishel-Machung et al. 1996, Shaffer and Ernst
1999, Whittecar and Daniels 1999, Stolt et al. 2000, Campbell et al. 2002, Bruland and
Richardson 2005, Daniels and Whittecar 2011). Several studies have found that
compacted soil reduces the survival and aboveground and belowground growth of
planted trees (Alberty et al. 1984, Clevland and Kjelgren 1994, Kozlowski 1999, SiegelIssem et al. 2005) similar to the results of this study. Clay concentrations influence bulk
density and have also been shown to negatively affect planted tree growth. Schaff et al.
(2003) found that S. nigra cuttings (posts) planted in fine-grained sediments (higher
silt/clay) compared to coarse-grained sediments in a restored streambank had lower
biomass accumulation and leaf area. They hypothesized that the fine-grained sediments
prevent root elongation and suggested that soil texture be evaluated prior to restoration.
Results from the present study similarly show that increased clay concentrations in
conjunction with higher bulk density in FLD reduced the survival and growth of all seven
species.
Several experiments have shown that decreased abundance of soil nutrients
decreases growth of trees. In a greenhouse experiment investigating the effect of flooding
and soil nutrients on T. distichum and Nyssa aquatica growth, Effler and Goyer (2006)
125
found that flooding in combination with low soil nutrients reduced growth, while
flooding in combination with fertilization lead to similar growth of trees grown without
flooding or fertilization. Day (1987) investigated the effects of flood frequency (no
flooding, intermittent flooding and continuous flooding) and nutrient enrichment (no
enrichment, nitrogen additions, phosphorus additions and N and P additions) on the
biomass production of Acer rubrum seedlings within a greenhouse. Continuous flooding
reduced biomass production; however, adding nutrients to the continuously flooded trees
increased stem and leaf production. Bailey et al. (2007) found that B. nigra planted in a
created wetland were larger when planted in areas with higher organic amendments that
increased the nutrient content of the soil. While the species and treatments may have
varied from previous studies, results from the present study similarly show that low soil
nutrient concentrations in combination with hydrologic and competitive stress will reduce
the survival and size of planted trees. This suggests that in order to ensure the return of
ecological structure and functions associated with planted trees in restored forested
wetlands, particular attention should be paid to the initial soil physical and chemical
characteristics.
Conclusions
From this study we conclude that species and stocktypes can be selected to match
environmental conditions and to maximize return of ecological structure, functions and
services to the landscape. These results can be applied to wetlands undergoing restoration
or enhancement that have a range of environmental conditions or to other afforestation or
reforestation projects, such as riparian buffers or uplands. Results from this study show
126
that primary successional species (especially S. nigra) planted using larger containerized
stock are more appropriate for harsh environmental conditions, such as areas with high
water tables, poor soil conditions or heavy herbaceous competition. These areas may
correspond to stream banks that experience flooding, post-agricultural wetland
restorations or other challenging environments. However, when attempting to restore
forested wetland habitat or any other ecosystem the diversity of planted species should be
considered.
In areas with less stressful environmental conditions (moderate water table, less
compacted and nutrient rich soil conditions and reduced herbaceous competition),
primary or secondary species could be expected to have similar survival rates. However,
due to the slow growth of secondary species, primary species are preferred for returning
their associated ecological structure and functions more quickly. In these less stressful
environmental conditions the less expensive bare root stocktype can be expected to
develop similar morphological structure as the more expensive gallon stocktype. The
tubeling stocktype does not appear to provide added benefit for its intermediate price.
Overall, from comparing differences among cells, the results from this study
suggest that initial environmental conditions can have a large influence on survival and
growth of planted trees. These results could be used for a variety of tree planting
situations besides forested wetland restoration including carbon sequestration projects,
wildlife habitat creation and other conservation projects.
127
Literature Cited
Acquaah, G. (2005). Horticulture: Principles and Practice (Third ed.). Upper Saddle
River, New Jersey: Pearson Education.
Adkins, C. R., Frank, S. D., Ward, N. A., & Fulcher, A. F. (2012). Birch - Betula spp. In
A. F. Fulcher, & S. A. White (Eds.), IPM for Select Deciduous Trees in
Southeastern US Nursery Production (2 ed.). Southern Nursery IPM working
group.
Alberty, C. A., Pellett, H. M., & Taylor, D. H. (1984). Characterization of Soil
Compaction at Construction Sites and Woody Plant Responses. Journal of
Environmental Horticulture, 2, 48-53.
Allen, A. W. (1987). Habitat suitability index models: barred owl. Biological Report,
United States Fish and Wildlife Service.
Allen, J. A., & Kennedy, H. E. (1989). Bottomland hardwood reforestation in the Lower
Mississippi Valley. Tech. rep., United States Department of the Interior - Fish and
Wildlife Service, National Wetlands Research Center, Southern Forest
Experiment Station, Stoneville, MS.
Atkinson, R., & Cairns Jr, J. (2001). Plant decomposition and litter accumulation in
depressional wetlands: functional performance of two wetland age classes that
were created via excavation. Wetlands, 21(3), 354-362.
Atkinson, R., Perry, J., & Cairns, J. J. (2005). Vegetation communities of 20-year-old
created depressional wetlands. Wetlands Ecology and Management, 13(4), 469478.
Atkinson, R., Perry, J., Noe, G., Daniels, W., & Cairns, J. J. (2010). Primary Productivity
in 20-year Old Created Wetlands in Southwestern Virginia. Wetlands, 30(2), 200210.
Acquaah, G. (2005). Horticulture: Principles and Practice (Third ed.). Upper Saddle
River, New Jersey: Pearson Education.
Adkins, C. R., Frank, S. D., Ward, N. A., & Fulcher, A. F. (2012). Birch - Betula spp. In
A. F. Fulcher, & S. A. White (Eds.), IPM for Select Deciduous Trees in
Southeastern US Nursery Production (2 ed.). Southern Nursery IPM working
group.
128
Alberty, C. A., Pellett, H. M., & Taylor, D. H. (1984). Characterization of Soil
Compaction at Construction Sites and Woody Plant Responses. Journal of
Environmental Horticulture, 2, 48-53.
Allen, A. W. (1987). Habitat suitability index models: barred owl. Biological Report,
United States Fish and Wildlife Service.
Allen, J. A., & Kennedy, H. E. (1989). Bottomland hardwood reforestation in the Lower
Mississippi Valley. Tech. rep., United States Department of the Interior - Fish and
Wildlife Service, National Wetlands Research Center, Southern Forest
Experiment Station, Stoneville, MS.
Atkinson, R., & Cairns Jr, J. (2001). Plant decomposition and litter accumulation in
depressional wetlands: functional performance of two wetland age classes that
were created via excavation. Wetlands, 21(3), 354-362.
Atkinson, R., Perry, J., & Cairns, J. J. (2005). Vegetation communities of 20-year-old
created depressional wetlands. Wetlands Ecology and Management, 13(4), 469478.
Atkinson, R., Perry, J., Noe, G., Daniels, W., & Cairns, J. J. (2010). Primary Productivity
in 20-year Old Created Wetlands in Southwestern Virginia. Wetlands, 30(2), 200210.
Bailey, D., Perry, J., & Daniels, W. (2007). Vegetation dynamics in response to organic
matter loading rates in a created freshwater wetland in southeastern Virginia.
Wetlands, 27(4), 936-950.
Bazzaz, F. (1979). The physiological ecology of plant succession. Annual Review of
Ecology and Systematics, 10, 351-371.
Beineke, W., & Perry, T. (1965). Genetic variation in ability to withstand transplanting
shock. 8th Southern Conference on Forest Tree Improvement, (pp. 106-109).
Savannah, GA.
Bergdolt, F., Prehmus, C., & Barham, J. (2005). An Evaluation of Wetland Mitigation
Site Compliance at the Washington State Department of Transportation. Tech.
rep., Washington State Department of Transportation, Olympia, WA.
129
Bergschneider, C. (2005). Determining an appropriate organic matter loading rate for a
created coastal plain forested wetland. Master's thesis, Virginia Polytechnic
Institute and State University, Blacksburg, VA, USA.
Bishel-Machung, L., Brooks, R., Yates, S., & Hoover, K. (1996). Soil properties of
reference wetlands and wetland creation projects in Pennsylvania. Wetlands,
16(4), 532-541.
Brady, N. C., & Weil, R. R. (2002). The Nature and Properties of Soil. Upper Saddle
River, NJ: Prentice Hall.
Brooks, R. P., Snyder, C., & Brinson, M. M. (2013). Aquatic Landscapes: The
Importance of Integrating Waters. In R. P. Brooks, & D. H. Wardrop (Eds.), MidAtlantic Freshwater Wetlands: Advances in Wetlands Science, Management,
Policy, and Practice. Springer.
Brown, S., & Veneman, P. (2001). Effectiveness of compensatory wetland mitigation in
Massachusetts, USA. Wetlands, 21(4), 508-518.
Bruland, G. L., & Richardson, C. J. (2005). Spatial variability of soil properties in
created, restored, and paired natural wetlands. Soil Science Society of America
Journal, 69(1), 273.
Bruland, G., & Richardson, C. (2004). Hydrologic gradients and topsoil additions affect
soil properties of Virginia created wetlands. Soil Science Society of America
Journal, 68(6), 2069-2077.
Burdett, A. N., Herring, L. J., & Thompson, C. F. (1984). Early growth of planted spruce.
Canadian Journal of Forest Research, 14(5), 644-651.
Burkett, V. R., Draugelis-Dale, R. O., Williams, H. W., & Schoenholtz, S. H. (2005).
Effects of flooding regime and seedling treatment on early survival and growth of
nuttall oak. Restoration Ecology, 13(3), 471-479.
Burns, R., & Honkala, B. (1990). Silvics of North America, Volume 2, Hardwoods.
Agriculture Handbook (Washington).
Campbell, D., Cole, C., & Brooks, R. (2002). A comparison of created and natural
wetlands in Pennsylvania, USA. Wetlands Ecology and Management, 10(1), 4149.
130
Carey, J. H. (1992). Quercus phellos. In Rocky Mountain Research Station, Fire Sciences
Laboratory (Ed.), Fire Effects Information System. United States Department of
Agriculture, Forest Service.
Carey, J. H. (1992a). Quercus palustris. In Rocky Mountain Research Station, Fire
Sciences Laboratory (Ed.), Fire Effects Information System. United States
Department of Agriculture, Forest Service.
Carlson, W. (1986). Root system considerations in the quality of loblolly pine seedlings.
Southern Journal of Applied Forestry, 10(2), 87-92.
Carlson, W. C., & Miller, D. E. (1990). Target Seedling Root System Size, Hydraulic
Conductivity, and Water Use During Seedling Establishment. In R. Rose, S. J.
Campbell, & T. D. Landis (Ed.), Target Seedling Symposium: Proceedings,
Combined Meeting of the Western Forest Nursery Associations. Roseburg, OR.
Chambers R., Fourqurean J. (1991) Alternative criteria for assessing nutrient limitation of
a wetland macrophyte (Peltandra virginica (L.) Kunth). Aquatic Botany 40,305320.
Chong, J.-H., Ward, N. A., Dunwell, W. C., & Chappell, M. R. (2012). Oak - Quercus
spp. In A. F. Fulcher, & S. A. White (Eds.), IPM for Select Deciduous Trees in
Southeastern US Nursery Production (2 ed.). Southern Nursery IPM working
group.
Clade, B. S. (1987). Comparison of tree basal area and canopy cover in habitat models:
Subalpine forest. Journal of Wildlife Management, 61(2), 326-335.
Cleveland, B., & Kjelgren, R. (1994). Establishment of six tree species on deep-tilled
minesoil during reclamation. Forest Ecology and Management, 68(2-3), 273-280.
Coladonato, M. (1992). Liquidambar styraciflua. In Rocky Mountain Research Station,
Fire Sciences Laboratory (Ed.), Fire Effects Information System. United States
Department of Agriculture, Forest Service.
Cole, C. A., & Brooks, R. P. (2000). A comparison of the hydrologic characteristics of
natural and created mainstem floodplain wetlands in Pennsylvania. Ecological
Engineering, 14(3), 221-231.
Cole, C., & Shafer, D. (2002). Section 404 wetland mitigation and permit success criteria
in Pennsylvania, USA, 1986--1999. Environmental Management, 30(4), 508-515.
131
Cole, C., Brooks, R., & Wardrop, D. (2001). Assessing the relationship between biomass
and soil organic matter in created wetlands of central Pennsylvania, USA.
Ecological Engineering, 17(4), 423-428.
Craft, C., & Casey, W. (2000). Sediment and nutrient accumulation in floodplain and
depressional freshwater wetlands of Georgia, USA. Wetlands, 20(2), 323-332.
Cypert, E., & Webster, B. S. (1948). Yield and use by wildlife of acorns of water and
willow oaks. The Journal of Wildlife Management, 12(3), 227-231.
Daniels, W. L., & Whittecar, G. R. (2011). Assessing soil and hydrologic properties for
the successful creation of non-tidal wetlands. In L. M. Vasilas, & B. L. Vasilas
(Eds.), A Guide to Hydric Soils in the Mid-Atlantic Region (Vol. 2.0, pp. 120136). Mid-Atlantic Hydric Soils Committee.
Daniels, W. L., Perry, J. E., & Whittecar, R. G. (2005). Effects of soil amendments and
other practices upon the success of the Virginia Department of Transportation’s
non-tidal wetland mitigation efforts. Final Contract Report, Virginia
Transportation Research Council, Charlottesville, VA.
Davis, M. A., Wrage, K. J., Reich, P. B., Tjoelker, M. G., Schaeffer, T., & Muermann, C.
(1999). Survival, growth, and photosynthesis of tree seedlings competing with
herbaceous vegetation along a water-light-nitrogen gradient. Plant Ecology,
145(2), 341-350.
Day, F. P. (1987). Effects of flooding and nutrient enrichment on biomass allocation in
Acer rubrum seedlings. American Journal of Botany, 74(10), 1541-1554.
Day, R., Doyle, T., & Draugelis-Dale, R. (2006). Interactive effects of substrate,
hydroperiod, and nutrients on seedling growth of Salix nigra and Taxodium
distichum. Environmental and Experimental Botany, 55(1-2), 163-174.
DeBerry, D., & Perry, J. (2004). Primary succession in a created freshwater wetland.
Castanea, 69(3), 185-193.
Denton, S. (1990). Growth rates, morphometrics and planting recommendations for
cypress trees at forested mitigation sites. In Webb (Ed.), Proceedings of the 17th
Annual Conference on Wetlands Restoration and Creation. Tampa, FL.
132
Dickerson, J. (2002). Pin Oak Quercus palustris Muenchh. Plant Fact Sheet. Tech. rep.,
United States Department of Agriculture, Natural Resources Conservation
Service.
Dickson, J. G., Conner, R. N., & Williamson, J. H. (1983). Snag retention increases bird
use of a clear-cut. The Journal of Wildlife Management, 47(3), 799-804.
Dixon, R. K., Garrett, H. E., Cox, G. S., & Pallardy, S. G. (1984). Mycorrhizae and
reforestation success in the oak hickory region. In M. L. Duryea, & G. N. Brown
(Eds.), Seedling physiology and reforestation success. Martinus Nijhoff/Dr W.
Junk Publishers.
Donovan, L., McLeod, K., Sherrod, K., & Stumpff, N. (1988). Response of woody
swamp seedlings to flooding and increased water temperatures. I. Growth,
biomass, and survivorship. American Journal of Botany, 75(8), 1181-1190.
Dosskey, M., & Bertsch, P. (1994). Forest sources and pathways of organic matter
transport to a blackwater stream: a hydrologic approach. Biogeochemistry, 24(1),
1-19.
Dugger, K. M., & Fredrickson, L. H. (1992). Life history and habitat needs of the wood
duck. Fish and Wildlife Leaflet, United States Department of the Interior, Fish
and Wildlife Service, Washington, DC.
Ebel, B. H., & Kormanik, P. P. (1966). Stictocephala militaris, a mebracid Homoptera)
associated with sweetgum, Liquidambar styraciflua. Annals of the Entomological
Society of America, 59(3), 600-601.
Effler, R., & Goyer, R. (2006). Baldcypress and water tupelo sapling response to multiple
stress agents and reforestation implications for Louisiana swamps. Forest Ecology
and Management, 226(1-3), 330-340.
Gamble, D. L., & Mitsch, W. J. (2006). Tree growth and hydrologic patterns in urban
forested mitigation wetlands. Tech. rep., Schiermeier Olentangy River Wetland
Research Park, School of Natural Resources, The Ohio State University.
Gardiner, E., Salifu, K., Jacobs, D., Hernandez, G., & Overton, R. (2007). Field
performance of Nuttall Oak on former agricultural fields: Initial effects of nursery
source and competition control. In L. E. Riley, R. Dumroese, & T. Landis (Ed.),
133
National proceedings: Forest and Conservation Nursery Associations. Fort
Collins, CO.
Godfrey, C. L. (1996). Reproductive biology and denning ecology of Virginia's exploited
black bear population. Master's thesis, Virginia Polytechnic Intitute and State
University, Blacksburg, VA.
Grossnickle, S. (2005). Importance of root growth in overcoming planting stress. New
Forests, 30(2), 273-294.
Grossnickle, S. C., & El-Kassaby, Y. A. (2015). Bareroot versus container stocktypes: a
performance comparison. New Forests.
Grovenburg, T. W., Jacques, C. N., Klaver, R. W., & Jenks, J. A. (2010). Bed site
selection by neonate deer in grassland habitats on the northern Great Plains. The
Journal of Wildlife Management, 74(6), 1250-1256.
Guyette, R. P., Rose-Marie, M., Kabrick, J., & Stambaugh, M. C. (2004). A Perspective
on Quercus Life History Characteristics and Forest Diturbance. In M. A. Spetich
(Ed.), Upland oak ecology symposium: history, current conditions, and
sustainability. Asheville, NC: U.S.
Hale, M. G., & Orcutt, D. M. (1987). The physiology of plants under stress. John Wiley
& Sons Ltd.
Hardin, K. I., & Evans, K. E. (1977). Cavity nesting bird habitat in the oak-hickory
forests -- a review. Gen. Teck. Rep., United States Department of Agriculture,
Forest Service, North Central Forest Experiment Station, St. Paul, MN.
Harlow, R. F., Shrauder, P. A., & Seehorn, M. E. (1975). Deer browse resources of the
Oconee National Forest. Research Paper, United States Department of
Agriculture, Forest Service, Southeastern Forest Experiment Station, Asheville,
NC.
Harmon, M., Franklin, J., Swanson, F., Sollins, P., Gregory, S., Lattin, J., et al. (1986).
Ecology of coarse woody debris in temperate ecosystems. Advances in Ecological
Research, 15(133), 302.
Henderson, D., Botch, P., Cussimanio, J., Ryan, D., Kabrick, J., & Dey, D. (2009).
Growth and mortality of pin oak and pecan reforestation in a constructed
134
wetland- Analysis with management implications. Science and Management
Technical Series, USDA Forest Service, Jefferson City, MO.
Hudson, I. H. (2010). The Effect of Adjacent Forests on Colonizing Tree Density in
Restored Wetland Compensation Sites in Virginia. Master's thesis, Christopher
Newport University.
Hunt, R. (1978). Plant Growth Analysis. Bedford Square, London: Edward Arnold.
Hunt, R. (1990). Basic growth analysis - Plant growth analysis for beginners. Broadwick
Street, London: Unwin Hyman, Ltd.
Hupp, C., Woodside, M., & Yanosky, T. (1993). Sediment and trace element trapping in
a forested wetland, Chickahominy River, Virginia. Wetlands, 13(2), 95-104.
Huston, M., & Smith, T. (1987). Plant succession: life history and competition. American
Naturalist, 130(2), 168-198.
Johnson, W. T., & Lyon, H. H. (1988). Insects that feed on trees and shrubs (Second ed.).
Comstock Publishing Associates.
Konopka, B., Pajtik, J., Seben, V., & Lukac, M. (2011). Belowground biomass functions
and expansion factors in high elevation Norway spruce. Forestry, 84(1), 41-48.
Kozlowski, T. T. (1999). Soil compaction and growth of woody plants. Scandinavian
Journal of Forest Research, 14, 596-619.
Kozlowski, T., & Davies, W. (1975). Control of water balance in transplanted trees.
Journal of Arboriculture, 1(1), 1-10.
Krinard, R. M., & Kennedy, J. H. (1987). Fifteen-year growth of six planted hardwood
species on Sharkey Clay Soil. Research Note, USDA Forest Service, New
Orleans, LA.
Latimer, J. G., & Close, D. (Eds.). (2014). Home Grounds and Animals: 2014 Pest
Management Guide. Virginia Cooperative Extension.
Lichvar, R. W. (2013). The National Wetland Plant List: 2013 Wetland Ratings.
Phytoneuron, 49, 1-241.
Lichvar, R. W., Melvin, N. C., Butterwick, M. L., & Kirchner, W. N. (2012). National
wetland plant list indicator rating definitions. Tech. rep., United States Army
Corps of Engineers, Engineer Research and Development Center.
135
MacArthur, R. H., & MacArthur, J. W. (1961). On bird species diversity. Ecology, 42(3),
594-598.
Martin, A. C., Zim, H. S., & Nelson, A. L. (1951). American Wildlife and Plants: A guide
to wildlife food habits: the use of trees, shrub, weed, and herbs by birds and
mammals of the United States (Second Edition ed.). McGraw-Hill BMcGraw-Hill,
Inc.
Matthews, J., & Endress, A. (2008). Performance criteria, compliance success, and
vegetation development in compensatory mitigation wetlands. Environmental
Management, 41(1), 130-141.
McCurry, J., Gray, M., & Mercker, D. (2010). Early Growing Season Flooding Influence
on Seedlings of Three Common Bottomland Hardwood Species in Western
Tennessee. Journal of Fish and Wildlife Management, 1(1), 11-18.
McLeod, K. (2000). Species selection trials and silvicultural techniques for the
restoration of bottomland hardwood forests. Ecological Engineering, 15, S35-S46.
McLeod, K. W., Reed, M. R., Moyer, B. P., & Ciravolo, T. G. (2006). Species selection
trials and silvicultural techniques for the restoration of bottomland hardwood
forests - A 10 year review. In K. F. Connor (Ed.), Proceedings of the 13th
biennial southern silvicultural research conference. Asheville, NC.
McLeod, K., & McPherson, J. (1973). Factors limiting the distribution of Salix nigra.
Bulletin of the Torrey Botanical Club, 100(2), 102-110.
McLeod, K., Reed, M., & Nelson, E. (2001). Influence of a willow canopy on tree
seedling establishment for wetland restoration. Wetlands, 21(3), 395-402.
Mitsch, W., Bernal, B., Nahlik, A., Mander, Zhang, L., Anderson, C., et al. (2012).
Wetlands, carbon, and climate change. Landscape Ecology, 28(4), 583-597.
Moore, L. (2002). Willow Oak Quercus phellos L. Plant Fact Sheet. Tech. rep., United
States Department of Agriculture, Natural Resources Conservation Service.
Moreno-Mateos, D., Power, M. E., Comin, F. A., & Yockteng, R. (2012). Structural and
functional loss in restored wetland ecosystems. PLoS Biology, 10(1).
Morgan, K., & Roberts, T. (2003). Characterization of wetland mitigation projects in
Tennessee, USA. Wetlands, 23(1), 65-69.
136
Morley, T. R. (2008). The functions of forested headwater wetlands in a New England
Landscape. Ph.D. dissertation, The University of Maine.
Mulholland, P., & Kuenzler, E. (1979). Organic carbon export from upland and forested
wetland watersheds. Limnology and Oceanography, 960-966.
National Research Counction (NRC). (2001). Compensating for wetland losses under the
Clean Water Act. Washington, DC: National Academies Press.
Nesom, G. (2009). Swamp white oak Quercus bicolor Willd. Plant Guide. Tech. rep.,
United States Department of Agriculture Natural Resources Conservation Service.
Niswander, S., & Mitsch, W. (1995). Functional analysis of a two-year-old created instream wetland: hydrology, phosphorus retention, and vegetation survival and
growth. Wetlands, 15(3), 212-225.
Nixon, C. M., McClain, M. W., & Russell, K. R. (1970). Deer food habits and range
characteristics in Ohio. The Journal of Wildlife Management, 34(4), 870-886.
Pallardy, S. G. (2008). Physiology of Woody Plants (Third ed.). Burlington, MA:
Academic Press, Inc.
Paul KI, Roxburgh SH, England JR, Ritson P, Hobbs T, Brooksbank K, Raison RJ,
Larmour JS, Murphy S, Norris J, Neumann C, Lewis T, Jonson J, Carter, JL,
McArthus G, Barton C, Rose B (2013) Development and testing of allometric
equations for estimating above-ground biomass of mixed-species environmental
plantings. Forest Ecology and Management 310:483-494
Pennington, M. R., & Walters, M. B. (2006). The response of planted trees to vegetation
zonation and soil redox potential in created wetlands. Forest Ecology and
Management, 233(1), 1-10.
Pinto, J. R., Marshall, J. D., Dumroese, R. K., Davis, A. S., & Cobos, D. R. (2012).
Photosynthetic response, carbon isotopic composition, survival, and growth of
three stock types under water stress enhanced by vegetative competition.
Canadian Journal of Forest Research, 42, 333-344.
Pinto, J., Marshall, J., Dumroese, R., Davis, A., & Cobos, D. (2011). Establishment and
growth of container seedlings for reforestation: A function of stocktype and
edaphic conditions. Forest Ecology and Management, 261, 1876-1884.
137
Rheinhardt, R. D., Brinson, M. M., Meyer, G. F., & Miller, K. H. (2012). Carbon storage
of headwater riparian zones in an agricultural landscape. Carbon Balance and
Management, 7(4).
Rheinhardt, R., Whigham, D., Khan, H., & Brinson, M. (2000). Vegetation of headwater
wetlands in the inner coastal plain of Virginia and Maryland. Castanea, 21-35.
Robb, J. (2002). Assessing wetland compensatory mitigation sites to aid in establishing
mitigation ratios. Wetlands, 22(2), 435-440.
Robbins, C. S., Dawson, D. K., & Dowell, B. A. (1989). Habitat area requirements of
breeding forest birds of the middle Atlantic states. Wildlife Monographs(103), 334.
Row, J. M., & Geyer, W. A. (2010). Black willow Salix nigra Marsh. Plant Guide, United
States Department of Agriculture Natural Resources Conservation Service.
Santamour, J. F., & Greene, A. (1986). European hornet damage to ash and birch trees.
Journal of Arboriculture (USA).
Sather, J. H., & Smith, R. D. (1984). An overview of major wetland functions and values.
Fort Collins, CO: United States Department of the Interior Fish and Wildlife
Service.
Schaff, S. D., Pezeshki, S. R., & Shields Jr, F. D. (2003). Effects of soil conditions on
survival and growth of black willow cuttings. Environmental Management, 31(6),
748-763.
Schweitzer, C., Gardiner, E., Stanturf, J., & Ezell, A. (1999). Methods to improve
establishment and growth of bottomland hardwood artificial regeneration. In J. W.
Stringer, & D. L. Loftis (Ed.), Proceedings of the 12th Central Hardwood Forest
Conference, Gen. Tech. Rep. SRS-24. Asheville, NC.
Shaffer, p. W., & Ernst, T. L. (1999). Distribution of SOM in freshwater wetlands in the
Protland, Oregon area. Wetlands, 19, 505-516.
Sharitz, R., Barton, C., & De Steven, D. (2006). Tree plantings in depression wetland
restorations show mixed success (South Carolina). Ecological Restoration, 24(2),
114-115.
138
Siegel-Issem, C. M., Burger, J. A., Powers, R. F., Ponder, F., & Patterson, S. C. (2005).
Seedling root growth as a function of soil density and water content. Soil Science
Society of America Journal, 69, 215-226.
Silberhorn, G. (1992, January). Sweet Gum. Wetland Flora, Virginia Institute of Marine
Science.
Simmons, M. E., Wu, X. B., & Whisenant, S. G. (2012). Responses of Pioneer and LaterSuccessional Plant Assemblages to Created Microtopographic Variation and Soil
Treatments in Riparian Forest Restoration. Restoration Ecology, 20(3), 369-377.
Snyder, S. A. (1992). Quercus bicolor. In Rocky Mountain Research Station, Fire
Sciences Laboratory (Ed.), Fire Effects Information System. United States
Department of Agriculture, Forest Service.
South, D., & Zwolinski, J. (1996). Transplant stress index: A proposed method of
quantifying planting check. New Forests, 13(1), 315-328.
Stanturf, J., Conner, W., Gardiner, E., Schweitzer, C., & Ezell, A. (2004). Recognizing
and overcoming difficult site conditions for afforestation of bottomland
hardwoods. Ecological Restoration, 22(3), 183.
Stanturf, J., Gardiner, E., Shepard, J., Schweitzer, C., Portwood, C., & Dorris Jr, L.
(2009). Restoration of bottomland hardwood forests across a treatment intensity
gradient. Forest Ecology and Management, 257(8), 1803-1814.
Stefanik, K., & Mitsch, W. (2012). Structural and functional vegetation development in
created and restored wetland mitigation banks of different ages. Ecological
Engineering, 39, 104-112.
Stolt, M., Genthner, M., Daniels, W., Groover, V., Nagle, S., & Haering, K. (2000).
Comparison of soil and other environmental conditions in constructed and
adjacent palustrine reference wetlands. Wetlands, 20(4), 671-683.
Streever, B. (1999). Examples of performance standards for wetland creation and
restoration in Section 404 permits and an approach to developing performance
standards. Tech. rep., United States Army Corps of Engineers, Wetlands
Research Program.
139
Sudol, M., & Ambrose, R. (2002). The US Clean Water Act and habitat replacement:
evaluation of mitigation sites in Orange County, California, USA. Environmental
Management, 30(5), 727-734.
Sullivan, J. (1993). Betula nigra. In Rocky Mountain Research Station, Fire Sciences
Laboratory (Ed.), Fire Effects Information System. United States Department of
Agriculture, Forest Service.
Sullivan, J. (1994). Platanus occidentalis. In Rocky Mountain Research Station, Fire
Sciences Laboratory (Ed.), Fire Effects Information System. United States
Department of Agriculture, Forest Service.
Taylor, T., Loewenstein, E., & Chappelka, A. (2004). Improving species composition in
mismanaged bottomland hardwood stands in western Alabama. In K. F. Connor
(Ed.), Proceedings of the 12th biennial southern silvicultural research
conference, (pp. 565-570). Asheville, NC.
Tesky, J. L. (1992). Salix nigra. In F. S. Rocky Mountain Research Station (Ed.), Fire
Effects Information System. United States Department of Agriculutre, Forest
Service.
Tiner, R. W., & Finn, J. T. (1986). Status and recent trends of wetlands in five MidAtlantic states: Delaware, Maryland, Pennsylvania, Virginia, and West Virginia.
Tech. rep., United States Department of the Interior, Fish and Wildlife Service,
Fish and Wildlife Enhancement, National Wetlands Inventory Project, Newton
Corner, MA.
Tiner, R., Swords, J., & Bergquist, H. (2005). Recent Wetland Trends in Southeastern
Virginia: 1994-2000. NWI Wetland Trends Report, United States Fish and
Wildlife Service, Northeast Region, Hadley, MA.
Twedt, D., & Wilson, R. (2002). Supplemental planting of early successional tree species
during bottomland hardwood afforestation. In K. W. Outcalt (Ed.), Proceedings of
the eleventh biennial southern silvicultural research conference, (pp. 358-364).
Asheville, NC.
United States Army Corps of Engineers (USACE) & United States Environmental
Protection Agency (USEPA) (2008). Compensatory Mitigation for Losses of
140
Aquatic Resources; Final Rule. Tech. rep., United States Army Corps of
Engineers and United States Environmental Protection Agency, Washington D.C.
United States Army Corps of Engineers (USACE) & Virginia Department of
Environmental Quality (VADEQ) (2004). Recommendations for wetland
compensatory mitigation: Including site design, permit conditions, performance
and monitoring criteria. Recomendations.
United States Department of Agriculture and Natural Resources Conservation Service,
Soil Survey Staff. Web Soil Survey. <http://websoilsurvey.nrcs.usda.gov/>
Accessed 2/23/2015.
United States Geological Survey (USGS) (1999). National Water Summary - Wetland
Resources: Virginia. Supply Paper, United States Geological Survey, USGS
National Center, Reston, VA.
Virginia Department of Environmental Quality (VADEQ) (2010). Mitigation Banking
Instrument Template. Template, Virginia Department of Environmental Quality,
Richmond, VA.
Van Dersal, W. R. (1938). Native woody plants of the United States, their erosion-control
and wildlife values. United States Department of Agriculture.
Washington State Department of Transportation (WSDOT) (2008). Woody Vegetation
Performance Criteria for Wetland Mitigation Sites in Washington. Tech. rep.,
Washington State Department of Transportation, Olympia, WA.
Walbridge, M. (1993). Functions and values of forested wetlands in the southern United
States. Journal of Forestry;(United States), 91(5).
Westcott, C. (2001). Westcott's Plant Disease Handbook (6th Edition ed.). (R. K. Horst,
Ed.) Norwell, Massachusetts: Kluwer Academic Publishers.
Westfall, P. H., Tobias, R. D., & Wolfinger, R. D. (2011). Multiple Comparisons and
Multiple Tests Using SAS®, Second Edition. Cary, NC: SAS Institute Inc.
Whigham, D., Chitterling, C., & Palmer, B. (1988). Impacts of freshwater wetlands on
water quality: a landscape perspective. Environmental Management, 12(5), 663671.
Whittecar, G., & Daniels, W. (1999). Use of hydrogeomorphic concepts to design created
wetlands in southeastern Virginia. Geomorphology, 31(1-4), 355-371.
141
Wigley, T. B., & Garner, M. E. (1988). Forest habitat use by the white-tailed deer inthe
arkansas coastal plain. Proceedings of Arkansas Academy of Science, 42, 96-98.
Winer, B. J., Brown, D. R., & Michels, K. M. (1991). Statistical principles in
experimental design (3rd Edition ed.). McGraw-Hill.
142
Tables
Table 2-1. Description of environmental parameters for the three experimental cells.
Numbers represent averages with associated standard deviations. Soil parameters
represent 44 samples taken in each cell.
Environmental Parameter
Ambient Cell
Saturated Cell
Flooded Cell
Hydrology
Recevied only precipitation
Kept saturated for a
minimum of 90% of the
growing season within the
root-zone (10 cm) of the
plantings and irrigated as
needed
Inundated above the root
crown for a minimum of
90% of each year
Soil Preparation
Disked and Tilled
Disked and Tilled
Excavated to a depth of 1
m to an existing clay layer
Herbaceous Vegetation
Control
Riding Lawnmower, Push
mower, weedwacker,
Glyphosate application
Riding Lawnmower, Push
mower, weedwacker,
Glyphosate application
None
Bulk Density (g/cm3)
Soil Percentage Sand
Soil Percentage Silt
Soil Percentage Clay
Soil Percentage Carbon
1.03 (0.11)
85.16 (6.16)
10.22 (5.48)
4.62 (1.25)
1.47 (0.37)
1.1 (0.13)
88.35 (4.38)
7.57 (3.12)
4.08 (1.5)
1.2 (0.4)
1.38 (0.14)
63.74 (10.05)
17.27 (6.44)
18.99 (6.64)
0.34 (0.12)
Soil Percentage Nitrogen
0.17 (0.04)
0.15 (0.04)
0.08 (0.03)
Soil Percentage Phosphorus
0.29 (0.08)
0.26 (0.08)
0.18 (0.04)
143
Table 2-2. Initial morphologies of planting stock and standard deviations. Same letters
denote no significant difference between stocktypes for each species (p>0.05).
Stem CrossStocktype sectional Area at
Groundline (cm2)
Betula nigra
Bare root
0.5 (0.7) b
Gallon
1 (1) a
Tubeling
0.6 (0.6) b
Liquidambar styraciflua Bare root
0.6 (0.6) b
Gallon
0.8 (0.8) a
Tubeling
0.6 (0.7) b
Platanus occidentalis
Bare root
0.4 (0.6) b
Gallon
1.1 (1.1) a
Tubeling
0.6 (0.6) b
Quercus bicolor
Bare root
0.7 (0.7) a
Gallon
0.7 (0.6) a
Tubeling
0.6 (0.6) a
Quercus palustris
Bare root
0.6 (0.6) b
Gallon
0.8 (0.7) a
Tubeling
0.5 (0.7) b
Quercus phellos
Bare root
0.5 (0.6) b
Gallon
1 (1) a
Tubeling
0.5 (0.5) b
Salix nigra
Bare root
0.5 (0.6) b
Gallon
1.2 (1.2) a
Tubeling
0.5 (0.6) b
Species
Height (cm)
Crown Diameter (cm)
50.6 (27.2) c
78.8 (48.4) a
61.1 (35.7) b
58.4 (37.8) ab
62.7 (33.9) a
50.1 (35.8) b
48.4 (27) b
83.6 (50.6) a
56.4 (29.3) b
52.2 (30.2) a
47.3 (28.1) a
51.8 (35.9) a
51.1 (33) b
66.7 (32.3) a
45 (34.5) b
55.2 (30.7) b
73.4 (44.5) a
54 (27.6) b
50.6 (28.8) b
76.7 (42) a
55.5 (34.7) b
6.2 (9.9) b
16.7 (18.1) a
9.2 (11.3) b
8.2 (11.1) b
14.6 (14.1) a
9.1 (11.1) b
6.6 (9.4) b
14.3 (13.5) a
9.3 (11.6) b
9.5 (9.6) a
9.9 (11) a
9.4 (11.9) a
8.7 (9.2) b
19.2 (18.6) a
6.4 (10.2) b
9.7 (11.2) b
21.6 (23.5) a
7.8 (9.5) b
7.2 (12.1) b
16.7 (15.1) a
7.6 (10.6) b
144
Table 2-3. Results of type 3 tests of fixed effects for species and stocktype for each
morphological variable over 5 years.
Stem Cross-sectional
Area at Groundline
Survival
Cell
AMB
SAT
FLD
Source of variation
Species
Stocktype
Species x Stocktype
Species
Stocktype
Species x Stocktype
Species
Stocktype
Species x Stocktype
DF
6
2
12
6
2
12
6
2
12
Wald Chi Square
29.03
21.60
57.00
18.86
16.22
21.40
120.89
4.53
44.77
Pr > F
0.0014
<0.001
<0.001
0.0038
<0.001
0.045
0.1036
<0.001
<0.001
F-value
1.52
33.66
8.22
13.26
26.41
3.67
86.01
79.34
3.42
Pr > F
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
Height
F-value
18.97
22.24
4.51
22.38
18.93
2.42
517.61
87.35
21.91
Pr > F
<0.001
<0.001
<0.001
<0.001
<0.001
0.0044
<0.001
<0.001
<0.001
Canopy Diameter
F-value
30.69
20.86
6.78
27.47
19.08
2.10
223.65
71.23
6.67
Pr > F
<0.001
<0.001
<0.001
<0.001
<0.001
0.0148
<0.001
<0.001
<0.001
145
Table 2-4. Results of type 3 tests of fixed effects for successional group and species for
each morphological variable over 5 years.
Stem Cross-sectional
area at Groundline
Survival
Cell
AMB
SAT
FLD
Source of variation
Succesional Group
Stocktype
Stocktype x Successional Group
Succesional Group
Stocktype
Stocktype x Successional Group
Succesional Group
Stocktype
Stocktype x Successional Group
DF
1
2
2
1
2
2
1
2
2
Wald Chi Square
0.0987
45.60
8.71
2.027
31.57
3.69
86.43
48.48
20.88
Pr > F
0.7534
<0.001
0.0128
0.1545
<0.001
0.158
<0.001
<0.001
<0.001
F-value
103.16
6.25
2.21
107.53
9.88
3.29
102.2
60.93
4.76
Pr > F
<0.001
0.002
0.1104
<0.001
<0.001
0.0379
<0.001
<0.001
0.0088
Height
F-value
61.31
15.67
1.23
77.63
20.37
0.6
179.25
336.23
10.57
Pr > F
<0.001
<0.001
0.2929
<0.001
<0.001
0.5513
<0.001
<0.001
<0.001
Canopy Diameter
F-value
59.06
23.83
1.37
60.14
24.90
0.74
39.19
146.19
1.65
Pr > F
<0.001
<0.001
0.254
<0.001
<0.001
0.4758
<0.001
<0.001
0.1923
146
Table 2-5. Results of type 3 tests of fixed effects for cell for each morphological variable
after 5 years.
Stem Cross-sectional
Area at Groundline
Source of
Stocktype Variation
Bare root
Cell
Gallon
Cell
Tubeling
Cell
L. styraciflua Bare root
Cell
Gallon
Cell
Tubeling
Cell
P. occidentalis Bare root
Cell
Gallon
Cell
Tubeling
Cell
Q. bicolor
Bare root
Cell
Gallon
Cell
Tubeling
Cell
Q. palustris
Bare root
Cell
Gallon
Cell
Tubeling
Cell
Q. phellos
Bare root
Cell
Gallon
Cell
Tubeling
Cell
S. nigra
Bare root
Cell
Gallon
Cell
Tubeling
Cell
Species
B. nigra
F-value
47.83
59.69
53.03
113.55
135.18
41.74
16.88
33.13
13.25
15.97
21.95
27.15
31.54
8.06
0.26
18.9
69.6
0.30
5.56
29.61
10.49
Pr > F
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
0.0010
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
0.0014
0.7742
<0.0001
<0.0001
0.5930
0.2872
<0.0001
0.0005
Height
F-value
144.90
276.75
209.13
319.96
357.83
102.38
24.67
103.85
19.48
33.89
64.39
64.44
34.96
76.54
0.29
70.58
119.7
0.92
24.18
91.29
56.54
Pr > F
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
0.1265
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
0.7524
<0.0001
<0.0001
0.3568
<0.0001
<0.0001
<0.0001
Canopy Diameter
F-value
151.03
383.96
195.23
249.77
507.01
144.86
17.02
174.63
16.17
40.81
81.5
92.74
47.55
53.71
0.61
71.9
112.69
0.95
21.96
83.05
46.03
Pr > F
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
0.0258
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
0.552
<0.0001
<0.0001
0.3493
<0.0001
<0.0001
<0.0001
147
Table 2-6. Number of species/measurement combinations that exhibited a particular
outcome when comparing stocktypes within each cell. The 28 species/measurement
combinations are 7 species paired with each of three morphological measurements
(CSAG, H, CD) and survival (e.g. B. nigra H, S. nigra survival, etc.) that were monitored
over 5 years. The outcomes (>,<,=) result from post-hoc comparisons of stocktypes (BR
vs GAL & TB vs GAL & BR vs TB) for each species/measurement combination. Percent
represents percentage of occurrence of each outcome for each group of stocktype posthoc comparisons (e.g. GAL>BR in 60.7% (17) of the 28 post-hoc BR vs. GAL
comparisons in the Ambient cell). Total represents sum of outcomes across all cells and
percent occurrence of outcomes for groups of post-hoc comparisons. See following tables
representing comparisons of stocktypes for each individual species/measurements
combination.
Outcome
BR < GAL
BR = GAL
BR > GAL
TB < GAL
TB = GAL
TB > GAL
BR = TB
BR > TB
BR < TB
Ambient
17 (60.7%)
11 (39.3%)
0 (0%)
22 (78.6%)
6 (21.4%)
0 (0%)
17 (60.7%)
9 (32.1%)
2 (7.1%)
Saturated
14 (50%)
14 (50%)
0 (0%)
21 (75%)
7 (25%)
0 (0%)
21 (75%)
7 (25%)
0 (0%)
Flooded
25 (89.3%)
2 (7.1%)
1 (3.6%)
26 (92.9%)
2 (7.1%)
0 (0%)
20 (71.4%)
7 (25%)
1 (3.6%)
Total
56 (66.7%)
27 (32.1%)
1 (1.2%)
69 (82.1%)
15 (17.9%)
0 (0%)
58 (69%)
23 (27.4%)
3 (3.6%)
148
Table 2-7. Number of stocktype/measurement combinations that exhibited a particular
outcome when comparing successional groups within each cell. The 12
stocktype/measurement combinations are 3 stocktypes paired with each of three
morphological measurements (CSAG, H, CD) and survival (e.g. BR H, GAL survival,
etc.) that were monitored over 5 years. The outcomes (>,<,=) result from comparison of
successional groups (primary vs. secondary) for each stocktype/measurement
combination. Percent represents percentage of occurrence of each outcome (e.g.
PRI>SEC in 75% (9) of the 12 successional group comparisons in the Ambient cell).
Total represents sum of outcomes across all cells and percent occurrence of outcomes for
successional group comparisons. See following graphs representing comparisons of
successional groups for each individual stocktype/measurement combination.
Outcome Ambient Saturated Flooded
Total
Pri>Sec
9 (75%)
9 (75%) 11 (91.7%) 29 (80.6%)
Pri=Sec 2 (16.7%) 2 (16.7%) 1 (8.3%) 5 (13.9%)
Pri<Sec 1 (8.3%) 1 (8.3%)
0 (0%)
2 (5.6%)
149
Table 2-8. Number of species/stocktype combinations that exhibited a particular outcome
when comparing cells for survival and morphological measurements (CSAG, CD, H)
after 5 years. The 21 species/stocktype combinations are 7 species paired with BR, GAL
and TB stocktypes (e.g. B. nigra BR, S. nigra TB, etc.). However, due to mortality all 21
combinations are not represented for all comparisons. The outcomes (>,<,=) result from
post-hoc comparisons of morphological measurements among cells (AMB vs. SAT,
AMB vs. FLD, SAT vs. FLD) for each species/stocktype combinations. Survival was not
compared statistically and represents absolute differences. Percent represents percentage
of occurrence of each outcome for each group of cell comparisons (e.g. AMB CSAG >
SAT CSAG in 42.9% (9) of the 21 post-hoc AMB vs. SAT comparisons). Total
represents sum of outcomes across survival and morphological measures and percent
occurrence of outcomes for cell groups of post-hoc comparisons. See following tables
representing comparisons of cells for each individual species/stocktype combination.
Outcome
% Survival
AMB>SAT
AMB=SAT
AMB<SAT
AMB>FLD
AMB=FLD
AMB<FLD
SAT>FLD
SAT=FLD
SAT<FLD
6 (28.6%)
1 (4.8%)
14 (66.7%)
16 (76.2%)
0 (0%)
5 (23.8%)
16 (76.2%)
0 (0%)
5 (23.8%)
Stem Cross-sectional
Area at Groundline
9 (42.9%)
12 (57.1%)
0 (0%)
16 (84.2%)
3 (15.8%)
0 (0%)
16 (84.2%)
3 (15.8%)
0 (0%)
Canopy Diameter
Height
Total
9 (42.9%)
12 (57.1%)
0 (0%)
17 (89.5%)
2 (10.5%)
0 (0%)
17 (89.5%)
2 (10.5%)
0 (0%)
15 (71.4%)
6 (28.6%)
0 (0%)
18 (94.7%)
1 (5.3%)
0 (0%)
17 (89.5%)
2 (10.5%)
0 (0%)
39 (46.4%)
31 (36.9%)
14 (16.7%)
67 (85.9%)
6 (7.7%)
5 (6.4%)
66 (84.6%)
7 (9%)
5 (6.4%)
150
Table 2-9. Species exhibiting each outcome within each cell. < and > indicate significant
difference in percent survival (See Figure 2-11). Total represents a count of how many
times each outcome occurred across all cells.
Outcome
BR < GAL
BR = GAL
Ambient
B. nigra, L. styraciflua, P. occidentalis,
Q. palustris, Q. phellos, S. nigra
Q. bicolor
Saturated
B. nigra, L. styraciflua, P. occidentalis,
Q. phellos, S. nigra
Q. bicolor, Q. palustris
Flooded
B. nigra, L. styraciflua, P. occidentalis,
Q. bicolor, Q. palustris, Q. phellos
S. nigra
BR > GAL
TB < GAL
TB = GAL
Total
17
4
0
B. nigra, L. styraciflua, Q. bicolor,
Q. palustris, Q. phellos, S. nigra
B. nigra, L. styraciflua, P. occidentalis,
L. styraciflua, P. occidentalis, Q. bicolor,
Q. bicolor, Q. palustris, Q. phellos, S. nigra
Q. palustris, Q. phellos
P. occidentalis
B. nigra, S. nigra
TB > GAL
18
3
0
BR = TB
B. nigra, Q. phellos
BR > TB
L. styraciflua, Q. bicolor, Q. palustris
BR < TB
P. occidentalis, S. nigra
B. nigra, P. occidentalis, Q. phellos,
S. nigra
L. styraciflua, Q. bicolor, Q. palustris
L. styraciflua, P. occidentalis,
Q. palustris, S. nigra
10
Q. bicolor, Q. phellos
8
B. nigra
3
151
Table 2-10. Species exhibiting each outcome for three stocktypes. < and > indicate
significant difference in percent survival (See Figure 2-11). Total represents a count of
how many times each outcome occurred across all stocktypes.
Outcome
Bare root
Gallon
Tubeling
Total
AMB>SAT
P. occidentalis
B. nigra, Q. palustris, S. nigra
P. occidentalis, S. nigra
6
AMB=SAT
Q. bicolor
AMB<SAT
B. nigra, L. styraciflua, Q. bicolor,
Q. palustris, Q. phellos, S. nigra
AMB>FLD
B. nigra, L. styraciflua, P. occidentalis,
Q. bicolor, Q. palustris, Q. phellos
L. styraciflua, P. occidentalis, Q. phellos
B. nigra, L. styraciflua, P. occidentalis,
Q. bicolor, Q. palustris, Q. phellos
1
B. nigra, L. styraciflua, Q. bicolor, Q. palustris,
Q. phellos
P. occidentalis, Q. bicolor, Q. palustris,
Q. phellos
AMB=FLD
AMB<FLD
SAT>FLD
16
0
S. nigra
B. nigra, L. styraciflua, P. occidentalis,
Q. bicolor, Q. palustris, Q. phellos
S. nigra
B. nigra, L. styraciflua, P. occidentalis,
Q. bicolor, Q. palustris, Q. phellos
B. nigra, L. styraciflua, S. nigra
P. occidentalis, Q. bicolor, Q. palustris,
Q. phellos
SAT=FLD
SAT<FLD
14
5
16
0
S. nigra
S. nigra
B. nigra, L. styraciflua, S. nigra
5
152
Table 2-11. Number of species exhibiting each outcome within each cell. < and >
indicate significant difference in stem cross-sectional area at groundline (CSAG) (See
Figure 2-12). Total represents a count of how many times each scenario occurred across
all cells.
Outcome
Ambient
Saturated
BR < GAL
B. nigra, Q. phellos, S. nigra
B. nigra, Q. palustris, Q. phellos
BR = GAL
L. styraciflua, P. occidentalis, Q. bicolor,
Q. palustris
L. styraciflua, P. occidentalis, Q. bicolor,
S. nigra
Flooded
B. nigra, L. styraciflua, P. occidentalis,
Q. palustris, Q. phellos, S. nigra
Q. bicolor
BR > GAL
TB < GAL
TB = GAL
BR > TB
BR < TB
12
9
0
B. nigra, L. styraciflua, Q. bicolor,
Q. palustris, Q. phellos
P. occidentalis, S. nigra
B. nigra, L. styraciflua, Q. bicolor,
Q. palustris, Q. phellos
B. nigra, L. styraciflua, P. occidentalis, Q.
bicolor, Q. palustris, Q. phellos, S. nigra
P. occidentalis, S. nigra
17
4
TB > GAL
BR = TB
Total
0
B. nigra, P. occidentalis, Q. palustris,
Q. phellos, S. nigra
L. styraciflua, Q. bicolor
B. nigra, L. styraciflua, P. occidentalis,
Q. palustris, Q. phellos, S. nigra
Q. bicolor
B. nigra, L. styraciflua, P. occidentalis,
Q. palustris, Q. phellos, S. nigra
Q. bicolor
17
4
0
153
Table 2-12. Number of species exhibiting each outcome for three stocktypes. < and >
indicate significant difference in stem cross-sectional area at groundline (CSAG) (See
Figure 2-12). Total represents a count of how many times each scenario occurred across
all stocktypes.
Outcome
Bare root
AMB>SAT
L. styraciflua, P. occidentalis, Q. palustris
AMB=SAT
B. nigra, Q. bicolor, Q. phellos, S. nigra
Gallon
B. nigra, P. occidentalis, Q. bicolor,
Q. palustris
L. styraciflua, Q. phellos, S. nigra
Tubeling
Total
P. occidentalis, Q. bicolor
9
B. nigra, L. styraciflua, Q. palustris,
Q. phellos, S. nigra
12
AMB<SAT
0
AMB>FLD
B. nigra, L. styraciflua, Q. bicolor,
Q. palustris, Q. phellos
AMB=FLD
S. nigra
B. nigra, L. styraciflua, P. occidentalis,
Q. bicolor, Q. palustris, Q. phellos, S. nigra
B. nigra, L. styraciflua, Q. bicolor, S. nigra
16
P. occidentalis, Q. palustris
3
AMB<FLD
SAT>FLD
SAT=FLD
SAT<FLD
0
B. nigra, L. styraciflua, Q. bicolor,
Q. palustris, Q. phellos, S. nigra
B. nigra, L. styraciflua, P. occidentalis,
Q. bicolor, Q. phellos, S. nigra
B. nigra, L. styraciflua, Q. bicolor, S. nigra
16
Q. palustris
P. occidentalis, Q. palustris
3
0
154
Table 2-13. Number of species exhibiting each outcome within each cell. < and >
indicate significant difference in crown diameter (See Figure 2-13). Total represents a
count of how many times each outcome occurred across all cells.
Outcome
Ambient
Saturated
Flooded
Total
B. nigra, L. styraciflua, P. occidentalis,
Q. bicolor, Q. palustris, Q. phellos, S. nigra
15
BR < GAL
B. nigra, Q. palustris, Q. phellos, S. nigra
B. nigra, Q. palustris, Q. phellos, S. nigra
BR = GAL
L. styraciflua, P. occidentalis, Q. bicolor
L. styraciflua, P. occidentalis, Q. bicolor
6
BR > GAL
0
TB < GAL
B. nigra, L. styraciflua, Q. bicolor,
Q. palustris, Q. phellos, S. nigra
TB = GAL
P. occidentalis
L. styraciflua, Q. bicolor, Q. palustris,
Q. phellos, S. nigra
B. nigra, L. styraciflua, P. occidentalis,
Q. bicolor, Q. palustris, Q. phellos, S. nigra
B. nigra, P. occidentalis
3
TB > GAL
BR = TB
BR > TB
BR < TB
18
0
B. nigra, P. occidentalis, Q. bicolor,
Q. phellos, S. nigra
L. styraciflua, Q. palustris
B. nigra, L. styraciflua, P. occidentalis,
Q. palustris, Q. phellos, S. nigra
Q. bicolor
B. nigra, L. styraciflua, P. occidentalis,
Q. palustris, Q. phellos, S. nigra
Q. bicolor
17
4
0
155
Table 2-14. Number of species exhibiting each outcome for three stocktypes. < and >
indicate significant difference in crown diameter (See Figure 2-13). Total represents a
count of how many times each outcome occurred across all stocktypes.
Outcome
Bare root
AMB>SAT
L. styraciflua, P. occidentalis, Q. bicolor,
Q. palustris
AMB=SAT
B. nigra, Q. phellos, S. nigra
Gallon
B. nigra, P. occidentalis, Q. bicolor,
Q. palustris
L. styraciflua, Q. phellos, S. nigra
Tubeling
Total
Q. bicolor
9
B. nigra, L. styraciflua, P. occidentalis,
Q. palustris, Q. phellos, S. nigra
AMB<SAT
AMB>FLD
AMB=FLD
0
B. nigra, L. styraciflua, Q. bicolor,
Q. palustris, Q. phellos
B. nigra, L. styraciflua, P. occidentalis,
Q. bicolor, Q. palustris, Q. phellos, S. nigra
S. nigra
B. nigra, L. styraciflua, P. occidentalis,
Q. bicolor, S. nigra
Q. palustris
AMB<FLD
SAT>FLD
SAT=FLD
SAT<FLD
12
17
2
0
B. nigra, L. styraciflua, Q. bicolor,
Q. palustris, Q. phellos, S. nigra
B. nigra, L. styraciflua, P. occidentalis,
Q. bicolor, Q. palustris, Q. phellos, S. nigra
B. nigra, L. styraciflua, Q. bicolor, S. nigra
17
P. occidentalis, Q. palustris
2
0
156
Table 2-15. Number of species exhibiting each outcome within each cell. < and >
indicate significant difference in height (See Figure 2-14). Total represents a count of
how many times each outcome occurred across all cells.
Outcome
Ambient
Saturated
BR < GAL
B. nigra, Q. palustris, Q. phellos, S. nigra
B. nigra, Q. phellos
BR = GAL
L. styraciflua, P. occidentalis, Q. bicolor
L. styraciflua, P. occidentalis, Q. bicolor,
Q. palustris, S. nigra
BR > GAL
TB < GAL
TB = GAL
B. nigra, L. styraciflua, Q. bicolor,
Q. palustris, Q. phellos
P. occidentalis, S. nigra
L. styraciflua, Q. bicolor, Q. palustris,
Q. phellos
Flooded
B. nigra, L. styraciflua, P. occidentalis,
Q. palustris, Q. phellos, S. nigra
BR > TB
BR < TB
12
8
Q. bicolor
1
B. nigra, L. styraciflua, P. occidentalis,
Q. bicolor, Q. palustris, Q. phellos, S. nigra
16
B. nigra, P. occidentalis, S. nigra
5
TB > GAL
BR = TB
Total
0
B. nigra, P. occidentalis, Q. bicolor,
Q. phellos, S. nigra
L. styraciflua, Q. palustris
B. nigra, P. occidentalis, Q. bicolor,
Q. phellos, S. nigra
L. styraciflua, Q. palustris
B. nigra, P. occidentalis, Q. phellos, S. nigra
14
L. styraciflua, Q. bicolor, Q. palustris
7
0
157
Table 2-16. Number of species exhibiting each outcome for three stocktypes. < and >
indicate significant difference in height (See Figure 2-14). Total represents a count of
how many times each outcome occurred across all stocktypes.
Outcome
AMB>SAT
AMB=SAT
Bare root
B. nigra, L. styraciflua, Q. bicolor,
Q. palustris, Q. phellos
P. occidentalis, S. nigra
Gallon
B. nigra, P. occidentalis, Q. bicolor,
Q. palustris, Q. phellos, S. nigra
L. styraciflua
Tubeling
Total
B. nigra, P. occidentalis, Q. bicolor, S. nigra
15
L. styraciflua, Q. palustris, Q. phellos
6
AMB<SAT
AMB>FLD
0
B. nigra, L. styraciflua, Q. bicolor,
Q. palustris, Q. phellos, S. nigra
B. nigra, L. styraciflua, P. occidentalis,
Q. bicolor, Q. palustris, Q. phellos, S. nigra
AMB=FLD
B. nigra, L. styraciflua, P. occidentalis,
Q. bicolor, S. nigra
Q. palustris
AMB<FLD
SAT>FLD
SAT=FLD
SAT<FLD
18
1
0
B. nigra, L. styraciflua, Q. bicolor,
Q. palustris, Q. phellos, S. nigra
B. nigra, L. styraciflua, P. occidentalis,
Q. bicolor, Q. palustris, Q. phellos, S. nigra
B. nigra, L. styraciflua, Q. bicolor, S. nigra
17
P. occidentalis, Q. palustris
2
0
158
Figures
Figure 2-1. A) The experimental site is located in the Mid-Atlantic Region of the United
States, in the Upper Coastal Plain Province of Virginia. B) The site is on an upland
terrace in the Virginia Department of Forestry, New Kent Forestry Center. C) The
hydrologically distinct cells (ambient (AMB), saturated (SAT) and flooded (FLD)) are 49
m x 95 m in size.
159
Figure 2-2. Soil bulk density across experimental site.
160
Figure 2-3. Percentage sand across experimental site
161
Figure 2-4. Percentage silt across experimental site.
162
Figure 2-5. Percentage clay across experimental site
163
Figure 2-6. Soil percentage carbon across experimental site.
164
Figure 2-7. Soil percentage nitrogen across experimental site.
165
Figure 2-8. Soil percentage phosphorus across experimental site.
166
Figure 2-9. Average crown diameter of stocktypes for three cells. Line represents mean
of seven species and ribbons represent 95% confidence interval.
167
Figure 2-10. Average height of successional groups for all stocktypes across cells. Line
represents mean and ribbons represent 95% confidence interval.
168
Figure 2-11. Simple effects model results for survival of stocktypes (lines) among species
(columns) and cells (rows). X-axis represents time since planting. For stocktype
comparisons, no stars represent no significant difference (p-value > 0.05) while *
indicates p-value ≤ 0.05, ** indicates p-value ≤ 0.01 and *** indicates p-value ≤ 0.001.
169
Figure 2-12. Simple effects model results for CSAG of stocktypes (lines) among species
(columns) and cells (rows). X-axis represents time since planting. Ribbons represent
95% confidence interval. Line represents mean. For stocktype comparisons, no stars
represent no significant difference (p-value > 0.05) while * indicates p-value ≤ 0.05, **
indicates p-value ≤ 0.01 and *** indicates p-value ≤ 0.001.
170
Figure 2-13. Simple effects model results for CD of stocktypes (lines) among species
(columns) and cells (rows). X-axis represents time since planting. Ribbons represent 95%
confidence interval. Line represents mean. For stocktype comparisons, no stars represent
no significant difference (p-value > 0.05) while * indicates p-value ≤ 0.05, ** indicates pvalue ≤ 0.01 and *** indicates p-value ≤ 0.001.
171
Figure 2-14. Simple effects model results for H of stocktypes (lines) among species
(columns) and cells (rows). X-axis represents time since planting. Ribbons represent 95%
confidence interval. Line represents mean. For stocktype comparisons, no stars represent
no significant difference (p-value > 0.05) while * indicates p-value ≤ 0.05, ** indicates pvalue ≤ 0.01 and *** indicates p-value ≤ 0.001.
172
Figure 2-15. Simple effects model results for survival (time until death) of successional
group (lines) among cells (columns) and stocktype (rows). Ribbons represent 95%
confidence interval. Line represents mean. For successional group comparisons, no stars
represent no significant difference (p-value > 0.05) while * indicates p-value ≤ 0.05, **
indicates p-value ≤ 0.01 and *** indicates p-value ≤ 0.001.
173
Figure 2-16. Simple effects model results for CSAG of successional group (lines) among
cells (columns) and stocktype (rows). Ribbons represent 95% confidence interval. Line
represents mean. For successional group comparisons, no stars represent no significant
difference (p-value > 0.05) while * indicates p-value ≤ 0.05, ** indicates p-value ≤ 0.01
and *** indicates p-value ≤ 0.001.
174
Figure 2-17. Simple effects model results for CD of successional group (lines) among
cells (columns) and stocktype (rows). Ribbons represent 95% confidence interval. Line
represents mean. For successional group comparisons, no stars represent no significant
difference (p-value > 0.05) while * indicates p-value ≤ 0.05, ** indicates p-value ≤ 0.01
and *** indicates p-value ≤ 0.001.
175
Figure 2-18. Simple effects model results for H of successional group (lines) among cells
(columns) and stocktype (rows). Ribbons represent 95% confidence interval. Line
represents mean. For successional group comparisons, no stars represent no significant
difference (p-value > 0.05) while * indicates p-value ≤ 0.05, ** indicates p-value ≤ 0.01
and *** indicates p-value ≤ 0.001.
176
CHAPTER 3: WOODY BIOMASS DEVELOPMENT OF SEVEN MIDATLANTIC SPECIES GROWN UNDER THREE HYDROLOGIC CONDITIONS
OVER 6 YEARS
177
Abstract
The primary goal of forested wetland creation and restoration is to replace or return
ecosystem structure and functions to the landscape. Production of plant biomass, of
which carbon is a large component, is an important ecosystem function that will develop
in successfully created and restored forested wetlands. Quantifying accumulation of
above and belowground biomass of planted trees is important in understanding this
development and may aid in quantification of global chemical cycles . Destructive
harvests and tissue elemental analysis of saplings (n=567) planted across a hydrologic
gradient were used to develop biomass estimation models for seven species common to
the Mid-Atlantic region of the United States. The model established that stem crosssectional diameter at groundline was an adequate predictor of total biomass. The model
was then applied to all living trees (n=1,258) to evaluate species and stocktype
performance over 6 years as well as differences in accumulation across a hydrologic
gradient. Early colonizing species accumulated more biomass than late succession
species. Larger stocktypes (1-gallon container) accumulated more biomass than smaller
stocktypes (bare root and tubeling) in areas with greater hydrologic stress and
accumulated biomass decreased with increasing hydrologic stress regardless of stocktype.
Aboveground dry woody tissue percentage carbon ranged from 44.2 to 48.5%; however,
the proportion of carbon in the woody tissues was not significantly different among the
seven species. Biomass in conjunction with elemental concentrations measured in this
study can be used to evaluate the development of restored or created forested wetlands
and to determine sapling contributions to these global chemical cycles.
178
Introduction
Production and accumulation (storage) of plant biomass, of which carbon is a
large component, is an important ecosystem function that develops in successfully
created and restored forested wetlands. In early successional created and restored forested
wetlands the majority of plant biomass is produced by herbaceous vegetation (Atkinson
et al. 2005, DeBerry & Perry 2012). However, as created and restored forested wetlands
develop, the production and accumulation of biomass shifts to perennial woody
vegetation (woody vines, shrubs, and trees) (Noon 1996, Odland 1997, Battaglia et al.
2002, DeBerry & Perry 2012, Mitsch et al. 2012), where biomass can remain long-term
in the boles, stems and roots. Therefore, the intended outcome of the creation or
restoration process of forested wetlands is that the majority of biomass/carbon
accumulated will be produced by long lived woody species, particularly trees.
If natural colonization by trees is insufficient in a created or restored forested
wetland, planting seeds, seedlings and/or saplings of desired woody species becomes
necessary (Clewell & Lea 1989, Hudson 2010). The nursery industry uses an assortment
of propagation techniques to produce a wide variety of tree seedling and sapling
stocktypes (e.g. bare root seedlings of various ages, tubelings or plugs, containerized,
balled and burlapped, live stakes, etc.). Very few studies have investigated how species
and stocktype selection influences planted tree survival and growth in created/restored
forested wetlands (Denton 1990); however, several studies have reported poor survival
and growth of planted trees in created/restored wetlands (Morgan & Roberts 2003,
Sharitz et al. 2006, Matthews & Endress 2008). Therefore, quantifying standing woody
biomass and production of planted trees is important in understanding the development of
179
ecosystem structure and functions in created and restored forested wetlands. Additionally,
accurate quantification of tree biomass, particularly saplings, is an important step in
understanding carbon dynamics across a range of forested ecosystems (including
wetlands) as saplings play a large, but yet unquantified, role in the global carbon cycle
(Temesgen et al. 2015).
Standing biomass of an individual tree is typically estimated by using
mathematical relationships between biomass and one or more morphological woody
vegetation characteristics, such as stem diameter and/or height (e.g. Jenkins et al. 2003).
This method is commonly referred to as dimensional analysis (Whittaker & Woodwell
1968) and relies on consistency in the correlations between the changes in relative
dimensions of parts of an organism with changes in overall size, referred to as allometry
(Gayon 2000, Stevens 2009). While allometric relationships may be expressed through
diverse mathematical formulations (Gould 1966), the mathematical relationships
describing the allometry of various characteristics of many organisms (including trees)
often conform to power laws (Niklas, 2004, Stevens 2009). Equations that relate biomass
and a morphological characteristic were developed and used for this study. These models
will be referred to as biomass estimation models (BEM) in this study, but have also been
referred to as allometric equations by others (Sileshi 2014).
Development of BEMs for seedlings and saplings is valuable because seedlings
and saplings can be a major component of understory (Gemborys 1974) and canopy gaps
of mature forests (Ehrenfeld 1980) and can be significant components of the vegetative
structure in early successional stages of many ecosystems, including abandoned
agricultural fields (Monette and Ware 1983) and recently restored (Hudson 2010) and
180
created wetlands (DeBerry & Perry 2012). Additionally, BEMs for seedlings and saplings
allow more accurate prediction of biomass during tree development, which may allow for
improved characterization of factors that affect development.
The purpose of this study was to construct BEMs relating total biomass (leafless
woody above and belowground coarse root dry biomass) to sapling stem cross-sectional
diameter at groundline for seven native Mid-Atlantic woody wetland species, and to
determine how stocktype, soil and hydrologic conditions influenced the amount of
biomass accumulated 6 years following planting. We anticipate that the results from this
study may enhance the probability of successfully returning ecosystem structure and
functions to created and restored forested wetlands through identification of appropriate
species and stocktypes for planting and by determining sapling biomass and carbon
accumulation across a hydrologic gradient.
Methods
Methods used to establish and measure the survival and morphology of saplings
planted in the experimental site are the same as Chapter 2.
Biomass Sampling
Since the species specific relationships between morphology and biomass change
during ontogeny and often in response to environmental conditions, aboveground
biomass (AGB) and belowground biomass (BGB) samples were taken in the winter of
2010-2011; AGB was also sampled from additional trees in late winter 2014. The
samples included trees from all planting times, cells and stocktypes to incorporate the
181
maximum variation in morphologies and biomass. A random subsample of saplings
(n=346) were removed in the winter of 2010-2011 to measure leafless AGB and coarse
root BGB, and a random subsample of trees (n=221) were removed in late winter of 2014
to measure leafless woody AGB. Sample size and average morphology for each species
and stocktype are presented in Table 3-1. In order to extract roots from the soil matrix a
variety of methods were used based on the size of the tree and planting location. Trees
removed from the FLD cell were removed by hand or with trowels and pitchforks. Soil
remaining on the roots was washed onsite prior to drying. Small trees (<0.5 m tall) were
removed using similar methods in AMB and SAT. For trees taller than 0.5 m,
approximately 0.1 m3 of soil was excavated around the main stem using a tree spade
(Dutchman Model 240o, Ontario, Canada) mounted on a skid steer loader (Bobcat S160,
Seoul, South Korea). Following excavation the soil was removed from the roots by hand
and with trowels and pitchforks. Any roots that were not excavated using the tree spade
were subsequently removed from the soil matrix by hand and with shovels, trowels and
pitchforks. All spreading and deep roots were followed to their terminus. Difficult to
remove soil around the roots was washed onsite prior to drying. While attempts were
made to capture all roots, this method excluded most fine roots smaller than 2 mm
diameter.
The complete above and belowground portions of the trees were separated and
placed in individual paper bags. Sampling occurred after leaf senescence and leaf
biomass was not measured; therefore BGB refers only to coarse roots and AGB refers to
stems and branches. All trees were solar dried on-site at approximately 50°C in
repurposed greenhouses until constant weight was obtained. The trees were weighed at
182
the end of the summer in 2011 and 2014 following complete drying. The root-to-shoot
ratio (r:s) was calculated for trees harvested in winter 2010-2011 where AGB and BGB
was harvested.
Subsamples of trunk and branches AGB (including bark) were collected from
trees that were harvested in early spring 2014 (n=103). Each subsample represented a
unique stem diameter class determined visually (e.g. If the main trunk was 5 cm
diameter, then a subsample was taken from the 4-5 cm diameter stem, 2-3 cm diameter
stem, 1 cm diameter and less than 1 cm diameter stem). Subsamples were manually
reduced in size using a small sledge hammer and then mechanically ground using a
Thomas Wiley Mill Model 4 and Mini-Mill (Thomas Scientific, Swedesboro, NJ). Using
a PE2400 CHNS/O Elemental Analyzer (Perkin Elmer, Massachusetts, USA) duplicate
dry ground samples were analyzed for percentage carbon and nitrogen elemental
concentrations, which were used to determine the carbon to nitrogen ratio (C:N).
Biomass Estimation Model Development
The relationship between the above (AGB) and belowground biomass (BGB) was
determined for each species from the 2011 samples (AGB and BGB sampled). The 2011
samples were pooled from all three cells due to lack of differences in model coefficients
when the AGB to BGB relationship was modeled for each cell separately. The following
non-linear equation below was used to determine the relationship between AGB and
BGB:
Y=aXb + ε
(Equation 1)
Y = BGB (kg)
183
X = AGB (kg)
a = model estimated intercept
b = model estimated scaling coefficient
ε = residual error term.
Using residual diagnostic plots, a heteroscedastic error structure (variance of BGB
increased with greater AGB) was observed. Because of heterogeneous variance, the
relationship between AGB and BGB was modeled using a generalized nonlinear leastsquares regression (Pinheiro & Bates 2000). The variance structure was modeled using
the power of the covariate (Packard 2014, Zuur et al. 2009) since this resulted in
homogenization of residuals and lowest Akaike information criterion (AIC) values.
Logarithmic transformations were not used because of criticisms associated with these
methods and the heterogeneous error structure associated with this dataset (Packard
2014). The resulting model was used to estimate the BGB of the 2014 samples (where
only AGB was harvested).
The same non-linear equation was used to determine the relationship between
total woody biomass and equivalent stem cross-sectional diameter at groundline (ESD):
Y=aXb + ε
(Equation 2)
Y= total woody biomass (AGB+BGB)
X = ESD
a = model estimated intercept
b = model estimated scaling coefficient
ε = residual error term.
184
All samples from 2011 and 2014 (n=567) were used to develop this relationship.
Using residual diagnostic plots, a heteroscedastic error structure (variance of biomass
increased with greater basal diameter) was observed. As above, generalized nonlinear
least-squares regression was used to fit the original untransformed data and the variance
structure was modeled using the power of the covariate structure (Packard 2014, Pinheiro
& Bates 2000, Zuur et al. 2009). Tree height and crown diameter were not included in the
BEM due to multicollinearity among the predictors, despite recent studies that suggests
they could be included (Picard et al. 2015).
Statistical Analysis
Pairwise t-test were used to compare root-to-shoot (r:s) ratios of trees that were
harvested in 2010-2011. Variance was estimated separately for each group, the Welch
modification to degrees of freedom was used and Holm’s p-value adjustment was used to
control the family-wise error rate. Percentage carbon (C) and nitrogen (N) and C:N ratio
of AGB subsamples from 2014 harvested trees were compared among species (and
species groups) within and across cells using the above methods.
Species specific BEMs were used to determine the biomass of living trees directly
following planting and after 6 years. Analysis was limited to trees that were planted in
2009. Using the same methods as above, differences in initial total woody biomass
(BGB+AGB directly following outplanting) and woody biomass after 6 years were
determined for stocktypes, species and species groups within and across cells. All alpha
values were set at 0.05. Data preparation and analysis were completed in R version 3.2.1
(R Core Team 2014).
185
Results
Root-to-Shoot Ratio
Sampled trees (n=567) ranged in equivalent stem cross-sectional diameter at
groundline (ESD) from 0.2 cm to 34.3 cm and heights from 4 cm to 1155 cm. The r:s
ratio of all trees harvested in 2010-2011 (aboveground leafless woody biomass and
coarse roots harvested) ranged from 0.0068 to 26.7 with an average of 1.82 (Standard
error (SE) = 0.117). There were significant differences in r:s among the seven species
within each cell (Table 3-2). Life history categories showed that secondary species had
greater r:s (mean of 2.26) compared to primary species (1.46) across all cells. When
comparing species across cells S. nigra had significantly higher r:s in FLD compared to
AMB and SAT and Q. phellos had significantly lower r:s in FLD than in SAT (Table
3-2).
Tissue Elemental Concentrations
There were no differences in percentage C among species within AMB, SAT and
FLD cells (Table 3-3). However, when combining all of the species into primary and
secondary groups, primary species had significantly greater percentage C than secondary
species. Percentage N of species was found to be significantly different within SAT and
FLD. Across the cells, only Q. phellos had greater N content in AMB than FLD. These
differences in species N content lead to the secondary successional species group having
significantly greater N (1.38% SE = 0.023) than primary successional species (1.27% SE
= 0.021) (Table 3-3). There were no differences in C:N ratio among species in AMB cell.
Across cells only Q. phellos had significantly greater C:N in FLD than AMB (Table 3-3).
186
Overall, the primary successional species group had significantly greater C:N ratio than
the secondary successional species group (Table 3-3).
Biomass Estimation Model Development
Using data from the 346 samples collected in 2011, a relationship between AGB
and BGB (coarse roots only) for each species was determined using Eq. 1. For each
species, samples were pooled across cells because the model estimated coefficients (a and
b) were not different when models were developed for each cell.
The resulting model fit the data well for each species based on the standard error
of the regression (Table 3-4). Primary species ranged from 0.544 (S. nigra) to 1.334 (L.
styraciflua) for the model derived intercept (a) and from 0.7394 (S. nigra) to 0.975 (P.
occidentalis) for the model derived exponent. The model derived intercept (a) for
secondary species ranged from 0.595 (Q. phellos) to 1.921 (Q. bicolor) and the model
derived exponent (b) ranged from 0.732 (Q. phellos) to 0.942 (Q. bicolor). The effect of
these ranges for the exponent (b) is evident when comparing the results of primary
species (Figure 3-1) and secondary species (Figure 3-2).
Using data from all 567 sampled trees the relationship between ESD and total
woody biomass was determined by using Eq. 2 for each species (Table 3-5). BGB of
trees harvested in 2014 (AGB only) was estimated using the model. The standard error of
the regression suggested that the models described the data well for all species (ranging
from 0.522 for P. occidentalis to 1.891 for Q. phellos) (Table 3-5). The model derived
intercept (a) ranged from 0.028 (P. occidentalis) to 0.032 (L. styraciflua) for primary
species and from 0.047 (Q. palustris) to 0.055 (Q. phellos) for secondary species. The
187
model derived exponent or scaling coefficient (b) ranged from 2.442 (B. nigra) to 2.788
(P. occidentalis) for primary species and the secondary species range was 2.455 (Q.
phellos) to 2.735 (Q. bicolor). The effect of these exponent (b) ranges on the amount of
biomass accumulated per centimeter stem diameter is evident when comparing the results
among primary species (Figure 3-3) and secondary species (Figure 3-4).
Biomass Estimation Model Implementation
Initial Biomass
Initial total biomass (model derived BGB+AGB directly following outplanting) of
all trees ranged from 0.0003 kg to 0.544 kg. Average initial biomass of each
species/stocktype combination is presented in Table 3-6. For all species there was no
difference in initial biomass between the BR and TB stocktypes. The GAL stocktype had
greater initial biomass than TB stocktype for all species except L. styraciflua and Q.
bicolor. Additionally, the GAL stocktype had greater initial biomass than BR for all
species except Q. bicolor (Table 3-6).
Accumulated Biomass
Stocktype Comparison
In the AMB cell the GAL had significantly greater final biomass than 3 species
planted as BR and 1 species planted as TB. Platanus occidentalis was the only species for
which both BR and TB final biomass was significantly greater than the GAL final
biomass in AMB. For 4 species, there was no difference in final biomass between the BR
188
and TB stocktypes, while the BR final biomass was significantly greater than TB final
biomass for 2 species (Table 3-7).
For 4 species in SAT, there was no difference in final biomass between the GAL
and BR and for 2 species there was no difference in final biomass between the GAL and
TB. However, the GAL had significantly greater final biomass than 2 species planted as
BR and 4 species planted as TB. Platanus occidentalis was the only species for which the
TB stocktype had significantly greater final biomass than the GAL. For 6 species there
was no difference in final biomass between the BR and TB (All BR P. occidentalis died
or were removed) (Table 3-7).
All species in FLD had no difference in final biomass among stocktypes except L.
styraciflua for which the GAL had significantly more final biomass than the TB. The lack
of statistically significant differences among stocktypes in FLD is a result of poor
survival and the stressful environmental conditions (Table 3-7).
Species Comparison
Average final biomass was calculated for each species incorporating all
stocktypes in order to make inferences about differences in species and cells, while
incorporating the maximum variation for each species (Table 3-8).
In the AMB there was no difference in final biomass among the three Quercus
species, and they had significantly less final biomass than P. occidentalis, B. nigra, L.
styraciflua, and S. nigra (listed in descending final biomass) (Table 3-8). In SAT, there
was no difference in final biomass among B. nigra, L. styraciflua, P. occidentalis and S.
nigra, and all 4 of those species had significantly greater biomass than the Quercus
189
species. Among the Quercus species, Q. phellos had significantly greater final biomass
than Q. palustris (Table 3-8). In FLD, only S. nigra had significantly greater final
biomass than L. styraciflua, P. occidentalis, and Q. bicolor (Table 3-8).
Species Group Comparison
Species groups (see Methods for description of primary and secondary species
groups) were compared within each cell to make broad inferences about the differences
between these groups. Primary species had significantly more final biomass than
secondary species in each cell (Table 3-9). Additionally, both primary and secondary
species groups had significantly more final biomass in AMB than SAT, which had
significantly more final biomass than FLD (Table 3-9).
Cell Comparison
The average final biomass for each species (incorporating each stocktype) was
compared among the cells in order to make inferences about how the environmental
conditions influenced the biomass accumulated for each species (Table 3-8). B. nigra, P.
occidentalis, Q. bicolor, Q. palustris and Q. phellos had significantly greater final
biomass in AMB than in SAT. All species had significantly greater biomass in AMB and
SAT than FLD.
190
Discussion
Root-to-Shoot Ratio
The most commonly reported literature value for r:s ratio in forests containing
many species, across latitudes, elevation and soil types is approximately 0.2 (Cairns et al.
1997, Mokany et al. 2006, Luo et al. 2012, Hui et al. 2014). The majorities of these ratios
were determined for large trees and were measured using partial root samples generalized
to forest stands. The r:s ratio in this study averaged 1.8 across all fully harvested
individual saplings. Similar investigations focusing on seedlings and saplings suggest that
they may have higher r:s ratios than large trees. Day (1987) showed that seedlings of
Acer rubrum grown in nutrient enrichment experiments have initially high (up to 3.22) r:s
ratio. Hui et al. 2014 showed that r:s ratios decrease with increasing diameter at breast
height (dbh) and Mokany et al. (2006) showed that stands with lower AGB had greater r:s
ratios. Luo et al. (2012) found that stands less than 20 years old had higher r:s ratios than
those greater than or equal to 20 years old. The r:s ratio from this study show that
saplings as opposed to larger trees, allocate more biomass belowground to coarse roots
than aboveground following outplanting. This suggests that belowground biomass is
important to consider when evaluating the biomass and carbon accumulation ecosystem
functions in restored/created wetlands.
The lower r:s ratio of primary successional species (excluding P. occidentalis, see
below) compared to secondary successional species suggests that this group of species
has the ability to rapidly accumulate belowground biomass and then shift allocation of
resources aboveground. These species are appropriate choices for restoration/creation of
wetlands for this reason among others.
191
While not statistically significant, P. occidentalis r:s declined from an average of
2.4 in AMB to 0.95 in SAT to 0.88 in FLD. This decline was similar to the decrease in
survival P. occidentalis exhibited across the cells. This suggests that P. occidentalis may
not be able to allocate resources effectively belowground under wetter hydrologic
conditions and may not be appropriate for planting in restored/created wetlands. The
opposite response occurred for S. nigra, where the r:s and survival increased moving
from AMB to SAT to FLD suggesting that S. nigra is able to effectively allocate
resources belowground in stressful hydrologic conditions and may be an appropriate
choice for planting in restored/created wetlands. The r:s increase of S. nigra in FLD is
contradictory to the decrease in r:s of T. distichum in continuously flooded treatments
found by Megonigal and Day (1992), but may be a response to the lower nitrogen
concentrations in the FLD. Several studies have found that r:s ratios are higher for a
variety of species grown in low nutrient conditions (Dickson and Broyer 1972, Birk and
Vitousek 1986, Jerbi et al. 2015).
Elemental Concentration
The woody tissue of primary species group had higher C:N (37.3) than secondary
species (34.1) mainly because of higher N concentrations in secondary species. Low
variation of C and N (and C:N ratios) within and across cells suggests that the species
used in this study did not alter the accumulation of these elements in response to the sites
hydrologic and soil elemental concentrations. The soil of FLD had lower nitrogen
concentrations, resulting from excavation of the topsoil and possibly from increased
denitrification from reduced soil oxygen concentrations. Other studies have shown that
192
soil N concentrations have a positive relationship with tissue N concentrations. Day
(1987) found that Acer rubrum grown under a variety of hydrologic conditions had
greater leaf N concentrations when N was added. The lack of differences in wood C and
N concentrations across the site suggest that these species are maintaining the balance
among these elements despite soil conditions, possibly through reduction in biomass
accumulation.
The wood carbon content from this study (average 46.6% of wood biomass) was
consistent with the range of wood carbon content found with temperate/boreal species
(43.4-55.6%) (Thomas & Martin 2012). The wood carbon content from this study can be
used to determine carbon accumulation of saplings in restored/created wetlands in this
region and demonstrates how important saplings are to returning this ecosystem function
in these wetlands.
Martin et al. (1998) investigated N content of different woody tissues in 10
Appalachian tree species. All 10 species had very low N concentration in the stem
heartwood, sapwood and branches (<4 mg/g) for all species. The low concentrations
found in the present study (average 1.3% of wood biomass) suggest that the woody tissue
of saplings is a small contributor to removal of N from the environment. However, the
long term storage of N in woody tissues is important across longer time scales and as
trees accumulate more biomass.
Biomass Estimation Model Development
The power-law relationship between BGB and ABG provide a means for nondestructive estimation of coarse root BGB of the seven species used in our study that is
193
more robust than r:s ratios. This relationship was developed for each species using
samples pooled from the three cells, because cell specific model derived coefficients did
not differ (confidence intervals overlapped). This may have resulted from the low number
of samples taken from each cell but is also evidenced in the lack of species statistical
differences in r:s ratios among cells.
Scaling coefficients from this relationship provide insights into how biomass
allocation changes as biomass increases. A scaling coefficient of less than 1 (as for most
species in this study) suggests that the r:s ratio decreases as saplings accumulate AGB.
Departure from an isometric relationship between AGB and BGB have been reported
across a range of habitats, species, spatial scales and time scales (Cheng & Niklas 2007,
Niklas 2005, Niklas 2004, Enquist & Niklas 2002, Yang & Luo 2011). Hui et al. 2014
showed that as dbh of trees increased, the r:s ratio decreased and the scaling coefficient
increased to approximately 1. Results from the current study suggest that inclusion of
saplings when determining biomass and carbon accumulation and allocation in
restored/created wetlands is important because they may be allocating more resources
belowground before shifting allocation of resources aboveground. L. styraciflua, P.
occidentalis and Q. bicolor have scaling coefficients approaching 1 suggesting they are
able to quickly allocate resources belowground following outplanting and then shift
allocation aboveground. These species may be appropriate for returning these important
ecosystem functions to restored/created wetlands.
The BEM developed in this study describing the relationship between stem
diameter and total biomass provides a non-destructive estimate of sapling biomass that is
useful for determining the development of this ecosystem function in restored/created
194
wetlands and other forested systems (riparian buffers, uplands, natural forested wetlands
etc.). Additionally, it provides an additional way to evaluate planted sapling performance.
Current BEMs for the species used in this study (see Jenkins et al. 2003, 2004, Chojnacky
et al. 2014) were constructed for large trees (dbh >2.5 cm) and only model AGB. The few
studies developing BEMs for saplings focused at the genus taxonomic level for Salix,
Betula and Quercus and most only sampled AGB (Tefler 1969, Roussopoulos & Loomis
1979, Smith & Brand 1983, Williams & McClenahem 1984). Comparisons with these
studies are further compounded by the methodological differences in stem measurements
(several sampled at 15 cm above soil surface) and model fitting (log base 10
transformations of biomass). Despite these differences, the estimated scaling coefficients
from this study (Table 3-5) were marginally higher than literature values for these
species. The inclusion of BGB in the present study is a potential reason for this but the
BEMs developed in this fashion provide a more robust model for estimating total
biomass for these species and for evaluating this ecosystem function in restored/created
wetlands.
Accumulated Biomass
Stocktype comparison
In FLD, the choice of stocktype for all species (except L. styraciflua) did not
affect the biomass accumulated after 6 years. This suggests that larger stocktypes are not
able to return this ecosystem function in very stressful environmental conditions;
however, it should be noted that the GAL stocktype did have increased survival
compared to the BR and TB.
195
Within SAT, the GAL stocktype had significantly greater biomass after 6 years
than the TB for B. nigra, L. styraciflua, Q. bicolor, and Q. phellos. This suggests that the
larger GAL stocktype may be a better selection than the TB in conditions similar to SAT.
In SAT the GAL exceeded the final biomass of BR for only B. nigra and Q. phellos. This
suggests that in these conditions the BR stocktype can accumulate a similar amount of
biomass over 6 years as the GAL for most species. In the AMB, there were more
differences among the stocktypes than in SAT and FLD. This suggests that when
environmental conditions are less stressful, smaller stocktypes are able to accumulate as
much or greater biomass than larger stocktypes.
Species Comparison
In addition to the stocktype, species selection was shown to influence biomass
accumulated after 6 years. Within FLD, S. nigra accounted for 55.6% of the total amount
of biomass accumulated and is the only species that did not experience substantial
mortality. This suggests that in very stressful created and restored wetland conditions S.
nigra may be a more appropriate choice to replace or return woody biomass to the
system. The only other species to have moderate success in FLD was B. nigra.
Species Group Comparison
In the SAT cell, primary species individually and as a group accumulated more
biomass than secondary species. Of the total biomass accumulated, the primary species
group accounted for an average of 92.4% and the secondary successional species group
accounted for 7.6% across all cells. This suggests that primary successional species as a
196
group are a better choice when the goal of creation and restoration is to replace lost
biomass. However, species and functional diversity of planting material should always be
considered.
Cell Comparison
Total biomass accumulated by all living trees after 6 years in AMB, SAT and
FLD was 20,077.2 kg, 9035.2 kg, and 195.8 kg respectively. Additionally, primary and
secondary species groups both had significantly more biomass in AMB than SAT and
FLD and significantly more biomass in SAT than FLD. The reduction in biomass in SAT
compared to AMB resulted from the increased hydrologic stress (saturation within the
root zone). The substantial reduction in biomass production in FLD resulted from
flooding above the root collar, reduced soil nitrogen and phosphorus concentrations,
increased clay concentrations, herbaceous competition and poor survival of planted trees.
Reduction in tree survival and biomass accumulation can be attributed to
prolonged saturated or flooded soil conditions that remove the plant available oxygen
from the soil pore space. The reduction in oxygen leads to a lack of aerobic respiration in
roots, which decreases the energy available for trees to maintain functions of existing
tissues (Hale & Orcutt 1987, Brady & Weil 2002). Many growth chamber, greenhouse,
mesocosm and field experiments have investigated the effect of hydrology on a multitude
of responses across many species of trees. While species specific responses may vary
(e.g. Taxodium distichum, mangroves) most species exhibit decreased survival and
biomass accumulation when grown under prolonged inundation. Niswander & Mitch
(1995) planted ten tree species (three of which were used in this study, B. nigra, L.
197
styraciflua, Q. palustris) across a hydrologic gradient in a created wetland. Similar to the
results in this study, they found that trees planted in shallow water died or were severely
stressed, and that trees planted in the wet meadow portion were able to survive and grow,
while trees planted in the upland section were the largest and had the densest foliage.
Pennington & Walters (2006) investigated growth and survival four species (two of
which were used in this study, Q. palustris and Q. bicolor) planted in three hydrologic
zones (wetland, transition, upland) of created perched wetlands. Again, similar to the
present study, trees grown in the transitions zone (high soil water availability with
oxidized root zone) had greater height growth and survival after 5 years than those
planted in the wetland zone (reduced oxygen in the root zone). This suggests that the
environmental conditions during and following restoration/creation of wetlands can
greatly influence the return of ecosystem functions such as biomass accumulation.
Conclusions
Models relating total biomass and stem cross-sectional diameter at groundline are
needed for saplings since most prior models excluded saplings and belowground biomass.
Models developed for larger diameter trees underestimate biomass of smaller trees even
with inclusion of sapling adjustments (Nelson et al. 2014). Additionally, accurate BEMs
are needed to make predictions about carbon storage and exchange in forest ecosystems.
The investigation of r:s ratios also provide important information about how saplings
transition from allocating biomass belowground to aboveground as they grow.
The results of investigating the biomass accumulated after 6 years for saplings
planted under different environmental conditions provides important information for
198
those attempting creation and restoration of any forested system, particularly for those
involving wetland conditions. When planting in stressful hydrologic, soil, and
competitive conditions, based on the results from FLD, S. nigra appears to be the best
choice among these species for returning woody biomass (followed by B. nigra).
However, it is always important to plant woody species that are found naturally in the
type of wetland that is being restored. Meaning, S. nigra and B. nigra would not be
appropriate choices for planting in restored wetlands types in which they do not naturally
occur.
In the FLD, the choice of stocktype influenced the survival but did not appear to
influence the biomass accumulated after 6 years as all stocktypes were heavily stressed.
When planting in less stressful wetland environmental conditions (moderate hydrologic
and competitive stress), the primary species used in this study are a better choice to return
woody biomass than secondary species. Secondary species (Quercus spp.) may be
important for additional ecological functions (such as acorn production) but may not be
an appropriate planting choice when rapid accumulation of biomass is the goal. The
choice of stocktype between GAL and TB appears to be important in areas of moderate
environmental stress, where larger stocktypes are a better choice for replacing biomass,
even after 6 years. However, in these conditions the BR was able to accumulate similar
amounts of biomass as the GAL, suggesting that it may be an acceptable alternative to the
larger more expensive stocktype. In areas with low environmental stress, the smaller
stocktype are able to equal and exceed the biomass accumulated by the larger stocktype.
In conclusion, the recommended species and stocktypes will help improve the
practice of wetland restoration and creation by enhancing the return of lost ecosystem
199
structure and functions which can be better evaluated using these new sapling biomass
estimation models.
200
Literature Cited
Atkinson RB, Perry J, Cairns JJ (2005) Vegetation communities of 20-year-old created
depressional wetlands. Wetlands Ecology and Management 13: 469-478
Bailey D, Perry J, Daniels, W (2007) Vegetation dynamics in response to organic matter
loading rates in a created freshwater wetland in southeastern Virginia. Wetlands
27: 936-950
Battaglia L, Minchin P, Pritchett D (2002) Sixteen years of old-field succession and
reestablishment of a bottomland hardwood forest in the Lower Mississippi
Alluvial Valley. Wetlands 22:1-17
Birk EM, Vitousek PM (1986) Nitrogen availability and nitrogen use efficiency in
loblolly pine stands. Ecology 64:69-79
Brady NC, Weil RR (2002) The Nature and Properties of Soil. Prentice Hall, Upper
Saddle River, New Jersey
Cairns M, Brown S, Helmer E, Baumgardner G (1997) Root biomass allocation in the
world's upland forests. Oecologia 111: 1-11
Cheng D-L, Niklas KJ (2007) Above-and below-ground biomass relationships across
1534 forested communities. Annals of Botany 99:95-102
Chojnacky DC, Heath LS, Jenkins, JC (2014) Updated generalized biomass equations for
North American tree species. Forestry 87:129-151
Clewell AF, Lea R (1989) Creation and restoration of forested wetland vegetation in the
southeastern United States. Pages 199-237 In: Kusler JA, Kentula ME (eds)
Wetland creation and restoration: The status of the science. United States
Department of Commerce, National Technical Information Service, Springfield,
Virginia.
Day FP (1987) Effects of flooding and nutrient enrichment on biomass allocation in Acer
rubrum seedlings. American Journal of Botany 74:1541-1554
DeBerry D, Perry J (2012) Vegetation dynamics across a chronosequence of created
wetland sites in Virginia, USA. Wetlands Ecology and Management 20:521-537
Denton S (1990) Growth rates, morphometrics and planting recommendations for cypress
trees at forested mitigation sites. Pages 24-32 In: Webb FJ (ed) Proceedings of the
201
17th Annual Conference on Wetlands Restoration and Creation, 10-11 May 1990.
Tampa, FL
Dickson, R and Broyer, T (1972) Effects of aeration, water supply, and nitrogen source
on growth and development of tupelo gum and bald cypress. Ecology 53:626-634
Ehrenfeld JG (1980) Understory response to canopy gaps of varying size in a mature oak
forest. Bulletin of the Torrey Botanical Club 107:29-41
Enquist BJ, Niklas KJ (2002) Global allocation rules for patterns of biomass partitioning
in seed plants. Science 295:1517-1520
Gayon J (2000) History of the concept of allometry. American Zoologist 40: 748-758
Gemborys S (1974) The structure of hardwood forest ecosystems of Prince Edward
County. Virginia Ecology 55:614-621
Gould SJ (1966) Allometry and size in ontogeny and phylogeny. Biological Reviews 41:
587-640
Hale MG, Orcutt, DM (1987) The physiology of plants under stress. John Wiley & Sons,
New York, New York
Hudson HW (2010) The effect of adjacent forests on colonizing tree density in restored
wetland compensation sites in Virginia. Master's thesis. Christopher Newport
University
Hui D, Wang J, Shen W, Le X, Ganter P, Ren H (2014) Near isometric biomass
partitioning in forest ecosystems of China. PloS one 9:e86550
Jenkins J, Chojnacky D, Heath L, Birdsey R (2003) National-scale biomass estimators for
United States tree species. Forest Science 49:12-35
Jenkins J, Chojnacky D, Heath L, Birdsey R (2004) Comprehensive database of diameter
based biomass regressions for North American Tree species. NE-319. United
States Department of Agriculture, Forest Service, Northeastern Research Station,
Burlington, Vermont
Jerbi A, Nissim WG, Fluet R, Labrecque M. (2015). Willow root development and
morphology changes under different irrigation and fertilization regimes in a
vegetation filter. BioEnergy Research 8:775-787.
202
Luo Y, Wang X, Zhang X, Booth TH, Lu F (2012) Root:shoot ratios across China’s
forests: forest type and climatic effects. Forest Ecology and Management 269:1925
Martin J, Kloeppel B, Schaefer T, Kimbler D, McNulty S (1998) Aboveground biomass
and nitrogen allocation of ten deciduous southern Appalachian tree species.
Canadian Journal of Forest Research 28:1648-1659
Matthews J, Endress A (2008) Performance criteria, compliance success, and vegetation
development in compensatory mitigation wetlands. Environmental Management
41:130-141
Megonigal, J., & Day, F. (1992). Effects of flooding on root and shoot production of bald
cypress in large experimental enclosures. Ecology, 73(4), 1182-1193.
Mitsch W, Zhang L, Stefanik K, Nahlik A, Anderson C, Bernal B, Song K (2012)
Creating wetlands: primary succession, water quality changes, and self-design
over 15 years. BioScience 62:237-250
Mokany K, Raison R, Prokushkin AS (2006) Critical analysis of root:shoot ratios in
terrestrial biomes. Global Change Biology 12:84-96
Monette R, Ware S (1983) Early forest succession in the Virginia coastal plain. Bulletin
of the Torrey Botanical Club 110:80-86
Morgan K, Roberts T (2003) Characterization of wetland mitigation projects in
Tennessee, USA. Wetlands 23:65-69
Nelson AS, Weiskittel AR, Wagner RG, Saunders MR (2014) Development and
evaluation of aboveground small tree biomass models for naturally regenerated
and planted species in eastern Maine, USA. Biomass and Bioenergy 68:215-227
Niklas KJ (2004) Plant allometry: is there a grand unifying theory? Biological Reviews
79:871-889
Niklas KJ (2005) Modelling below-and above-ground biomass for non-woody and woody
plants. Annals of Botany 95:315-321
Niswander S, Mitsch W (1995) Functional analysis of a two-year-old created in-stream
wetland: hydrology, phosphorus retention, and vegetation survival and growth.
Wetlands 15:212-225
203
Noon K (1996) A model of created wetland primary succession. Landscape and Urban
Planning 34:97-123
Odland A (1997) Development of vegetation in created wetlands in western Norway.
Aquatic Botany 59:45-62
Packard GC (2014) Multiplicative by nature: logarithmic transformation in allometry.
Journal of Experimental Zoology Part B: Molecular and Developmental Evolution
322:202-207
Pennington MR, Walters MB (2006) The response of planted trees to vegetation zonation
and soil redox potential in created wetlands. Forest Ecology and Management
233:1-10
Picard N, Rutishauser E, Ploton P, Ngomanda A, Henry M (2015) Should tree biomass
allometry be restricted to power models? Forest Ecology and Management
353:156-163
Pinheiro JC, Bates DM (2000) Mixed-effects models in S and S-PLUS. Springer, New
York, New York
R Core Team (2014) R: a language and environment for statistical computing. R
Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0.
http://www.R-project.org/ (accessed 8 Jan 2016)
Roussopoulos PJ, Loomis, RM (1979) Weight and didimension properties of shrubs and
small trees of the Great Lakes conifer forest. NC-178 United States Department of
Agriculture Forest Service, St Paul, Minnesota
Sileshi GW (2014) A critical review of forest biomass estimation models, common
mistakes and corrective measures. Forest Ecology and Management 329:237-254
Smith WB, Brand GJ (1983) Allometric biomass equations for 98 species of herbs,
shrubs, and small trees. NC-299. United States Department of Agriculture Forest
Service, North Central Forest Experiment Station, St Paul, Minnesota
Soil Survey Staff, United States Department of Agriculture, Natural Resources
Conservation Service. Web Soil Survey. http://websoilsurvey.sc.egov.usda.gov
(accessed 23 February 2015)
Stevens CF (2009) Darwin and Huxley revisited: the origin of allometry. Journal of
Biology 8:1-7
204
Telfer ES (1969) Weight-diameter relationships for 22 woody plant species. Canadian
Journal of Botany 47:1851-1855
Temesgen H, Affleck D, Poudel K, Gray A, Sessions J (2015) A review of the challenges
and opportunities in estimating above ground forest biomass using tree-level
models. Scandinavian Journal of Forest Research 30:326-335
Thomas SC, Martin AR (2012) Carbon content of tree tissues: a synthesis. Forests 3:332352
Wagner R, Ter-Mikaelian M (1999) Comparison of biomass component equations for
four species of northern coniferous tree seedlings. Annals of Forest Science
56:193-199
Whittaker R, Woodwell G (1968) Dimension and production relations of trees and shrubs
in the Brookhaven Forest, New York. The Journal of Ecology 56:1-25
Williams R, McClenahen J (1984) Biomass prediction equations for seedlings, sprouts,
and saplings of ten central hardwood species. Forest Science 30:523-527
Yang Y, Luo Y (2011) Isometric biomass partitioning pattern in forest ecosystems:
evidence from temporal observations during stand development. Journal of
Ecology 99:431-437
Zuur A, Ieno EN, Walker N, Saveliev AA, Smith GM (2009) Mixed effects models and
extensions in ecology with R. Springer, New York, New York.
205
Tables
Table 3-1. Morphological averages (standard deviation in parentheses) of trees
destructively harvested in 2011 and 2014. Species specific and successional group values
are averaged across three cells.
n
Cell
Ambient
Ambient
Ambient
Ambient
Ambient
Ambient
Ambient
Saturated
Saturated
Saturated
Saturated
Saturated
Saturated
Saturated
Flooded
Flooded
Flooded
Flooded
Flooded
Flooded
Flooded
Species
Betula nigra
Liquidambar styraciflua
Platanus occidentalis
Quercus bicolor
Quercus palustris
Quercus phellos
Salix nigra
Betula nigra
Liquidambar styraciflua
Platanus occidentalis
Quercus bicolor
Quercus palustris
Quercus phellos
Salix nigra
Betula nigra
Liquidambar styraciflua
Platanus occidentalis
Quercus bicolor
Quercus palustris
Quercus phellos
Salix nigra
Species Specific
Betula nigra
Liquidambar styraciflua
Platanus occidentalis
Quercus bicolor
Quercus palustris
Quercus phellos
Salix nigra
Successional Groups
Primary
Secondary
2011
14
16
17
17
15
18
16
14
17
18
16
17
18
17
15
16
18
14
20
16
17
2011
43
49
53
47
52
52
50
2011
195
151
2014
12
13
12
11
12
12
16
12
12
15
9
12
10
12
9
9
9
7
4
4
9
2014
33
34
36
27
28
26
37
2014
140
81
Equivalent Stem CrossSectional Diameter at
Groundline (cm)
2011
2014
3.55 (1.99) 16.13 (9.85)
3.19 (2.1) 8.86 (2.26)
2.46 (1.4) 15.99 (6.03)
1.26 (0.65) 4.29 (2.17)
1.19 (0.64) 3.64 (2.43)
1.47 (1.03) 5.21 (2.32)
2.04 (1.18) 7.91 (4.17)
2.44 (1.26) 14.34 (7.87)
1.48 (0.85) 7.93 (2.35)
1.73 (0.46) 8.73 (4.41)
1.16 (0.47) 3.61 (2.23)
1.21 (0.5) 3.03 (1.19)
1.21 (0.7) 4.71 (2.13)
1.5 (0.64) 10.17 (6.08)
1.29 (0.75) 2.5 (1.01)
1.13 (0.72) 1.87 (0.92)
0.98 (0.54) 1.69 (1.07)
1.04 (0.48) 1.7 (1.15)
0.99 (0.59) 1.58 (0.71)
0.93 (0.51) 1.45 (0.99)
1.86 (1.1) 3.77 (2.5)
2011
2014
2.4 (1.67) 11.76 (9.42)
1.92 (1.61) 6.68 (3.56)
1.71 (1.07) 9.39 (7.06)
1.16 (0.54) 3.39 (2.18)
1.12 (0.58) 3.09 (1.87)
1.21 (0.8) 4.44 (2.42)
1.8 (1)
7.63 (5.08)
2011
2014
1.94 (1.37) 8.83 (6.8)
1.16 (0.65) 3.62 (2.21)
Height (cm)
2011
189.86 (96.49)
137.44 (92.4)
127.29 (70.27)
47.71 (26.97)
65.33 (38.88)
95.72 (82.23)
106.25 (36.72)
119.29 (67.03)
62.59 (33.32)
100 (31.74)
46.25 (22.91)
52.82 (32.15)
69.94 (46.33)
86.82 (45.85)
90.4 (47.2)
49.38 (30.79)
51.06 (37.99)
35.21 (16.98)
51.15 (33.8)
54.38 (39.1)
88 (31.68)
2011
132.19 (82.48)
82.71 (69.62)
92.13 (57.71)
43.49 (23.14)
55.79 (34.68)
74.08 (60.93)
93.44 (38.8)
2011
98.93 (65.4)
58.26 (44.64)
2014
617.5 (219.03)
465 (123.03)
764.67 (244.1)
239.55 (129.26)
235.42 (202.69)
331.67 (159.71)
520.31 (257)
520.63 (202.59)
362.5 (94.64)
442.81 (200.62)
151.11 (91.71)
158.75 (78.45)
276.8 (108.99)
434.83 (236.97)
101.56 (34.61)
72.67 (40.59)
66.33 (43.76)
54.43 (56.89)
51.25 (19)
43.5 (22.25)
100.22 (32.1)
2014
441.56 (278.15)
324.97 (185.71)
455.98 (327.18)
162.07 (124.34)
176.25 (152.62)
266.23 (159.99)
390.41 (272.25)
2014
403.43 (273.42)
200.41 (151.63)
Canopy Diameter (cm)
2011
448.75 (176.18)
301.4 (67.69)
500.01 (153.04)
160.27 (83.39)
139.09 (87.75)
203.88 (90.14)
232.16 (92.65)
380.98 (126.39)
239.66 (59.4)
268.55 (109.55)
130.03 (78.43)
123.59 (63.75)
189.86 (93.3)
304.48 (139.15)
52.59 (26.91)
35.33 (25.04)
19.07 (14.1)
35 (36.49)
22.83 (14.82)
33.65 (25.64)
86.33 (78.14)
2011
83.68 (78.94)
45.03 (50.04)
38.33 (37.28)
23.87 (17.07)
28.04 (23.22)
43.91 (36.64)
58.13 (52.75)
2011
55.09 (57.7)
32.21 (28.36)
2014
150.61 (92.83)
89.71 (63.57)
69.58 (47.39)
30.52 (22.92)
39.73 (26.95)
59.83 (45.27)
78.76 (69.13)
73.26 (53.93)
31.86 (21.03)
37.85 (13.6)
27.17 (10.43)
32.55 (22.65)
43.51 (27.45)
47.96 (36.28)
30.95 (20.58)
14.34 (12.94)
9.31 (9.11)
12.03 (6.06)
15.44 (13.65)
26.46 (27.57)
48.89 (45.82)
2014
316.06 (209.88)
209.18 (122.17)
283.33 (215.3)
117.71 (86.81)
115.84 (79.79)
172.3 (102.3)
220.14 (133.27)
2014
256.34 (178.56)
134.59 (92.46)
206
Table 3-2. Average above (AGB) and belowground biomass (BGB), total biomass, and
root-to- shoot ratio (R:S) (standard deviation in parentheses) of trees destructively
harvested in 2011 and 2014. Species specific and successional group values are averaged
across three cells.
Cell
Ambient
Ambient
Ambient
Ambient
Ambient
Ambient
Ambient
Saturated
Saturated
Saturated
Saturated
Saturated
Saturated
Saturated
Flooded
Flooded
Flooded
Flooded
Flooded
Flooded
Flooded
Species
Betula nigra
Liquidambar styraciflua
Platanus occidentalis
Quercus bicolor
Quercus palustris
Quercus phellos
Salix nigra
Betula nigra
Liquidambar styraciflua
Platanus occidentalis
Quercus bicolor
Quercus palustris
Quercus phellos
Salix nigra
Betula nigra
Liquidambar styraciflua
Platanus occidentalis
Quercus bicolor
Quercus palustris
Quercus phellos
Salix nigra
Species Specific
Betula nigra
Liquidambar styraciflua
Platanus occidentalis
Quercus bicolor
Quercus palustris
Quercus phellos
Salix nigra
Successional Groups
Primary
Secondary
BGB (kg)
2011
2014*
0.52 (0.51) 12.29 (14.44)
0.6 (0.77) 8.96 (6.74)
0.29 (0.38) 48.26 (43.36)
0.08 (0.12) 2.99 (3.08)
0.09 (0.15) 1.15 (1.61)
0.12 (0.19) 1.53 (1.24)
0.08 (0.12) 1.94 (1.99)
0.22 (0.43) 7.01 (6.05)
0.08 (0.07) 5.47 (3.61)
0.05 (0.05) 7.83 (7.4)
0.04 (0.03) 1.63 (2.23)
0.06 (0.1) 0.64 (0.7)
0.07 (0.11) 1.21 (1.04)
0.04 (0.05) 3.22 (3.47)
0.03 (0.04) 0.1 (0.11)
0.02 (0.03) 0.11 (0.15)
0.01 (0.01) 0.07 (0.08)
0.02 (0.03) 0.23 (0.52)
0.02 (0.03) 0.07 (0.07)
0.02 (0.02) 0.08 (0.08)
0.11 (0.16) 0.16 (0.13)
2011
2014*
0.25 (0.42) 7.05 (10.4)
0.23 (0.5) 5.39 (5.79)
0.12 (0.24) 19.37 (32.43)
0.05 (0.08) 1.82 (2.55)
0.05 (0.1) 0.77 (1.18)
0.07 (0.13) 1.18 (1.15)
0.08 (0.12) 1.92 (2.58)
2011
2014*
0.16 (0.36) 8.46 (18.56)
0.06 (0.11) 1.25 (1.79)
AGB (kg)
2011
2014
0.69 (0.86) 33.14 (46.26)
0.62 (0.79) 7.56 (6.02)
0.23 (0.39) 42.08 (38.56)
0.04 (0.06) 1.65 (1.78)
0.04 (0.07) 1.36 (2.18)
0.16 (0.3) 4.15 (4.46)
0.13 (0.2) 6.74 (9.2)
0.2 (0.26) 15.81 (16.27)
0.05 (0.05) 4.49 (3.09)
0.08 (0.1) 6.53 (6.28)
0.02 (0.01) 0.88 (1.25)
0.05 (0.1) 0.63 (0.81)
0.05 (0.09) 3.05 (3.34)
0.05 (0.05) 13.58 (18.66)
0.04 (0.06) 0.09 (0.12)
0.02 (0.02) 0.07 (0.1)
0.02 (0.03) 0.05 (0.06)
0.01 (0.01) 0.12 (0.27)
0.02 (0.03) 0.04 (0.05)
0.03 (0.04) 0.07 (0.08)
0.13 (0.36) 0.21 (0.23)
2011
2014
0.3 (0.57) 17.83 (31.7)
0.23 (0.53) 4.5 (5.04)
0.11 (0.24) 16.76 (28.63)
0.02 (0.04)
1 (1.45)
0.03 (0.07) 0.86 (1.56)
0.08 (0.19) 3.1 (3.84)
0.1 (0.24) 7.37 (12.94)
2011
2014
0.18 (0.42) 11.55 (22.81)
0.05 (0.12) 1.62 (2.68)
Total Biomass (kg)
2011
2014*
1.21 (1.34) 45.43 (60.68)
1.22 (1.45) 16.53 (12.76)
0.52 (0.75) 90.34 (81.93)
0.12 (0.18) 4.64 (4.86)
0.13 (0.22) 2.51 (3.79)
0.28 (0.48) 5.68 (5.7)
0.21 (0.32) 8.67 (11.17)
0.42 (0.65) 22.82 (22.3)
0.12 (0.11) 9.96 (6.7)
0.13 (0.12) 14.36 (13.69)
0.06 (0.04) 2.51 (3.48)
0.11 (0.2) 1.27 (1.5)
0.12 (0.19) 4.26 (4.38)
0.09 (0.09) 16.8 (22.11)
0.07 (0.1) 0.19 (0.23)
0.04 (0.05) 0.18 (0.25)
0.03 (0.04) 0.12 (0.15)
0.03 (0.03) 0.35 (0.79)
0.04 (0.06) 0.12 (0.13)
0.04 (0.06) 0.15 (0.15)
0.24 (0.51) 0.37 (0.35)
2011
2014*
0.55 (0.96) 24.87 (42.05)
0.45 (0.98) 9.88 (10.83)
0.22 (0.47) 36.13 (61.06)
0.07 (0.11)
2.82 (4)
0.09 (0.17) 1.63 (2.74)
0.15 (0.32) 4.28 (4.98)
0.18 (0.35) 9.29 (15.5)
2011
2014*
0.34 (0.74) 20.01 (39.55)
0.11 (0.22) 2.88 (4.09)
R:S Ratio
2011
0.97 (0.64) bc; y
1.67 (0.93) ac; y
2.4 (2.48) abc; y
2.74 (1.39) a; y
2.24 (0.44) a; y
2.51 (2.7) abc; yz
0.76 (0.26) b; z
2.78 (6.89) abc; y
1.74 (0.63) a; y
0.95 (0.49) bc; y
2.67 (1.32) a; y
2.35 (1.75) ab; y
2.19 (1.04) a; y
0.86 (0.44) c; z
0.95 (0.48) b; y
1.42 (0.5) ab; y
0.88 (0.5) b; y
2.04 (1.03) a; y
2.62 (4.36) ab; y
1.06 (0.46) ab; z
2.29 (2.03) ab; y
2011
1.55 (3.96) ab
1.61 (0.71) b
1.39 (1.59) b
2.51 (1.28) a
2.42 (2.85) ab
1.95 (1.8) ab
1.31 (1.39) b
2011
1.46 (2.17) b
2.29 (2.1) a
* BGB and total biomass were calculated using the non-linear BGB~AGB relationship
developed from the 2011 samples.
a – d letters represent significant differences (p < 0.05) among species within cells
y – z letters represent significant differences (p < 0.05) among cells within species
207
Table 3-3. Average percentage carbon, nitrogen and C:N ratio of subsamples from trees
destructively harvested in 2014. Species specific and successional group values are
averaged across three cells. Values in parentheses represent standard deviation.
n
Cell
Ambient
Ambient
Ambient
Ambient
Ambient
Ambient
Ambient
Saturated
Saturated
Saturated
Saturated
Saturated
Saturated
Saturated
Flooded
Flooded
Flooded
Flooded
Flooded
Flooded
Flooded
Species
Betula nigra
Liquidambar styraciflua
Platanus occidentalis
Quercus bicolor
Quercus palustris
Quercus phellos
Salix nigra
Betula nigra
Liquidambar styraciflua
Platanus occidentalis
Quercus bicolor
Quercus palustris
Quercus phellos
Salix nigra
Betula nigra
Liquidambar styraciflua
Platanus occidentalis
Quercus bicolor
Quercus palustris
Quercus phellos
Salix nigra
Species Specific
Betula nigra
Liquidambar styraciflua
Platanus occidentalis
Quercus bicolor
Quercus palustris
Quercus phellos
Salix nigra
Successional Groups
Primary
Secondary
2014
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
4
4
5
2014
15
15
15
15
14
14
15
2014
60
43
Percentage
Percentage
Carbon
Nitrogen
2014
2014
47.44 (0.82) a; y 1.27 (0.29) a; y
46 (0.2) a; y
1.2 (0.1) a; y
46.36 (1.02) a; y 1.13 (0.08) a; y
46.42 (0.21) a; y 1.46 (0.22) a; y
46.3 (0.37) a; y
1.37 (0.11) a; y
46.38 (0.36) a; y 1.37 (0.11) a; y
46.03 (0.58) a; y
1.48 (0.2) a; y
47.16 (0.48) a; y 1.15 (0.1) cd; y
46.19 (0.67) a; y 1.27 (0.08) abcd; y
46.91 (0.56) a; y 1.17 (0.06) d; y
46.22 (0.36) a; y 1.55 (0.15) ab; y
46.36 (0.78) a; y 1.45 (0.07) ab; y
46.13 (0.41) a; y 1.26 (0.02) ad; yz
46.34 (0.54) a; y 1.39 (0.04) bc; y
47.72 (0.53) a; y 1.3 (0.09) ab; y
45.84 (1.05) a; y 1.23 (0.14) ab; y
47.34 (1.19) a; y 1.22 (0.08) ab; y
46.22 (0.76) a; y 1.42 (0.09) a; y
47.3 (1.0) a; y
1.29 (0.13) ab; y
46.16 (1.15) a; y 1.17 (0.05) b; z
46.98 (1.1) a; y
1.44 (0.1) a; y
2014
2014
47.44 (0.63) a
1.24 (0.18) bc
46.01 (0.69) b
1.23 (0.11) bc
46.87 (0.98) ab
1.18 (0.08) bc
46.29 (0.47) b
1.48 (0.16) a
46.6 (0.82) ab
1.37 (0.12) ab
46.23 (0.64) b
1.27 (0.11) bc
46.45 (0.83) b
1.44 (0.13) a
2014
2014
46.69 (0.94) a
1.27 (0.16) b
46.37 (0.66) b
1.38 (0.15) a
C:N Ratio
2014
38.69 (7.67) a; y
38.42 (3.08) a; y
41.16 (3.51) a; y
32.3 (4.31) a; y
34.09 (2.74) a; y
33.97 (2.59) a; z
31.48 (4.15) a; y
41.19 (3.11) ab; y
36.4 (2.22) abbc; y
40.07 (1.92) a; y
30.09 (3.03) cd; y
32.09 (1.43) c; y
36.74 (0.48) ad; yz
33.38 (0.54) bc; y
36.93 (2.62) ab; y
37.78 (4.19) ab; y
38.79 (1.84) a; y
32.57 (2.3) b; y
37.16 (4.6) ab; y
39.55 (1.86) a; y
32.71 (2.72) ab; y
2014
38.94 (4.98) ab
37.53 (3.14) ab
40.01 (2.56) a
31.65 (3.28) c
34.25 (3.5) bc
36.55 (2.88) b
32.52 (2.79) c
2014
37.25 (4.47) a
34.09 (3.75) b
a – d letters represent significant differences (p < 0.05) among species within cells
y – z letters represent significant differences (p < 0.05) among cells within species
208
Table 3-4. Results of fitting Eq. 1 (Y=aXb + ε) from saplings destructively harvested in
2011. Y=Total dry belowground biomass (kg). X=Total dry aboveground biomass
(excluding leaves) (kg).
Primary Successional Species
Betula nigra
Liquidambar styraciflua
Platanus occidentalis
Salix nigra
a
0.787
1.334
1.272
0.544
a SE
0.109
0.184
0.480
0.162
b
0.817
0.948
0.975
0.739
b SE SE of the Regression
0.091
0.439
0.031
0.509
0.092
1.296
0.093
0.578
Secondary Successional Species
Quercus bicolor
Quercus palustris
Quercus phellos
1.921
1.019
0.595
0.480
0.219
0.127
0.942
0.831
0.732
0.059
0.064
0.055
0.264
0.210
0.451
209
Table 3-5. Results of fitting Eq. 1 (Y=aXb + ε) to saplings destructively harvested in 2011
and 2014. Y=Total dry biomass (excluding leaves) (kg). X= Equivalent stem crosssectional diameter at groundline (ESD) (cm).
Primary Successional Species
Betula nigra
Liquidambar styraciflua
Platanus occidentalis
Salix nigra
a
0.032
0.032
0.028
0.029
a SE
0.005
0.010
0.003
0.007
b
2.442
2.751
2.789
2.519
b SE
0.067
0.149
0.049
0.099
SE of the Regression
0.667
0.908
0.522
0.911
Secondary Successional Species
Quercus bicolor
Quercus palustris
Quercus phellos
0.047
0.047
0.055
0.007
0.004
0.007
2.735
2.502
2.455
0.113
0.133
0.172
0.571
1.016
1.891
210
Table 3-6. Average initial biomass estimates for each species/stocktype combination.
Same letters represent no significant difference among stocktype for each species
(p>0.05). n represents initial number of trees planted.
Species
Betula nigra
Stocktype
Bare root
Gallon
Tubeling
Liquidambar styraciflua Bare root
Gallon
Tubeling
Platanus occidentalis
Bare root
Gallon
Tubeling
Quercus bicolor
Bare root
Gallon
Tubeling
Quercus palustris
Bare root
Gallon
Tubeling
Quercus phellos
Bare root
Gallon
Tubeling
Salix nigra
Bare root
Gallon
Tubeling
n
137
105
94
111
112
109
75
112
73
125
105
128
127
116
96
170
107
93
112
109
123
Initial
Biomass (kg)
0.021
0.046
0.025
0.024
0.040
0.029
0.018
0.058
0.025
0.047
0.048
0.037
0.035
0.061
0.032
0.039
0.083
0.035
0.020
0.055
0.022
Standard
Deviation
0.037
0.051
0.029
0.034
0.052
0.049
0.034
0.075
0.034
0.061
0.056
0.057
0.048
0.065
0.046
0.055
0.102
0.041
0.030
0.073
0.028
Within Species
Comparison
b
a
b
b
a
ab
b
a
b
a
a
a
b
a
b
b
a
b
b
a
b
211
Stocktype
Bare root
Gallon
Tubeling
Liquidambar styraciflua Bare root
Gallon
Tubeling
Platanus occidentalis
Bare root
Gallon
Tubeling
Quercus bicolor
Bare root
Gallon
Tubeling
Quercus palustris
Bare root
Gallon
Tubeling
Quercus phellos
Bare root
Gallon
Tubeling
Salix nigra
Bare root
Gallon
Tubeling
Betula nigra
Species
14
36
8
28
36
5
24
32
29
38
34
21
32
35
5
24
30
6
0
33
12
n
Final
Biomass (kg)
36.88
64.26
43.12
41.64
36.68
14.40
151.74
71.13
219.11
4.32
5.31
3.11
3.17
6.61
0.56
2.72
10.07
2.97
NA
18.16
28.47
Ambient
Standard
Deviation
26.72
41.29
31.35
23.34
24.01
10.23
123.38
79.44
153.15
4.39
4.74
3.18
2.89
8.32
0.96
3.27
8.09
3.78
NA
20.57
28.69
a
a
Within Species
Comparison
b
a
ab
a
a
b
a
b
a
a
a
a
b
a
c
b
a
b
24
35
22
25
36
14
0
35
20
35
36
25
28
38
14
34
32
23
12
33
15
n
Final
Biomass (kg)
18.43
40.45
27.51
31.23
35.76
19.77
NA
20.12
54.77
2.53
3.72
1.45
1.97
3.00
1.25
2.65
6.68
2.99
39.77
19.34
32.87
Saturated
Standard
Deviation
12.24
24.19
17.76
20.79
20.50
12.28
NA
23.78
68.63
2.66
5.02
1.57
1.78
2.82
2.01
2.45
4.22
2.96
37.52
25.65
48.37
b
a
ab
a
b
a
a
a
b
a
b
a
a
a
Within Species
Comparison
b
a
b
ab
a
b
5
29
22
10
26
16
0
4
0
7
11
0
2
2
0
1
7
0
34
36
28
n
Final
Biomass (kg)
1.11
1.24
0.41
0.32
0.55
0.12
NA
0.31
NA
0.59
0.29
NA
0.02
1.27
NA
0.09
0.83
NA
1.28
1.22
0.76
Flooded
Standard
Deviation
1.90
2.24
1.00
0.55
0.80
0.21
NA
0.24
NA
0.72
0.37
NA
0.01
1.42
NA
NA
0.76
NA
1.72
1.85
0.85
a
a
a
a
a
a
a
Within Species
Comparison
a
a
a
ab
a
b
Table 3-7. Average biomass accumulated 6 years following planted estimated for each
stocktype/species combination. Same letter represent no significance difference among
stocktype for each species within each cell (p>0.05). n represents count of live trees after
6 years.
212
Table 3-8. Average biomass accumulated 6 years following planted estimated for each
species (incorporating all stocktypes). Average biomass of species is compared within
cells and cells are compared for each species. Same letters represent no significant
difference (p<0.05). n represents count of live trees after 6 years.
Ambient
Betula nigra
Liquidambar styraciflua
Platanus occidentalis
Quercus bicolor
Quercus palustris
Quercus phellos
Salix nigra
Saturated
Betula nigra
Liquidambar styraciflua
Platanus occidentalis
Quercus bicolor
Quercus palustris
Quercus phellos
Salix nigra
Flooded
Betula nigra
Liquidambar styraciflua
Platanus occidentalis
Quercus bicolor
Quercus palustris
Quercus phellos
Salix nigra
58
69
85
93
72
60
45
Average Total
Biomass (kg)
54.74
37.08
144.38
4.41
4.66
6.42
20.91
Standard
Deviation
38.53
23.79
135.28
4.32
6.40
7.15
23.13
Within cell, species
comparison
b
c
a
e
e
e
d
Within species, cell
comparison
x
x
x
x
x
x
x
81
75
55
96
80
89
60
30.41
31.27
32.72
2.70
2.34
4.19
26.81
21.54
20.00
47.92
3.64
2.44
3.78
35.34
a
a
a
bc
c
b
a
y
x
y
y
y
y
x
56
52
4
18
4
8
98
0.90
0.37
0.31
0.41
0.65
0.73
1.11
1.83
0.64
0.24
0.53
1.09
0.75
1.58
ab
b
b
b
ab
ab
a
z
y
z
z
z
z
y
n
213
Table 3-9. Average biomass accumulated 6 years following planted estimated for primary
and secondary groups (incorporating all species and stocktypes). Average biomass of
species groups are compared within cells and cells are compared for each species group.
Same letters represent no significant difference (p<0.05). n represents count of live trees
after 6 years.
Ambient
Primary
257
Secondary 225
Average Total
Biomass (kg)
73.72
5.03
Standard
Deviation
95.76
5.90
Within cell, species
group comparison
a
b
Within species group,
cell comparison
x
x
Primary
271
Secondary 265
30.32
3.09
31.36
3.46
a
b
y
y
Primary
210
Secondary 30
0.86
0.53
1.50
0.67
a
b
z
z
n
Saturated
Flooded
a – b letters represent differences among species within cells
y – z letters represent differences among cells within species
214
Figures
Figure 3-1. Results of fitting Eq. 1 (Y=aXb + ε) for primary successional species
destructively harvested in 2011. Y=Total dry belowground biomass (coarse roots) (kg)
and X= Total dry aboveground biomass (kg) (excluding leaves) (See Table 3-4 for results
of fitting model).
215
Figure 3-2. Results of fitting Eq. 1 (Y=aXb + ε) for secondary successional species
destructively harvested in 2011. Y=Total dry belowground biomass (coarse roots) (kg)
and X= Total dry aboveground biomass (excluding leaves) (kg) (See Table 3-4 for results
of fitting model).
216
Figure 3-3. Results of fitting Eq. 1 (Y=aXb + ε) for primary successional species
destructively harvested in 2011 and 2014. Y=Total biomass (kg) and X=Equivalent stemcross sectional diameter at groundline (cm) (See Table 3-5 for results of fitting model).
217
Figure 3-4. Results of fitting Eq. 1 (Y=aXb + ε) for secondary successional species
destructively harvested in 2011 and 2014. Y=Total biomass (kg) and X=Equivalent stemcross sectional diameter at groundline (cm) (See Table 3-5 for results of fitting model).
218
CHAPTER 4: EFFECT OF HYDROLOGIC CONDITIONS ON ABSOLUTE AND
RELATIVE GROWTH RATES OF SEVEN WETLAND TREE SPECIES
219
Abstract
Biomass accumulation rates (growth rates) are influenced by many factors and have not
been documented for many wetland tree species. Growth rates can be useful in
determining the development of created/restored forested wetlands. Successfully
restoring these wetlands requires that trees grow and accumulate woody biomass. The
purpose of this study was to investigate growth rates of seven Mid-Atlantic wetland tree
species (four early successional, three late successional) in response to planting along a
controlled hydrologic gradient. Biomass was determined using species specific models
and cumulative, absolute (AGR), and relative (RGR) growth rates were determined over
6 years using functional-derived nonlinear models. Growth rates at standardized masses
were compared among species and successional groups across the hydrologic gradient.
Fast growing early successional species obtained more biomass than late successional
species and therefore had increased AGR and RGR rates. At 11 kg, P. occidentalis had
the greatest AGR (36 g/day) and RGR (3.1 g/g/day) in the least stressful hydrologic
treatment. Increased hydrologic stress reduced biomass and biomass accumulation rates;
however, moderate hydrologic stress did not reduce AGR or RGR as the selected species
are typically found in wetland systems. In the most severe hydrologically stressed
conditions, S. nigra had the greatest AGR (0.27 g/day) and RGR (2.8 g/g/day) at 0.1 kg
standardized biomass. Overall, these results provide needed information about the
biomass accumulation rates of planted saplings that are often not accounted for when
studying the development of restored forested wetlands.
220
Introduction
Growth is generally defined as an irreversible change in a measurable quantity
(Hunt 1990, Jørgensen et al. 2000). While this definition is simple, a multitude of
methods are designed to describe growth. Common measurable quantities can include
size, form, or number (Hunt 1982). These measurable quantities can be applied to a vast
range of living and non-living systems from individual bacterium to human populations
and from crystals all the way to the entire universe. For the purpose of this paper, growth
will focus on groups of planted sapling’s increase in total woody biomass.
Many researchers (Blackman 1919, Evans 1972, Causton and Venus 1981, Hunt
1982, Paine et al. 2012 (See Pommerening and Muszta 2016 for review) developed
methods to address questions specific to plant growth analysis: through time particular
methods have proved more valuable than others and general trends became apparent.
Increases in plant mass through time (referred to as cumulative growth) tends to follow
the same general sigmoidal pattern that is found elsewhere in nature (Weiner and Thomas
2001, Pommerening and Muszta 2016). Plants start with little to no detectable increases
in mass, then mass increases rapidly. As plants reach maturity, the increase in mass slows
(reaches an asymptote) and finally ceases (Hunt 1990). This pattern of growth is
exhibited by many types of plants; however, the magnitude and symmetry of the curve
can change substantially. Further studies have suggested that biomass (or carbon)
accumulation of trees may not reach an asymptote (Muller-Landau et al. 2006, Sillett
2010, Stephenson et al. 2014).
From this general sigmoidal pattern, two biomass growth rates can be determined.
The first is termed absolute growth rate (AGR) and is defined as the change in mass over
221
a given time interval (e.g. g/day). Following the above sigmoidal pattern example, AGR
will increase rapidly, peak, rapidly decline, and then slowly taper off. AGR yields
important information, but presents challenges when comparing growth of individuals of
different starting masses (Hunt 1982, Hunt 1990, Rees et al. 2010, Pommerening et al.
2016), which is the focus of this study.
Relative growth rate (RGR) can be used to compare individuals of different sizes
when calculated correctly (Rose et al. 2009). Relative growth rate is defined as the
increase in mass per unit mass per unit time (e.g. g/g/day) or growth per unit mass.
Relative growth rate is referred to as an ‘efficiency index’ because it measures the
efficiency of plant material to produce new material (Blackman 1919). It is often
compared to the rate of compound interest earned on capital in the financial world
(Blackman 1919, Hunt 1990). Relative growth rates of trees decline through time as trees
grow larger (ontogenetic drift) and because of self-shading, increased allocation to
structural (non-photosynthetic) components, declines in leaf area ratio, and reduced
nutrient availability (Evans 1972, Hunt 1982, South 1995, Rees et al. 2010, Paine et al.
2012, Philipson et al. 2012). Therefore, comparisons of RGR among species or
experimental treatments must account for differences in mass because high values of
RGR can occur when plants are either smaller in size or because they are growing faster
(Turnbull et al. 2008, Rees et al. 2010, Turnbull et al. 2012).
Tree growth can be positively or negatively influenced by hydrologic conditions
depending on the species and other environmental factors (Kozlowski 1984b). Very few
studies, however, measured total biomass (aboveground and belowground) and calculated
AGR and RGR for individual seedlings or saplings grown across a hydrologic gradients
222
(Tang and Kozlowski 1982a,b). Several studies have investigated morphological growth
(basal area, stem diameter, tree rings, height, leaf mass, etc.) of trees in different
hydrologic regimes and found positive and negative responses (Malecki et al. 1983,
Mitsch and Rust 1984, Keeland et al. 1997, Kabrick et al. 2005, Anderson and Mitsch
2008, McCurry et al. 2010, Rodríquez-González et al. 2010, Kabrick et al. 2012, Smith et
al. 2013). However, a large proportion of studies fail to account for small trees that have
a small stem diameter at breast height (dbh) or are not tall enough to have a dbh at all
(Das 2012). In recently restored wetlands and other developing ecosystems, accounting
for these trees is important because they are contributing substantially to the overall
ecosystem structure and function.
The purpose of this study was to determine biomass growth rates (AGR and
RGR) of seven species from the Mid-Atlantic region planted across a hydrologic
gradient.
Methods
Methods used to establish and measure the survival and morphology of saplings
planted in the experimental site are the same as Chapter 2.
Biomass Estimation Model
Species specific biomass estimation models (BEM) were used to determine
biomass of individual saplings for all sample periods over 6 years (See Chapter 3 for
BEM development).
223
Differences in initial and final biomass for each species in each cell were
determined using pairwise t-tests. Variance was estimated separately for each group, the
Welch modification to degrees of freedom was used and Holm’s p-value adjustment was
used to control the family-wise error rate. All alpha values were set at 0.05.
Growth Rate Calculations
The methods used to calculate AGR and RGR have been debated extensively
(Hunt 1982, Hunt 1990, South 1991, South 1995, Hoffmann and Poorter 2002, Shimojo
et al. 2002, MacFarlane and Kobe 2006, Paine et al. 2012, Matsushita et al. 2015,
Pommerening and Muszta 2015, 2016). Nonlinear function-derived growth rates
presented in Paine et al. (2012) were used to calculate AGR and RGR as a function of
mass. These methods of calculating growth rates were used because the repeated biomass
estimations made over 6 years allowed for use of these function-derived growth rates that
better capture the temporal dynamics of growth than the traditional (classical) growth rate
calculations and because of the mathematical support provided for these calculations.
To reduce heterogeneous variance, biomass was transformed using the natural
logarithm. The monomolecular function (also known as Mitscherlich function (Zeide
1993)) was fit to the transformed data for each species within each cell using the standard
selfStart function, SSasymp, in the NLME package (Pinheiro and Bates 2000) in R
version 3.2.1 (R Core Team 2015) (Figure 4-2). The following equation was used to
model the natural logarithm of biomass:
yt = K-e-rt(K-M0)
(Equation 1)
yt = Natural logarithm of biomass at time t
224
K = Upper asymptote (model derived)
e = Euler’s number
r = Rate constant (model derived)
t = Time
M0 = Y intercept (model derived)
This function was selected because the saplings were undergoing rapid increases
in biomass and produced lower Akaike information criterion (AIC) values than other
functions when modelling the natural logarithm of biomass. The custom function
(output.mono.nls) provided by Paine et al. (2012) was used to determine AGR and RGR
and associated confidence intervals as a function of biomass. The following equation was
used to determine AGRm (AGR at a specific biomass, variables same as Eq. 1):
AGRm = (re-rt(k-M0))eyt
(Equation 2)
The following equation was used to determine RGRm (RGR at a specific biomass,
variables same as Eq. 1)
RGRm=r(K-yt)
(Equation 3)
Comparing Growth Rates
Several authors suggest comparing species (or treatments) AGR and RGR at
particular biomass amounts (Turnbull et al. 2008, Rose et al. 2009, Paul-Victor et al.
2010, Rees et al. 2010, Philipson et al. 2012, Turnbull et al. 2012). This is referred to as
size corrected or size standardized growth rates. Most commonly this size standardization
is applied to RGR (SRGR); however it can also be applied to AGR (SAGR). This
225
technique is useful because AGR and RGR trajectories can intersect as experimental units
grow larger (Larocque and Marshall 1993, Hautier et al. 2010, Paine et al. 2012).
A standard size was selected at which to compare AGR and RGR for individual
species in PRI and SEC groups and differences in rates between AMB and SAT. The size
selected represented the maximum biomass obtained by all species in a group from both
AMB and SAT. For PRI, 11 kg was selected for comparison and for SEC, 1 kg was
selected. Due to the low growth in FLD, the maximum size obtained by all species was
0.1 kg, which was used to compare all species. Importantly, species in these cells
obtained these respective biomasses as different times.
Results
Biomass Cumulative Growth
Initial biomass varied among the species (Table 4-1), ranging from 30 g (Standard
Deviation (SD) = 42) for B. nigra to 50 g (SD = 72) for Q. phellos. Biomass increased
through time for most species in AMB and SAT, while most species exhibited little
cumulative growth in FLD (Figure 4-1). As a result, most species had greater biomass
after 6 years in AMB than SAT and FLD (Table 3-6 and Figure 4-1). Additionally, there
were differences in final biomass among species within each cell (Table 3-6 and Figure
4-1). In the AMB and SAT, PRI species had greater biomass than SEC species. Salix
nigra had greater final biomass in FLD than L. styraciflua, P. occidentalis, and Q. bicolor
(Table 3-6).
226
Absolute Growth Rate (AGR)
For most species, AGR increased as sapling size increased (Figure 4-3). For
several species, AGR started to reach an asymptote as mass increased; S. nigra in FLD
reached a peak AGR and then started to decline.
Absolute growth rate differed among PRI and SEC species in AMB. At a standard
total biomass of 11 kg, P. occidentalis had the greatest AGR (36 g/day), followed by both
B. nigra (26 g/day) and L. styraciflua (23 g/day) (Table 4-2). Salix nigra had the lowest
AGR at 11 kg (15 g/day) of PRI species in AMB while Q. phellos had the greatest AGR
at 1 kg total biomass (2.0 g/day) followed by Q. bicolor (1.8 g/day) and Q. palustris (1.6
g/day) in AMB. The 95% confidence intervals (CI) overlapped for Q. bicolor and Q.
palustris, suggesting that their growth rates may not be significantly different (Table 4-2).
Liquidambar styraciflua had the greatest AGR at 11 kg biomass (34 g/day) in
SAT. However, CI for L. styraciflua overlapped with S. nigra (27 g/day) and B. nigra (26
g/day). Platanus occidentalis had the lowest AGR (22 g/day). Quercus phellos had
greater AGR at 1 kg (2.0 g/day) than Q. bicolor (1.4 g/day) and Q. palustris (1.3 g/day).
The 95% CI overlapped for Q. bicolor and Q. palustris (Table 4-2).
Salix nigra had the greatest AGR in FLD at 0.1 kg biomass (0.27 g/day). As a
result of poor survival and dieback, CI of L. styraciflua (0.14 kg/day), B. nigra (0.12
g/day) and P. occidentalis (0.10 g/day) were overlapping. Similarly, the CI for Q. phellos
(0.14 g/day), Q. bicolor (0.09 g/day) and Q. palustris (0.08 g/day) also overlapped at 0.1
kg standard biomass (Table 4-2).
Species AGR were compared between AMB and SAT using 11 kg
standardization. Both L. styraciflua and S. nigra increased AGR in SAT compared to
227
AMB, while B. nigra AGR remained the same (Table 4-2). AGR for P. occidentalis
declined 38.9% in SAT compared to AMB. Both Q. bicolor and Q. palustris had
decreased AGR in SAT compared to AMB, while Q. phellos AGR remained similar. The
saplings planted in FLD had lower total biomass after 6 years (Table 3-6) suggesting that
their AGR were consistently lower than AMB and SAT.
Relative Growth Rate (RGR)
RGR of all species declined rapidly as saplings developed (Figure 4-4). Several
species in FLD (P. occidentalis, Q. bicolor, and Q. phellos) exhibited negative RGR due
to mortality and die back.
Platanus occidentalis had the greatest RGR at 11 kg (3.1 g/g/day) in AMB,
followed by B. nigra (2.4 g/g/day) and L. styraciflua (2.1 g/g/day). The 95% CI
overlapped for B. nigra and L. styraciflua (Table 4-2). Salix nigra had the lowest RGR at
11 kg (1.3 g/g/day) of PRI species. Quercus phellos had the greatest RGR in AMB at 1
kg total biomass (2.0 g/g/day) followed by Q. bicolor (1.8 g/g/day) and Q. palustris (1.6
g/g/day). The 95% confidence intervals (CI) overlapped for Q. phellos and Q. bicolor as
well as for Q. bicolor and Q. palustris suggesting that their growth rates may not be
significantly different (Table 4-2).
In SAT, L. styraciflua RGR was greatest (3.0 g/g/day) followed by S. nigra (2.5
g/g/day) and B. nigra (2.3 g/g/day). The CI of S. nigra and B. nigra overlapped. Platanus
occidentalis had the lowest RGR (2.0 g/g/day) of PRI species in SAT at 11 kg biomass.
Quercus phellos had greater RGR at 1 kg (2.0 g/g/day) than Q. bicolor (1.4 g/g/day) and
228
Q. palustris (1.3 g/g/day). There was overlap of CI between the Q. bicolor and Q.
palustris (Table 4-2).
When comparing PRI species in FLD, S. nigra had the greatest RGR (2.8
g/g/day). The remaining species, L. styraciflua (1.4 g/g/day), B. nigra (1.2 g/g/day), and
P. occidentalis (0.99 g/g/day) had overlapping 95% CI, suggesting that there was no
difference in RGR among these species. Similarly, the CI for Q. phellos (1.4 g/g/day), Q.
bicolor (0.09 g/g/day) and Q. palustris (0.08 g/g/day) also overlapped at 0.1 kg standard
biomass (Table 4-2).
When comparing the RGR of PRI species between AMB and SAT, L. styraciflua
and S. nigra exhibited increases in SAT compared to AMB, while B. nigra RGR
remained the same and P. occidentalis RGR decreased in SAT (Table 4-2). Quercus
bicolor RGR decreased in SAT compared to AMB and Q. palustris and Q. phellos RGR
remained similar (Table 4-2).
When observing the entire RGR curves for B. nigra, L. styraciflua, Q. phellos and
S. nigra the saplings in SAT had lower initial RGR than AMB. However, at the end of
the study, RGR for these species in SAT was greater than AMB (Figure 4-4).
Discussion
Biomass Cumulative Growth
Biomass increased through time for most species in AMB and SAT, but most
species planted in FLD had little cumulative growth. As discussed in Chapter 3, this most
likely resulted from flooding above the root collar, reduced soil nitrogen and phosphorus
concentrations, increased clay concentrations, and herbaceous competition. The reduction
229
in the amount of biomass accumulated as a result of increased hydrologic stress has been
confirmed for P. occidentalis (Tang and Kozlowski 1982a), S. nigra (Donovan et al.
1988, Pezeshki et al. 1998), Taxodium distichum (Mitsch and Ewel 1979), and various
flood intolerant species (Kozlowski 1984b). However, T. distichum is unique in its
adaptations to low oxygen conditions resulting from soil flooding, and other studies have
found that the biomass accumulated under stressful hydrologic conditions can be similar
to the amount when planted in less stressful hydrologic condition (Megonigal and
Day1992, Dickinson 2007). However, the AGB and BGB allocation patterns of T.
distichum have been found to shift based on hydrologic conditions (lower BGB
production in flooded conditions) (Megonigal and Day 1992). Salix nigra has been found
to be very tolerant of hydrologic stress (Hook 1984) and Day et al. (2006) showed stem
biomass accumulation to be similar in flooded and drained conditions.
Growth rates were compared among species within their respective groups (PRI
and SEC) because of the differences in life history strategies between these two groups
(Bazzaz 1979). Additionally, the biomass accumulated after 6 years by the SEC species
was significantly lower than PRI (Table 3-9). PRI had greater biomass accumulated than
SEC in AMB and SAT, while in FLD all species had less biomass accumulated than S.
nigra.
Absolute Growth Rate (AGR)
The increase in AGR as sapling biomass increased is the result of greater ability
to capture resources (light, nutrients, water, etc.) with most of the resources being
allocated to biomass accumulation as opposed to maintenance and reproduction. Many of
230
the species had not yet reached a peak in AGR where biomass accumulation is at the
maximum rate at the time of sampling. S. nigra in FLD appears to have reached a
maximum and AGR and started to decline; however, this could be the result of
continuing mortality of stems and with additional regrowth this trend could change in the
future.
When comparing species AGR at standardized sizes, different species had the
greatest AGR in each cell, reflecting their different abilities to tolerate hydrologic stress.
Platanus occidentalis had the greatest AGR in AMB, L. styraciflua had the greatest AGR
in SAT, and S. nigra had the greatest AGR in FLD. Quercus phellos had the greatest
AGR in AMB and SAT. Absolute growth rate was not different among SEC species in
the FLD. The greater AGR exhibited by L. styraciflua, S. nigra, and Q. phellos suggests
that they may be appropriate for planting in restored wetlands or other
afforestation/reforestation projects where rapid accumulation of biomass is desired.
Both L. styraciflua and S. nigra increased AGR in SAT compared to AMB. Salix
nigra has been shown to be sensitive to dry conditions (McLeod and McPherson 1973,
Hook 1984, Schaff et al. 2003), suggesting a reason for the lower AGR in AMB.
Although not investigated in this study, the increased size of the trees (particularly P.
occidentalis and B. nigra) in AMB may have caused competition for resources, which
may have reduced the AGR of L. styraciflua and S. nigra. Platanus occidentalis AGR
declined 38.9% from AMB to SAT suggesting that it has reduced tolerance for
hydrologic stress as suggested by Tang and Kozlowoski (1982a). In FLD, P. occidentalis
AGR was the lowest for PRI species, further suggesting that it has a lower tolerance for
hydrologic stress than the other species.
231
AGR of Q. phellos was similar in AMB and SAT and was greater than the other
SEC species. Both Q. bicolor and Q. palustris had slightly lower AGR in SAT than
AMB. Quercus phellos may be more tolerant of hydrologic stress than Q. bicolor and Q.
palustris. All three Quercus spp. are considered moderately to somewhat flood tolerant
and are typically found on poorly drained soils (Whitlow and Harris 1979, Hook 1984,
Burns and Honkala 1990a).
Relative Growth Rate (RGR)
Biomass RGR declined rapidly as saplings increased biomass, possibly due to
self-shading, increased allocation to structural (non-photosynthetic) components, declines
in leaf area ratio, and reduced nutrient availability (Evans 1972, Hunt 1982, South 1995,
Rees et al. 2010, Paine et al. 2012, Philipson et al. 2012). Since the species had different
initial sizes, several authors suggest comparing species RGR at standardized mass
(Turnbull et al. 2008, Rose et al. 2009, Paul-Victor et al. 2010, Rees et al. 2010, Philipson
et al. 2012, Turnbull et al. 2012). However, appropriateness of this procedure is still
being actively debated (Pommerening and Muszta 2016). Comparison of RGR and AGR
for species and cells at standardized masses yielded similar results suggesting that this
method may be appropriate and provides additional validation to making planting
recommendations based on both measures.
Species that had the greatest biomass RGR were the same species that had the
greatest biomass AGR across the cells suggesting that these species also had the greatest
growth efficiencies. Platanus occidentalis had the greatest RGR in AMB, L. styraciflua
had the greatest RGR in SAT, and S. nigra had the greatest RGR in FLD. Quercus
232
phellos had the greatest RGR in AMB and SAT and in FLD there was no difference in
RGR among SEC species. These species may be appropriate for planting in restored
wetlands or other afforestation/reforestation projects where efficient accumulation of
biomass is desired.
Both L. styraciflua and S. nigra exhibited greater RGR in SAT compared to
AMB. This suggests that their growth efficiencies may be reduced in drier conditions.
Platanus occidentalis RGR was lower in SAT than in AMB, suggesting that growth
efficiency is decreased even in moderate hydrologic stress. Consistent with the AGR
results, Q. phellos RGR was similar in AMB and SAT, while Q. bicolor and Q. palustris
had slightly lower RGR in SAT than AMB. This again suggests that Q. phellos may be
more tolerant of hydrologic stress than Q. bicolor and Q. palustris.
Betula nigra, L. styraciflua, Q. phellos and S. nigra saplings in SAT initially had
lower initial RGR than those in the AMB. After 6 years, RGR in SAT was greater than
saplings in the AMB for these species. This suggests their growth efficiency was initially
higher in AMB and decreased rapidly as biomass increased compared to the saplings
grown in SAT (slower decrease in RGR). Similar overlap has been found for grass
species grown with and without parasitic plants (Hautier et al. 2010). While not the focus
of this study, the rapid decreases in growth efficiency in AMB (dropping below SAT),
may suggest that a resource(s) is becoming limited in AMB. This resource limitation may
be the result of competition similar to the findings of Larocque and Marshall (1993).
Further investigations are needed to elucidate the effect of competition on sapling RGR.
233
Conclusions
Those species that have greater AGR and RGR under hydrologic stress (L.
styraciflua, S. nigra, and Q. phellos) may be better suited for planting in restored
wetlands as they are able to effectively and efficiently accumulate biomass. These species
have morphological, physiological, and life history traits (e.g. shallow roots, adventitious
roots, hypertrophied lenticels, and seed germination under water) which allow them to
grow in soils where oxygen may be limited (Hook 1984).
The novel growth rates presented here can be used to accurately determine woody
tissue carbon accumulation rates (from concentrations in Chapter 3). This study illustrates
the importance of calculating AGR and RGR across a range of biomass amounts as both
of these rates change as saplings grow and suggests that further studies are needed to
investigate how resource limitation can influence these rates.
234
Literature Cited
Anderson, C., & Mitsch, W. (2008). Tree Basal Growth Response to Flooding in a
Bottomland Hardwood Forest in Central Ohio. JAWRA Journal of the American
Water Resources Association, 44(6), 1512-1520.
Bazzaz, F. (1979). The physiological ecology of plant succession. Annual Review of
Ecology and Systematics, 10, 351-371.
Blackman, V. (1919). The compound interest law and plant growth. Annals of Botany(3),
353.
Britt, J. R., Mitchell, R. J., Zutter, B. R., South, D. B., Gjerstad, D. H., & Dickson, J. F.
(1991). The influence of herbaceous weed control and seedling diameter on six
years of loblolly pine growth--a classical growth analysis approach. Forest
Science, 37(2), 655-668.
Burns, R., & Honkala, B. (1990). Silvics of North America, Volume 2, Hardwoods.
Agriculture Handbook (Washington).
Causton, D. & Venus, J. 1981. The Biometry of Plant Growth. Edward Arnold. London.
Das, A. (2012). The effect of size and competition on tree growth rate in old-growth
coniferous forests. Canadian Journal of Forest Research, 42(11), 1983-1995.
Day, R., Doyle, T., & Draugelis-Dale, R. (2006). Interactive effects of substrate,
hydroperiod, and nutrients on seedling growth of Salix nigra and Taxodium
distichum. Environmental and Experimental Botany, 55(1-2), 163-174.
Dickinson, S. B. (2007). Influences of soil amendments and microtopography on
vegetation at a created tidal freshwater swamp in southeastern Virginia. Master's
thesis, Virginia Polytechnic Institute and State University.
Donovan, L., McLeod, K., Sherrod, K., & Stumpff, N. (1988). Response of woody
swamp seedlings to flooding and increased water temperatures. I. Growth,
biomass, and survivorship. American Journal of Botany, 75(8), 1181-1190.
Evans, G. C. (1972). The Quantitative Analysis of Plant Growth. Berkley and Los
Angeles: University of California Press.
Farmer Jr., R. E. (1980). Comparative analysis of first-year growth in six deciduous tree
species. Canadian Journal of Forestry, 10, 35-41.
235
Hautier, Y., Hector, A., Vojtech, E., Purves, D., & Turnbull, L. A. (2010). Modelling the
growth of parasitic plants. Journal of Ecology, 98(4), 857-866.
Hoffmann, W., & Poorter, H. (2002). Avoiding bias in calculations of relative growth
rate. Annals of Botany, 90(1), 37.
Hook, D. (1984). Waterlogging tolerance of lowland tree species of the South. Southern
Journal of Applied Forestry, 8(3), 136-149.
Hunt, R. (1982). Plant Growth Curves - The functional approach to plant growth
analysis. Edward Arnold.
Hunt, R. (1990). Basic growth analysis - Plant growth analysis for beginners. Broadwick
Street, London: Unwin Hyman, Ltd.
Jørgensen, S. E., Patten, B. C., & Straškraba, M. (2000). Ecosystems emerging:: 4.
growth. Ecological Modelling, 126(2), 249-284.
Kabrick, J. M., Dey, D. C., Van Sambeek, J., Wallendorf, M., & Gold, M. A. (2005). Soil
properties and growth of swamp white oak and pin oak on bedded soils in the
lower Missouri River floodplain. Forest Ecology and Management, 204(2), 315327.
Keeland, B., Conner, W., & Sharitz, R. (1997). A comparison of wetland tree growth
response to hydrologic regime in Louisiana and South Carolina. Forest Ecology
and Management, 90(2-3), 237-250.
Kozlowski, T. (1984). Responses of woody plants to flooding. In T. Kozlowski (Ed.),
Flooding and Plant Growth. Academic Press, Inc.
Larocque, G. R., & Marshall, P. L. (1993). Evaluating the impact of competition using
relative growth rate in red pine (Pinus resinosa Ait.) stands. Forest ecology and
management, 58(1), 65-83.
MacFarlane, D., & Kobe, R. (2006). Selecting models for capturing tree-size effects on
growth resource relationships. Canadian Journal of Forest Research, 36(7), 16951704.
Malecki, R., Lassoie, J., Rieger, E., & Seamans, T. (1983). Effects of long-term artificial
flooding on a northern bottomland hardwood forest community. Forest Science,
29(3), 535-544.
236
Matsushita, M., Takata, K., Hitsuma, G., Yagihashi, T., Noguchi, M., Shibata, M., et al.
(2015). A novel growth model evaluating age--size effect on long-term trends in
tree growth. Functional Ecology.
McCurry, J., Gray, M., & Mercker, D. (2010). Early Growing Season Flooding Influence
on Seedlings of Three Common Bottomland Hardwood Species in Western
Tennessee. Journal of Fish and Wildlife Management, 1(1), 11-18.
McLeod, K., & McPherson, J. (1973). Factors limiting the distribution of Salix nigra.
Bulletin of the Torrey Botanical Club, 100(2), 102-110.
Megonigal, J., & Day, F. (1992). Effects of flooding on root and shoot production of bald
cypress in large experimental enclosures. Ecology, 73(4), 1182-1193.
Mitsch, W., & Ewel, K. (1979). Comparative biomass and growth of cypress in Florida
wetlands. American Midland Naturalist, 101(2), 417-426.
Mitsch, W., & Rust, W. (1984). Tree growth responses to flooding in a bottomland forest
in northeastern Illinois. Forest Science, 30(2), 499-510.
Muller-Landau, H. C., Condit, R. S., Chave, J., Thomas, S. C., Bohlman, S. A.,
Bunyavejchewin, S., et al. (2006). Testing metabolic ecology theory for
allometric scaling of tree size, growth and mortality in tropical forests. Ecology
Letters, 9(5), 575-588.
Paine, C. E., Marthews, T., Vogt, D., Purves, D., Rees, M., Hector, A., et al. (2012). How
to fit nonlinear plant growth models and calculate growth rates: an update for
ecologists. Methods in Ecology and Evolution, 3(2), 245-256.
Paul-Victor, C., Züst, T., Rees, M., Kliebenstein, D. J., & Turnbull, L. A. (2010). A new
method for measuring relative growth rate can uncover the costs of defensive
compounds in Arabidopsis thaliana. New Phytologist, 187(4), 1102-1111.
Pennington, M. R., & Walters, M. B. (2006). The response of planted trees to vegetation
zonation and soil redox potential in created wetlands. Forest Ecology and
Management, 233(1), 1-10.
Philipson, C. D., Saner, P., Marthews, T. R., Nilus, R., Reynolds, G., Turnbull, L. A., et
al. (2012). Light-based Regeneration Niches: Evidence from 21 Dipterocarp
Species using Size-specific RGRs. Biotropica, 44(5), 627-636.
237
Pinheiro, J. C., & Bates, D. M. (2000). Mixed-effects models in S and S-PLUS. Springer.
Pommerening, A., & Muszta, A. (2015). Methods of modelling relative growth rate.
Forest Ecosystems, 2(1), 1-9.
Pommerening, A., & Muszta, A. (2016). Relative plant growth revisited: Towards a
mathematical standardisation of separate approaches. Ecological Modelling, 320,
383-392.
R Core Team (2014) R: a language and environment for statistical computing. R
Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0.
http://www.R-project.org/ (accessed 8 Jan 2016)
Rees, M., Osborne, C. P., Woodward, F. I., Hulme, S. P., Turnbull, L. A., & Taylor, S. H.
(2010). Partitioning the components of relative growth rate: how important is
plant size variation? The American Naturalist, 176(6), E152--E161.
Rose, K. E., Atkinson, R. L., Turnbull, L. A., & Rees, M. (2009). The costs and benefits
of fast living. Ecology Letters, 12(12), 1379-1384.
Shimojo, M., Asano, Y., Ikeda, K., Ishiwaka, R., Shao, T., Sato, H., et al. (2002). Basic
Growth Analysis and Symmetric Properties of Exponential Function with Base e.
J. Fac. Agr., 47(1), 55-60.
Sillett, S., Van Pelt, R., Koch, G., Ambrose, A., Carroll, A., Antoine, M., et al. (2010).
Increasing wood production through old age in tall trees. Forest Ecology and
Management, 259(5), 976-994.
Smith, M. C., Anthony Stallins, J., Maxwell, J. T., & Van Dyke, C. (2013). Hydrological
shifts and tree growth responses to river modification along the Apalachicola
River, Florida. Physical Geography, 34(6), 491-511.
South, D. (1991). Testing the hypothesis that mean relative growth rates eliminate sizerelated growth differences in tree seedlings. New Zealand Journal of Forestry
Science, 21(2/3), 144-164.
South, D. (1995). Relative Growth Rates: A Critique. South African Forest Journal, 173,
43-48.
Stephenson, N., Das, A., Condit, R., Russo, S., Baker, P., Beckman, N., et al. (2014).
Rate of tree carbon accumulation increases continuously with tree size. Nature,
507, 90-93.
238
Tang, Z. C., & Kozlowski, T. T. (1982). Physiological, morphological, and growth
responses of Platanus occidentalis seedlings to flooding. Plant and Soil, 66,
243:255.
Tang, Z. C., & Kozlowski, T. T. (1982). Some physiological and growth responses of
Betula papyrifera seedlings to flooding. Physiologia Plantarum, 55(4), 415-420.
Turnbull, L. A., Paul-Victor, C., Schmid, B., & Purves, D. W. (2008). Growth rates, seed
size, and physiology: do small-seeded species really grow faster. Ecology, 89(5),
1352-1363.
Turnbull, L. A., Philipson, C. D., Purves, D. W., Atkinson, R. L., Cunniff, J.,
Goodenough, A., et al. (2012). Plant growth rates and seed size: a re-evaluation.
Ecology, 93(6), 1283-1289.
Whitlow, T. H., & Harris, R. W. (1979). Flood tolerance in plants: a state-of-the-art
review. Technical Report E-79-2, Department of Environmental Horticulture,
University of California.
239
Tables
Table 4-1. Initial biomass with standard deviation in parentheses. Same letters represent
no significant difference (p<0.05). N represents number of planted saplings.
Species
Betula nigra
Liquidambar styraciflua
Platanus occidentalis
Quercus bicolor
Quercus palustris
Quercus phellos
Salix nigra
N
339
332
264
361
342
378
351
Initial Biomass (g)
30 (42) c
31 (46) c
36 (58) abc
43 (58) b
43 (55) ab
50 (72) b
31 (50) ac
240
Quercus bicolor
Quercus palustris
Quercus phellos
Secondary Species
1
1
1
Mass (kg)
11
Betula nigra
11
Liquidambar styraciflua
11
Platanus occidentalis
11
Salix nigra
Primary Species
AMB
1.8 (1.6-1.9)
1.6 (1.3-1.7)
2.0 (1.8-2.2)
AGR (g/day)
26 (24-29)
23 (21-26)
36 (34-38)
15 (11-18)
1.8 (1.6-1.9)
1.6 (1.4-1.7)
2.0 (1.9-2.1)
1
1
1
RGR (g/g/day) Mass (kg)
11
2.4 (2.2-2.5)
11
2.1 (1.9-2.2)
11
3.1 (3.0-3.2)
11
1.3 (1.1-1.5)
SAT
1.4 (1.0-1.5)
1.3 (0.9-1.5)
2.0 (1.7-2.1)
AGR (g/day)
26 (22-28)
34 (28-37)
22 (17-26)
27 (20-31)
FLD
1.4 (1.2-1.5)
1.3 (1.1-1.4)
2.0 (1.8-2.1)
0.1
0.1
0.1
0.09 (0-0.10) 0.92 (-0.64-0.98)
0.08 (0.01-0.1) 0.80 (0.30-0.90)
0.14 (0-0.15) 1.4 (-0.10-1.5)
RGR (g/g/day) Mass (kg) AGR (g/day) RGR (g/g/day)
0.12 (0.10-0.14) 1.2 (1.1-1.3)
0.1
2.3 (2.2-2.5)
1.4 (-1.0-1.5)
0.14 (0-0.16)
0.1
3.0 (2.8-3.2)
0.10 (0.03-0.11) 0.99 (0.54-1.1)
0.1
2.0 (1.7-2.1)
0.27 (0.25-0.30) 2.8 (2.7-2.9)
0.1
2.5 (2.2-2.7)
Table 4-2. Species AGR and RGR at standard masses. Values in parentheses represent
±95% confidence intervals.
241
Figures
Figure 4-1. Average predicted biomass of seven species (columns) in each cell (rows).
Shading represents 95% confidence intervals around mean.
242
Figure 4-2. Natural log transformed biomass of seven species (columns) in each cell
(rows). Red line represents the results of fitting the monomolecular function.
243
Figure 4-3. Absolute growth rate (AGR) of seven species (columns) in each cell (rows) as
a function of mass. Shading represents 95% confidence intervals around mean.
244
Figure 4-4. Relative growth rate (RGR) of seven species (columns) in each cell (rows) as
a function of mass. Shading represents 95% confidence intervals around mean.
245
CHAPTER 5: DEVELOPING AN ECOLOGICAL PERFORMANCE STANDARD
FOR WOODY VEGETATION IN COMPENSATORY MITIGATION
WETLANDS OF VIRGINIA
246
Abstract
Compensatory wetland mitigation sites (CMS) are required to replace ecological
structures, functions, and services lost during permitted impacts. Meeting established
ecological performance standards (EPS) is the current legislative-mandated method for
determining if CMS are meeting this goal. However, existing EPSs often are not adequate
measures of ecological functions and services. The purpose of this study was to develop a
woody EPS for use in forested CMS in Virginia that is a suitable indicator of woody
biomass accumulation, an important ecosystem function. The existing woody EPSs in
Virginia (stem density, height growth, canopy cover) were found to be inadequate
representations of sapling biomass accumulation. Using data from the previous chapters,
a minimum stem cross-sectional area measured at groundline (CSAG) was recommended
as an additional EPS. The recommended EPS is that CMS have greater than 5.2 m2/ha
CSAG at the end of the 5th year following construction. Sapling CSAG provides a better
representation of sapling biomass accumulation than the existing EPSs in Virginia and is
more appropriate to determine whether forested CMS are reaching their intended goal of
replacing lost ecosystem structure, functions and services.
247
Introduction
Wetlands are restored for a variety of reasons, including but not limited to:
compensation for permitted impacts to existing wetlands (wetland compensatory
mitigation); state and federal goals (Chesapeake 2000 Agreement); re-establishing bird
habitat (Ducks Unlimited); and agricultural easements (Agricultural Conservation
Easement Program). Despite varying motivations for restoration, projects implicitly seek
to return lost ecological structures, functions and services to the landscape. Wetlands
restored to compensate for permitted impacts often have additional goals they are
required to meet as a result of the legal framework they fall within.
Wetland impacts are regulated by the 1972 Federal Water Pollution Control Act,
which was amended and renamed the Clean Water Act (CWA) in 1977 (33 U.S.C. 1251
et seq.). Wetland impacts are specifically regulated under Section 404 of the CWA (33
U.S.C 1344) which authorized the United States Army Corps of Engineers (USACE)
under the direction of the United States Environmental Protection Agency (USEPA) to
issue permits regulating the discharge of dredged or fill material into “waters of the
United States”, which include wetlands (40 CFR Part 230.1).
Mitigation was first defined in the 1978 Council on Environmental Quality (CEQ)
Regulations for Implementing the Procedural Provisions of the National Environmental
Policy Act (NEPA – 42 U.S.C 4321) (40 CFR 1500-1508). In this context, mitigation
refers to avoidance, minimization, rectification, reduction over time, or compensation for
impacts to any type of environment. The wetland mitigation sequence, in descending
order of priority, emphasizes avoidance, minimization and compensation for wetland
impacts and was established in the 1990 Memorandum of Agreement between the
248
Environmental Protection Agency and the Department of the Army. This sequence was
devised to allow for permitted wetland impacts while maintaining the physical, biological
and chemical integrity of the waters of the United States and remaining in compliance
with the CWA. If impacts to existing wetlands are determined unavoidable, four methods
of compensation have been deemed acceptable: 1) restoration of converted wetlands, 2)
creation of new wetlands, 3) enhancement of existing wetlands, 4) preservation of
existing wetlands (listed in descending order of priority (33 CFR PART 332, USACE and
USEPA 2008).
Ecological Performance Standards
Wetlands that are created, restored, enhanced, or preserved to offset permitted
impacts are often referred to as compensatory mitigation sites (CMS). These CMS are
required to replace ecological structures, functions and services lost during permitted
impacts (33 CFR PART 332.3). The current legislative-mandated method for determining
if a CMS is developing into the desired wetland type and is providing the expected
ecological functions is through meeting project specific ecological performance standards
(EPS) (aka: success criteria, success standards or release criteria) (33 CFR PART 332.5,
USACE and USEPA 2008). Performance standards for all CMS are required to be clear,
objective, verifiable, based on the best available science and able to be assessed in a
practicable manner (33 CFR PART 332.5, USACE and USEPA 2008).
Ecological performance standards are developed on a project-by-project basis
(USACE and USEPA 2008). While EPS vary across USACE districts, the majority are
based on soil, vegetation, and hydrologic indicators from the 1987 Federal Delineation
249
Manual (Environmental Laboratory 1987) and subsequent editions (Breaux and
Serefiddin 1999, Streever 1999, Matthews and Endress 2008, USACE 2008). Additional
EPSs may include survival of planted stock, specific density (or cover) of herbaceous or
woody plants, and limitation of exotic and nuisance plants (Breaux and Serefiddin 1999,
Streever 1999, Matthews and Endress 2008, USACE 2008).
Most EPSs are structural measurements of the vegetative community and/or
physical environment (Wilson and Mitsch 1996) and are not direct measurements of
wetland functions or services (Mitsch and Wilson 1996, Streever 1999, NRC 2001).
Additionally, many EPSs are not adequate indicators of wetland functions or services
(Kentula et al. 1992). Cole (2002) suggested that measurement of herbaceous cover (a
common EPS) may not serve as an accurate indicator of the replacement of wetland
functions. Therefore, meeting site specific EPSs does not guarantee that wetland
functions and services are being replaced (the overall goal of compensatory mitigation)
(Matthews and Endress 2008).
Virginia EPS
In Virginia, the USACE Norfolk District and the Virginia Department of
Environmental Quality (VADEQ) (2004) provide recommended hydrologic, soil, and
vegetation EPSs for CMS. The VADEQ (2010) also provides a template outlining
recommended EPSs specifically for mitigation banks. However, these only serve as
guidelines and specific EPSs are developed for each project. These EPSs are designed to
ensure that the CMS qualifies as a wetland using the indicators from the 1987 USACE
Wetland Delineation Manual (and subsequent supplements) (Environmental Laboratory
250
1987). To that end, the hydrologic and soil EPS require satisfying their respective criteria
in the 1987 USACE Wetland Delineation Manual. Monitoring and compliance reports are
required and the current timing in Virginia is to monitor for six years over a 10 year
period (typically years 1, 2, 3, 5, 7, and 10) (USACE Norfolk District and VADEQ
2004).
Vegetation EPSs are separated into requirements for herbaceous and woody
strata. For both strata, 50% of all the dominant plant species are required to be
Facultative (FAC, occurs in wetland and non-wetlands) Facultative Wetland (FACW,
usually occurs in wetlands) or Obligate (OBL, almost always occurs in wetlands)
(Lichvar 2013). Additionally, areal herbaceous coverage greater than 50% is required in
emergent wetland areas after one growing season. Undesirable species (including
invasive species and other “weedy” species that may compete with planted saplings) in
both strata are required to be quantified and controlled if necessary (USACE Norfolk
District and VADEQ 2004).
Specific survival rates of planted herbaceous and woody vegetation may be
required in some CMS in Virginia. The particular percentage survival required varies by
project; however, planted woody vegetation can be required to exceed 80% survival
(Mike Rolband, Personal Communication). In the woody stratum, a stem density of both
naturally colonizing and planted trees of 495-990 stems/ha is required until the canopy
cover is 30% or greater (USACE Norfolk District and VADEQ 2004). However, in
practice most CMS in Virginia are required to have greater than 990 stems/ha (Mike
Rolband, Personal Communication).
251
Possibly because of limited information on growth rates of saplings in created or
restored wetlands, few states have established EPSs related to sapling growth in CMS
(Denton 1990, Niswander and Mitsch 1995, Streever 1999, Gamble and Mitsch 2006,
Pennington and Walters 2006). Virginia is unique in that it has established a height
growth EPS for mitigation banks in particular. This EPS states that until the canopy
coverage exceeds 30%, average height of all woody stems (including planted and
colonizing trees) in each cell, field or block must have an average increase in height of
10% by both the 5th and 10th year following construction. An alternative method for
meeting this EPS requires average tree height must be ≥ 1.5 m in the 5th year following
construction and 3.0 m in the 10th year following construction (VADEQ 2010). However,
few projects have implemented this height-growth EPS (Mike Rolband, personal
communication).
In Virginia, the majority of wetland impacts are to non-tidal forested wetlands due
to their geographic extent and drier hydrologic conditions (Tiner and Finn 1986, USGS
1999, Tiner et al. 2005, VADEQ 2014). Subsequently forested wetlands have the most
compensation area (VADEQ 2014). Therefore, it is important that the woody vegetation
EPS in particular be a direct measure of, or proven indicator of, ecological functions that
can be compared to those found in natural (reference) systems. They must be quantifiable
(in a clear, repeatable, and rapid manner) and ecologically relevant to the region.
The purpose of this study is to determine if the existing woody vegetation EPSs in
Virginia are related to woody biomass accumulation (ecological function) and to propose
additional woody vegetation EPSs that are representative of this ecological function.
252
Methods
Methods used to establish and measure the survival and morphology of saplings
planted in the experimental site are the same as Chapter 2.
Existing EPS Relationship to Biomass Accumulation
Species specific biomass estimation models (BEM) were used to determine the
total biomass of saplings living after 7 years. The BEMs were developed in Chapter 3 and
relate ESD to total sapling biomass (aboveground biomass (AGB) and belowground
biomass (BGB)). The woody EPSs for Virginia were directly compared to planted
sapling biomass accumulation.
Development of Additional Woody EPS
The relationship between ESD and H and ESD and CD for saplings living after 7
years was determined using simple linear regressions. Individual relationships were
developed for each species. An additional EPS was developed using stem cross-sectional
area at groundline (CSAG).
Results
Sapling Survival EPS (>80% survival) and Stem Density EPS (>990 stems/ha)
After 7 years average percentage survival and stem density in SAT was the
highest (66.4% - 1140 stems/ha) followed by AMB (59.0% - 1012 stems/ha) and FLD
(27.3% - 449 stems/ha). Greater survival and higher stem density in SAT did not result in
biomass greater than AMB (Table 5-1). There was a positive linear relationship between
253
survival and biomass accumulated after 5 years in the FLD (r2 = 0.253, p-value = 0.02),
but not in the AMB (r2 = 0.035, p-value = 0.204) or SAT (r2 = 0.003, p-value = 0.8)
(Figure 5-1).
Height Growth EPSs (>10%/year or > 1.5 m after 5 years)
After 5 years the average percentage change in average height was greatest in
AMB (54.4%), followed by SAT (43.7%) and FLD (8.8%). Similarly, the average height
after 5 years was greatest in AMB (4.8 m) followed by SAT (3.2 m) and FLD (0.9 m).
FLD was the only cell that did not satisfy both height growth EPSs. Greater percentage
change in height and greater average height in AMB did reflect greater biomass than SAT
or FLD (Table 5-1). For example the AMB had 54.4% change in height and 38.8 Mg/ha
of biomass, while FLD had 8.8% change in height and only 0.55 Mg/ha. There was a
positive linear relationship between the 5-year average percentage change in height and
biomass accumulated after 5 years in the AMB (r2 = 0.376, p-value = 0.003), but not in
the SAT (r2 = 0.426, p-value = 0.396) or FLD (r2 = 0.039, p-value = 0.393) (Figure 5-2).
Canopy Cover EPS (>30% Canopy Closure)
If tree canopy cover exceeds 30%, stem density and height growth EPSs are no
longer required (USACE Norfolk District 2004, VADEQ 2010). In AMB and SAT cells,
30% canopy closure occurred in years 3 and 4 respectively. After 7 years, AMB had the
greatest canopy coverage (165.4%), followed by SAT (120.6%) and FLD (4.9%). Values
can exceed 100% due to overlapping canopies. The FLD treatment failed to satisfy the
30% canopy closure EPS. The greater canopy cover in AMB corresponded to greater
254
biomass than SAT or FLD (Table 5-1). There was a significant positive relationship
between percentage canopy cover and average biomass accumulated after 7 years in the
AMB (r2 = 0.492, p-value < 0.001), SAT (r2 = 0.442, p-value = 0.001 ), and FLD (r2 =
0.353, p-value = 0.005 ) (Figure 5-3).
Development of Additional Woody EPS
Equivalent stem cross-sectional diameter at groundline (ESD) was found to have
a strong relationship with total AGB and BGB accumulated in individual saplings
(Chapter 3). Additionally, there were statistically significant positive relationships (pvalue < 0.001) between ESD and H for PRI (Figure 5-4) and SEC (Figure 5-5) groups.
There were statistically significant positive relationships (p-value <0.001) between ESD
and CD for PRI (Figure 5-6) and SEC (Figure 5-7) species groups. Quercus bicolor
tended to have the lowest coefficient of determination for both relationships which may
indicate non-linear relationships in life history strategies and resource allocation patterns.
Stem cross-sectional area at groundline (CSAG) is equivalent to ESD, requires
fewer steps to calculate and can be summed across areas (m2/unit area). These
characteristics suggest that total CSAG/unit area is a robust and appropriate EPS because
it provides a strong indication of an ecological function (biomass accumulation) and can
be compared to a natural (reference) wetland.
Total CSAG was greatest in AMB followed by SAT and then FLD at the end of 7
years (Table 5-2). Based on the total CSAG /unit area results from FLD, an appropriate
minimum CSAG EPS would be greater than 5.2 m2/ha at the end of the 5th year
following construction (Table 5-2). This minimum CSAG EPS corresponds to ~3%
255
canopy cover and 0.55 Mg/ha total biomass in FLD. Based on the results from SAT, an
appropriate maximum CSAG EPS would be 70.1 m2/ha by the 5th year. This corresponds
to a stage in site development in which 30% canopy closure occurred and when total
biomass reached 14.35 Mg/ha. Additional EPSs could be developed based on CSAG
from other points in time during the study.
Discussion
The existing EPSs are not strong predictors of sapling biomass accumulation,
which is an important ecosystem function necessary for successful forested headwater
CMS. These existing EPS may be useful predictors/indicators of other ecosystem
functions and sapling characteristics not investigated here. However, the recommended
CSAG has the potential to be a more robust EPS for reasons discussed below.
Sapling Survival and Stem Density EPS
The relationship between sapling survival (or stem density) and average biomass
accumulated was significantly positive in the FLD cell only. However, the percentage of
the biomass accumulation variation explained in the FLD by the linear model was low
(r2=0.253). The slope of the regression line in the AMB and SAT was not significantly
different than zero and the amount of variation explained by the linear model was low in
both cells. An example from AMB and SAT can illustrate what this low amount of
variation explained represents. Two species/stocktypes combinations had 100% survival
in AMB (B. nigra GAL and Q. bicolor GAL) and two combinations had 100% survival
in SAT (L. styraciflua GAL and Q. bicolor GAL). The average biomass accumulated by
256
Q. bicolor in AMB and SAT were 5.25 kg and 5.80 kg respectively; the average biomass
accumulated by B. nigra and L. styraciflua was 92.43 kg and 50.58 kg respectively. This
large variation in biomass may result from differences in species life history traits among
primary and secondary species, as described by Bazzaz (1979), and may also explain why
survival was not correlated with biomass accumulation. Additionally, high stem densities
with low biomass may result from saplings that grow tall with little accumulation in the
main stem or saplings that die back and re-sprout. Several of the individual plantings in
this study expressed these growth habits. While survival and stem density are important
metrics to track during restoration, the CSAG EPS does take into account these variables
and would indicate if survival and stem density were decreasing through time. Overall,
survival and stem density are not strongly related to biomass accumulation and may not
be appropriate EPSs.
Height Growth EPSs
The significant positive relationship between average percentage change in height
and average biomass in AMB and the greater amount of biomass variation explained by
the linear model (r2=0.376) suggests that height growth is a more suitable EPS than
survival and stem density in less stressful sites with more sapling growth. However, in
the SAT and FLD, there was not a significant relationship between height growth and
biomass accumulation. This suggests that in sites with moderate or high hydrologic stress
where growth is limited, height growth is not a good predictor of biomass accumulation.
While, height growth is relatively easy to quantify, the existing calculations hide
variation associated with individual growth and have the potential to undervalue larger
257
trees (with slower height growth) or trees that follow different growth patterns. For
example, S. nigra tend to be multi-stemmed and expand their crown diameter as opposed
to accumulating height. The height growth calculation does not account for this
difference in resource allocation and would not provide accurate representations of
biomass accumulation for this species. Additionally, measuring heights of individual
saplings in CMS is challenging when the saplings are large and the canopies are dense
(Mike Rolband, personal communication; and anecdotal evidence from this study).
The positive relationship between ESD and H show there is support for the
indirect connection between this EPS and biomass accumulation. However, the more
robust relationship between ESD and biomass suggest that using CSAG (equivalent to
ESD) would provide a stronger measure of ecosystem function than height.
Canopy Cover EPS
Canopy cover can be quantified using a variety of techniques (visual estimation,
spherical densiometer, hemispherical photography, remote sensing, etc.). Standardized
procedures that can be consistently and efficiently applied are necessary for an
appropriate EPS. In this study, canopy cover was quantified by taking three
measurements of crown diameter of individual saplings. While time consuming, this
allows for a robust and unbiased estimation of canopy cover.
The significant positive relationship between canopy cover and biomass
accumulation in all cells and the greater amount of biomass variation explained by the
linear model suggest that canopy cover is more appropriate than both height growth and
sapling survival. A positive relationship between crown diameter and ESD suggests that
258
crown diameter has an indirect weak relationship with biomass accumulation. Canopy
cover has been used as an indicator of other ecological functions or overall ecosystem
development (Rheinhardt et al. 2009). Therefore, of the three existing metrics it has the
best support as an appropriate EPS for use in CMS but standardized procedures that are
rapid and reproducible are needed.
Development of Additional Woody EPS
The recommended CSAG EPS for headwater wetland CMS in VA is that CSAG
should exceed 5.2 m2/ha after 5 years and 6.4 m2/ha after 7 years. By measuring the
CSAG earlier than 5 years, it is possible to determine if a CMS is developing toward
meeting this EPS and taking corrective actions if it is not.
Stem cross-sectional area at groundline is the most appropriate measure to be used
for an EPS for several reasons. The most compelling reason is that ESD (or CSAG) had
significant positive relationships explaining a large amount of the variation associated
with total biomass accumulation (AGB and BGB) for all species used in this study
(Chapter 3).
The relationship between CSAG and biomass is particularly important, because
woody biomass (and by extension of the relationship, CSAG) has been shown to be a
good indication that other wetland functions and services are occurring. For example,
when investigating six wetland functions, Cole (2002) suggested that stem area may be a
better predictor of six main wetland ecosystem functions than percent herbaceous
vegetation cover.
259
In addition to the relationship with biomass, ESD (or CSAG) had significant
positive relationships with H and CD for all species in this study with high coefficients of
determination. Other studies have suggested that measures of stem diameter (diameter at
breast height (dbh) in particular) have positive relationship with other morphological
measurements (H, trunk taper, volume, stem surface area, leaf surface, etc.) (Whittaker
and Woodwell 1968, Niklas 1995). Quercus bicolor had the lowest coefficient of
determination when relating ESD to H (r2 = 0.5563) and CD (r2 = 0.6115) of all seven
species. This suggests that the relationship between these morphologies for Q. bicolor
may be related in a non-linear fashion. If this is the case it suggests that H and CD
increase for Q. bicolor was rapid relative to ESD (taller and wider crowns, with smaller
stems) initially and as ESD increases the rate of H and CD growth decreases (stems
become larger, while H and CD slowly increase). Therefore, Q. bicolor may have a
different growth strategy (e.g. allocating more resources to height initially and then
allocating more resources to the main stem) than the other Quercus species used in this
study, similar to the results of Guyette et al. (2004).
Sapling GA also has methodological advantages compared to the existing EPSs.
Measuring the stem diameter at the groundline allows for the inclusion of saplings that
have not yet attained a height of 1.37 (common height for dbh measurements). When
trees achieve 1.37 m in height and before they develop a distinct main stem (trunk) it is
difficult to determine where to measure dbh, which is not an issue with CSAG. However,
using GA precludes the use of traditional comparisons of forests and wetlands that
mainly focus on dbh. Measuring dbh on multi-stem saplings presents even more of a
problem since energy needed for height growth is usually re-directed to multiple stem
260
growth. However, measuring groundline diameter is simple for multi-stemmed saplings
since each stem that originates belowground is measured individually.
A challenge associated with measuring groundline diameter is root swelling and
buttressing. It is recommended that measurements be taken directly above any root
swelling or buttressing (defined as stem diameter > 10% larger than stem above
swelling), where the stem diameter becomes constant. Irregularly shaped stems should be
measured using a dbh tape or by averaging multiple groundline diameter measurements.
Measurements should be made in the fall, after the stems have completed the majority of
their growth.
The recommended CSAG EPS should be used in conjunction with the existing
woody EPS to confirm that this new standard is adequately representing the other EPS. If
this CSAG EPS is found to be effective, it has the potential to eliminate some or all of the
existing woody EPSs in Virginia. Since this is the first attempt at developing a CSAG
EPS, additional measurements in restored/created forested CMS wetlands are needed to
confirm that it is set at an appropriate level. Total CSAG can also be compared to natural
(reference wetlands); however, most stem area measurements in natural sites are
determined at breast height (basal area). Additionally, CSAG (or ESD) can be used to
estimate total biomass for comparisons or calculation of carbon accumulation, another
important ecosystem function (Chapter 3).
In conclusion, CSAG is easily measured in the field, easily calculated in the
office, strongly related to biomass accumulation, related to other morphological
measures, comparable to natural reference wetlands and should be considered as an
additional EPS in Virginia. Inclusion of this metric will hopefully lead to better
261
evaluation of new and existing forested headwater CMS and ultimately more successful
restorations/creations of this important ecosystem.
262
Literature Cited
Bailey, D., Perry, J., & Daniels, W. (2007). Vegetation dynamics in response to organic
matter loading rates in a created freshwater wetland in southeastern Virginia.
Wetlands, 27(4), 936-950.
Bazzaz, F. (1979). The physiological ecology of plant succession. Annual Review of
Ecology and Systematics, 10, 351-371.
Breaux, A., & Serefiddin, F. (1999). Validity of performance criteria and a tentative
model for regulatory use in compensatory wetland mitigation permitting.
Environmental Management, 24(3), 327-336.
Cole, C. (2002). The assessment of herbaceous plant cover in wetlands as an indicator of
function. Ecological Indicators, 2(3), 287-293.
Council on Environmental Quality (CEQ) (1978). National Environmental Policy Act –
Regulation: Implementation of the Procedural Provisions. Tech. rep., Council on
Environmental Quality.
Denton, S. (1990). Growth rates, morphometrics and planting recommendations for
cypress trees at forested mitigation sites. In F. Webb (Ed.), Proceedings of the
17th Annual Conference on Wetlands Restoration and Creation, (pp. 24-32).
Tampa, FL.
Environmental Laboratory. (1987). Corps of Engineers Wetlands Delineation Manual.
Vicksburg, MS.
Gamble, D. L., & Mitsch, W. J. (2006). Tree growth and hydrologic patterns in urban
forested mitigation wetlands. Tech. rep., Schiermeier Olentangy River Wetland
Research Park, School of Natural Resources, The Ohio State University.
Guyette, R. P., Rose-Marie, M., Kabrick, J., & Stambaugh, M. C. (2004). A Perspective
on Quercus Life History Characteristics and Forest Diturbance. In M. A. Spetich
(Ed.), Upland oak ecology symposium: history, current conditions, and
sustainability. Asheville, NC: U.S.
Kentula, M. E., Brooks, R. P., Gwin, S. E., Holland, C. C., Sherman, A. D., & Sifneos, J.
C. (1992). Wetlands: an approach to improving decision making in wetland
restoration and creation. (A. J. Hairston, Ed.) Island Press.
263
Lichvar, R. W. (2013). The National Wetland Plant List: 2013 Wetland Ratings.
Phytoneuron, 49, 1-241.
Lichvar, R. W., Melvin, N. C., Butterwick, M. L., & Kirchner, W. N. (2012). National
wetland plant list indicator rating definitions. Tech. rep., United States Army
Corps of Engineers, Engineer Research and Development Center.
Matthews, J., & Endress, A. (2008). Performance criteria, compliance success, and
vegetation development in compensatory mitigation wetlands. Environmental
Management, 41(1), 130-141.
Mitsch, W., & Wilson, R. (1996). Improving the success of wetland creation and
restoration with know-how, time, and self-design. Ecological Applications, 6(1),
77-83.
National Research Council (2001). Compensating for wetland losses under the Clean
Water Act. Washington, DC: National Academies Press.
Niklas, K. J. (1995). Size-dependent allometry of tree height, diameter and trunk-taper.
Annals of Botany, 75(3), 217-227.
Niswander, S., & Mitsch, W. (1995). Functional analysis of a two-year-old created instream wetland: hydrology, phosphorus retention, and vegetation survival and
growth. Wetlands, 15(3), 212-225.
Pennington, M. R., & Walters, M. B. (2006). The response of planted trees to vegetation
zonation and soil redox potential in created wetlands. Forest Ecology and
Management, 233(1), 1-10.
Rheinhardt, R., McKenney-Easterling, M., Brinson, M., Masina-Rubbo, J., Brooks, R.,
Whigham, D., et al. (2009). Canopy Composition and Forest Structure Provide
Restoration Targets for Low-Order Riparian Ecosystems. Restoration Ecology,
17(1), 51-59.
Streever, B. (1999). Examples of performance standards for wetland creation and
restoration in Section 404 permits and an approach to developing performance
standards. Tech. rep., United States Army Corps of Engineers, Wetlands
Research Program.
Tiner, R. W., & Finn, J. T. (1986). Status and recent trends of wetlands in five MidAtlantic states: Delaware, Maryland, Pennsylvania, Virginia, and West Virginia.
264
Tech. rep., United States Department of the Interior, Fish and Wildlife Service,
Fish and Wildlife Enhancement, National Wetlands Inventory Project, Newton
Corner, MA.
Tiner, R., Swords, J., & Bergquist, H. (2005). Recent Wetland Trends in Southeastern
Virginia: 1994-2000. NWI Wetland Trends Report, United States Fish and
Wildlife Service, Northeast Region, Hadley, MA.
United States Department of Agriculture (USDA) Natural Resources Conservation
Service (NRCS) Soil Survey Staff. Web Soil Survey. Accessed 2/23/2015
http://websoilsurvey.sc.egov.usda.gov
United States Environmental Protection Agency (USEPA) & United States Army. (1990).
Memorandum of Agreement; The Determination of Mitigation Under the Clean
Water Act Section 404(b)(1) Guidelines. Memorandum of Agreement, United
States Environmental Protection Agency and United States Army.
United States Geological Survey (USGS) (1999). National Water Summary - Wetland
Resources: Virginia. Supply Paper, United States Geological Survey, USGS
National Center, Reston, VA.
United States Army Corps of Engineers Norfolk District & Virginia Department of
Environmental Quality (2004). Recommendations for wetland compensatory
mitigation: Including site design, permit conditions, performance and monitoring
criteria. Recomendations.
United States Army Corps of Engineers (USACE) (2008). Final Environmental
Assessment, Finding of No Significant Impact, and Regulatory Analysis for the
Compensatory Mitigation Regulation. Tech. rep., Directorate of Civil Works
Operations and Regulatory Community of Practice.
United States Army Corps of Engineers, & United States Environmental Protection
Agency. (2008). Compensatory Mitigation for Losses of Aquatic Resources; Final
Rule. Tech. rep., United States Army Corps of Engineers and United States
Environmental Protection Agency, Washington D.C.
Virginia Department of Environmental Quality (VADEQ) (2010). Mitigation Banking
Instrument Template. Template, Virginia Department of Environmental Quality,
Richmond, VA.
265
Virginia Department of Environmental Quality (VADEQ) (2014). Virginia 305(b)/303(d)
Water Quality Integrated Report. Tech. rep., Virginia Department of
Environmental Quality, Richmond, VA.
Whittaker, R., & Woodwell, G. (1968). Dimension and production relations of trees and
shrubs in the Brookhaven Forest, New York. The Journal of Ecology, 56(1), 1-25.
Wilson, R., & Mitsch, W. (1996). Functional assessment of five wetlands constructed to
mitigate wetland loss in Ohio, USA. Wetlands, 16(4), 436-451.
266
Tables
Table 5-1. Total biomass (aboveground and belowground) accumulation (Mg/ha) of all
species/stocktype combinations across the 7-year study.
Ambient
Saturated
Flooded
2009
0.26
0.13
0.12
2010
1.73
0.47
0.37
2011
8.05
2.49
0.19
2012
18.49
7.52
0.30
2013
38.25
14.35
0.55
2014
42.77
19.25
0.42
2015
60.61
28.81
0.65
267
Table 5-2. Total CSAG (m2/ha) across the 7-year study.
Ambient
Saturated
Flooded
2009
2.9
1.7
1.6
2010
12.0
4.7
3.7
2011
39.6
18.7
2.2
2012
72.7
43.1
3.2
2013
125.1
70.1
5.2
2014
145.8
97.2
4.7
2015
187.3
129.6
6.4
268
Figures
Figure 5-1. Relationship between percentage survival and average biomass accumulated
after 7 years for individual species/stocktype combinations. Points represent
species/stocktype combinations. Line represents simple linear regression and shading
represents 95% confidence interval.
269
Figure 5-2. Relationship between 5-yr average percentage change in height and average
biomass after 5 years. Points represent species/stocktype combinations.
270
Figure 5-3. Relationship between canopy cover and average biomass after 7 years. Points
represent species/stocktype combinations. Line represents linear regression.
271
Figure 5-4. The relationship between ESD and H after 7 years for primary species. Line
represents linear regression.
272
Figure 5-5. The relationship between ESD (cm) and H (cm) after 7 years for secondary
species. Line represents linear regression.
273
Figure 5-6. The relationship between ESD (cm) and CD (cm) after 7 years for primary
species. Line represents linear regression.
274
Figure 5-7. The relationship between ESD (cm) and CD (cm) after 7 years for secondary.
Line represents linear regression.
275
CHAPTER 6: ECONOMIC ANALYSIS AND OVERALL CONCLUSIONS
276
Abstract
Ecological restoration is a multibillion dollar industry and research focusing on the
economics of restoration is needed to provide better prioritization of finite financial and
ecological resources. Maximizing the ecological benefits of forested wetland restoration
based on financial investments is necessary because of the costs required in restoring
these systems ($98,842/ha to $192,742/ha). The purpose of this study was to investigate
the amount of woody biomass of seven Mid-Atlantic wetland tree species (four early
successional, three late successional) returned after 5 years based on initial investment.
Trees were planted using three stocktypes (bare root, tubeling and 1-gallon container)
across a hydrologic gradient to investigate the reduction in planting cost that could be
achieved in restoring forested wetlands. The greatest biomass return on initial investment
was achieved by P. occidentalis tubelings (39.05 kg/$1) in the least hydrologically
stressed treatment. In the most stressful conditions (flooding above the root collar with
herbaceous competition) S. nigra and B. nigra had the greatest biomass returns on initial
investments (average 0.26 kg/$1 for all 3 stocktypes) and an initial density of 520
stems/ha is recommended for planting of these species. Of the species planted in this
study, S. nigra and B. nigra are the best choices for maximizing biomass accumulation
based on initial costs when restoring forested wetlands. In addition to these findings,
overall conclusions of this dissertation are provided in this chapter.
277
Economic Analysis
Economics of restoration is a relatively new area of study that is becoming more
widespread (Blignaut et al. 2014). Investigating restoration economics is necessary since
restoration has become a multibillion dollar industry (Woodworth 2006). This field
broadly encompasses ecosystem valuation, financing restoration, cost-benefit analysis of
restoration practices, and economic/socioeconomic effects of restoration. Studying the
economics of restoration can lead to better prioritization of finite resources (financial and
ecological), which ultimately helps achieve restoration goals.
Since there is finite capital available for each forested wetland restoration project,
managers must seek to maximize their ecological returns on their financial investments.
However, few studies provided information on the methods that will restore ecosystems
at the lowest cost (Robbins and Daniels 2012). Substantial financial investments are
required when restoring forested wetlands, that vary substantially based on site specific
conditions and goals (King and Bohlen 1994, 1995). Cost estimates are difficult to
acquire because restoration work is primary performed by consultants who publish
infrequently or view this information as proprietary (Holl and Howarth 2000, Robbins
and Daniels 2012). Published total costs to restore 1 ha of freshwater forested wetland
range from $98,842/ha to $192,742/ha (King and Bohlen 1994, 1995, Zentner et al. 2003,
Daniels et al. 2005). These total costs represent many different aspects of the restoration
process, including planning, permitting, materials, construction, maintenance, and
monitoring (Zentner et al. 2003). Of this total cost, tree planting can represent a
substantial proportion but can be offset by using volunteers to help plant (Zentener et al.
2003).
278
To recommend cost effective planting material for returning woody biomass to
restored forested wetlands, biomass produced per dollar invested (standardized to 1 ha)
was determined for the species/stocktype combinations used in this study using the
following equation.
𝑅𝑒𝑡𝑢𝑟𝑛 𝑜𝑛 𝑖𝑛𝑣𝑒𝑠𝑡𝑚𝑒𝑛𝑡 (𝑘𝑔⁄$1) =
𝑡𝑜𝑡𝑎𝑙 𝑏𝑖𝑜𝑚𝑎𝑠𝑠 𝑝𝑟𝑜𝑑𝑢𝑐𝑒𝑑 𝑎𝑓𝑡𝑒𝑟 5 𝑦𝑒𝑎𝑟𝑠 (𝑘𝑔⁄ℎ𝑎)
𝑖𝑛𝑖𝑡𝑖𝑎𝑙 𝑖𝑛𝑣𝑒𝑠𝑡𝑚𝑒𝑛𝑡 ($⁄ℎ𝑎)
These methods are similar to those applied by Kimball et al. (2015). This robust
metric incorporates initial planting costs, survival, and biomass accumulation after 5
years and represents a return (biomass) on the initial investment (planting costs). Initial
planting costs included purchase price for the material, installation cost and
miscellaneous costs (Table 6-1). It is important to note that greater biomass per dollar
does not always represent species/stocktype combinations that have greater biomass
production. Greater biomass returns on initial investments with lower overall biomass
accumulation can result from species/stocktype combinations that have very low initial
planting costs and moderate biomass accumulation.
In AMB, the return on investment ranged from 0 kg/$1 (S. nigra BR had no
survivors) to 39.05 kg/$1 (P. occidentalis TB) and the average was 5.51 kg/$1 (Table
6-2). There were many species/stocktypes combinations that would be economically
sensible to plant in upland sites that have minimal herbaceous competition (11
combinations had >1 kg/$1). The species/stocktype combination with the greatest return
on initial investment was P. occidentalis (TB and BR), followed by L. styraciflua BR and
B. nigra (GAL and BR) (Table 6-2). The majority of the Quercus spp. (secondary
successional species (SEC)) had lower returns on investment than primary successional
species (PRI) because of their slow biomass accumulation rates (Chapter 4). However, Q.
279
bicolor (BR) had greater return on investment than several primary successional species,
mainly due to the low initial cost (Table 6-1). The BR and TB stocktypes had greater
returns on investment than the GAL for PRI. In comparison, the BR and GAL had greater
returns than the TB for SEC (Table 6-2).
In SAT, the return on investment ranged from 0 kg/$1 (P. occidentalis BR had no
surviving individuals) to 5.66 kg/$1 and the average was 1.67 kg/$1 (Table 6-2). Similar
to AMB, there were many species/stocktype combinations that represent good returns on
initial investments (10 combinations had >1 kg/$1). The species/stocktype with the
greatest return on investment was S. nigra BR, followed by L. styraciflua (BR), P.
occidentalis TB and B. nigra (BR, GAL, TB). The majority of SEC had lower returns on
investment than PRI; however, Q. bicolor (BR) had greater return on investment than L.
styraciflua TB and S. nigra GAL, mainly due to the low initial cost (Table 6-1). Similar
to AMB, the BR and TB in SAT had greater returns on investment than the GAL for PRI,
while the BR and GAL had greater returns than the TB for SEC (Table 6-2).
The average return on investment in FLD (0.08 kg/$1) was lower than AMB (5.51
kg/$1) and SAT (1.67 kg/$1) reflecting the stressful environmental conditions. The return
on investment ranged from 0 kg/$1 (7 species/stocktype combinations had 1 or 0
survivors) to 1.08 kg/$1 (Table 6-2). Salix nigra BR was the only species/stocktype
combination to have >1 kg/$1 return on investment and S. nigra TB and GAL were the
2nd and 3rd ranked species/stocktype combinations (Table 6-2). The majority of SEC had
very low returns on investment and Q. bicolor BR and Q. palustris GAL were the only
combinations with greater than 0.02 kg/$1. The BR and TB stocktypes had higher returns
280
on investment than GAL for PRI, while the GAL had the greater returns than the BR and
TB for SEC.
Initial Planting Density
The above return on investment metric was based on the initial density used in
this study which may not be the density that is appropriate for planting saplings in
restored wetlands. To recommend an initial planting density and subsequent planting
costs, the stem cross-sectional area at groundline (CSAG) ecological performance
standard (EPS) proposed in Chapter 5 (>5.2 m2/ha by year 5) was used as a target.
The initial density required to meet the CSAG EPS was determined for each
species/stocktype combination within FLD. This initial density represents planting each
combination exclusively and should be adjusted based on the number of
species/stocktype combinations used in practice. Additionally, planting a mixture of
species is recommended because of the increased ecological benefits associated with
greater species diversity in restored wetlands (Allen 1997, Naeem 2006, Aerts and
Honnay 2011). Greater biodiversity in a variety of ecosystems has also been shown to
provide services that support human well-being (Palumbi et al. 2009). The recommended
density was determined using the following equation:
5.2 (𝑚2⁄ℎ𝑎 )
𝑅𝑒𝑐𝑜𝑚𝑚𝑒𝑛𝑑𝑒𝑑 𝑑𝑒𝑛𝑠𝑖𝑡𝑦 (𝑠𝑡𝑒𝑚𝑠⁄ℎ𝑎 ) = 𝑖𝑛𝑖𝑡𝑖𝑎𝑙 𝑑𝑒𝑛𝑠𝑖𝑡𝑦(𝑠𝑡𝑒𝑚𝑠⁄ℎ𝑎 ) ∗ (
)
5𝑡ℎ 𝑦𝑒𝑎𝑟 𝐶𝑆𝐴𝐺 (𝑚2 ⁄ℎ𝑎 )
This calculation assumes that the survival and growth trajectories remain the same if
planting density was increased.
Based on the results from FLD, the minimum initial densities recommended for S.
nigra BR, GAL, TB, and B. nigra GAL (the top four species/stocktype combinations
281
from the economic analysis) was 303, 371, 463, and 519 stems/ha respectively (Table
6-3). These are the initial densities required to meet the recommended CSAG EPS. Based
on the average cost of these planting materials, this represents expenditures of
approximately $4000/ha for planting and installation of these species/stocktype
combinations. Due to slow growth and/or poor survival, minimum initial density
recommended increases dramatically beyond the above recommended species/stocktypes.
For example, initial density required to meet CSAG EPS for Q. palustris GAL (top
recommended Quercus spp.) is 3967 stems/ha (Table 6-3). This increased density leads to
a substantial increase in planting costs ($40,000/ha), which is not financially feasible.
Overall Conclusions
I provided recommendations in this dissertation regarding selection of species and
stocktypes for planting in restored forested wetlands based on survival, morphological
development, total biomass accumulated, biomass accumulation rates, and economic
costs. Based on all of these factors, I recommend that in restored wetlands (which
typically have stressful hydrologic and soil conditions) early successional species
(particularly S. nigra and B. nigra) be planted at higher densities (520 stems/ha) than
secondary successional species (particularly Q. bicolor and Q. palustris) which should be
planted to ensure diverse forest development which mimics natural succession. However,
the diversity of species selected for planting should reflect the diversity found in the
wetland type that is being restored.
Less expensive stocktypes (BR and TB) should be used for primary species and
BR and GAL stocktypes used for secondary species based on their biomass returns on
282
initial investment. However, larger stocktypes do have greater survival probabilities in
stressful conditions and should be considered if funding allows. In afforestation and
reforestation projects where hydrologic stress is not a concern, the smaller less expensive
stocktypes could be used because they accumulated equivalent biomass for less initial
costs than larger stocktypes.
My second recommendation is that the ecological performance standards for
forested compensatory wetland mitigation sites in Virginia need to be re-evaluated by the
responsible state agencies. I have provided an additional performance standard based on
stem cross-sectional area at groundline that requires greater than 5.2 m2/ha of woody
stems be achieved by the 5th year following construction. I have demonstrated that this
performance standard can better evaluate the ecological function of woody biomass
accumulation in compensatory sites than the existing performance standards and show
how this standard could be easily incorporated into the monitoring of these sites.
In conclusion, I hope that this dissertation will contribute to increasing the success
of restoring forested headwater wetlands, which are very important to protecting and
enhancing the health of our environment, and will contribute to the understanding of how
these species respond to environmental stressors.
283
Literature Cited
Aerts, R., & Honnay, O. (2011). Forest restoration, biodiversity and ecosystem
functioning. BMC Ecology, 11(1), 29.
Allen, J. A. (1997). Reforestation of bottomland hardwoods and the issue of woody
species diversity. Restoration Ecology, 5(2), 125-134.
Blignaut, J., Aronson, J., & Wit, M. (2014). The economics of restoration: looking back
and leaping forward. Annals of the New York Academy of Sciences.
Daniels, W. L., Perry, J. E., & Whittecar, R. G. (2005). Effects of soil amendments and
other practices upon the success of the Virginia Department of Transportation’s
non-tidal wetland mitigation efforts. Final Contract Report, Virginia
Transportation Research Council, Charlottesville, VA.
Holl, K. D., & Howarth, R. B. (2000). Paying for restoration. Restoration Ecology, 8(3),
260-267.
Kimball, S., Lulow, M., Sorenson, Q., Balazs, K., Fang, Y.-C., Davis, S. J., et al. (2015).
Cost-effective ecological restoration. Restoration Ecology, 23(6), 800-810.
King, D., & Bohlen, C. (1994). A technical summary of wetland restoration costs in the
continental United States. Tech. rep., University of Maryland, Solomons,
Maryland.
King, D., & Bohlen, C. (1995). The Cost of Wetland Creation and Restoration. Tech.
rep., University of Maryland, Solomons, MD.
Naeem, S. (2006). Biodiversity and ecosystem functioning in restored ecosystems:
extracting principles for a synthetic perspective. In D. A. Falk, M. A. Palmer, & J.
B. Zedler (Eds.), Foundations of Restoration Ecology. Island Press.
Palumbi, S. R., Sandifer, P. A., Allan, J. D., Beck, M. W., Fautin, D. G., Fogarty, M. J.,
Halpern, B. S., Incze, L. S., Leong, J.-A., Norse, E., Stachowicz, J. J., & Wall, D.
H. (2009). Managing for ocean biodiversity to sustain marine ecosystem services
Frontiers in Ecology and the Environment, 7, 204-211.
Robbins, A. S., & Daniels, J. M. (2012). Restoration and economics: a union waiting to
happen? Restoration Ecology, 20(1), 10-17.
Woodworth, P. (2006). What price ecological restoration? In putting a price tag on
endangered species and degraded ecosystems, ecologists and economists have
284
joined forces to formulate a new rationale for environmental issues: restoring
natural capital. The Scientist, 20(4), 38-40.
Zentner, J., Glaspy, J., & Schenk, D. (2003). Wetland and riparian woodland restoration
costs. Ecological Restoration, 21(3), 166-173.
285
Stocktype
Bare root
Gallon
Tubeling
Bare root
Gallon
Tubeling
Bare root
Gallon
Tubeling
Bare root
Gallon
Tubeling
Bare root
Gallon
Tubeling
Bare root
Gallon
Tubeling
Bare root
Gallon
Tubeling
Species
Betula nigra
Betula nigra
Betula nigra
Liquidambar styraciflua
Liquidambar styraciflua
Liquidambar styraciflua
Platanus occidentalis
Platanus occidentalis
Platanus occidentalis
Quercus bicolor
Quercus bicolor
Quercus bicolor
Quercus palustris
Quercus palustris
Quercus palustris
Quercus phellos
Quercus phellos
Quercus phellos
Salix nigra
Salix nigra
Salix nigra
Location
Clinton, NC
Native Roots Nursery
Clinton, NC
Native Roots Nursery
Clinton, NC
Native Roots Nursery
Clinton, NC
Native Roots Nursery
Clinton, NC
Native Roots Nursery
Clinton, NC
Native Roots Nursery
Warren County Nursery McMinnville, TN
Clinton, NC
Native Roots Nursery
Atlantic, VA
Against the Wind Nursery
Clinton, NC
Native Roots Nursery
Clinton, NC
Native Roots Nursery
Clinton, NC
Native Roots Nursery
Clinton, NC
Native Roots Nursery
Clinton, NC
Native Roots Nursery
Clinton, NC
Native Roots Nursery
Clinton, NC
Native Roots Nursery
Clinton, NC
Native Roots Nursery
Atlantic, VA
Against the Wind Nursery
Warren County Nursery McMinnville, TN
Columbus, NJ
Pinelands Nursery
Atlantic, VA
Against the Wind Nursery
Nursery
0.65
3.25
1.00
0.65
3.25
1.00
0.56
3.25
1.00
0.65
3.25
1.00
0.65
3.25
1.00
0.65
3.25
1.00
0.48
7.95
1.00
1.00
5.00
1.75
1.00
5.00
1.75
1.00
5.00
1.75
1.00
5.00
1.75
1.00
5.00
1.75
1.00
5.00
1.75
1.00
5.00
1.75
0.25
2.00
1.25
0.25
2.00
1.25
0.25
2.00
1.25
0.25
2.00
1.25
0.25
2.00
1.25
0.25
2.00
1.25
0.25
2.00
1.25
1.90
10.25
4.00
1.90
10.25
4.00
1.81
10.25
4.00
1.90
10.25
4.00
1.90
10.25
4.00
1.90
10.25
4.00
1.73
14.95
4.00
Material Price Installation Misc. Cost Total Cost
($/Tree)
Cost ($/Tree) ($/Tree)
($/Tree)
Tables
Table 6-1. Source of each species and stocktype and associated costs. Installation and
miscellaneous costs were obtained from Wetland Studies and Solutions, Inc. and
represent prices in 2012. Material price is based on 2008 purchase price. Miscellaneous
costs include mulch, agriform fertilizer, shipping, and terrasorb. No mulch, fertilizer or
terrasorb were used in this study, but these products are commonly used in practice.
286
Stocktype
Tubeling
Bare root
Bare root
Gallon
Bare root
Gallon
Gallon
Bare root
Tubeling
Tubeling
Gallon
Bare root
Gallon
Gallon
Bare root
Gallon
Tubeling
Tubeling
Tubeling
Tubeling
Bare root
Species
Platanus occidentalis
Platanus occidentalis
Liquidambar styraciflua
Betula nigra
Betula nigra
Platanus occidentalis
Liquidambar styraciflua
Quercus bicolor
Betula nigra
Salix nigra
Salix nigra
Quercus palustris
Quercus phellos
Quercus bicolor
Quercus phellos
Quercus palustris
Quercus bicolor
Liquidambar styraciflua
Quercus phellos
Quercus palustris
Salix nigra*
Ambient Cell
Species
39.05
Salix nigra
37.93 Liquidambar styraciflua
9.53 Platanus occidentalis
6.08
Betula nigra
4.94
Betula nigra
4.83
Betula nigra
2.63 Liquidambar styraciflua
2.06
Salix nigra
2.05 Platanus occidentalis
1.73
Quercus bicolor
1.10 Liquidambar styraciflua
0.81
Salix nigra
0.67
Quercus palustris
0.60
Quercus phellos
0.47
Quercus phellos
0.46
Quercus bicolor
0.36
Quercus phellos
0.32
Quercus palustris
0.12
Quercus bicolor
0.02
Quercus palustris
0
Platanus occidentalis*
kg/$1
Bare root
Bare root
Tubeling
Bare root
Tubeling
Gallon
Gallon
Tubeling
Gallon
Bare root
Tubeling
Gallon
Bare root
Bare root
Gallon
Gallon
Tubeling
Gallon
Tubeling
Tubeling
Bare root
Stocktype
Saturated Cell
Species
5.66
Salix nigra
5.66
Salix nigra
4.43
Salix nigra
3.35
Betula nigra
2.88
Betula nigra
2.78 Liquidambar styraciflua
1.88
Quercus bicolor
1.68
Betula nigra
1.53
Quercus palustris
1.20 Liquidambar styraciflua
0.91
Quercus bicolor
0.89
Quercus phellos
0.49 Liquidambar styraciflua
0.46 Platanus occidentalis
0.41
Quercus bicolor**
0.28
Quercus palustris**
0.22
Quercus phellos**
0.18 Platanus occidentalis*
0.11 Platanus occidentalis*
0.07
Quercus palustris*
0
Quercus phellos*
kg/$1
Bare root
Tubeling
Gallon
Gallon
Tubeling
Gallon
Bare root
Bare root
Gallon
Bare root
Gallon
Gallon
Tubeling
Gallon
Tubeling
Bare root
Bare root
Bare root
Tubeling
Tubeling
Tubeling
Stocktype
Flooded Cell
1.08
0.27
0.09
0.08
0.05
0.03
0.02
0.02
0.02
0.02
0.01
0.01
0.01
0.003
0
0
0
0
0
0
0
kg/$1
Table 6-2. Species/stocktype combinations return on investment ranked within each cell.
* No surviving individuals
** 1 individual surviving
287
Table 6-3. Initial densities required to meet 5.2 m2/ha CSAG EPS in FLD. ND represents
species/stocktype combinations for which initial density was not determined due to 1 or 0
surviving individuals.
Species
Stocktype
Salix nigra
Salix nigra
Salix nigra
Betula nigra
Liquidambar styraciflua
Betula nigra
Quercus palustris
Quercus phellos
Quercus bicolor
Betula nigra
Liquidambar styraciflua
Platanus occidentalis
Liquidambar styraciflua
Quercus bicolor
Quercus bicolor**
Quercus palustris**
Quercus phellos**
Platanus occidentalis*
Platanus occidentalis*
Quercus palustris*
Quercus phellos*
Bare root
Gallon
Tubeling
Gallon
Gallon
Tubeling
Gallon
Gallon
Gallon
Bare root
Bare root
Gallon
Tubeling
Bare root
Tubeling
Bare root
Bare root
Bare root
Tubeling
Tubeling
Tubeling
Initial
Density
(stems/ha)
303
371
463
519
1607
1972
3967
4247
4785
9315
10126
10986
11252
12366
ND
ND
ND
ND
ND
ND
ND
* No surviving individuals
** 1 individual surviving
288
VITA
Herman Wesley Hudson III
Wes was born in Richmond, VA on January 26, 1985. After graduating from Varina High
School in Henrico, VA in 2003, he went on to earn a B.S. in Biology (concentration in
Ornamental Horticulture) from Christopher Newport University in 2008. He completed a
M.S. in Environmental Science with Dr. Robert Atkinson at CNU in May 2010 focusing
on natural tree colonization into post agricultural restored wetlands in Southeastern
Virginia. He entered the Ph.D. program at the Virginia Institute of Marine Science,
College of William & Mary under graduate advisor Dr. Jim Perry in the fall of 2010. He
was a NSF GK-12 PERFECT Graduate Fellow during 2013-2014 and taught at the
Chesapeake Bay Governor School. He will graduate in May 2016 with a Ph.D. in Marine
Science.
289