WETLANDS, Vol. 29, No. 2, June 2009, pp. 520–526
’ 2009, The Society of Wetland Scientists
FIRE IN FLOODPLAIN FORESTS IN THE SOUTHEASTERN USA: INSIGHTS
FROM DISTURBANCE ECOLOGY OF NATIVE BAMBOO
Paul R. Gagnon
Louisiana State University
Department of Biological Sciences
202 Life Sciences Building
Baton Rouge, Louisiana, USA 70803
Current address: University of Florida
Department of Wildlife Ecology and Conservation
110 Newins-Ziegler Hall
Gainesville, Florida, USA 32611
E-mail: [email protected]
Abstract: Floodplain forests in the southeastern USA have recently been the focus of intensive
restoration efforts after centuries of human-caused decline. Many of these restored forests appear to
suffer from systemic problems arising from the altered disturbance regime in modern southeastern
floodplains. Increasing evidence suggests that fire may be an occasional but important ecosystem
component missing from these forests. Most relevant literature mentions fire only in passing, if at all; the
literature that does discuss fire is typically either speculative or draws heavily from other ecosystems. This
article develops the hypothesis that fire has been an important and recurrent disturbance in southeastern
alluvial floodplains for millennia. It first synthesizes research indicating that the expansive
monodominant bamboo stands (called canebrakes) once common throughout these floodplain forests
were likely fire-obligate and might therefore be used as indicators of recurrent fires. It then examines prehistoric, historic, and recent evidence of fire in bottomland forests from both natural and human sources.
Finally, it places these findings into ecological context, proposes an integrated study by which future
research might clarify the ecological role of fire in southeastern floodplain forests, and addresses some
implications for management.
Key Words: Arundinaria gigantea, bottomland hardwood forests, canebrakes, multiple disturbance
interactions, fire ecology, hurricanes, tornados, windstorms
INTRODUCTION
water quality maintenance, flood control, and
habitat for diverse wildlife species, BLH have
increasingly been the focus of reforestation efforts
since then (Stanturf et al. 2001, King et al. 2005,
Wilson et al. 2007). As a result, the amount of land
reforested with BLH tree species has recently
increased substantially, and additional reforestation
projects are planned (King et al. 2006).
Reforestation cannot ensure fully functioning
restoration of these bottomlands (King et al.
2005). Reasoned and sometimes intensive efforts
are necessary for effective restoration because of the
extent to which humans have changed underlying
ecological processes throughout the BLH ecosystem
(Stanturf et al. 2001, Wilson et al. 2007). To return
full function, it is necessary to restore the relevant
disturbance regime (Wilson et al. 2007). Large
disturbances in BLH included windstorms, ice
storms, droughts, floods, and fires (King et al.
2005, Wilson et al. 2007). After flooding, fire
Bottomland hardwood forests (BLH) in the
southeastern U.S. are rebounding after two centuries
of radical alteration to alluvial floodplains. Prior to
European settlement, these forests covered perhaps
40–50 million hectares of what is now the southeastern United States (The Nature Conservancy
1992). That amount has been reduced by 60%
overall, much more in certain regions like the lower
Mississippi River alluvial valley (Noss et al. 1995,
Stanturf et al. 2001). During the 19th and 20th
centuries, land-clearing for industrial agriculture,
intensive logging and river channelization combined
to take a heavy toll on BLH forests (King et al.
2005, Saikku 2005). When high soy prices drove
another round of forest clearing in the 1970s, area of
BLH hit an all-time low (Stanturf et al. 2001).
Because of the many products and ecosystem
services BLH provide, including valuable timber,
520
Gagnon, FIRE IN BOTTOMLAND HARDWOOD FORESTS
regimes are most likely to have been fundamentally
altered, and fire may be the most often overlooked
ecological and evolutionary force in BLH.
Systemic problems are recognized with regenerating BLH that arise from their altered disturbance
regimes. Wilson et al. (2007) note that ‘‘Reduced
levels of disturbance acting in concert with unsustainable forest management practices have resulted
in homogeneous, closed canopy forests with little
structural diversity or understory vegetation.’’ Canopy structure in modern BLH is altered, as
evidenced by poor regeneration of several shadeintolerant species valued for their timber and mast, a
critical food source for wildlife. Recent research
indicates that flooding alone will not ensure the
regeneration of young bottomland oaks (King and
Antrobus 2005), which need large or recurrent gaps
to reach the forest canopy (Oliver et al. 2005,
Holladay et al. 2006, Collins and Battaglia 2007).
There is also evidence that oak-dominated forests
historically established after fire (Aust et al. 1985,
Hodges 1998), and that certain other bottomland
assemblages were maintained by fire (e.g., Hughes
1957, Ewel 1998, Gagnon and Platt 2008). Most
discussions in the literature about fire in BLH are
either speculative or extrapolate from other systems (e.g., Sharitz et al. 1992, Myers and Van Lear
1998).
In this article I examine the hypothesis that fire
has been an important and recurrent disturbance in
southeastern alluvial floodplains for millennia. I first
synthesize research indicating that expansive monodominant bamboo stands (called canebrakes) once
common throughout BLH were likely fire-obligate,
and might therefore be used as indicators of
recurrent fires. Next, I examine pre-historic, historic,
and recent evidence of fire in BLH, from both
natural and human sources. Finally, I put these
findings into ecological context, propose an integrated study by which future research might clarify
the ecological role of fire in BLH, and discuss some
related management implications.
CANEBRAKES AS
DISTURBANCE-OBLIGATE
MONODOMINANT COMMUNITIES
Canebrakes are monodominant stands of Arundinaria bamboo, a genus endemic to the United States.
Of the three recognized species of Arundinaria, it is
A. gigantea (Walt.) Muhl. (giant cane or river cane)
that frequently occurs on higher ground within
southeastern alluvial floodplains (Judziewicz et al.
1999, Triplett et al. 2006). Culms of A. gigantea can
exceed 6 m in height and 3 cm in diameter; they
521
attain their largest size in the most fertile soils
(Saikku 2005, Stewart 2007). Giant cane today is a
common understory and edge component in BLH
(Marsh 1977, Kellison et al. 1998, Judziewicz et al.
1999). Like other woody bamboos, A. gigantea is
thought to be semelparous, meaning that after
decades of vegetative growth, it flowers once and
then dies (Judziewicz et al. 1999). Nowadays giant
cane grows in small patches along forest edges or
diffusely under forest canopy (Marsh 1977, Gagnon
et al. 2007). Early European explorers describe vast,
dense stands of giant cane frequently growing on
levees and ridges in southeastern alluvial floodplains
(Platt and Brantley 1997, Stewart 2007). John James
Audubon describes how canebrakes were typically
too thick to traverse even on foot unless one pushed
through hunched-over and backwards (Platt et al.
2002a).
Floods influenced giant cane and other species in
BLH over time. Floods occurred during most years
in southeastern floodplains (Lentz 1931, Hodges
1998, Kellison et al. 1998). Their impact on
vegetation varied greatly depending on local hydrology and microtopography (Hodges 1998, Kellison et
al. 1998, Saikku 2005). On natural levees and ridges
where giant cane was common, floods were occasional, intermittent events (Kellison et al. 1998).
Giant cane could tolerate these intermittent floods
(Cirtain et al. 2004), although not prolonged
inundation, and may have benefited from the new
space and sediment that accompanied floods (Gagnon and Platt 2008).
Violent storms also shaped giant cane populations
in BLH. Tornados, hurricanes, severe thunderstorms, and ice storms all damaged the canopy of
BLH and thereby increased light levels in the
understory. Some of these events snapped single
branches or tipped-over individual trees; others
removed forest canopy over large areas (Ker 1816,
Sharitz et al. 1992, Gagnon et al. 2007). In the high
light environment of large, wind-generated forest
gaps, rates of new culm production in A. gigantea
could be twice that under forest canopy (Gagnon et
al. 2007).
Droughts may have interacted with windstormdamaged landscapes to create conditions favorable
for fires. Windstorm and fire interactions have been
documented for other southeastern ecosystems
(Platt et al. 2002b, Liu et al. 2008), and could have
also been important in BLH. Fires are reported in
BLH during drought years in large re-vegetating
gaps (Lentz 1931, Kaufert 1933). Over ecological
time, droughts in BLH were probably linked to
ENSO weather patterns, which have cycled on a 3–7
year interval for the last 130,000 years (Tudhope et
522
al. 2001, Beckage et al. 2003). High frequency of
lightning strikes has been pervasive across the
southeastern U.S. (Platt 1999). Lightning from
spring storms following a dry period could have
ignited fires in windstorm-damaged areas (Myers
and Van Lear 1998, Gagnon and Platt 2008).
Human-caused fires probably augmented the number of fire events affecting giant cane in southeastern
alluvial floodplains.
Stand structure in Arundinaria bamboos is largely
a function of disturbance regime (Hughes 1957,
Gagnon et al. 2007, Gagnon and Platt 2008). Over a
few years, stands of Arundinaria growing in large
gaps gradually decline, perhaps a result of selfcrowding or self-shading (Hughes 1957, 1966,
Gagnon and Platt 2008). A new disturbance like
fire will typically reinvigorate declining Arundinaria
stands (Hughes 1957, 1966, Wright and Bailey
1982). Without periodic disturbances to open
growing space, stands of giant cane will decline
and trees of BLH will eventually overtop them
(Wright and Bailey 1982, Gagnon et al. 2007),
precluding the formation of expansive monodominant canebrakes like those described by so many
early European explorers (Platt and Brantley 1997).
In open-grown stands of giant cane, fire clears
competing vegetation and accelerates new culm
production while eliminating senescing older culms
(Gagnon and Platt 2008). The link between fire and
canebrake-like stand structure (i.e., dense and
expansive monodominant bamboo) in Arundinaria
is tight enough that historical canebrakes can serve
as indicators of recurrent fires (Wright and Bailey
1982, Frost 2000).
FIRE DURING THE PRE-HISTORIC PAST
Evidence suggests that fires recurred in southeastern alluvial floodplains for millennia. Palynological
records of herbaceous Chenopodiaceae and Amaranthus pollen in the Yazoo River floodplain
indicate open, potentially burned areas were present
in BLH forests since the last glaciation (Saikku
2005). In the Western Lowlands of the Mississippi
River alluvial valley, Arundinaria thickets first
appeared in the pollen record around 9,500 years
ago (Royall et al. 1991, King et al. 2005). Beginning
approximately 5,000 years ago, climatic patterns
became like modern ones (Saikku 2005), so lightning
would have been frequent across the landscape. The
pollen record indicates a concurrent shift from oakto pine-dominated plant communities along the
northern Gulf Coast, meaning fire frequencies likely
increased on the landscape level (Delcourt and
Delcourt 1987, King et al. 2005).
WETLANDS, Volume 29, No. 2, 2009
Native Americans inhabited and set fires in BLH
for thousands of years (Komarek 1974, Sharitz et al.
1992, Saikku 2005, Mann 2006). Evidence of Native
American occupancy in the lower Mississippi River
alluvial valley first appears 12,000 years ago (Saikku
2005, King et al. 2005). Human populations
fluctuated in the floodplains over time, and these
early inhabitants survived by hunting different types
of large and small animals, and by gathering oak/
hickory mast and various understory flora (Saikku
2005, King et al. 2005). During the early Archaic
period (9,000 to 5,000 years ago), drier climatic
conditions developed, and more people moved into
the bottomlands as resources grew scarcer in
adjacent uplands (Saikku 2005). Modern climatic
patterns during the middle Archaic period (5,000–
2,500 years ago) coincided with intensifying exploitation of riverine resources, growth of interregional
trade, and development of cultigens and ceramic
containers (Saikku 2005). The development of
Poverty Point in northeastern Louisiana (3,600–
2,600 years ago) is clear evidence that large human
populations were living in permanent habitations
within and alongside southeastern floodplains during that period (Saikku 2005, Mann 2006). Over
time, Native Americans grew increasingly reliant on
horticulture (Saikku 2005, King et al. 2005). Cornbased agriculture was introduced in the lower
Mississippi River alluvial valley around 400 A.D.
and quickly became an important means of subsistence (Saikku 2005). Human activities caused
intensifying changes in flora and fauna during this
period, mainly from their use of fire and stone
wood-cutting tools to clear patches of BLH for
garden plots (King et al. 2005).
The Mississippian culture arose around 900 A.D.
and lasted until the first Europeans arrived (Saikku
2005). These people were reliant on corn-based
agriculture and on fire for clearing land. Sedentary
riverine villages with sociopolitical organization
were located around ceremonial centers that included large temple mounds. For their fields, people
preferred the natural levees on recently abandoned
channels because the sandy loam there was easily
tilled using hand tools, and because these high
grounds were the last to flood (Saikku 2005). When
fields lost fertility, people cleared new ones by
girdling trees and setting fires (Morse and Morse
1983, Saikku 2005). Their primary game animal was
the white-tailed deer (Odocoileus virginianus Zimm.),
and tribes maximized the edge habitat that deer
prefer by setting fires (Saikku 2005). Analyses of
faunal remains indicate decreasing arboreal species
and increasing edge species during this period
(Saikku 2005).
Gagnon, FIRE IN BOTTOMLAND HARDWOOD FORESTS
The first Europeans brought waves of epidemics
that killed perhaps 80–90% of Native American
populations, and caused their societies to collapse
(Platt and Brantley 1997, Saikku 2005, Mann 2006).
Native American occupation of the lower Mississippi River alluvial valley ended during the first half
of the 18th century after two centuries of epidemics
and slave-raiding. Reforestation occurred naturally
after that, and the area remained an occasional
hunting ground for dispersed remnant survivors and
for upland tribes (Platt and Brantley 1997, Saikku
2005).
FIRE AND EUROPEAN SETTLEMENT
Accounts by European explorers of southeastern
bottomlands strongly suggest that recurrent presettlement fires had shaped that habitat. There are
numerous descriptions of large treeless areas in the
floodplains (Platt and Brantley 1997, Saikku 2005,
Stewart 2007). For example, early French accounts
and 19th century surveyor field notes both mention
large grasslands or ‘‘prairies’’ on high ground within
the Yazoo floodplain (Saikku 2005). Without fire or
some other recurrent space-opening disturbance,
southeastern alluvial floodplains would have quickly
succeeded into BLH forests (Kuchler 1964, Sharitz
et al. 1992).
According to many early accounts, canebrakes
were pervasive on alluvial high ground in the
Southeast (Platt and Brantley 1997). They were so
common that any mention of canebrakes was
typically off-hand (Stewart 2007). For example,
Thomas Nutall, described how along the Mississippi
River, ‘‘vast tracts of cane land occur in the bends’’
(Saikku 2005). Travelers often sought canebrakes
for evening camps because Arundinaria was a highly
nutritious fodder for their horses (Wright and Bailey
1982). The naturalist William Bartram made note of
canebrakes as he documented vegetation in the
southeastern U.S. between 1773 and 1777 (Harper
1958). He mentions ‘‘cane’’ (Arundinaria) dozens of
times in his ‘‘Travels,’’ often in reference to
expansive canebrakes (Harper 1958). The ubiquity
of canebrakes in southeastern floodplains was a
strong indicator of recurrent fires (Wright and
Bailey 1982, Frost 2000).
European settlers used fire to clear land, and
they sought out canebrakes for their fields. The
first wave of settlers drove their herds of cattle,
horses and pigs into the bottomlands, where they
feasted on nutritious canebrakes (Platt and Brantley
1997, Stewart 2007). During a second wave, which
peaked between 1840 and 1860 in the Mississippi
River alluvial valley, would-be cotton growers
523
targeted cane lands as indicators of rich soil and
relative safety from flooding (Platt and Brantley
1997, Saikku 2005). Planters systematically eradicated most canebrakes by chopping the culms,
pulling up the rhizomes, and then burning several
days later once the cane was dry (Stewart 2007).
They cleared forested lands by girdling trees and
burning, and their fires frequently escaped (Saikku
2005).
FIRE AND THE RECENT PAST
Alteration of BLH accelerated after the Civil War.
In the late 19th century, intensive logging of
southern forests began in earnest (Sharitz et al.
1992, Saikku 2005). Ditching and levee building
accelerated shortly thereafter, drastically changing
the natural flood regime (King et al. 2005, Saikku
2005). Millions of hectares were logged and then
permanently converted to cotton, corn, and soy
fields (King et al. 2005, Saikku 2005). By the middle
of the 20th century, cyclical flooding in many BLH
had ceased entirely. Remaining BLH forests were
highly fragmented and had greatly altered hydrology.
Accounts from several early forest researchers
describe how fires occurred in BLH forests in the
recent past. G. H. Lentz, a forester from the New
York College of Forestry at Syracuse on sabbatical
with the Southern Forest Experiment Station
(Saikku 2005), describes a recurrent ‘‘fire problem’’
in Mississippi bottomland forests in 1931:
The summer of 1924 was characterized by an insufficient amount of rainfall throughout the lower Mississippi Delta. A dry fall and winter followed. Many of
the usually moist and wet areas in the bottomlands had
dried out by the early spring of 1925 and many serious
fires resulted. Although foresters and timbermen
generally discount the fire problem in the southern
hardwoods, a real problem exists… The past summer,
1930, was decidedly dry and was followed by a fall and
winter also quite dry. Many sloughs and brakes usually
filled with water have been dry for several months. The
spring rains have not been able to wet down the litter
and debris and conditions in the woods are just right
for fires to spread.
Lentz describes fires in a stand that had been
clearcut three years prior. He notes that the fire was
hot enough to kill virtually all advanced regeneration and to severely damage several full-size trees.
He notes other fires burning simultaneously in ‘‘both
cut and uncut’’ stands of timber. He concludes by
saying that ‘‘If timber growing on an appreciable
scale is to be practiced in the bottomlands, the
524
WETLANDS, Volume 29, No. 2, 2009
solution of the fire problem is going to be one of the
first essential steps.’’
F. H. Kaufert (1933) followed Lentz’s report with
a study of fire scars in BLH of the Mississippi River
alluvial valley in Louisiana. Kaufert examined over
500 stumps from three separate logged tracts and
documented recurrent fire damage predating the
Civil War. He states that ‘‘Information obtained
from many ring counts indicates that widespread
burning of the bottoms occurred in the years 1898–
99, 1911–12, 1916–17, and 1924–25,’’ and lesswidespread fires in 1931–32, an interval that ranges
from 5 to 13 years. Ironically, Kaufert states, ‘‘It
seems probable…that widespread burning of the
bottoms did not occur until about 1890, for old
settlers recall the time (before 1895) when the
bottoms were veritable canebrakes,’’ and says that,
according to the ‘‘oldtimers’’ in the area, ‘‘Cane is
killed by a single burn.’’ This anecdotal information
directly contradicts experimental studies of fire
effects on Arundinaria performed later, all of which
conclude that cane benefits from periodic fire
(Hughes 1957, 1966, Wright and Bailey 1982,
Gagnon and Platt 2008). If anything, the presence
of canebrakes might more accurately be considered
evidence that fires did occur. Kaufert attests to the
pyrogenic nature of cane when he says that
according to these ‘‘oldtimers,’’ fires in the past
were often more severe than the more recent fires
because ‘‘dry cane made excellent fuel.’’ He says that
‘‘since the passing of the cane the principal fire
hazard has been leaf litter.’’ He goes on to decry
great damage done to both merchantable timber and
young trees by rot that enters via fire scars. He states
flatly that, ‘‘To eliminate the serious losses caused by
decay and to insure (sic) good stocking in young
stands, fire must be kept out.’’
FIRE, WINDSTORM AND
DROUGHT INTERACTIONS
With today’s modified BLH as our reference, it
can be difficult to conceive that fires were likely once
a part of this floodplain disturbance regime. Land
managers have suppressed fire for decades after its
detrimental effects to timber became well publicized
(Lentz 1931, Kaufert 1933, Toole and McKnight
1956). Suppressing fires is not difficult now because
modern second-growth forests are virtually fireproof – their dense canopies reduce air movement
and shade the understory, minimizing plant growth
and fuel build-up (Wilson et al. 2007). Fires no
longer travel across the landscape as when pyrogenic
pine-filled uplands lay adjacent to the floodplains
(Platt 1999). And canebrakes that were conduits to
fire-spread in BLH have been eradicated in these
fragmented forests (Kaufert 1933, Platt and Brantley
1997).
Accounts of Lentz (1931) and Kaufert (1933)
inform our thinking with several points about the
history of fire in BLH. First, while many of the fires
they describe occurred in large clear-cut gaps,
tornados, hurricanes, and ice-storms similarly created large gaps in these forests over ecological time
(Sharitz et al. 1992, Gagnon et al. 2007, Liu et al.
2008). Second, they describe fires that occurred
several years after the gap-creating disturbances.
Because of the multi-year lag between logging and
drought, these fires would have spread through
dense and continuous regenerating vegetation rather
than through long-decomposed logging slash. This is
consistent with the multi-year lag in hurricane/fire
interactions described by Liu et al. (2008) from the
Alabama Gulf Coast. Third, Lentz and Kaufert
describe fires that occurred in late winter or spring
after prolonged drought had dried sloughs that
might have otherwise served as fire-breaks. Spring is
typically the natural fire season throughout the
coastal southeastern U.S., including where Kaufert
and Lentz made their observations in Louisiana,
because fuel conditions are favorable and lightning
is prevalent (Olson and Platt 1995, Platt 1999).
However, natural fires are also possible later in the
growing season whenever lightning coincides with
dry fuels (Foti and Glenn 1992). Although Lentz
and Kaufert suggest that the fires were humancaused, the pyrogenic power of lightning was often
underestimated at the time, and it is likely that the
true ignition sources were often unknown. Lentz
(1931) notes that early spring rainstorms were
insufficient to moisten the dry leaf litter; these same
storms likely came with lightning. Fourth, the
interval between fires described by Kaufert (1933)
ranges from five to 13 years, which fit closely the fire
interval suggested by Hughes (1957, 1966) to keep
Arundinaria stands healthy.
There are some important qualifiers for this fire
hypothesis. Fires would not have burned in BLH
with the frequency or continuity of adjacent upland
pine savannas (Platt 1999). Fires would have ranged
from creeping low-intensity fires in hardwood leaf
litter, to hot and violent conflagrations that
consumed canebrakes and threatened individual
trees. Certain areas, especially the ‘‘prairies’’ and
canebrakes on higher ground, might have burned
during most droughts, while adjacent areas would
have been virtually fire-proof just like modern
second-growth BLH. Native Americans probably
increased the frequency and pervasiveness of fires in
the bottomlands, but when conditions were right
Gagnon, FIRE IN BOTTOMLAND HARDWOOD FORESTS
and disturbances coincided, some areas almost
certainly burned without human input. The true
extent to which fires influenced BLH remains an
open question, but surely the answer is ‘‘somewhat’’
rather than ‘‘not at all’’ as is now commonly
practiced by modern BLH land-managers.
A multi-pronged study could clarify the extent to
which fires shaped BLH. The prevalence of A.
gigantea remains in southeastern bottomlands could
serve as an indicator of fire regime. Unfortunately,
palynological information is sparse because Arundinaria pollen is virtually indistinguishable from that
of many other grasses (K-b. Liu, Louisiana State
University Department of Oceanography and
Coastal Sciences, personal communication). It has
recently been shown that phytoliths (precipitating
microscopic plant silica bodies) from A. gigantea are
unique and could well serve the same purpose (Lu
and Liu 2003). An integrated study of macroscopic
charcoal, microscopic charcoal, and Arundinaria
phytoliths could clarify the extent to which fires
shaped given bottomland sites ranging back several
millennia (K-b. Liu, personal communication).
Bottomland hardwood forests likely burned
during the confluence of other interacting disturbances. This is similar to findings by Liu et al. (2008)
of interactions between hurricanes and fires on the
Alabama Gulf Coast. In BLH, the stage is set for
fire when effects of windstorms or other large
canopy disturbances coincide in subsequent years
with prolonged droughts. Natural fires could have
occurred in the following scenario: 1) A powerful
windstorm or other disturbance opens a large
canopy gap. 2) Over a few years, the gap fills with
dense, regenerating vegetation. 3) Eventually,
drought renders this new vegetation flammable and
increases landscape connectivity for potential fire. 4)
With the onset of spring thunderstorms, lightning
strikes cause fires in these large, highly flammable
canopy gaps. 5) Canebrakes and adjoining upland
pine savannas facilitate fire spread into other parts of
BLH. 6) Giant cane and other pyrogenic grasses
thrive after burning, producing a flammable fuel bed
that can burn again during the next drought. As
canebrakes expand over time (Gagnon et al. 2007),
they increase the area beyond the initial canopy
disturbance that is susceptible to burning in subsequent droughts, and thereby set up a positive
feedback (Gagnon and Platt 2008). In this way once
a large canopy disturbance initiates the sequence,
canebrakes and other pyrogenic assemblages within
BLH could essentially be self-sustaining.
Management plans for comprehensive restoration
efforts should consider fire-maintained plant assemblages within BLH. These pyrogenic plant systems
525
likely burned during drought conditions and may
not have burned otherwise. Such drought conditions
might be outside management prescriptions for
controlled burning; therefore, land managers should
give consideration to how such areas might be
burned within prescription. Areas containing pyrogenic species like A. gigantea could be cut and then
burned two weeks later, before new culms resprout.
Selected clearcuts or existing large blowdowns
containing Arundinaria could be candidate sites for
prescribed burning.
ACKNOWLEDGMENTS
I thank Kevin Robertson and Loretta Battaglia
for ideas that inspired this paper. Kam-biu Liu and
Mart Stewart made valuable suggestions as I
prepared the manuscript. I thank Heather Passmore
and Jonathan Myers for their helpful comments on
early drafts, and Wylie Barrow, Sammy King and an
anonymous reviewer for suggestions on the later
manuscript. The following were instrumental for my
dissertation research in giant cane, which underpins
this synthesis: guidance from Bill Platt; logistical
support from Kenny Ribbeck, Tommy Tuma, and
the Louisiana Department of Wildlife and Fisheries;
funding from US EPA STAR Fellowship #
U916181 and research grants from the J. B.
Johnston Science Foundation and the American
Bamboo Society.
LITERATURE CITED
Aust, W. M., J. D. Hodges, and R. L. Johnson. 1985. The origin,
growth and development of pure, even-aged stands of
bottomland oak. p. 163–170. In E. Shoulders (ed.) Proceedings
of the Third Biennial Southern Silvicultural Research Conference. USDA Forest Service, New Orleans, LA. GTR-SO-54.
Beckage, B., W. J. Platt, M. G. Slocum, and B. Panko. 2003.
Influence of the El Niño Southern Oscillation on fire regimes in
the Florida Everglades. Ecology 84:3124–3130.
Cirtain, M. C., S. B. Franklin, and S. R. Pezeshki. 2004. Effects of
nitrogen and moisture regimes on Arundinaria gigantea (Walt.)
Muhl. seedling growth. Natural Areas Journal 24:251–57.
Collins, B. and L. L. Battaglia. 2007. Oak regeneration in
southeastern bottomland hardwood forest. Forest Ecology and
Management. doi:10.1016/j.foreco.2007.09.023.
Delcourt, P. A. and H. R. Delcourt. 1987. Long-term Forest
Dynamics of the Temperate Zone: A Case Study of LateQuaternary Forests in Eastern North America. SpringerVerlag, New York, NY, USA.
Ewel, K. C. 1998. Pondcypress swamps. p. 405–20. In M. G.
Messina and W. H. Conner (eds.) Southern Forested Wetlands:
Ecology and Management. Lewis Publishers, New York, NY,
USA.
Foti, T. L. and S. M. Glenn. 1992. The Ouachita Mountain
landscape at the time of settlement. p. 49–65. In L. Hedrick and
D. Henderson (eds.) Conference Proceedings: Restoring of old
growth forests in the Interior Highlands of Arkansas and
Oklahoma. Ouachita National Forest, Winrock International
Institute for Agricultural Development.
526
Frost, C. C. 2000. Studies in landscape fire ecology and
presettlement vegetation of the southeastern United States.
Ph.D. dissertation, University of North Carolina. Chapel Hill,
NC, USA.
Gagnon, P. R., W. J. Platt, and E. B. Moser. 2007. Response of a
native bamboo [Arundinaria gigantea (Walt.) Muhl.] in a winddisturbed forest. Forest Ecology and Management 241:
288–294.
Gagnon, P. R. and W. J. Platt. 2008. Multiple disturbances
accelerate clonal growth in a potentially monodominant
bamboo. Ecology 89:612–18.
Harper, F. 1958. The Travels of William Bartram: Naturalist’s
Edition. Yale University Press, New Haven, CT, USA.
Hodges, J. D. 1998. Minor alluvial floodplains. p. 325–41. In M.
G. Messina and W. H. Conner (eds.) Southern Forested
Wetlands: Ecology and Management. Lewis Publishers, New
York, NY, USA.
Holladay, C. A., C. Kwit, and B. Collins. 2006. Woody
regeneration in and around aging southern bottomland
hardwood forest gaps: Effects of herbivory and gap size.
Forest Ecology and Management 223:218–25.
Hughes, R. H. 1957. Response of cane to burning in the North
Carolina coastal plain. North Carolina Agricultural Experiment Station. Bulletin 402.
Hughes, R. H. 1966. Fire ecology of canebrakes. p. 149–58. In E.
V. Komarek (ed.) Proceedings of Fifth Annual Tall Timbers
Fire Ecology Conference, Tall Timbers Research Station,
Tallahassee, FL, USA.
Judziewicz, E. J., L. G. Clark, X. Londoño, and M. J. Stern.
1999. American Bamboos. Smithsonian Institute Press, Washington, DC, USA.
Kaufert, F. H. 1933. Fire and decay injury in the southern
bottomland hardwoods. Journal of Forestry 31:64–67.
Kellison, R. C., M. J. Young, R. R. Braham, and E. J. Jones.
1998. Major alluvial floodplains. p. 291–323. In M. G. Messina
and W. H. Conner (eds.) Southern Forested Wetlands: Ecology
and Management. Lewis Publishers, New York, NY, USA.
Ker, H. 1816. Travels Through the Western Interior of the United
States, From the Year 1808 up to the Year 1816. Elizabethtown, NJ. In M. Saikku (ed.) 2005. This Delta, This Land: An
Environmental History of the Yazoo-Mississippi Floodplain.
The University of Georgia Press, Athens, GA, USA.
King, S. L. and T. J. Antrobus. 2005. Relationships between gap
makers and gap fillers in an Arkansas floodplain forest. Journal
of Vegetation Science 16:471–80.
King, S. L., J. P. Shepard, K. Ouchley, J. A. Neal, and K.
Ouchley. 2005. Bottomland hardwood forests: past, present,
and future. p. 1–17. In L. H. Fredrickson, S. L. King, and R.
M. Kaminski (eds.) Ecology and management of bottomland
hardwood systems: the state of our understanding. University
of Missouri-Columbia. Gaylord Memorial Laboratory Special
Publication No. 10. Puxico, Missouri, USA.
King, S. L., D. J. Twedt, and R. R. Wilson. 2006. The role of the
wetland reserve program in conservation efforts in the
Mississippi River alluvial valley. Wildlife Society Bulletin
34:914–20.
Komarek, E. V. 1974. Effects of fire on temperate forests and
related ecosystems: southeastern United States. p. 1–272. In T.
T. Kozlowski and C. E. Algren (eds.) Fire and Ecosystems.
Academic Press, New York, NY, USA.
Kuchler, A. W. 1964. Potential natural vegetation of the
conterminous United States. American Geographical Society,
Special Publication Number 36.
Lentz, G. H. 1931. Forest fires in the Mississippi bottomlands.
Journal of Forestry 29:831–32.
Liu, K.-B., H. Lu, and C. Shen. 2008. A 1200-year proxy record
of hurricanes and fires from the Gulf of Mexico coast: Testing
the hypothesis of hurricane–fire interactions. Quaternary
Research 69:29–41.
Lu, H. and K-b. Liu. 2003. Phytoliths of common grasses in the
coastal environments of southeastern USA. Estuarine, Coastal
and Shelf Science 58:587–600.
WETLANDS, Volume 29, No. 2, 2009
Marsh, D. L. 1977. The taxonomy and ecology of cane,
Arundinaria gigantea (Walter) Muhlenberg. Ph.D. Dissertation.
University of Arkansas, Little Rock, USA.
Mann, C. C. 2006. 1491: New Revelations of the Americas Before
Columbus. Vintage Books, New York, NY, USA.
Morse, D. F. and P. A. Morse. 1983. Archaeology of the Central
Mississippi Valley. Academic Press, New York, NY, USA.
Myers, R. K. and D. H. Van Lear. 1998. Hurricane-fire
interactions in coastal forests of the south: a review and
hypothesis. Forest Ecology and Management 103:265–76.
Noss, R. F., E. T. LaRoe III, and J. M. Scott. 1995. Endangered
ecosystems of the United States: a preliminary assessment of
loss and degradation. Biological Report 28. US Department of
Interior National Biological Service, p. 58. Washington, DC.
Oliver, C. D., E. C. Burkhardt, and D. A. Skojac. 2005. The
increasing scarcity of red oaks in Mississippi River floodplain
forests: Influence of the residual overstory. Forest Ecology and
Management 210:393–414.
Olson, M. S. and W. J. Platt. 1995. Effects of habitat and growing
season fires on resprouting shrubs in longleaf pine savannas.
Vegetatio 119:101–18.
Platt, S. G. and C. G. Brantley. 1997. Canebrakes: an ecological
and historical perspective. Castanea 62:8–21.
Platt, S., C. Brantley, and T. Rainwater. 2002a. Canebrakes:
Bamboo forests of the Southeast. Wild Earth 12:39–45.
Platt, W. J. 1999. Southeastern pine savannas. p. 23–51. In R. C.
Anderson, J. S. Fralish, and J. Baskin (eds.) The Savanna,
Barren, and Rock Outcrop Communities of North America.
Cambridge University Press, Cambridge, England.
Platt, W. J., B. Beckage, R. F. Doren, and H. H. Slater. 2002b.
Interactions of large-scale disturbances: prior fire regimes and
hurricane mortality of savanna pines. Ecology 83:1566–72.
Royall, P. D., P. A. Delcourt, and H. R. Delcourt. 1991. Late
Quaternary peleoecology and paleoenvironments of the central
Mississippi alluvial valley. Geological Society of America
Bulletin 103:157–70.
Saikku, M. 2005. This Delta, This Land: An Environmental
History of the Yazoo-Mississippi Floodplain. The University of
Georgia Press, Athens, GA, USA.
Sharitz, R. R., L. R. Boring, D. H. Van Lear, and J. E. Pinder III.
1992. Integrating ecological concepts with natural resource
management of southern forests. Ecological Applications 2(3):
226–37.
Stanturf, J. A., S. H. Schoenholtz, C. J. Schweitzer, and J. P.
Shepard. 2001. Achieving restoration success: myths in
bottomland hardwood forests. Restoration Ecology 9(2):
189–200.
Stewart, M. A. 2007. From king cane to king cotton: razing cane
in the old South. Environmental History 12:59–79.
The Nature Conservancy. 1992. Restoration of the Mississippi
River Alluvial Plain as a Functional Ecosystem. The Nature
Conservancy, Baton Rouge, LA, USA.
Toole, E. R. and J. S. McKnight. 1956. Fire effects in southern
hardwoods. Fire Control Notes 17(3):1–4.
Triplett, J. K., A. S. Weakley, and L. G. Clark. 2006. Hill cane
(Arundinaria appalachiana), a new species of bamboo (Poaceae:
Bambusoideae) from the southern Appalachian Mountains.
SIDA 22(1):79–85.
Tudhope, A. W., C. P. Chilcott, M. T. McCulloch, E. R. Cook, J.
Chappell, R. M. Ellam, D. W. Lea, J. M. Lough, and G. B.
Shimmield. 2001. Variability in the El Niño-Southern Oscillation through a glacial–interglacial cycle. Science 291:1511–17.
Wright, H. A. and A. W. Bailey. 1982. Fire Ecology: United
States and Southern Canada. John Wiley and Sons, New York,
NY, USA.
Wilson, R., K. Ribbeck, S. King, and D. Twedt (eds.)
Restoration, Management, and Monitoring of Forest Resources in the Mississippi Alluvial Valley: Recommendations for
Enhancing Wildlife Habitat. Lower Mississippi Valley Joint
Venture Forest Resource Conservation Working Group,
Vicksburg, MS, USA.
Manuscript received 2 March 2008; accepted 31 October 2008.