WATER USE EFFICIENCY AND IRRIGATION RESPONSE OF COTTON
CULTIVARS GROWN ON SUB-SURFACE DRIP IN WEST TEXAS
by
MICHAEL CHASE SNOWDEN, B.S.
A Thesis
In
CROP SCIENCE
Submitted to the Graduate Faculty
of Texas Tech University in
Partial Fulfillment of
the Requirements for
the Degree of
MASTER OF SCIENCES
Approved
Glen L. Ritchie
Chair of Committee
Thomas L. Thompson
Kevin Mulligan
Peggy Gordon Miller
Dean of the Graduate School
May, 2012
©2012
Chase Snowden
All Rights Reserved
Texas Tech University, Michael Chase Snowden, May 2012 ACKNOWLEDGMENTS
It is hard to know where to begin in acknowledging people who played a part in
helping me get to this point in my life. Dr. Thompson, Thank you for seeing enough in
me to allow me to pursue a graduate degree here at Texas Tech. While you decided to
choose a different career path and change universities, I still have learned a lot from you
over the past three years. I am also grateful for you and Dr. Bednarz collaborating to find
this project that kept me in school after my original project fell through.
I am also very blessed to have such an amazing wife who always had words of
encouragement through the trying times as well as the smiles and kisses to spare during
triumphant periods. Darlin without your love and support, this chapter of my life would
have been a lot harder to get through, I love you so much and am thankful for all you do
for me.
Mom, Dad, Machew and grandparents, while ya’ll were sometimes out of the
loop about what was going on at different phases of grad school, ya’ll were always there
to show support, always concerned about my success and continually proud of me, and I
am grateful to have such a strong family base to stand on.
Dr. Ritchie, oh Dr. Ritchie, there could not have been a better advisor to come in
and guide me through the remaining time in my graduate degree. I have enjoyed every
instance that we have been able to interact, whether it be running stats, collecting data, or
playing tennis together, I have enjoyed it all and have learned a lot from you in the
process. It has been through your guidance that I have been as successful as I have in my
time here at Texas Tech.
ii Texas Tech University, Michael Chase Snowden, May 2012 This work was also made a lot easier and enjoyable through the help of my fellow
cohorts during the long summer days of data collection and harvesting. I know all of ya’ll
particularly enjoyed installing and uninstalling my moisture sensors. Tyler, Heath, Matt,
Fulvio, Luke, Evan, Curtis, Jon, and Bablu otherwise known as the Crop Physiology
Crew, I will cherish the camaraderie we had together and will never forget the good times
we had both in and out of the field.
iii Texas Tech University, Michael Chase Snowden, May 2012 TABLE OF CONTENTS
ACKNOWLEDGMENTS ............................................................................................................. ii
ABSTRACT ................................................................................................................................... vi
LIST OF TABLES ...................................................................................................................... viii
LIST OF FIGURES ...................................................................................................................... ix
LIST OF ABBREVIATIONS ...................................................................................................... xi
I. INTRODUCTION ...................................................................................................................... 1
Cotton Production in Texas ..................................................................................................... 1
Water in the Texas High Plains ............................................................................................... 2
Irrigation Scheduling with Evapotranspiration Estimates ....................................................... 2
Water Use Efficiency ............................................................................................................... 4
Yield ........................................................................................................................................ 5
Cultivar Differences ................................................................................................................. 6
Boll distribution ....................................................................................................................... 7
II. MATERIALS AND METHODS ............................................................................................. 9
Irrigation .................................................................................................................................. 9
In-season Data Collection ...................................................................................................... 10
Harvest ................................................................................................................................... 11
Data Analysis ......................................................................................................................... 11
III. RESULTS AND DISCUSSION............................................................................................ 12
Irrigation ................................................................................................................................ 12
Soil Moisture Content ............................................................................................................ 13
Lint Yield ............................................................................................................................... 15
Water Use Efficiency ............................................................................................................. 17
Boll Distribution .................................................................................................................... 18
IV. CONCLUSIONS .................................................................................................................... 40
iv Texas Tech University, Michael Chase Snowden, May 2012 LITERATURE CITED ............................................................................................................... 42
APPENDIX ................................................................................................................................... 45
v Texas Tech University, Michael Chase Snowden, May 2012 ABSTRACT
The High Plains Aquifer is the source of nearly all agriculture irrigation water in
the Texas High Plains, and its resources are being depleted due to withdrawals that
greatly exceed recharge. Decreasing water availability has led to research on water use
requirements of most agronomic crops, including yield and quality impacts of deficit
irrigation. Some drought-tolerant crops such as cotton (Gossypium hirsutum L.), can
adapt well to deficit irrigation. The objectives of this study were to i) evaluate the water
use efficiency and boll distribution patterns of cotton cultivars at varying levels of subsurface drip irrigation from severe-deficit to fully irrigated, ii) compare growth and yield
characteristics between cultivars at varying irrigation levels, and iii) determine yield
stability under deficit irrigation in West Texas. Yield, water use efficiency and boll
distribution were compared during 2010 and 2011 for cultivars DP 0912, DP 0924, DP
0935, DP 1028, DP 1032, DP 1044, and FM 9160. In 2010, FM 9160, DP 1044, and DP
0912 had the three highest average yields and water use efficiencies. DP1044 and
FM9160 performed very well under deficit irrigation. In 2011 cultivar DP1044 again was
a top performer along with DP0935 and DP0924. Average yield ranges of 1077 to 1256
kg ha-1for 2010 and 958 to 1074 kg ha-1 for 2011 were common to those produced in
West Texas. Water use efficiency was also common for West Texas with ranges of 0.23
to 0.27 for 2010 and 0.17 to 0.19 kg m-3 for 2011. Boll distribution patterns varied
significantly between cultivars and within irrigation treatments. Three cultivars (DP1044,
FM 9160 and DP 0935) increased fruit production near the top of the plants in response
vi Texas Tech University, Michael Chase Snowden, May 2012 to irrigation, and also had good yield and yield stability; their yield patterns may be
favorable for limited water conditions in the Texas High Plains.
vii Texas Tech University, Michael Chase Snowden, May 2012 LIST OF TABLES
1
Monthly Climatic Data for 2010 and 2011 Daily Mean Values Compared
with 30-year Mean Data for Lubbock .......................................................................22
1A
Fiber Quality measurements by high volume instrument (HVI) for
micronaire, staple length, uniformity, and strength in 2010 for all irrigation
treatments and cultivars ............................................................................................46
2A Fiber Quality measurements by high volume instrument (HVI) for
micronaire, staple length, uniformity, and strength in 2011 for all irrigation
treatments and cultivars ............................................................................................47
viii Texas Tech University, Michael Chase Snowden, May 2012 LIST OF FIGURES
1 Total water applied in millimeters for three irrigation treatments in 2010. ...................23 2 Cumulative GDD by days after planting for 2010 and 2011. ........................................24
3 Total water applied in millimeters to four irrigation treatments in 2011.......................25
4 Soil volumetric water content for each irrigation treatment in 2010 .............................26
5 Soil volumetric water content for each irrigation treatment in 2011 ............................27
6 Lint yield expressed in kg ha-1 for each cultivar in 2010. Means followed by
same letter do not differ significantly. Different letters indicate significant
differences at (Pcrit = 0.05) using Fisher’s protected means separation. ......................28
7 Lint yield expressed in kg ha-1 for each cultivar within each irrigation
treatment for 2010. Bars containing the same letters do not differ
significantly within each irrigation treatment, ns indicates no significance. ...............29
8 Lint yield expressed in kg ha-1 for each cultivar in 2011. Means followed by
same letter do not differ significantly. Different letters indicate significant
differences at (Pcrit = 0.05) using Fisher’s protected means separation. ......................30
9 Lint yield expressed in kg ha-1 for each cultivar within each irrigation
treatment for 2011. Bars containing the same letters do not differ
significantly within each irrigation treatment, ns indicates no significance. ...............31
10 WUE expressed as kg m-3 for each cultivar in 2010. Means followed by
same letter do not differ significantly. Different letters indicate significant
differences at (Pcrit = 0.05) using Fisher’s protected means separation. ......................32
11 WUE expressed in kg m-3 for each cultivar within each irrigation treatment
for 2010. Bars containing the same letters do not differ significantly within
each irrigation treatment, ns indicates no significance. ...............................................33
ix Texas Tech University, Michael Chase Snowden, May 2012 12 WUE expressed as kg m-3 for each cultivar in 2011. Means followed by
same letter do not differ significantly. Different letters indicate significant
differences at (Pcrit = 0.05) using Fisher’s protected means separation. ......................34
13 WUE expressed in kg m-3 for each cultivar within each irrigation treatment
for 2011. Bars containing the same letters do not differ significantly within
each irrigation treatment, ns indicates no significance. ..............................................35
14 Number of first position bolls plant-1 by node for each cultivar averaged
over each irrigation treatment ......................................................................................36
15 First position bolls node-1 plant-1 produced over severe-deficit treatment by
node for DP0935 (a), DP1044 (b), and FM9160 (c) cultivars. Bars indicate
LSD values for each node. ...........................................................................................37
16 First position bolls node-1 plant-1 produced over severe-deficit treatment by
node for DP0912 (a), DP 0924 (b), DP1028 (c), DP1032 (d). Bars indicate
LSD values for each node. ...........................................................................................38
17 Second position bolls node-1 plant-1 produced over severe-deficit treatment
by node for FM 9160 (a), DP 1028 (b), and DP 1032 (c) cultivars. Bars
indicate LSD values for each node. .............................................................................39
1A Second position bolls node-1 plant-1 produced over severe-deficit treatment
by node for DP 0912 (1A), DP 0924 (1B), and DP 0935 (1C), and DP 1044
(1D) cultivars. Bars indicate LSD values for each node. .............................................48 x Texas Tech University, Michael Chase Snowden, May 2012 LIST OF ABBREVIATIONS
DAP, Days after planting
ETc, Crop evapotranspiration
ETo, Reference evapotranspiration
GDD, Growing degree days
HVI, High volume instrument
Kc, Crop coefficient
LEPA, low energy precision application
SDI, Sub-surface drip irrigation
WUE, Water use efficiency
xi Texas Tech University, Michael Chase Snowden, May 2012 CHAPTER I
INTRODUCTION
Cotton Production in Texas
The Texas High Plains annually produces 25% of the entire U.S. cotton crop, and
cotton is the leading cash crop in Texas. The cotton production region within a 130-km
radius around Lubbock, TX is semi-arid, and about half of the area is non-irrigated, with
yields that vary with rainfall (Wanjura et al., 2002). Irrigated cotton yields for Lubbock
County and the 25-county area around Lubbock have been increasing since 1986, due to
the availability of irrigation water (Wanjura et al., 2002). The dependence of crop yields
on water supply is becoming a critical issue with the limited water resources in the Texas
High Plains.
In the 2007 USDA census of Agriculture, Lubbock County was one of only six
counties in the United States that sold more than $100 million in cotton. Average yields
for Lubbock County increased from 334 to 1107 kg ha-1 from 2000 to 2007 and have
decreased to 610 kg ha-1 in 2011(NASS-USDA, 2012). These trends are similar to cotton
production in Texas and the United States as a whole. The United States produced higher
averages than both Texas and the Lubbock County, except in 2007, ranging from 705 to
968 kg ha-1.
Decreasing water availability has led to research on water use requirements of
most agronomic crops and yield and quality impacts of deficit irrigation. Some droughttolerant crops, including cotton (Gossypium hirsutum L.), can adapt well to deficit
irrigation (Basal et al., 2009). Under good management deficit irrigation can result in
1 Texas Tech University, Michael Chase Snowden, May 2012 substantial water savings (Kirda, 2002) and maximize yields per unit of water (Basal et
al., 2009).
Water in the Texas High Plains
The Ogallala (High Plains) aquifer is one of the largest aquifer systems in the
world, containing 3.5 billion acre-feet of water and underlying 174,000 square miles
across South Dakota, Nebraska, Wyoming, Colorado, Kansas, Oklahoma, New Mexico,
and Texas (Torell et al., 1990). The aquifer is a closed basin in which withdrawals greatly
exceed recharge. This has resulted in severe decline of groundwater levels since pumping
was commenced from the aquifer (Colaizzi et al., 2008). Over 90% of the Ogallala
withdrawals in the Texas High Plains are for agriculture irrigation, so it is essential that
water be used as efficiently as possible to maintain irrigated crop production. The
reversion to non-irrigated farming in the West Texas area will result in a significant
amount of total cotton production in Texas, and the United States to be lost.
Efficient irrigation water use is particularly important in the Southern High Plains
of Texas, because rainfall events are erratic and vary considerably from year to year.
Many producers are installing more efficient systems, such as low energy precision
application (LEPA) sprinklers and subsurface drip irrigation (SDI) to increase water use
efficiency (WUE) and yield compared to conventional irrigation methods (Cetin and
Bilgel, 2002; Mateos et al., 1991; Hodgson et al., 1992).
Irrigation Scheduling with Evapotranspiration Estimates
Several methods of irrigation, including low energy precision application (LEPA)
sprinklers and subsurface drip irrigation (SDI) are being used to increase WUE and yield;
2 Texas Tech University, Michael Chase Snowden, May 2012 however, the challenge still remains of determining irrigation rates within a specific
system. Determining rates of irrigation for specific environments is difficult, because
environmental factors and soil type will affect the amount of irrigation that is needed by
the crop. One solution is the use of a reference measure of evapotranspiration (ETo),
calculated from a standardized reference surface under ambient environmental conditions
(Walter et al., 2000). Relating ET to a specific surface allows ET to be transferable across
different environments. Therefore, the only factors affecting ETo are climatic, including
temperature, solar radiation, relative humidity, and wind speed, which can be computed
from weather data using the FAO Penman-Monteith equation (Allen et al., 1998).
A crop coefficient (Kc) can be applied by expressing the ratio between crop ET
(ETc) and the reference ET. This allows for estimates of crop water use at specific growth
stages during the growing season. The single crop coefficient outlined in the Allen et al.
(1998) approach combines crop transpiration and soil water evaporation into one value
and accounts for crop growth stage by assuming a low demand at the beginning of the
season, increasing demand during crop development, maximum water use at mid-season,
and sharply declining transpiration late in the season.
Allen et al. (1998) expanded on this by adding a second Kc equation to account
for crop transpiration, soil evaporation, and water stress. Tolk and Howell (2001) tested
the single Kc and dual Kc models on sorghum grown in the Texas High Plains and found
that the dual Kc model improved water use predictions over the single Kc. Howell et al.
(2004) used the dual Kc model to establish crop coefficients for cotton in the Texas High
Plains based on the accumulation of heat over the growing season as growing degree days
(GDD), instead of a model solely based on days after planting (DAP). This procedure has
3 Texas Tech University, Michael Chase Snowden, May 2012 been used to more accurately estimate and relate crop water use over a variety of
locations and growing seasons (Howell et al., 2004).
Water Use Efficiency
Depletion of water resources has prompted improvements in WUE, not only
through improved irrigation methods but also enhanced management practices (Jenkins et
al., 1990). There are many ways to improve WUE by management practices, including
tillage, irrigation scheduling, cultivar advances, and crop selection for area (Howell et al.,
2004) . Irrigation can also be an effective means to improve WUE through increased crop
yields, especially in semiarid and arid environments like the Texas High Plains (Howell,
2001). The most recent research incorporates simulation models to address all scenarios
involved with improving WUE (Evett and Tolk, 2009; Baumhardt et al., 2009); however,
these models do not examine differences between cultivars.
Definitions for WUE are desirable, and for full irrigated applications are generally
accurate (Evett and Tolk, 2009). However, under deficit and dryland conditions,
definitions are less accurate because field practices vary from dryland to irrigated
agriculture, making it difficult to standardize WUE. Field practices include, but are not
limited to soil fertility, pest management, planting density, and row spacing (Howell,
2001), and all of these can affect yield. Defined physiologically, WUE is the unit of dry
matter produced per unit of water used (Viets, 1962). In production, this definition is
often summarized as the unit yield per unit of irrigation, because yield is easier to
measure than in-season biomass changes. Several studies have shown WUE to increase
with yield and irrigation at decreasing marginal returns (Howell et al., 2004; and Grismer,
4 Texas Tech University, Michael Chase Snowden, May 2012 2002). In contrast, Dagdelen et al. (2009) reported that WUE increased with deficit
irrigation, and that the more severe the deficit, the more efficiently the crop converted
water to yield. Basal et al. (2009) in Turkey also found that WUE was inversely
proportional to irrigation applied from 100% to 75% of soil water depletion. WUE was
also higher in these studies than those produced by Howell et al. (2004).
Yield
Irrigation has proven to increase yields for Lubbock county and the 25-county
area around Lubbock which have been increasing since 1986 (Wanjura et al., 2002).
Irrigation and total water of 58 cm and 74 cm, respectively, described in Wanjura et al.
(2002), achieve maximum yield in the West Texas area. Total seasonal water application
by DeTar (2008) for the San Joaquin Valley of California produced similar values to
Wanjura et al. (2002), with a critical level of 65.4 cm to achieve maximum yield.
Wanjura also concluded that a 5% reduction of water below the critical level would
produce a 4.6% reduction in yield. Wanjura and Upchurch (2000) compared yield
reduction differences with deficit irrigation between corn and cotton and concluded that
the level of water deficit that decreased yield by 60% in corn only resulted in a 17% yiled
decrease in cotton. This difference in yield reductions could be attributed to differences
between an indeterminate and determinate crop. Harvestable portions of corn decrease
more when water deficit stress affects critical growth stages compared to cotton which
aborts or produces less fruit.
It is well established that yield decreases as water deficit increases. Dagdelen et
al. (2009) and Basal et al. (2009) found no yield reductions when reducing irrigation from
5 Texas Tech University, Michael Chase Snowden, May 2012 a 100% to 75% of soil water depletion. Total yield in Dagdelen et al. (2009) decreased as
deficit increased, with severe water stress resulting in over 50% yield reduction compared
to the fully irrigated crop.
Irrigation treatments for Basal et al. (2009) were set up the same as described in
Dagdelen et al. (2009); however, decreases in yield were less across all irrigation
treatments. Decreasing water application by 25%, 50%, and 75% reduced yield by 8%,
20-30%, and 40-45%, respectively. Other studies have also demonstrated varying
decreases in seed yield under water-stress (Cook and El-Zik, 1993; Lopez et al., 1995;
Saranga et al., 1998; Pettigrew, 2004; Dagdelen et al., 2006; DeTar, 2008).
Cultivar Differences
Deficit irrigation has been reported to significantly affect WUE and yield;
however, differences between yields and WUE among cotton cultivars are less
documented. Detar (2008) compared the yield and WUE of two cultivars and observed
that both cultivars had similar responses to irrigation rate. There were no significant
differences in yield or WUE at any of the six application rates, which ranged from 33%
to 144% of Class A pan evaporation. Basal et al. (2009) investigated fiber quality
parameters for two cultivars, and concluded that deficit irrigation of 75% of soil water
depletion affected and had differing effects on fiber quality between cultivars. These
results indicate the 75% deficit irrigation effect on fiber quality was determined by cotton
genotype, and cotton could be irrigated with the 75% deficit level without reducing lint
yield in regions where water is limited. Pettigrew (2004) evaluated lint yield differences
6 Texas Tech University, Michael Chase Snowden, May 2012 between eight different cultivars at irrigated and dryland moisture treatments and
concluded that the lint yield response was similar among genotypes at each water level.
Boll distribution
Irrigation studies have documented how cotton growth and yield are altered by
moisture deficit. However, much less is known about how boll distribution and other
components of yield are affected. Changes in fruiting habits can affect in-season plant
growth, including plant maturity. Ritchie et al. (2009) concluded that irrigation method
had a significant effect on boll distribution. They compared overhead and SDI irrigation
methods and concluded that SDI produced more bolls at the base of the plant, and fewer
bolls at the top compared to overhead irrigation, resulting higher yields with SDI.
Pettigrew (2004) also compared boll distribution characteristics between irrigated and
dryland cotton under furrow irrigation. He concluded that irrigation alters both the
horizontal and vertical distribution of bolls. Yield reduction by moisture deficit stress was
primarily due to reduction in number of bolls and less lint per seed, while increases in
yield by irrigation were due to additional bolls being set at higher nodes and more distal
positions on the plant.
The purpose of this study was to determine differences between WUE, yield, and
boll distribution among cotton cultivars with different irrigation treatments, and
determine which cultivars performed better under deficit irrigation. The main objectives
for this research included: (i) evaluating the WUE and boll distribution of cotton cultivars
at different rates of SDI; (ii) comparing yield, WUE and boll distribution between
cultivars at varying irrigation levels; and (iii) determining the cultivar with best yield
7 Texas Tech University, Michael Chase Snowden, May 2012 stability under deficit irrigation to reduce irrigation water use without reducing yield in
West Texas.
8 Texas Tech University, Michael Chase Snowden, May 2012 CHAPTER II
MATERIALS AND METHODS
Research was conducted for two years at the Texas Tech Research Farm in New
Deal, Texas. The soil is a Pullman clay loam (fine, mixed, superactive, thermic Torrertic
Paleustoll). The experimental design was a split plot design with three replications, with
irrigation as main plot and cultivar as split plot. Seven cotton cultivars were examined:
DP 0912 B2RF, DP 0924 B2RF, DP 0935 B2RF, DP 1028 B2RF, DP 1032 B2RF,
DP1044 B2RF, and FM 9160 B2F.
Planting dates were May 26, 2010 and May 17, 2011. The seeding rate was 18
seeds per linear meter, and row spacing was 102 cm. Plots were four beds wide and 10.6
m long in 2010 and 12.2 m long in 2011. Fertility, weed control, and field production
practices were based on extension recommendations in West Texas.
Irrigation
The SDI system at New Deal consists of drip tape under every raised bed on 102
cm centers, 20-24 cm below the surface with 61 cm emitter spacing. The irrigation
treatments in 2010 were severe-deficit, moderate-deficit, and fully irrigated and 2011 had
severe-deficit, moderate-deficit, mild-deficit, and fully irrigated treatments. Irrigations
were applied daily using the SDI system. Irrigation treatments were initiated at first
square and adjusted weekly based on changes in weekly evapotranspiration readings.
Irrigation was monitored using the crop evapotranspiration (ETc) equation described in
Allen et al. (1998).
ETC= ETo x Kc
9 (1)
Texas Tech University, Michael Chase Snowden, May 2012 ETo is the reference evapotranspiration determined by weather data inputs into
the Penman-Montieth equation from Allen et al. (1998). Kc is the single crop coefficient
derived from Fig. (4) in Howell et al. (2004).
In-season Data Collection
Each cultivar was planted in four rows, but data collection and harvest were
performed in the middle two rows. Emergence was measured at or before the 4-leaf stage
by counting plants in a 4 m section on each data row.
During the season, plant height and total nodes were measured weekly on
cultivars DP 1044 and DP 0912, with initial, mid-season, and final measurements
conducted on all cultivars. Final plant measurements included plant height, total nodes,
node of first fruiting branch, node with cracked boll, and node of uppermost harvestable
boll on five consecutive plants from every plot in the trial. In 2011, boll distribution was
also determined on five consecutive undamaged plants from every plot by recording bolls
present by fruiting site on each individual plant.
Soil moisture was monitored weekly using neutron probe readings for soil
volumetric water content, beginning at pin-head square and continuing until cutout on
cultivars DP 0912 and DP 1044 at all irrigation levels on two replicates. Measurements
were conducted to a depth of 100 cm in 20-cm increments. The data gathered by the
neutron probe were used to calculate a water balance each week to ensure accuracy of
irrigation treatments throughout the season.
Three soil cores were collected at planting for the 2011 season to determine soil
volumetric water content. Samples were collected using a tractor-mounted hydraulic
10 Texas Tech University, Michael Chase Snowden, May 2012 Giddings probe to a maximum depth of 120 cm, due to underlying calcic horizon. The
collected samples were divided into 0-30 cm, 30-60 cm, and deeper than 60 cm. The soil
was oven-dried for 48 hours, and bulk density and volumetric water content were
determined for each depth to establish a baseline water level for season-long monitoring
of water balance.
Harvest
Yield was determined from the two middle rows of each plot, and 1.5 kg fiber
sub-samples were also collected at the time of harvest, ginned by Monsanto, and
analyzed using a high volume instrument (HVI) at the Fiber and Biopolymer Research
Institute in Lubbock, TX. WUE was determined as lint yield (expressed in g m-2), divided
by the depth of total water applied.
Data Analysis
Yield and WUE averages over all irrigation treatments were analyzed for each
cultivar and irrigation treatment. Boll distribution patterns were also analyzed by
irrigation treatment and cultivar in the study.
Boll numbers were compared between cultivars by fruiting node and position to
determine boll distribution on five consecutive plants in each plot. Analysis for boll
distribution characteristics was conducted on each node independently, averaged over all
irrigation treatments and by changes in boll patterns with increasing irrigation for each
cultivar. All data were analyzed using SAS proc GLIMMIX, with random effects defined
to deal with the fixed and random effects for a split plot design, as discussed by Littell et
al. (2006)
11 Texas Tech University, Michael Chase Snowden, May 2012 CHAPTER III
RESULTS AND DISCUSSION
Irrigation
In 2010, most of the total water available to the crop came as rainfall, with 330
mm of rainfall falling during the growing season. Rainfall amounts greater than 13 mm
occurred 28, 106, and 148 DAP. Irrigation amounts for the treatments were 28, 121, and
206 mm applied to the severe deficit, moderate deficit, and fully irrigated treatments
respectively (Fig. 1). The total water applied for the fully irrigated treatment
corresponded to 71% of ETc replacement, with the moderate-deficit and severe-deficit
treatments having 60% and 48% ETc replacement respectively. One explanation for the
fully irrigated treatment only achieving 71% of ETc replacement could be attributed to
the effective timing of the rainfall during the season.
There were 1181 growing degree days (GDD°C) in 2010 (Fig. 2), similar to the
environmental conditions described by Howell et al., (2004), allowing for similar Kc
values observed in the study to be applied over the 2010 growing season. Other climatic
data for 2010 and 2011, including temperature, rainfall, radiation load, and wind speed,
are shown in Table 1.
The 2011 season was extremely hot and dry, and total water applied for 2011 was
higher than 2010, although the growing season was 25 days shorter. Virtually all of the
available water came from irrigation, in contrast to 2010. Total in-season rainfall in 2011
was 46 mm, with no rainfall amounts resulting over 13 mm in one event. The total
amount of irrigation applied in 2011 was 355 mm for the severe-deficit, 473 mm for the
12 Texas Tech University, Michael Chase Snowden, May 2012 moderate-deficit, 527 mm for the mild-deficit, and 638 mm for the fully irrigated
treatment (Fig. 3). The ETc replacement percentages for all three deficit treatments in
2011 were similar to the three treatments in 2010 with 48% for the severe-, 63% for the
moderate-, and 69%, for the mild-deficit irrigation. The fully irrigated treatment resulted
in an ETc replacement of 83%.
Cumulative GDDs were much higher in 2011 than 2010, due to the hot
temperature and high winds experienced during the growing season. Total GDD for 2011
was 1415°C, which was higher than those experienced in Howell et al. (2004). Therefore,
Kc values had to be interpolated to account for an additional accumulation of 280°C
GDD. However, these values are not unheard of for the Southern High Plains. Wanjura et
al. (2002) experienced GDD values as high as 1576°C in one year of a 12 year study,
with a point of maximum yield range between 1092 and 1368 heat units (HU) on surface
drip irrigated cotton at the Texas Agriculture Experiment Station in Lubbock, TX.
Soil Moisture Content
Volumetric water contents (θv) for the irrigation treatment in 2010 are shown in
Fig. 4. Lighter colors represent areas of high water content and darker areas indicate
decreasing water content. During the season, moisture deeper in the soil decreased as the
crops extracted stored water in the profile. This was particularly apparent in the severedeficit treatment. The region with the highest soil moisture was around 40 cm depth of
each irrigation treatment, next to the drip tape. The fully irrigated treatment in 2010
consistently had a θv of near 0.28 at the 40 cm depth throughout the season with
13 Texas Tech University, Michael Chase Snowden, May 2012 decreases down to 0.24 to 0.22 near 80 cm depth. The lowest volumetric water content
was at 120 DAP between 80 and 100 cm.
The moderate-deficit treatment followed the same trend as the fully irrigated, but
with greater soil water depletion with increasing depth and DAP. The θv at the 40 cm
depth ranged from 0.28 at the beginning of measurements to 0.22 by the end of the
season. The lowest moisture content for the moderate-deficit treatment was 0.19, which
occurred between 80 and 100 cm near the end of the season. The severe-deficit treatment
had the lowest soil water content throughout the profile compared to the other treatments
with the highest θv of 0.24 at the 40 cm depth and significant decreases at DAP 95 in the
lower portions of the profile. Soil water was depleted to 0.18 on DAP 100 at the 90 cm
depth and decreased to a depth of 80 cm by the end of the season.
Soil water content for the irrigation treatments in 2011 followed a different
pattern throughout the soil profile than in 2010 (Fig. 5). The fully irrigated treatment had
a θv of 0.22 that persisted to 60 cm throughout the season. Significant decreases in soil
moisture began 90 DAP and continued throughout the season with the decreases to 0.16
occurring after DAP 100 between 80 and 100 cm. The mild-deficit treatment had similar
soil moisture at the 40 cm depth and greater soil water depletion at the lower profile
depths compared to the fully irrigated treatment. θv decreased to 0.16 between 70 and 100
cm, 80 DAP and continued to decrease.
The moderate-deficit treatment did not follow the same soil water depletion trends
as seen in the higher irrigation treatments. The highest θv was at the 40 cm depth with
0.26 which decreased to 0.21 by the end of the season. Moisture depletion at deeper
depths in the profile was not as severe as the mild-deficit and fully irrigated treatments
14 Texas Tech University, Michael Chase Snowden, May 2012 with the lowest moisture content of 0.20 at the 80 to 100 cm depth after 100 DAP. θv of
0.22 were still evident at 85 DAP at this depth, which is a much higher soil moisture
content than the higher irrigated treatments at this depth. The severe-deficit treatment
followed a similar trend to the moderate-deficit treatment except with greater soil water
depletion at the soil surface and 80 cm depths in the profile. Highest θv was at the 40 cm
depth with 0.22 at the beginning of the season that decreased to 0.20 by the end of the
season. Beginning 80 DAP, soil water at the surface was decreased to 0.18 θv which
reached 30 cm from the surface by the end of the season. Large decreases in soil moisture
did not begin till 85 DAP were soil moisture was 0.22 between 70 and 100cm and
continued to decrease. θv reached its lowest point of 0.18 at the 80 cm depth 100 DAP.
Decreases in θv deeper in the soil were attributed to increased zone of root
exploration and increased transpiration in the mild-deficit treatment and the fully
irrigated treatment. The moderate-and severe-deficit treatments resulted in almost all
water uptake occurring near the drip tape.
These results are comparable to Whitaker et al. (2008) who looked at soil
moisture tension for SDI. Total water applied for the study was also similar however;
timing and frequency or irrigation applications differed.
Lint Yield
Lint yield averages over all irrigation treatments for 2010 were 1074 to 1256 kg
ha-1 (Fig. 6). FM 9160 and DP 1044 had the two highest averages and were significantly
different from all other cultivars except DP 0912. Yields increased with water
application.
15 Texas Tech University, Michael Chase Snowden, May 2012 In the severe-deficit treatment, FM 9160 produced the highest yield at 711 kg ha-1
followed by DP 1044 with 655 kg ha-1. DP 1044 and FM 9160 were also the top
performers in the moderate-deficit treatment, producing 1435 and 1341 kg ha-1
respectively. In the fully irrigated treatment, DP 0912 had the highest yields, with 1743
kg ha-1, the highest in the study. FM 9160 and DP1044 followed with 1714 and 1621 kg
ha-1 respectively. Lint yields for all cultivars within each irrigation treatment for 2010 are
shown in Fig. 7. The average yield for the severe-deficit treatment was 606 kg ha-1. The
moderate-deficit treatment increased yields by 634 kg ha-1 to result in an average of 1240
kg ha-1 and was the only irrigation treatment to produce significant results with a P value
of (0.02). Cultivar DP 1044 was significantly different from all other cultivars except FM
9160 for this treatment. The yield increase from the moderate-deficit to the fully irrigated
treatment was 367 kg ha-1 and had an average of 1607 kg ha-1.
Lint yield averages in 2011 were lower than 2010 ranging from 958 to 1074 kg
ha-1 (Fig. 8). The highest average yields were produced by DP 0935 and DP 1044, with
the DP 0935 average differing significantly from all other cultivars except DP 1044. DP
0935 produced the highest yields in the severe-, mild-deficit, and fully irrigated
treatments with 595, 1274, and 1537 kg ha-1 respectively. DP 1032 produced the highest
yield in the moderate-deficit treatment with 910 kg ha-1. Yield values for the two years of
the study do not differ greatly from those produced in other studies in the area (Wanjura
et al., 2002; Howell et al., 2004). Lint yields for each irrigation treatment are in Fig. 9.
Yields increased with irrigation; however the only treatment that had significant cultivar
difference was the severe-deficit, which had an average yield of 539 kg ha-1. In this
treatment DP 0935 significantly differed from cultivars FM 9160 and DP 0912. The
16 Texas Tech University, Michael Chase Snowden, May 2012 moderate-deficit treatment increased yields by 330 kg ha-1 over the severe-deficit with an
average of 869 kg ha-1. DP 0935 and DP 1044 produced yields that were 100 kg ha-1 over
the other cultivars; however these differences were not significant against the other
cultivars in the treatment. The mild-deficit treatment increased yields by 304 over the
moderate-deficit treatment and the fully irrigated treatment increased yields 300 kg ha-1
over the moderate-deficit. The average yields were 1173 and 1473 kg ha-1 for the milddeficit and fully irrigated treatments respectively.
Water Use Efficiency
The average WUE was 0.23 to 0.27 kg m-3 in 2010 (Fig. 10). FM 9160 had the
highest WUE for the severe-deficit treatment, (0.20 kg m-3). DP 1044 had the highest
WUE in the moderate deficit treatment (0.32 kg m-3), and DP 0912 had the highest WUE
in the fully irrigated treatment (0.33 kg m-3). WUE values for all cultivars within each
irrigation treatment are shown in Fig. 11. The severe-deficit treatment had an average
WUE of 0.17 kg m-3 while the moderate-deficit treatment increased to 0.28 kg m-3. The
fully irrigated produced the highest average of 0.30 kg m-3. The moderate-deficit
treatment was the only irrigation treatment that resulted in significant differences
between cultivars, with DP 1044 having significantly higher WUE than all other cultivars
except FM 9160.
Average WUE were lower in 2011, with a range of 0.17 to 0.19 kg m-3 (Fig. 12).
DP 0935 had the highest WUE in the severe-deficit, mild-deficit, and fully irrigated
treatments with 0.15, 0.23 and 0.22 kg m-3, respectively. In the moderate-deficit
treatment, DP 1032 had the highest WUE, with 0.17 kg m-3. Values for WUE in 2011 are
17 Texas Tech University, Michael Chase Snowden, May 2012 similar to those reported by Howell et al., (2004); however WUE values for the 2010
season were numerically greater. DeTar 2008 also produced similar WUE averages on
six irrigation treatments over four years from 0.13 to 0.22 kg m-3 for California. These
values are much lower than those produced by Dagdelen et al. (2006) in Turkey, where
WUE values of 0.60 to 0.74 kg m-3 were common.
Figure 13 shows WUE values for individual irrigation treatments in 2011. The
only treatment that resulted in significant differences among cultivars was the severedeficit treatment. Within the severe-deficit treatment DP 0935 had a significantly higher
WUE value than cultivars FM 9160 and DP 0912. The average WUE for the severedeficit treatment was 0.13 kg m-3, compared to 0.17 kg m-3 with moderate-deficit. The
mild-deficit and fully irrigated treatments collectively resulted in an average WUE of
0.21 kg m-3.
Boll Distribution
Differences in number of first position bolls per plant produced for each cultivar
can be seen in Fig. 14, where boll distribution patterns are averaged over all irrigation
treatments. The graph represents the percentage of plants that produce a boll at each
node. Nodes 5 through 7 show similar numbers of bolls per plant for each cultivar;
however at node 8 patterns begin to separate. At this growth stage, FM 9160 began to
produce fewer bolls than all other cultivars in the study. This pattern persisted through
node 11 at which point FM 9160 did not show the sharp decline in boll production at
higher nodes in the plant like the other cultivars. This resulted in more bolls being
produced at higher portions in the plant than all other cultivars except DP 1044. DP 1044
18 Texas Tech University, Michael Chase Snowden, May 2012 produced the most bolls per plant in the middle and upper portions of the plant (node 9
through 15), which was significantly higher than all other cultivars except DP 0935.
Turnout percentage demonstrated a weakness for DP 1044, because DP 0935 had the
highest average yield for 2011, although DP 1044 produced the highest seed cotton yield
and had the high boll production in upper portions of the plant.
Both irrigation and cultivar had significant effects on first position boll production
and retention. Fig. 15 shows the three highest yielding cultivars in 2011, and their
differing boll distribution patterns with irrigation. Each graph represents the change in
boll node-1 plant-1 over the severe-deficit treatment with increasing irrigation. As shown
in Fig. 15a, boll production change for DP 0935 decreased by 5 to 15% in lower portions
of the plant (node 9 and below) for each irrigation treatment over severe-deficit. There
were significant increases of 10 to 45% in bolls produced at higher portions of the plant
(node 10 and above) with increasing irrigation. DP 1044 (Fig. 15b) followed a similar
pattern, with boll reduction in lower portions (node 7 and below) of 5 to 15% in the
moderate- and mild-deficit treatments only. Boll production higher on the plant (node 8
and above) was higher than DP 0935 with increases of 21 to 55% with increasing
irrigation.
Boll production for FM 9160 (Fig. 15c) followed a very unique pattern with no
reductions in lower portions of the plant, but rather two peaks in bolls produced at nodes
7 and 13 for each irrigation treatment and similar patterns as irrigation was increased
compared to variable patterns by DP 0935 and DP 1044. The bimodal distribution of
increased fruit for FM 9160 suggest that this cultivar had two periods of time where it
was either particularly sensitive to stress, or prone to produce additional fruit. 19 Texas Tech University, Michael Chase Snowden, May 2012 Figure 16 (a, b, c, d) shows the change in first position bolls per node per plant as
irrigation increased for the remaining cultivars in the study. DP 0912 (Fig. 16a) was
similar to both DP 0935 and DP 1044; boll production decreased at lower portions of the
plant (nodes 8 and below) and increased at upper portions of the plant (nodes 9 and
above). Decreases in boll production ranged from 10 to 22% while boll production at
upper portions of plant increased from 20 to 42% with increasing irrigation. This trend
was observed for the moderate- and mild-deficit treatments for the DP 0924 (Fig. 16b);
however the fully irrigated treatment resulted in increased boll production for DP 0924,
and slightly less boll production at the top of the plant than mild-deficit treatment. DP
1028 (Fig. 16 c) and DP 1032 (Fig. 16d) both have varying distribution patterns that do
not follow any other cultivars. These cultivars have less boll production in middle
portions of the plant than all other cultivars with increases of only 25 to 35 percent for the
DP 1032 and DP 1028 respectively. This increase of 35% for the DP 1028 occurs
between nodes 7 and 10 and declines in the upper portions. For DP 1032, boll production
gradually increased up to node 11in the fully irrigated treatment with 25% more boll
production over severe-deficit before declining.
These results add to the understanding of differences in boll distribution patterns
observed by Ritchie et al. (2009) by demonstrating that even within a single irrigation
method, SDI, cultivars and irrigation rates have significant effects on changes in boll
distribution patters. These results also confirm findings by Pettigrew. (2004) that
irrigation produces bolls higher and at more distal positions on the plant. This study also
demonstrates that individual cultivar boll patterns are unique and are not consistent as
additional water is applied.
20 Texas Tech University, Michael Chase Snowden, May 2012 Boll distribution patterns for additional second position bolls with irrigation are
represented in Fig. 17Figure 17, for cultivars FM 9160 (Fig. 17Figure 17a), DP 1028 (Fig.
17b), and DP 1032 (Fig 17c). These cultivars produced unique patterns for second
position bolls that differed from all other cultivars. Each irrigation treatment for FM 9160
produced very similar boll numbers as the severe-deficit treatment with the highest
increase of only 10% for each irrigation treatment over severe-deficit treatment. DP 1028
produced most of its yield in the mild-deficit treatment at the second position with a
gradual increase of bolls produced up to 35% at node 7 before declining. This is the only
cultivar in the study to produce such a pattern. DP 1028 produced very similar boll
production in the fully irrigated treatment as the severe-deficit, and the moderate-deficit
treatment had reductions in bolls produced that decreased from node 5 to 8 by 15%
before increasing to boll production rates similar to the severe-deficit. DP 1032 showed a
similar trend only not as extreme as DP 1028 with gradual increases in boll production in
the mild-deficit of 20% up to node 8 before declining. Decreases in boll production were
seen in the moderate-deficit treatment with a gradual decline of 15% up to node 8 before
production began to increase. The fully irrigated treatment had some increase in number
of bolls produced but only an 18% occurring at node 11 before declining.
21 Texas Tech University, Michael Chase Snowden, May 2012 Table 1. Monthly Climatic Data for 2010 and 2011 Daily Mean Values Compared
with 30-year Mean Data for Lubbock
Month 2010 May June July August September October November Average 2011 May June July August September October November† Average 30‐year Average‡ May June July August September October November Average Avg. Temp (°C) 20.8 27.2 25.2 26.8 23.8 17.6 10.4 21.7 Max Temp (°C) 28.1 34.4 30.3 34.0 31.1 26.3 19.4 29.1 Min Temp (°C) 13.6 20.0 20.1 19.6 16.6 8.8 1.5 14.3 Rain (mm) 29 65 181 34 24 66 2 57 Wind (m s‐1) 4.5 3.9 2.9 2.5 2.9 2.6 3.3 3.2 Tot. Radiation
(MJ m‐2 day‐1) 24.0 26.1 22.6 24.5 19.6 16.7 13.6 21.0 21.5 29.9 30.0 29.9 22.1 16.9 10.4 23.0 21.0 25.2 26.8 26.1 22.1 16.4 9.9 21.1 30.6 38.2 37.3 37.3 30.1 24.9 17.9 30.9 28.8 32.6 33.8 32.9 29.2 24.0 17.6 28.4 12.5 21.6 22.7 22.6 14.1 8.8 2.8 15.0 13.3 25.2 19.8 19.2 14.9 8.8 2.2 14.8 1 0 1 9 32 34 7 12 58 77 49 49 64 49 22 53 4.8 4.7 2.9 2.5 2.7 3.4 3.7 3.5 28.7 29.2 27.0 25.1 19.3 16.9 12.7 22.7 NA NA* † Cotton was harvested on October 7, 2011. ‡ Based on weather.gov (National Weather Service) data for average temp, rain, and 30yr average (1981‐
2010) for Lubbock, with Wind and Radiation for 2010 and 2011 from weather station data at New Deal. * NA, Data not avaliable
22 Texas Tech University, Michael Chase Snowden, May 2012 Figure 1. Total water applied in millimeters for three irrigation treatments in 2010.
23 Texas Tech University, Michael Chase Snowden, May 2012 o
Growing Degree-Days ( C)
1800
2010
2011
1600
1400
1200
1000
800
600
400
200
0
0
20
40
60
80
100
120
140
160
180
DAP
Figure 2. Cumulative GDD by days after planting for 2010 and 2011.
24 200
Texas Tech University, Michael Chase Snowden, May 2012 Figure 3. Total water applied in millimeters to four irrigation treatments in 2011.
25 Texas Tech University, Michael Chase Snowden, May 2012 Depth (cm)
-20
-30
-40
-50
-60
-70
-80
-90
Severe-Deficit
-100
-20
-30
-40
-50
-60
-70
-80
-90 Moderate-Deficit
-100
-20
-30
-40
-50
-60
16
-70
16
18
18
20
-80
20
22
22
-90
24
Fully Irrigated
24
-100
80 85 90 95 100 105 110 115
Days after Planting
Figure 4. Soil volumetric water content for each irrigation treatment in 2010
by increasing depth and days after planting.
26 Texas Tech University, Michael Chase Snowden, May 2012 Depth (cm)
-20
-30
-40
-50
-60
-70
-80
-90
-100
-20
-30
-40
-50
-60
-70
-80
-90
-100
-20
-30
-40
-50
-60
-70
-80
-90
-100
-20
-30
-40
-50
-60
-70
-80
-90
-100
Severe-Deficit
18
20
22
24
26
28
Moderate-Deficit
Mild-Deficit
18
20
22
24
26
28
Fully Irrigated
60
70
80
90
Days After Planting
100
Figure 5. Soil volumetric water content for each irrigation treatment in 2011
by increasing depth and days after planting.
27 Texas Tech University, Michael Chase Snowden, May 2012 1400
Lint Yield, kg ha
-1
1200
ab
b
b
b
b
a
a
1000
800
600
400
200
DP 0912 DP 0924 DP 0935 DP 1028 DP 1032 DP 1044 FM 9160
Figure 6. Lint yield expressed in kg ha-1 for each cultivar in 2010. Means followed
by same letter do not differ significantly. Different letters indicate significant
differences at (Pcrit = 0.05) using Fisher’s protected means separation.
28 Texas Tech University, Michael Chase Snowden, May 2012 2000
1800
Lint yield kg ha
-1
1600
1400
ns
DP 0912
DP 0924
DP 0935
DP 1028
DP 1032
DP 1044
FM 9160
a
bb b ba b
ccc c
c
1200
1000
800
ns
600
400
200
0
Severe-Deficit
Moderate-Deficit
Fully Irrigated
Figure 7. Lint yield expressed in kg ha-1 for each cultivar within each irrigation
treatment for 2010. Bars containing the same letters do not differ significantly, ns
indicates no significance.
29 Texas Tech University, Michael Chase Snowden, May 2012 Figure 8. Lint yield expressed in kg ha-1 for each cultivar in 2011. Means followed
by same letter do not differ significantly. Different letters indicate significant
differences at (Pcrit = 0.05) using Fisher’s protected means separation.
30 Texas Tech University, Michael Chase Snowden, May 2012 1600
1400
Lint yield kg ha
-1
1200
1000
DP 0912
DP 0924
DP0935
DP 1028
DP 1032
DP 1044
FM 9160
ns
ns
ns
800
600
400
c
aab
a
a
b a bb c
200
0
Severe-Deficit Moderate-Deficit
Mild-Deficit
Fully Irrigated
Figure 9. Lint yield expressed in kg ha-1 for each cultivar within each irrigation
treatment for 2011. Bars containing the same letters do not differ significantly, ns
indicates no significance.
31 Texas Tech University, Michael Chase Snowden, May 2012 0.30
WUE kg m-3
0.25
ab
a
b
b
b
a
b
0.20
0.15
0.10
0.05
0.00
DP0912
DP0924
DP0935
DP1028
-3
DP1032
DP1044
FM9160
Figure 10. WUE expressed as kg m for each cultivar in 2010. Means followed by
same letter do not differ significantly. Different letters indicate significant
differences at (Pcrit = 0.05) using Fisher’s protected means separation.
32 Texas Tech University, Michael Chase Snowden, May 2012 0.35
0.30
WUE kg m-3
0.25
0.20
DP 0912
DP 0924
DP 0935
DP 1028
DP 1032
DP 1044
FM 9160
bb
cc c
c
b
c
aa
b
ns
ns
0.15
0.10
0.05
0.00
Severe-Deficit
Moderate-Deficit
Fully Irrigated
Figure 11. WUE expressed in kg m-3 for each cultivar within each irrigation
treatment for 2010. Bars containing the same letters do not differ significantly, ns
indicates no significance.
33 Texas Tech University, Michael Chase Snowden, May 2012 0.25
0.20
WUE kg m
-3
cd
bc
a
bc
bc
ab
cd
0.15
0.10
0.05
0.00
DP 0912 DP 0924 DP 0935 DP 1028 DP 1032 DP 1044 FM 9160
Figure 12. WUE expressed as kg m-3 for each cultivar in 2011.
Means followed by same letter do not differ significantly. Different letters indicate
significant differences at (Pcrit = 0.05) using Fisher’s protected means separation.
34 Texas Tech University, Michael Chase Snowden, May 2012 0.25
WUE kg m-3
0.20
DP 0912
DP 0924
DP 0935
DP 1028
DP 1032
DP 1044
FM 9160
0.15
c
ns
ns
ns
a a a ab
ab
b
bc
0.10
0.05
0.00
Severe-Deficit
Moderate-Deficit
Mild-Deficit
Fully Irrigated
Figure 13. WUE expressed in kg m-3 for each cultivar within each irrigation
treatment for 2011. Bars containing the same letters do not differ significantly, ns
indicates no significance.
35 Texas Tech University, Michael Chase Snowden, May 2012 0.8
Bolls per Plant
ns
* *
DP0912B2RF
DP 0912
DP0924B2RF
DP 0924
DP0935B2RF
DP 0935
DP1028B2RF
DP 1028
DP 1032
DP1032B2RF
DP 1044
DP1044B2RF
FM 9160
FM9160B2F
*
0.6
**
ns
**
0.4
ns
**
0.2 ns
*
* Significant at P=0.05
** Significant at P=0.01
0.0
4
6
8
10
12
14
ns
16
ns ns
18
Node
Figure 14. Number of first position bolls plant-1 by node
for each cultivar averaged over each irrigation treatment.
36 20
Texas Tech University, Michael Chase Snowden, May 2012 0.6
A
DP 0935
0.4
1st Position
**
** **
**
*
-1 plant-1-1Over Severe-Deficit
AdditionalAdditional
Bolls nodeBolls
Treatment
node plant-1 Over Severe-Deficit
0.2
**
*
0.0
-0.2
0.6
**
DP 1044
**
1st Position
0.4
B
**
**
**
0.2
*
0.0
-0.2
0.6
C
FM 9160
1st Position
0.4
**
*Significant at P=0.05
**Significant at P=0.01
**
**
**
0.2
*
0.0
Severe-Deficit
Moderate-Deficit
Mild-Deficit
Fully Irrigated
-0.2
6
8
10
12
14
16
18
Node
-1
Figure 15. First position bolls node plant-1 produced over severe-deficit treatment
by node for DP0935 (a), DP1044 (b), and FM9160 (c) cultivars. Bars indicate LSD
values for each node.
37 Texas Tech University, Michael Chase Snowden, May 2012 0.6
A
DP 0912
1st Position
**
0.4
** **
**
**
0.2
**
**
0.0
-0.2
0.6
B
DP 0924
1st Position
0.4
Additional Bolls node-1 plant-1 Over Severe-Deficit Treatment
0.2
* *
** ** ** **
**
**
0.0
-0.2
0.6
C
DP 1028
1st Position
0.4
* *
*
*
0.2
0.0
-0.2
0.6
DP 1032
D
1st Position
0.4
** **
0.2
*
0.0
Severe-Deficit
Moderate-Deficit
Mild-Deficit
Fully Irrigated
-0.2
6
8
*Significant at P=0.05
**Significant at P=0.01
10
12
Node
-1
14
16
18
Figure 16. First position bolls node-1 plant produced over severe-deficit treatment
by node for DP0912 (a), DP 0924 (b), DP1028(c), DP1032 (d). Bars indicate LSD
values for each node.
38 Texas Tech University, Michael Chase Snowden, May 2012 0.6
A
FM 9160
0.4
2nd Position
Additional Bolls node-1 plant-1 Over Severe-Deficit Treatment
0.2
*
0.0
-0.2
0.6
B
DP 1028
2nd Position
0.4
**
0.2
**
**
*
0.0
-0.2
0.6
C
DP 1032
0.4
2nd Position
0.2
*
*Significant at P=0.05
**Significant at P=0.01
** *
*
0.0
Severe-Deficit
Moderate-Deficit
Mild-Deficit
Fully Irrigated
-0.2
6
8
10
12
14
16
18
Node
-1
Figure 17. Second position bolls node plant-1 produced over severe-deficit
treatment by node for FM 9160 (a), DP 1028 (b), and DP 1032 (c) cultivars. Bars
indicate LSD values for each node.
39 Texas Tech University, Michael Chase Snowden, May 2012 CHAPTER IV
CONCLUSIONS
Total water applied was higher in 2011 than in 2010, due to 2010 being
abnormally wet and 2011being abnormally dry. ETc percentages for the three irrigation
treatments in 2010 were similar to the three deficit treatments in 2011 with the fully
irrigated treatment having the highest ETc of the trial.
There were several notable cultivars for yield and WUE for both years of the
study. In 2010, FM 9160, DP 1044, and DP 0912 had the highest average yields and
WUE. FM 9160 and DP 1044 performed very well in the deficit treatments for 2010. In
2011, DP 0935, DP 1044 and DP 0924 had the highest average yields. DP 0935 and DP
1044 performed very well in the deficit treatments. Despite having the highest seed yield
and boll numbers, DP 1044 had the second highest yield, due to low turnout
Boll distribution patterns averaged over all irrigation treatments showed that DP
1044 had the highest number of bolls produced in upper portions of the plant in 2011. FM
9160 produced fewer bolls in the middle of the plant than any other cultivar however did
not have the sharp decline in boll production in the middle of the plant like the other
cultivars. Boll distribution patterns were also highly variable between cultivars and
within irrigation treatments. DP 0935 and DP 1044 had reduced boll production in lower
portions of the plant and significant increases in boll production at higher portions of the
plant compared to severe-deficit treatments. FM 9160 had a very unique distribution
pattern which did not have declines in bolls produced in lower portions of the plant
compared to severe-deficit like most of the other cultivars. Instead there were two peaks
40 Texas Tech University, Michael Chase Snowden, May 2012 in bolls produced at nodes 7 and 13 with patterns that were consistent with one another as
irrigation increased which also differed from all other cultivars. DP 1028 also had a
unique boll production pattern with most of the yield in the for the mild-deficit treatment
produced at the second position which differs from all other cultivars except DP 1032;
however is less pronounced than DP 1028.
Three cultivars (DP1044, FM 9160 and DP 0935) increased fruit production near
the top of the plants in response to irrigation, and also had good yield and yield stability;
their yield patterns may be favorable for limited water conditions in the Texas High
Plains.
41 Texas Tech University, Michael Chase Snowden, May 2012 LITERATURE CITED
Allen, R. G., L. S. Periera, D. Raes, and M. Smith. 1998. Crop evapotranspiration:
Guidelines for computing crop water requirements. Irrig. and Drain. Paper No. 56.
Rome, Italy: United Nations, Food and Agric. Org.
Baumhardt, R.L., S. A. Staggenborg, P. H. Gowda, P. D. Colaizzi, and T. A. Howell.
2009. Modeling irrigation management strategies to maximize cotton lint yield
and water use efficiency. Agron. J. 101:460-468.
Basal, H., N. Dagdelen, A. Unay and E. Yilmaz. 2009. Effects of Deficit Drip Irrigation
Ratios on Cotton(Gossypium hirsutum L.) Yield and Fibre Quality. J. Agron. and
Crop Sci. 195:19–29.
Colaizzi, P. D., P.H. Gowda, T.H. Marek, and D.O. Porter. 2008. Irrigation in the Texas
High Plains: A brief history and potential reductions in demand. Irrig. and Drain.
DOI: 10.1002/ird.418.
Cook, C. G., and K. M. El-Zik. 1993. Fruiting and lint yield of cotton cultivars under
irrigated and non-irrigated conditions. Field Crops Res. 33:411–421.
Cetin, O., and L. Bilgel. 2002. Effects of different irrigation methods on shedding and
yield of cotton. Agric. Water Manag. 54:1-15.
Dagdelen, N., E. Yilmaz, F. Sezgin, and T. Gurbuz. 2006. Water-yield relation and water
use efficiency of cotton (Gossypium hirsutum L.) and second crop corn (Zea mays
L.) in western Turkey. Agric. Water Manag. 82:63–85.
Dagdelen, N. H. Basal, E. Yılmaz, T. Gurbuz, S. Akcay. 2009. Different drip irrigation
treatments affect cotton yield, water use efficiency and fiber quality in western
Turkey. Agric. Water Manag. 9 6:111–120.
DeTar, W.R. 2008. Yield and growth characteristics for cotton under various irrigation
treatments on sandy soil. Agric. Water Manag. 9 5:69 –76.
Evett, S.R., and J. A. Tolk. 2009. Introduction: Can water use efficiency be molded well
enough to impact crop management. Agron. J. 101:423-425.
Grismer, M.E. 2002. Regional cotton lint yield, ETc, and water value in Arizona and
California. Agric. Water Manag. 54:227-242.
Hodgson, A.S., G.A. Constable, G.R. Duddy, I.G. Danieles. 1992. A comparison of drip
and furrow irrigated cotton on a cracking clay soil. Soil Ferti. 55:2
42 Texas Tech University, Michael Chase Snowden, May 2012 Howell, T.A. 2001. Enhancing water use efficiency in irrigated agriculture. Agron. J.
93:281-289.
Howell, T.A., S. R. Evett, J. A. Tolk, and A. D. Schneider. 2004. Evapotranspiration of
full-, deficit-irrigated, and dryland cotton on the northern Texas high plains. J. of
Irrig. and Drain. Eng. 130:277-285.
Jenkins, J.N., W.L. Parrott, J.C. McCarty Jr. 1990. Effectiveness of fruiting sites in
cotton: yield. Crop Sci. 30:365-369.
Littell, R. C., G. A. Milliken, W.W. Stroup, R.D. Wolfinger,O. Schabenberberet. 2006.
SAS for Mixed Models, 2nd Ed. Cary, NC, SAS Publishing
Lopez, M. J., J. C. Gutierrez, and E. O. Leidi. 1995. Selection and characterization of
cotton cultivars for dry land production in the southwest of Spain. Eur. J. Agric.
4:119–126.
Kirda, C. 2002. Deficit irrigation scheduling based on plant growth stages showing water
stress tolerance. Deficit Irrigation Practices. Water report, 22:3–10.
Mateos, L., J. Beregena, F. Orgaz, J. Diz, and E. Fereres. 1991. A comparison between
drip and furrow irrigation in cotton at two levels of water supply. Agric. Water
Manag. 19:313-324.
Pettigrew, W.T. 2004. Moisture deficit effects on cotton lint yield, yield components, and
boll distribution. Agron. J. 96:377-383.
Ritchie, G.L., J.R. Whitaker, C.W. Bednarz, and J.E. Hook. 2009. Subsurface drip and
overhead irrigation: a comparison of plant boll distribution in upland cotton.
Agron. J. 101:1336-1344
Saranga, Y., I. Flash, and D. Yakir. 1998. Variation in wateruse efficiency and its relation
to carbon isotope ratio in cotton. Crop Sci. 38:782–787.
Tolk, J. A., and T. A. Howell. 2001. Measured and simulated evapotranspiration of grain
sorghum with full and limited irrigation in three High Plains soils. Trans. ASAE,
44:1553-1558.
Torell, L. A., J. D. Libbin, M. D. Miller. 1990. The market value of water in the Ogallala
Aquifer. Land Economics. 66:163-175.
Viets Jr., F.G. 1962. Fertilizers and the efficient use of water. Adv. Agron. 14:223-264.
43 Texas Tech University, Michael Chase Snowden, May 2012 Walter, I.A., R.G. Allen, R. Elliott, M.E. Jensen, D. Itenfisu, B. Mecham, T.A. Howell, S.
Snyder, P. Brown, S. Echings, T. Spofford, M. Hattendorf, R.H. Cuenca, J.L.
Wright, D. Martin. 2000. ASCE’s standardized reference evapotranspiration
equation. In: Proceedings of Fourth National Irrigation Symposium, ASAE,
Phoenix, AZ, USA, 14–16 November.
Wanjura, D.F., and D.R. Upchurch. 2000. Canopy temperature characteristics of corn and
cotton water status. Trans ASAE. 43:867-875.
Wanjura, D.F., D. R. Upchurch, J. R. Mahan, and J. J. Burke. 2002. Cotton yield and
applied water relationships under drip irrigation. Argic. Water Manag. 55:217237.
Whitaker, J.R., G.L. Ritchie, C.W. Bednarz and C.I Mills. 2008. Cotton subsurface drip
and overhead irrigation efficiency, maturity, yield and quality. Agron. J.
100:1763-1768.
44 Texas Tech University, Michael Chase Snowden, May 2012 APPENDIX
45 Texas Tech University, Michael Chase Snowden, May 2012 Table 1A. Fiber Quality measurements by high volume instrument (HVI) for micronaire,
staple length, uniformity, and strength in 2010 for all irrigation treatments and cultivars.
Cultivar Fiber parameter Severe‐
Moderate‐
Fully LSD Deficit Deficit Irrigated (0.05) DP 0912 Micronaire 4.75 4.56 4.24 NS Staple Length (mm) 27.3 28.6 29.2 NS Uniformity (%) 80.5 81.3 ab 83.0 a 1.4 b† Strength (g/tex) 27.2 28.9 30.1 NS DP 0924 Micronaire 4.25 4.29 4.03 NS Staple Length (mm) 27.3 28.4 28.8 NS NS Uniformity (%) 80.2 82.1 80.6 Strength (g/tex) 28.2 28.1 28.3 NS DP 0935 Micronaire 4.63 4.70 4.03 NS Staple Length (mm) 27.2 b 27.9 b 29.2 a 1.0 Uniformity (%) 81.4 81.5 82.6 NS Strength (g/tex) 28.1 27.6 27.4 NS DP 1028 Micronaire 4.85 4.28 4.20 NS 28.6 28.3 NS Staple Length (mm) 27.6 Uniformity (%) 80.4 80.9 81.7 NS Strength (g/tex) 27.4 29.3 27.6 NS DP 1032 Micronaire 4.88 4.56 4.08 NS Staple Length (mm) 27.8 29.4 28.9 NS Uniformity (%) 82.5 82.5 81.8 NS Strength (g/tex) 29.0 30.8 29.2 NS 4.83 3.99 4.19 NS DP 1044 Micronaire Staple Length (mm) 26.8 28.7 28.7 NS Uniformity (%) 81.4 80.5 81.2 NS Strength (g/tex) 28.5 30.1 28.2 NS FM 9160 Micronaire 4.82 a 4.44 b 4.17 b 0.20 Staple Length (mm) 27.6 28.3 29.3 NS Uniformity (%) 81.6 81.6 81.1 NS Strength (g/tex) 28.5 29.6 28.8 NS †Horizontal means followed by the same letter within a row are not significant (p=0.05) 46 Texas Tech University, Michael Chase Snowden, May 2012 Table 2A. Fiber Quality measurements by high volume instrument (HVI) for micronaire,
staple length, uniformity, and strength in 2011 for all irrigation treatments and cultivars.
Cultivar Fiber parameter Severe‐
Moderate‐ Mild‐
Fully LSD Deficit Deficit Deficit Irrigated (0.05) DP 0912 DP 0924 DP 0935 DP 1028 DP 1032 DP 1044 FM 9160 Micronaire Staple Length (mm) Uniformity (%) Strength (g/tex) Micronaire Staple Length (mm) Uniformity (%) Strength (g/tex) Micronaire Staple Length (mm) Uniformity (%) Strength (g/tex) Micronaire Staple Length (mm) Uniformity (%) Strength (g/tex) Micronaire Staple Length (mm) Uniformity (%) Strength (g/tex) Micronaire Staple Length (mm) Uniformity (%) Strength (g/tex) Micronaire Staple Length (mm) Uniformity (%) Strength (g/tex) 4.18 b† 24.8 b 79.2 b 25.2 d 4.28 c 25.0 b 79.7 b 27.0 b 4.45 25.3 b 79.7 26.7 b 4.64 25.9 c 80.4 b 27.1 4.44 b 25.9 79.1 26.3 4.41 25.1 b 79.2 b 26.8 b 4.06 b 26.7 b 80.7 27.2 4.76 a 25.7 a 80.0 ab 27.2 c 4.74 b 25.4 b 80.0 b 26.9 b 4.78 25.6 b 80.5 27.5 ab 4.84 26.3 bc 81.1 ab 28.5 4.67 a 26.6 79.8 28.3 4.76 26.4 a 80.6 a 28.9 a 4.48 a 27.1 b 81.1 28.5 5.00 a 25.9 a 80.7 a 28.2 b 4.93 ab 26.0 a 80.5 a 28.7 a 4.82 25.7 b 79.7 28.1 ab 4.79 26.9 b 82.1 a 28.9 4.74 a 26.5 79.9 29.1 4.60 26.7 a 80.7 a 28.4 ab 4.54 a 27.3 ab 81.5 28.8 4.91 a 26.4 a 80.7 a 29.2 a 4.98 a 26.7 a 81.4 a 30.0 a 4.59 26.6 a 80.7 28.9 a 4.62 27.7 a 82.3 a 29.1 4.64 a 27.3 80.7 29.4 4.59 26.9 a 81.0 a 29.6 a 4.27 ab 28.2 a 82.2 29.3 † Horizontal means followed by the same letter within a row are not significant (p=0.05) 47 0.23 0.8 1.0 0.6 0.16 1.0 0.5 1.0 NS 0.5 NS 1.1 NS 0.8 0.9 NS 0.08 NS NS NS NS 0.8 0.4 1.1 0.22 1.0 NS NS Texas Tech University, Michael Chase Snowden, May 2012 0.6
A
DP 0912
2nd Position
0.4
* *
0.2
*
*
0.0
-0.2
Additional Bolls node-1 plant-1 Over Severe-Deficit Treatment
0.6
DP 0924
B
DP 0935
C
2nd Position
0.4
0.2
0.0
-0.2
0.6
0.4
2nd Position
* **
0.2
**
0.0
**
-0.2
0.6
D
DP 1044
*Significant at P=0.05
**Significant at P=0.01
2nd Position
0.4
* **
0.2
*
*
0.0
Severe-Deficit
Moderate-Deficit
Mild-Deficit
Fully Irrigated
-0.2
6
8
10
12
14
16
18
Node
Figure 18A. Second position bolls node-1 plant-1 produced over severe-deficit
treatment by node for DP 0912 (A), DP 0924 (B), and DP 0935 (C), and DP 1044 (D)
cultivars. Bars indicate LSD values for each node.
48