Published online 5 September 2006
Published in Agron J 98:1359-1366 (2006)
DOI: 10.2134/agronj2006.0042
© 2006 American Society of Agronomy
677 S. Segoe Rd., Madison, WI 53711 USA
Production Papers
Yield–Water Supply Relationships of Grain Sorghum and Winter Wheat
Loyd R. Stonea,* and
Alan J. Schlegelb
a Dep. of Agron., Throckmorton Hall, Kansas State Univ., Manhattan, KS 66506-5501
b KSU Southwest Res.-Ext. Cent., Route 1, Box 148, Tribune, KS 67879-9774
* Corresponding author (stoner{at}ksu.edu)
Received for publication February 9, 2006.
 |
ABSTRACT
|
|---|
Agricultural sustainability in the USA's west-central Great Plains depends on efficient use of water—the primary yield-limiting factor in the region. With perennially water-short status, the efficient capture and storage of precipitation in soil, and the yield responsiveness of crops to water, are emphasized. Our objective was to quantify grain sorghum [Sorghum bicolor (L.) Moench] and winter wheat (Triticum aestivum L.) yield responses to stored soil water and precipitation by using data gleaned from research conducted from 1973 to 2004 near Tribune, KS on Ulysses silt loam (fine-silty, mixed, superactive, mesic Aridic Haplustolls) and Richfield silt loam (fine, smectitic, mesic Aridic Argiustolls) soils. Soil water content was measured gravimetrically to the 183-cm depth at crop emergence. Grain yield was related to available soil water at emergence (ASWe) (increased 221 kg ha–1 cm–1 in sorghum and 98 kg ha–1 cm–1 in wheat). Grain yield was also related to in-season precipitation (ISP) (increased 164 kg ha–1 cm–1 in sorghum and 83 kg ha–1 cm–1 in wheat). From response-surface analyses, 63% (sorghum) and 70% (wheat) of variations in grain yield were explained by variations in ASWe and ISP. In data sorted by tillage, yield response to water supply (WS) was greater with no-till than with conventional tillage in both crops (184 vs. 129 kg ha–1 cm–1 in sorghum; 138 vs. 86 kg ha–1 cm–1 in wheat). This finding supports the concept that less tillage and more residue lead to more efficient use during the growing season of ASWe and ISP.
Abbreviations: ASWe, available soil water at emergence • ASWp, available soil water at planting • CT, conventional, stubble-mulch (sweep) tillage • ISP, in-season precipitation • NT, no-till • RT, reduced tillage • SF, sorghum–fallow • SS, continuous sorghum • WF, wheat–fallow • WS, water supply • WSF, wheat–sorghum–fallow • WUE, water use efficiency • WW, continuous wheat
|
INTRODUCTION
|
|---|
WATER SHORTAGE is the primary factor limiting crop production in the USA's west-central Great Plains, and agricultural sustainability depends on efficient use of water resources. Grain sorghum and winter wheat are the primary dryland crops of the region, and corn (Zea mays L.) is the primary irrigated crop (Kansas Agric. Stat., 2004). Precipitation is limited and sporadic and has provided the impetus for the development and role of irrigation in the region. Mean annual precipitation ranges from 35 to 50 cm across the region (HPRCC, 2005), which is only 60 to 85% of the water requirement (evapotranspiration) needed for full production of sorghum and wheat (Doorenbos and Kassam, 1979). Therefore, yields of dryland crops are limited and variable.
Irrigation in the region largely depends on groundwater, with the Ogallala formation of the High Plains aquifer being the main source (McGrath and Dugan, 1993). Water-level declines in the Ogallala started to occur soon after the beginning of extensive groundwater irrigation, and declines from predevelopment (about 1950) to 2003 of 15 m or more are widespread in parts of eastern Colorado, western Kansas, and southwestern Nebraska, with some declines >45 m (McGuire, 2004). With water-level declines, well yields are reduced and pumping costs are increased by the additional lift.
As farmers consider reduced irrigation amounts, limited irrigation in otherwise dryland rotations, or more intensive dryland rotations compared with the more traditional crop–fallow, water use efficiency (WUE) and the selection of appropriate crops are critical factors in successful cropping systems. Improved WUE can be gained by decreased tillage, increased residue, reduced length of fallow periods, contour farming, furrow dikes, control of plowpans, crop selection, and use of appropriate crop rotations (Nielsen et al., 2005). With the limited conditions of precipitation and declining water levels of the Ogallala, every effort should be made to increase the efficiency of use of all water sources.
As one or more of the water sources for crop production (irrigation, stored soil water, and ISP) are reduced, additional relative importance is placed on efficient use of water from the remaining sources. As irrigation decreases, and as dryland rotations become more intensive, additional emphasis is placed on the efficient capture and storage of precipitation in soils and on understanding the yield responsiveness of crops to water. Economics is a factor in considering management practices that influence WUE through the influence on production costs incurred by use of the practice and the influence on production returns realized through the relationships between crop yield and WS.
Data on crop yield and ASWe have been collected during the past 30 yr in research conducted near Tribune, KS. These data allow an assessment of crop yield–WS relationships in the west-central Great Plains, information that can be used in considering the economic viability of water-conserving practices. Our objective was to examine crop yield–WS component responsiveness of the primary crops (winter wheat and grain sorghum) of dryland and limited-irrigation cropping systems of the region. Specifically, we aimed to quantify the yield-enhancing value of ASWe and ISP, by using the 30 yr of data collected near Tribune.
|
MATERIALS AND METHODS
|
|---|
Data on crop yield and ASWe were gleaned from a number of studies conducted during 1973 to 2004 near Tribune, KS, USA (38°28' N, 101°45' W; 1101 m elev.). Rainfed-only research was conducted on the Dryland Field and irrigation research on the Irrigation Field, the two fields separated by 11.6 km. Soils on both fields formed on upland plains in calcareous loess and are deep and well drained. Ulysses silt loam is the main soil type on the Irrigation Field, and Richfield silt loam is the main soil type on the Dryland Field (Gwin et al., 1974). Similar soils occupy about 2.34 million ha of the west-central Great Plains (Aandahl, 1982). Soil water content at the upper limit of availability is 68.5 and 65.0 cm in the 183-cm Richfield and Ulysses profiles, respectively. Water content at the lower limit of availability is 32.5 and 29.0 cm in the 183-cm Richfield and Ulysses profiles, respectively (Soil Survey Staff, 1966; Stone et al., 1987). The region has a semiarid, continental climate with mean annual temperature of 11.3°C and a frost-free growing season of 160 d (first week of May to second week of October). Long-term (1912–2004) mean annual precipitation is 42.3 cm (HPRCC, 2005).
Crop yield and soil water data were gleaned from dryland tests where crop rotations were wheat–fallow (WF), sorghum–fallow (SF), wheat–sorghum–fallow (WSF), continuous wheat (WW), and continuous sorghum (SS). The rotations allow one crop in 2 yr (WF and SF), two crops in 3 yr (WSF), and annual cropping (WW and SS). In the 30 yr of studies, various levels of tillage and/or herbicides, ranging from conventional, stubble-mulch (sweep) tillage (CT, tillage exclusively) to reduced tillage (RT, herbicides and tillage) to no-till (NT, herbicides exclusively), were used during noncrop periods for weed control. Sweep plows (blades) and offset disk (occasionally used) were the primary tillage tools and operated at depths of 8 to 15 cm. Sweep plows ran below the soil surface, cutting off the roots of weeds, and reduced the amount of stubble by about 10% per operation. The offset disk mixed stubble and weeds into the soil and reduced the amount of stubble by about 50% per operation. Irrigation test data were used if treatments received only preplant irrigation—to decrease the likelihood of using data from treatments with significant profile drainage or introducing measurement questions involved with irrigation.
All studies used in building data sets had a randomized complete-block design with three or four replications, and as recommended by Gomez and Gomez (1984), treatment means were used in analyses. Soil water content was determined at crop emergence by gravimetric sampling, with one core per plot taken in 30.5-cm-depth increments to 183 cm. Gravimetric water content was multiplied by soil bulk density to convert to water content by volume. Profile water amount at the lower limit of availability was subtracted from the total equivalent depth of water in the 183-cm profile to yield ASWe. Daily precipitation during the year was measured at the Dryland Field with a standard rain gauge, with snow recorded as liquid equivalent. Daily precipitation from 1 April to 30 September was measured at the Irrigation Field with a standard rain gauge. Precipitation during 1 October to 31 March at the Irrigation Field was assumed to be the same as that measured at the Dryland Field.
Wheat was planted in mid-to-late September, and sorghum was planted in late May to mid-June. Cultivars planted differed over the years but were ones that performed well in Kansas State University performance tests at Tribune. Wheat was planted at 40 to 60 kg ha–1 (row spacing of 19.0 to 35.6 cm, changing over the years with cropping system and equipment design) in dryland cropping systems and at 100 kg ha–1 (row spacing of 25.4 cm) with preplant irrigation. Sorghum was planted at a row spacing of 76.2 cm. Planting rates of sorghum were about 75 thousand and 150 thousand seeds ha–1 with dryland and preplant irrigation, respectively. Plots were combine-harvested in late June to early July for wheat and in late October to early November for sorghum. Harvest width was 1.9 m for wheat and 1.5 m for sorghum; harvest length was a minimum of 12 m. Grain yield of both crops was adjusted to 125 g kg–1 water on a moist-mass basis.
Data sets used in analyses contained 142 and 253 data lines for sorghum and wheat, respectively. Each data line contained grain yield, ASWe, and ISP (15 September–14 June for wheat and 15 June–14 September for sorghum). Water supply was calculated as the sum of ASWe and ISP. Data from years with catastrophic yield losses due to freeze or hail were not used. Data were used if catastrophic yield losses were due to WS conditions. Descriptive information for the two data sets is in Table 1.
View this table:
[in this window]
[in a new window]
|
Table 1. Descriptive information on grain sorghum and winter wheat data sets gleaned from research near Tribune, KS from 1973 through 2004.
|
|
The association between grain yield and each of ASWe, ISP, and WS were described by use of regression analyses with procedures by SAS (SAS Inst., 1985). For each combination of crop yield and independent (water) variable, PROC STEPWISE was used with the first- and second-degree forms of the water variable. Equations were selected from the PROC STEPWISE analysis if all regression coefficients were significant (P < 0.05). If an equation selected was not linear, simple linear regression was performed by using PROC GLM. The linear form was used to compare our yield–water associations with those in the literature.
|
RESULTS AND DISCUSSION
|
|---|
An analysis of crop yield response in a long-term study involves possible change in yield potential due to genetic improvement over the life of the study. To evaluate this possibility, we used data from irrigated cultivar performance tests conducted at Tribune. The means of all test entries in a year for wheat (1974 to 2004 with 25 yr of data; none due to hail in 1987 and 2002, and no testing from 1998 to 2001) and sorghum (1973 to 2003 with 26 yr of data; none due to early freeze in 1983 and 1997, and due to hail in 1974, 1995, and 2002) were used with linear regression (SAS Inst., 1985) to evaluate yield change over time. From winter wheat test data:
 | [1] |
with r2 = 0.023, P = 0.472, and mean yield over years of 4101 kg ha–1. From grain sorghum test data:
 | [2] |
with r2 = 0.008, P = 0.660, and mean yield over years of 7532 kg ha–1. In Eq. [1] and [2], Y is grain yield in kg ha–1 at 125 g kg–1 water on a moist-mass basis, and X is year. From an analysis using 61 yr (1930–1990) of data from numerous locations in the USA, sorghum and wheat yields increased by about 50 and 30 kg ha–1 yr–1, respectively, due to genetic improvement and improved farming practices (Eghball and Power, 1995). With the model significance levels obtained in our analyses of data from a 30-yr span at one location, no confidence is placed on there having been a change in sorghum or wheat yield potential over the years of our study. Therefore, in our efforts to quantify yield response to water, we did not make adjustments in yield for an increasing yield trend that results from genetic improvement.
Sorghum yield responses to ISP (15 June–14 September) and ASWe were determined by stepwise regression, with results summarized in Table 2. The complete sorghum data set was used, with no sorting by rotation or tillage (Table 1). In the PROC STEPWISE analysis, the first- and second-degree forms of ASWe and ISP, and their interaction products, were the independent variables. Regression Eq. [2.2] of Table 2 was used to plot the sorghum response surface (Fig. 1
). From the response-surface analysis, 63% of the variability in sorghum yield was explained by variations in ISP and ASWe. There was no significant interaction between ASWe and ISP in their influences on sorghum grain yield.
View this table:
[in this window]
[in a new window]
|
Table 2. Results from stepwise regression analysis of sorghum grain yield associated with available soil water at emergence and 15 June to 14 September precipitation, Tribune, KS, 1973–2003.
|
|
View larger version (49K):
[in this window]
[in a new window]
|
Fig. 1. Grain sorghum yield associated with water-supply components (available soil water at emergence and in-season precipitation).
|
|
Wheat grain yield responses to ISP (15 September–14 June) and ASWe were determined by stepwise regression in the same fashion as with sorghum, and results are summarized in Table 3. Regression Eq. [3.5] of Table 3 was used to plot the wheat grain yield response surface (Fig. 2
). From the response-surface analysis, 70% of the variability in wheat yield was explained by variations in ISP and ASWe, with grain yield significantly influenced by interaction between ASWe and ISP. With dryland winter wheat in Colorado, Nielsen et al. (2002) found grain yield was significantly influenced by interaction between available soil water at planting (ASWp) and precipitation; yield response (increase) to increasing ASWp was greater with wetter than with drier precipitation conditions.
View this table:
[in this window]
[in a new window]
|
Table 3. Results from stepwise regression analysis of winter wheat grain yield associated with available soil water at emergence and 15 September to 14 June precipitation, Tribune, KS, 1974–2004.
|
|
View larger version (51K):
[in this window]
[in a new window]
|
Fig. 2. Winter wheat yield associated with water-supply components (available soil water at emergence and in-season precipitation).
|
|
Slopes of grain yield response surfaces of Fig. 1 (sorghum) and 2 (wheat) were determined by taking the first partial derivatives of equations describing the surfaces, where Y, W, and P represent grain yield (kg ha–1), ASWe (cm), and ISP (cm), respectively. For sorghum, differentiating Eq. [2.2] of Table 2 with respect to W yielded:
 | [3] |
and differentiating with respect to P yielded:
 | [4] |
For wheat, differentiating Eq. [3.5] of Table 3 with respect to W yielded:
 | [5] |
and differentiating with respect to P yielded:
 | [6] |
Sorghum yield increased in response to increased ASWe at a rate of 208 kg ha–1 cm–1 at all ISP levels (Eq. [3] and Fig. 1). Sorghum yield increase per centimeter increase in ISP increased as ISP increased (Eq. [4] and Fig. 1) but did not vary as ASWe varied. Precipitation has an increasing effectiveness in producing sorghum grain as ISP increases (Fig. 1). As ISP increases, sorghum stand establishment and survival, and the efficacy of herbicides and fertilizers improves. With adequate moisture, sorghum progresses at more normal rates than if under water stress; during vegetative growth, sorghum has the ability to go into physiological dormancy under water-stress conditions and then rebound after water conditions improve (Bennett et al., 1990). A more normal growth progression assists sorghum in avoiding freeze damage in the fall. As total rain increases, there is greater chance of having rain events large enough to infiltrate soil and be available for plant uptake and of having reduced evaporative conditions of high air temperature, high global radiation, and low relative humidity. With dryland sorghum in this region, profile drainage and runoff are not consistent, significant events.
Wheat yield increase per centimeter increase in ASWe increased as ASWe and ISP increased (Eq. [5] and Fig. 2). Increased ASWe causes improved wheat germination, seedling emergence, stand establishment and vigor, and winter survival, which create a better stand with greater potential yield (Paulsen, 1987). The improved stands caused by increased ASWe are more effective at producing grain in response to increased ASWe and ISP. Also, increased ASWe and increased soil water at the beginning of spring growth enable more efficient use of ISP and limited irrigation, in part because winter wheat is able to develop a deep root system that can maintain water extraction during water-stress periods (Musick et al., 1994). Wheat yield increase per centimeter increase in ISP decreased as ISP increased and increased as ASWe increased (Eq. [6] and Fig. 2). Increased moisture in spring can increase susceptibility to diseases (Krupinsky et al., 2002) and lodging (Carter, 1987; Musick and Porter, 1990), which decreases the effectiveness of ISP in producing grain as ISP increases (Fig. 2).
Mean ISP was 16.7 and 26.4 cm for sorghum and wheat data sets, respectively (Table 1). The distribution of ISP at Tribune over 92 crop seasons is given in Fig. 3
for sorghum and wheat. For sorghum, ISP was 
10.7, 16.3, and 23.9 cm in 80, 50, and 20% of the seasons, respectively. For wheat, ISP was 17.0, 23.8, and 33.0 cm in 80, 50, and 20% of the seasons, respectively. Mean ASWe was 19.3 and 16.6 cm for the full sorghum and wheat data sets, respectively (Table 1). In sorghum with no irrigation, ASWe was 11.4, 19.3, and 24.5 cm for 80, 50, and 20%, respectively, of the 119 data points. In wheat with no irrigation, ASWe was 8.6, 16.8, and 21.6 cm for 80, 50, and 20%, respectively, of the 242 data points. These historical values of ISP and ASWe can be used with the equations of yield response (Eq. [2.2], Table 2, sorghum; Eq. [3.5], Table 3, wheat) and slopes (Eq. [3]
–[6]) to explore yield responses to water in the Great Plains and to assess production risks. For example, with 15 cm of ASWe in wheat, Eq. [3.5] indicates an 80, 50, and 20% chance of obtaining grain yield of at least 1404, 2136, and 2651 kg ha–1, respectively.
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 3. Percentage of 92 sorghum (15 June–14 September) and wheat (15 September–14 June) seasons with precipitation greater than or equal to the listed amount of seasonal precipitation for Tribune, KS.
|
|
Much of the literature on dryland crop yield response to water in the Great Plains consists of results from simple regression analyses in which yield was the dependent variable and one of ASWe, ASWp, ISP, or WS (ASWe or ASWp plus ISP) was the independent variable. Crop yield response to water is altered by latitudinal position in the Great Plains and surface-residue conditions. Yield response to water decreases north to south in the Great Plains, primarily because of the increasingly greater evaporative demand of climate with decreasing latitude (Musick et al., 1994; Nielsen et al., 2002). Increased surface residue provides for more efficient use of ISP (Unger, 1978). To aid comparisons with literature, regression analyses were performed with WS components and yield.
Associations between grain yield and ISP are presented in Fig. 4
(sorghum) and 5 (wheat). Sorghum yield increased 164 kg ha–1 cm–1 of ISP. The association between wheat grain yield and ISP was best described as curvilinear (Fig. 5
), with the linear equation also significant:
 | [7] |
with P < 0.0001 and r2 = 0.258, where Y is yield (kg ha–1) and X is ISP in centimeters. The slope of winter wheat yield vs. ISP was 55 kg ha–1 cm–1 in WF and WSF rotations with CT during 1943 to 1956 at Bushland, TX (Army et al., 1959). Our greater slope of 83 kg ha–1 cm–1 (Eq. [7]) was from a mix of CT, RT, and NT systems; from further north in the Great Plains; and with more modern crop management than that in Army et al. (1959).
Associations between grain yield and ASWe of the 183-cm profile are presented in Fig. 6
(sorghum) and 7 (wheat). Sorghum grain vs. ASWe had a linear slope of 221 kg ha–1 cm–1. With CT at Bushland, TX, grain sorghum yield increased at 170 kg ha–1 cm–1 of ASWp (180-cm soil profile) (Jones and Hauser, 1975). Our association between wheat grain and ASWe was best described as curvilinear (Fig. 7
), with the linear equation also significant:
 | [8] |
with P < 0.0001 and r2 = 0.322, where Y is yield (kg ha–1) and X is ASWe (cm). Winter wheat grain yield response to water stored in soil at planting was 113, 106, 72, 70, 65, and 51 kg ha–1 cm–1 in plots managed uniformly at Huntley, MT; North Platte, NE; Colby, Garden City, and Hays, KS; and Woodward, OK, respectively (Johnson, 1964). Tillage treatments included plowing, so residue levels were likely quite low in some plots. A wheat yield slope of 95 kg ha–1 cm–1 of ASWp was calculated by Nielsen et al. (2002) with data from Garden City, KS (Norwood, 2000). The Norwood (2000) data included CT and NT systems with residue from corn, sorghum, sunflower (Helianthus annuus L.), or soybean [Glycine max (L.) Merr.] preceding wheat planting. Wheat data at Tribune included three tillage systems (CT, RT, and NT) and three crop rotations (WW, WF, and WSF), and grain yield increase in response to ASWe was 98 kg ha–1 cm–1 (Eq. [8]).
Water-supply components (ISP and ASWe) were summed, and the associations between grain yield and WS are in Fig. 8
(sorghum) and 9 (wheat). The linear regression slope of sorghum yield related to WS was 166 kg ha–1 cm–1. Wheat grain yield response to WS was best described as curvilinear (Fig. 9
), with the linear equation also significant:
 | [9] |
with P < 0.0001 and r2 = 0.638, where Y is yield (kg ha–1) and X is WS (cm). From 19 yr of winter wheat in Montana, grain yield vs. WS (ISP plus ASWp) slope was 129 kg ha–1 cm–1 (Brown and Carlson, 1990) compared with the slope at Tribune of 100 kg ha–1 cm–1 (Eq. [9]). Seven of the eight zero-yield data points in Fig. 9 were from the 2001–2002 wheat season, the driest (9.2 cm of precipitation) in the 92 seasons of record. Even with WS of 32 to 33 cm, with the WS being primarily from ASWe and with the limited ISP, the crop could not be sustained and grain could not be produced.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 8. Grain sorghum yield associated with water supply [available soil water (ASW) at emergence plus in-season precipitation].
|
|
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 9. Winter wheat yield associated with water supply [available soil water (ASW) at emergence plus in-season precipitation].
|
|
Associations between sorghum yield and WS with CT (Section A) and NT (Section B) are presented in Fig. 10
. The yield–WS slope was 129 kg ha–1 cm–1 with CT and 184 kg ha–1 cm–1 with NT. Associations between wheat yield and WS with CT (Section A) and NT (Section B) are given in Fig. 11
. In each group of Fig. 11, the response was best described as curvilinear, and each group also had a significant linear relationship. The linear equation for CT is:
 | [10] |
with P < 0.0001 and r2 = 0.651, where Y is yield (kg ha–1) and X is WS (cm). The linear equation for NT is:
 | [11] |
with P < 0.0001 and r2 = 0.831, where Y is yield (kg ha–1) and X is WS (cm). For sorghum and wheat, we determined whether the regression coefficients (slopes) from the simple linear regression analyses of grain yield vs. WS from the two tillage systems were statistically the same by using a t test (Gomez and Gomez, 1984). Within each crop, slopes were significantly greater (P = 0.001) for NT than for CT (184 vs. 129 kg ha–1 cm–1 in sorghum, Fig. 10; 138 vs. 86 kg ha–1 cm–1 in wheat, Eq. [11] and [10]).
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 10. Grain sorghum yield associated with water supply (available soil water at emergence plus in-season precipitation) for dryland conventional tillage (Section A) and no-till (Section B) treatment groups.
|
|
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 11. Winter wheat yield associated with water supply (available soil water at emergence plus in-season precipitation) for dryland conventional tillage (Section A) and no-till (Section B) treatment groups.
|
|
Residue (mulch) increases the amount of precipitation stored in soil during fallow and, thus, available at planting of the subsequent crop; residue also causes more efficient use of ISP. With CT for weed control in Texas, sorghum grain yield response to ASWp was 170 kg ha–1 cm–1 (Jones and Hauser, 1975). In the same region, when wheat residue was added after winter wheat harvest, the sorghum crop that followed had a mean grain yield response to ASWp of 230 kg ha–1 cm–1 (Unger, 1978). With residue amounts expected to be NT > RT > CT, residue levels will cause the WUE during the growing season to be NT > RT > CT. The grain-yield WUE of winter wheat at Akron, CO increased from 69 kg ha–1 cm–1 in WF CT to 75 kg ha–1 cm–1 in WF NT to 84 kg ha–1 cm–1 in wheat–corn–fallow NT (Nielsen et al., 2005). At Garden City, KS, WUE of corn, sunflower, grain sorghum, and soybean were all numerically greater with NT than with CT versions of winter wheat–spring crop–fallow rotations: although only the NT results for corn and sunflower were significantly greater (Norwood, 1999).
The greater ability of NT, compared with CT, to retain precipitation received during fallow and to have more water stored in the soil profile for the next crop has been quantified in a number of studies in the Great Plains (Table 4). In cropping systems that produce minimal amounts of residue or where residue cover is destroyed, reduced storage of water during fallow and reduced yield response to WS during the growing season are expected. With RT and sunflower in three of five crop rotations at Akron, CO, the slope of winter wheat grain yield vs. ASWp was 79 kg ha–1 cm–1 (Nielsen et al., 1999). From a later study with NT and no sunflower in crop rotations at Akron, the slope of wheat grain yield vs. ASWp was 141 kg ha–1 cm–1 from a grouping of average to wet years (Nielsen et al., 2002). In our data from Tribune, there was better yield response to WS with NT, compared with CT, both in sorghum (Fig. 10) and wheat (Fig. 11).
View this table:
[in this window]
[in a new window]
|
Table 4. Additional water gain during fallow with no-till, compared with conventional tillage, of various rotations and locations in the western Great Plains.
|
|
|
CONCLUSIONS
|
|---|
We assembled data sets from the past 30 yr at Tribune where grain sorghum and winter wheat grain yields, ASWe, and ISP had been measured. Grain yield was significantly related to ASWe (increased 221 kg ha–1 in sorghum and 98 kg ha–1 in wheat, per centimeter of additional ASWe). Grain yield was significantly related to ISP (increased 164 kg ha–1 in sorghum and 83 kg ha–1 in wheat, per centimeter of additional ISP). From response–surface analyses, 63% of the variation in sorghum grain yield and 70% of the variation in wheat grain yield was explained through variations in ASWe and ISP. In sorghum, there was no significant interaction between ASWe and ISP in their influence on grain yield; in wheat there was. When treatment data were sorted by tillage, grain yield response to WS was significantly greater with NT than with CT in both crops (184 vs. 129 kg ha–1 cm–1 in sorghum; 138 vs. 86 kg ha–1 cm–1 in wheat).
That NT is superior to CT in capturing and retaining precipitation during fallow for use by the subsequent crop in the Great Plains is well established. This paper presents evidence that less tillage and more residue also provide benefit through more efficient use during the growing season of ASWe and ISP. Yield response to WS components can aid producers as they consider the yield–economic impact of tillage and residue management in selection of tillage systems and crop rotations.
|
ACKNOWLEDGMENTS
|
|---|
This work was supported in part by funds from the USDA-ARS under Agreement no. 58-6209-3-018.
|
NOTES
|
|---|
Contrib. 06-227-J, Kansas Agric. Exp. Stn.
|
REFERENCES
|
|---|
- Aandahl, A.R. 1982. Soils of the Great Plains: Land use, crops, and grasses. Univ. of Nebraska Press, Lincoln.
- Army, T.J., J.J. Bond, and C.E. Van Doren. 1959. Precipitation–yield relationships in dryland wheat production on medium to fine textured soils of the Southern High Plains. Agron. J. 51:721–724.[Abstract/Free Full Text]
- Bennett, W.F., B.B. Tucker, and A.B. Maunder. 1990. Modern grain sorghum production. Iowa State Univ. Press, Ames.
- Brown, P.L., and G.R. Carlson. 1990. Grain yields related to stored soil water and growing season rainfall. Special Rep. 35. Montana Agric. Exp. Stn., Montana State Univ., Bozeman.
- Carter, D.L. 1987. Water relations and irrigation. p. 390–395. In E.G. Heyne (ed.) Wheat and wheat improvement. 2nd ed. Agron. Monogr. 13. ASA, CSSA, and SSSA, Madison, WI.
- Doorenbos, J., and A.H. Kassam. 1979. Yield response to water. Irrig. and Drain. Pap. 33. FAO, Rome.
- Eghball, B., and J.F. Power. 1995. Fractal description of temporal yield variability of 10 crops in the United States. Agron. J. 87:152–156.[Abstract/Free Full Text]
- Gomez, K.A., and A.A. Gomez. 1984. Statistical procedures for agricultural research. 2nd ed. John Wiley & Sons, New York.
- Gwin, R.E., Jr., O.W. Bidwell, R.C. Angell, and G. Muilenburg. 1974. Making the most of soil, water, climate in west-central Kansas. Bull. 577. Kansas Agric. Exp. Stn., Kansas State Univ., Manhattan.
- [HPRCC] High Plains Regional Climate Center. 2005. Historical data summaries [Online]. Available at www.hprcc.unl.edu/products/historical.htm (verified 26 June 2006).
- Johnson, W.C. 1964. Some observations on the contribution of an inch of seeding-time soil moisture to wheat yield in the Great Plains. Agron. J. 56:29–35.[Medline]
- Jones, O.R., and V.L. Hauser. 1975. Runoff utilization for grain production. p. 277–283. In G.W. Frasier (ed.) Proc. Water Harvesting Symp., Phoenix, AZ. 26–28 Mar. 1974. Publ. ARS W-22. USDA, Washington, DC.
- Jones, O.R., and T.W. Popham. 1997. Cropping and tillage systems for dryland grain production in the Southern High Plains. Agron. J. 89:222–232.[Abstract/Free Full Text]
- Kansas Agricultural Statistics. 2004. Kansas farm facts 2004 [Online]. Available at www.nass.usda.gov/ks/ffacts/2004/ffacts.htm (verified 26 June 2006). USDA Natl. Agric. Stat. Serv., Washington, DC.
- Krupinsky, J.M., K.L. Bailey, M.P. McMullen, B.D. Gossen, and T.K. Turkington. 2002. Managing plant disease risk in diversified cropping systems. Agron. J. 94:198–209.[Abstract/Free Full Text]
- McGrath, T., and J.T. Dugan. 1993. Water-level changes in the High Plains Aquifer—predevelopment to 1991. U.S. Geol. Survey Water-Resour. Investigations Rep. 93–4088. U.S. Geol. Survey, Lincoln, NE.
- McGuire, V.L. 2004. Water-level changes in the High Plains Aquifer, predevelopment to 2003 and 2002 to 2003. U.S. Geol. Survey Fact Sheet FS-2004–3097. U.S. Geol. Survey, Lincoln, NE.
- Musick, J.T., O.R. Jones, B.A. Stewart, and D.A. Dusek. 1994. Water–yield relationships for irrigated and dryland wheat in the U.S. Southern Plains. Agron. J. 86:980–986.[Abstract/Free Full Text]
- Musick, J.T., and K.B. Porter. 1990. Wheat. p. 597–638. In B.A. Stewart and D.R. Nielsen (ed.) Irrigation of agricultural crops. Agron. Monogr. 30. ASA, CSSA, and SSSA, Madison, WI.
- Nielsen, D.C., R.L. Anderson, R.A. Bowman, R.M. Aiken, M.F. Vigil, and J.G. Benjamin. 1999. Winter wheat and proso millet yield reduction due to sunflower in rotation. J. Prod. Agric. 12:193–197.
- Nielsen, D.C., P.W. Unger, and P.R. Miller. 2005. Efficient water use in dryland cropping systems in the Great Plains. Agron. J. 97:364–372.[Abstract/Free Full Text]
- Nielsen, D.C., M.F. Vigil, R.L. Anderson, R.A. Bowman, J.G. Benjamin, and A.D. Halvorson. 2002. Cropping system influence on planting water content and yield of winter wheat. Agron. J. 94:962–967.[Abstract/Free Full Text]
- Norwood, C.A. 1992. Tillage and cropping system effects on winter wheat and grain sorghum. J. Prod. Agric. 5:120–126.
- Norwood, C.A. 1999. Water use and yield of dryland row crops as affected by tillage. Agron. J. 91:108–115.[Abstract/Free Full Text]
- Norwood, C.A. 2000. Dryland winter wheat as affected by previous crops. Agron. J. 92:121–127.[Abstract/Free Full Text]
- Paulsen, G.M. 1987. Wheat stand establishment. p. 384–389. In E.G. Heyne (ed.) Wheat and wheat improvement. 2nd ed. Agron. Monogr. 13. ASA, CSSA, and SSSA, Madison, WI.
- SAS Institute Inc. 1985. SAS/STAT guide for personal computers. 6th ed. SAS Inst. Inc., Cary, NC.
- Smika, D.E. 1990. Fallow management practices for wheat production in the Central Great Plains. Agron. J. 82:319–323.[Abstract/Free Full Text]
- Smika, D.E., and G.A. Wicks. 1968. Soil water storage during fallow in the Central Great Plains as influenced by tillage and herbicide treatments. Soil Sci. Soc. Am. Proc. 32:591–595.
- Soil Survey Staff. 1966. Soil survey laboratory data and descriptions for some soils of Kansas. USDA-SCS Soil Survey Investigations Rep. 4. U.S. Gov. Print. Office, Washington, DC.
- Stone, L.R., R.E. Gwin, Jr., P.J. Gallagher, and M.J. Hattendorf. 1987. Dormant-season irrigation: Grain yield, water use, and water loss. Agron. J. 79:632–636.[Abstract/Free Full Text]
- Unger, P.W. 1978. Straw-mulch rate effect on soil water storage and sorghum yield. Soil Sci. Soc. Am. J. 42:486–491.[Abstract/Free Full Text]
TOP
ABSTRACT
INTRODUCTION
