| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Crop, Soil, and Environmental Sciences, Univ. of Arkansas, 276 Altheimer Drive, Fayetteville, AR 72704 USA
b Dep. of Biological and Agricultural Engineering, Northeast Research and Extension Center, Univ. of Arkansas, P.O. Box 48, Keiser, AR 72351 USA
lpurcell{at}comp.uark.edu
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
Abbreviations: BM, biomass • CGR, crop growth rate • DAE, days after emergence • HI, harvest index • LI, light interception • MG, maturity group • tb , lost time
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
Early maturity groups have not been used when sown late in the season because inadequate canopy development generally occurs at recommended populations (Kane and Grabau, 1992). Early-maturing cultivars have a shorter period of vegetative development than full-season cultivars, but the length of the seed-fill phase is about the same as conventional cultivars (Egli et al., 1978; Egli, 1993; Kane and Grabau, 1992). Use of early-maturing cultivars places time constraints on soybean production systems, and growers frequently lack information on the population densities and BM thresholds needed to exploit yield potential.
Short-season cultivars may produce inadequate leaf area to fully utilize available light during flowering and seed fill (Board et al., 1992). Complete LI maximizes potential daily photosynthesis (Shibles and Weber, 1966), and photosynthesis per unit ground area is proportional to grain yield (Wells et al., 1982). Hence, achieving a full canopy cover during a shorter growing season may be a strategy to increase yields. Narrow inter-row spacing (Board and Hall, 1984) and decreased intra-row spacing are management options that lessen the time required for complete LI. To ensure this, we revisited population density recommendations and accepted ideas of how a soybean canopy functions in yield formation in a time-constrained system.
One approach for optimizing yield is to maximize BM, because yield is determined by the product of BM and harvest index (HI) (Sinclair, 1986; Board et al., 1990). Harvest index is constant in most circumstances (Spaeth et al., 1984). Therefore, maximizing BM should produce the greatest yield response (Egli et al., 1987). This relationship holds for vegetative crops, such as forages (where HI = 1), and for grain crops which have an asymptotic yield relationship over a large range of population densities (Willey and Heath, 1969). Very high populations in some crops, including soybean, may decrease HI because of lodging or barren plants (Weber et al., 1966).
The accumulation of BM represents the cumulative factors affecting growth during the life of the crop. After emergence, growth of an unstressed crop increases exponentially (Fig. 1 ; Phase I), followed by an approximately linear growth phase (II) until maximum BM is reached (Phase III). Total biomass decreases during senescent phases (IV and V). Goudriaan and Monteith (1990) named the generalized growth curve for Phases I and II an expolinear function. Extrapolation of the linear portion of the growth-response line in Fig. 1 to the x axis (time) results in the parameter tb, or lost time (Goudriaan and Monteith, 1990). Lost time represents the time required to achieve canopy closure and linear crop growth rate (CGR).
|
Short growing seasons present serious time limitations on crop growth, in which the soybean crop needs to establish and maximize canopy coverage rapidly to exploit available light. We propose increasing population density as a major way to reduce lost time and to increase LI in order to maximize yield in time-constrained cropping systems. The specific objectives of this research were to: (i) evaluate increased plant populations as a strategy for management in a time-constrained production system under irrigated and nonirrigated conditions; and (ii) explore relationships among yield, LI, BM, tb, and CGR in a time-constrained production system.
|
Materials and methods |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
|
|
The irrigated treatment received water from an overhead irrigator when the estimated soil-moisture deficit reached 50 mm on the basis of an estimated cumulative evaporative demand (Cahoon et al., 1990). Weather variables were recorded by an automatic weather station in both years.
Data Collection and Calculation
Developmental stages were recorded every 14 d during vegetative growth and every 7 d during reproductive growth in 1997, and in 1998 at 2-d intervals throughout the season. Plots were considered to be in a particular growth stage, as described by Fehr and Caviness (1977), when 25% of the plants had reached a developmental stage. Above-ground BM was sampled at approximately 14-d intervals from a 1-m2 area of bordered rows beginning at growth stage V4 and continuing to R6, and samples were dried and weighed.
Crop growth rate and tb were calculated by regressing BM against time in the linear portion of BM accumulation. A linear regression was made for each plot using data that appeared linear (between days after emergence [DAE] 29 and 75 in 1997, and between DAE 37 and 78 in 1998). Linear regression indicated that at least 90% of the plots had R2 values of 0.90 or greater. The equation coefficients of intercept and slope were used to predict the x-intercept (where BM = 0), which served as our estimate of tb. Goudriaan and Monteith (1990) derived tb from BM samples when LI was greater than 95%. Our data included six BM samples per growing season, and, although our BM accumulation data were highly linear, maximum LI was much less than 95% for some treatment combinations. Our interpretation of tb will, therefore, include data where full LI was not achieved for low populations, and tb from low population plots would be underestimated.
Light interception was measured between 1100 and 1400 h on the same day as BM sampling, at approximately 10- to 14-d intervals during the growing season. A line quantum sensor (Li-Cor LI-191SA, Lincoln, NE), 1 m in length, which measured photosynthetically active radiation (PAR, µmol m-2 s-1) was first held above the canopy and then two measurements were made from each plot at the soil surface (Board et al, 1992). To achieve a representative measurement, we placed the sensor initially at a plant base perpendicular to the row for the first reading. The next reading was made approximately 0.25 m away from the initial placement within the same row, and a third measurement (if in a low density plot) was made at a random distance from the initial reading while remaining in the same row. Mean LI values from this technique for low population plots gave results not statistically different from LI values averaged from 100-mm distances down the plot (Ball and Purcell, 1997, unpublished results).
Light interception (LI, %) was calculated as follows:
 |
Yield was harvested from a bordered 20-m2 section of each plot with a plot combine. Yield was expressed at 130 g kg-1 seed moisture.
Experimental Design
In 1997 and 1998, the experiment was a multiple split-plot arrangement of treatments in a randomized complete block design, with four replications treated as the block factor. The main experimental unit was irrigation (irrigated and nonirrigated), with subunits of row spacing. The row spacing unit was further divided into a sub-subunit of cultivar, and the final split was population density (Tables 1 and 2). Treatment structure was 2 irrigation regimes x 3 row spacings x 2 cultivars x 5 population densities x 4 replications = 240 plots for each of 2 yr. Data were analyzed by year and row spacing in accordance with a general linear model (SAS Inst., 1989). Row spacing was not used as a class variable since populations within each row spacing were unique. Population density did not differ statistically over irrigation regime and cultivar within a year. Therefore, population was used as a class variable within a row spacing. Block was considered a random effect, and irrigation, cultivar, and population were considered as fixed effects. Relationships between yield and growth components were assessed from the Pearson product-moment correlation statistic.
|
Results and discussion |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
|
0.05, data not shown). Nonirrigated and irrigated yield generally increased as population density increased. For the indeterminate cultivar A4922, yield was still increasing at the highest population. In 1997, Manokin grain yields at higher populations within each row spacing tended to decline (Table 1), an effect attributed to lodging. At a low population density, the determinate Manokin had greater branching and, hence, higher yielding capabilities than did A4922. For example, in 1997 for the 0.19-m row spacing, yield of Manokin at a density of 12 plants m-2 was 67% greater than A4922 for the irrigated treatment and 58% greater for the nonirrigated treatment (Table 1).
Discounting lodging, higher yields came from treatments having the highest population densities. A4922 had the higher yield potential of the two cultivars used in our study when grown at populations exceeding 80 plants m-2. In both years, yields of cultivars were similar in irrigated 0.19-m rows at recommended populations (30–40 plants m-2). Maximum yield for A4922, however, occurred at greater than recommended populations, and yield was greater for A4922 than for Manokin (1997: 329 versus 289 g m-2; 1998: 332 versus 314 g m-2). Yield potentials, therefore, may differ among cultivars within a maturity group, and may require field testing at a range of population densities to determine the optimum density for a cultivar in a particular environment.
Light Interception and Lost Time
In a short season, population recommendations designed for a full-season crop may result in a crop stand unable to maximize light use. To maximize yield potential, the crop must produce sufficient leaf area to intercept light completely as early as possible. In 1997, all population densities at 22 plants m-2 and above for irrigated 0.19-m row treatments had maximum LI values above 90% (Table 1). Manokin, the determinate cultivar, had more branches at low populations resulting in slightly higher LI values than A4922. For both cultivars in 1997, densities at 22 plants m-2 and above in the 0.19-m row spacing and irrigated treatment had LI values greater than 90%. Nonirrigated soybean from 0.19-m row spacing in 1997 reached a LI > 90% for densities of 64 plants m-2 or greater. For the 0.19-m row spacing nonirrigated treatments in 1998, only the highest population density of 91 plants m-2 had an LI approaching 90% (Table 2). To achieve >90% LI for the irrigated treatment in 0.19-m rows in 1998, A4922 required a population of 91 plants m-2 while Manokin needed 42 plants.
Maximum LI in the growing season varied according to population density (Tables 1 and 2). Irrigated 0.19-m row treatments had a range of maximum LI from 100 to 85% for 1997 data, and 93 to 55% in 1998 (Fig. 3A) . Moreover, the time required for linear crop growth to begin was associated with maximum LI. The range of tb values varied from 16 to 27 d in 1997, and 22 to 37 d in 1998. Excessive rainfall and lower light intensities due to cloudy conditions in the early part of the 1998 season were associated with lengthening tb by up to 10 d. When looking at the subset of treatments in Fig. 3A, in which maximum LI was >90%, tb ranged from 14 to 26 d across treatments. This indicates that up to 12 d difference in achieving >90% LI capabilities existed due to population density. As density increased, the time taken to produce a canopy capable of intercepting >90% light decreased. In 1998 most treatments did not reach 90% LI. At these low populations, even at the tb of 35 d, two data points still had maximum LI of <70%. Plants from plots with low LI values failed to reach their full vegetative potential between R2 and R5. The range in tb values shows that lower populations took more time to reach linear growth, and, if populations were particularly low, maximum LI was considerably less than 90%.
|
In the 0.57-m row spacing treatments, tb values ranged from 18 to about 35 d in both irrigated and nonirrigated conditions (Fig. 4A and B) , and there was a close association between maximum LI and tb. Maximum LI varied from 100 to 38% regardless of irrigation treatment, although eight treatment combinations achieved >90% LI for irrigated conditions compared with only four treatment combinations for nonirrigated conditions. Despite a tb range close to those from 0.19-m row treatments (15–36 d), at maximum LI the tb value was generally greater in 0.57-m rows when compared with the 0.19-m row spacing. This is in agreement with similar yield increases associated with greater light interception duration due to sowing in narrow rows (Board et al., 1992). Although the range of tb values was similar for irrigated and nonirrigated treatments, the maximum LI values were less for nonirrigated than irrigated treatments at similar tb values. In nonirrigated conditions, the crop may have also experienced compounded limitations. Among these limitations are slower soybean crop growth (reduced CGR) under drought stress and lesser rates of leaf area expansion (Sinclair et al., 1987). For nonirrigated treatments at a given population density, a lower maximum LI than the irrigated counterpart resulted.
|
There are limits to the maximum population attainable for any given row spacing. As row spacing increases, intraplant distance (within row) decreases as population is increased until seeds are virtually sown side by side. A way to achieve high populations is to utilize narrow row spacing or equidistant planting patterns. Current Arkansas sowing density recommendations are 26 plants m-2 for 0.95-m rows, 26 plants m-2 for 0.57-m rows, and 35 plants m-2 for 0.15-m rows (Cooperative Extension Service, University of Arkansas, 1990). For A4922 in 1997 (Table 1) and both cultivars in 1998 (Table 2), densities exceeding current recommendations were clearly needed to achieve canopy closure and a maximum LI value of at least 90%. Under less optimum growing conditions, it is likely that even higher population densities would be required to reach 90% LI. Cooper (1989) found that high population in a low-yielding environment was neutral for yield response, whereas high population increased yield with favorable weather. Plant population may, therefore, be a strategy for optimizing yield in areas where intermittent drought is common.
Relationships between Yield and Growth Parameters
From Tables 1 and 2, we found the maximum LI in a growing season to be greatly different among populations and between years and irrigation treatments. Inter-relationships between yield and growth parameters were explored with simple correlations. Maximum LI for given treatment combinations was closely associated with maximum BM, with correlation coefficients ranging from 0.85 to 0.96 (Tables 3 and 4)
. Given that HI changed very little in our study in response to population density or irrigation regime (Ball et al., 2000), the amount of BM produced was strongly associated with final yield, and correlation values ranged from 0.59 to 0.94 (Tables 3 and 4).
|
|
As row spacing increased, maximum population density was limited by the distance between plants. Wide rows had a narrower range of population densities represented and, presumably, more plant competition within the row (limited rooting area, above-ground canopy spatial limitations, etc). The 0.95-m row showed the highest correlations between variables, reflecting that maximum LI, maximum BM, tb, and CGR were all limiting yield (data not shown). Lost time showed the strongest relationships with other parameters when rows became wider, and population density and LI were limiting. Sowing in a more equidistant configuration, such as in 0.19-m rows, permitted plants a geometrical advantage because of uniform distribution, leading to rapid canopy coverage.
Yield and Associated Lost Time
In Tables 3 and 4, the data indicated that yield was inversely related to tb. As LI approaches 100% early in the season, BM accumulation becomes linear and tb is minimized (Fig. 1). In a time-constrained crop, CGR, BM, and tb are dependent primarily upon early establishment of high LI. Light interception by a canopy can then be further broken down into the length of time the crop grows with LI greater than 85 to 90% (Monteith, 1977), and measured as LI duration (Board et al., 1992). Therefore, under nonstressed conditions, the BM produced is proportional to the cumulative amount of intercepted radiation (Monteith, 1977).
The inverse relationship between yield and tb is illustrated by Fig. 5 and 6 for 0.19 m and 0.57-m row treatments, respectively. For the irrigated treatments in 0.19-m rows, there appeared to be a threshold whereby decreasing tb to less than 25 d gave no further increase in yield (Fig. 5A). For the irrigated treatments in 0.57-m rows, decreasing tb gave corresponding increases in yield over the entire range of tb values (Fig. 6A). Similar responses for tb and yield were observed for 0.95-m row spacing (data not shown). Under nonirrigated conditions, a relationship between yield and tb was apparent for the 0.19 m (Fig. 5B) and 0.57 m (Fig. 6B) row spacing, but data were more variable than for irrigated treatments. Under nonirrigated conditions, water availability certainly limited yield. The association of yield with tb under nonirrigated conditions, however, indicates that having a complete canopy under intermittent drought provides the crop the capacity to utilize light energy effectively when conditions become more favorable for the production of BM and seed.
|
|
|
Conclusions |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
1. Longer times (tb) taken to reach full light interception. This results in a shorter time spent in the linear phase of BM accumulation resulting in a BM limitation. Note that excessive lost time reduces the time spent in the approximately linear growth because early maturing cultivars are selected for their short vegetative growth phases and the length of the reproductive phase is essentially fixed.
2. Lower rate of crop growth because of environmental stress during complete LI (the slope of the BM accumulation/time graph at the linear phase). Although having a high plant population and complete LI may not increase CGR under conditions of water-deficit stress, it provides the crop the capacity to utilize light energy efficiently should the stress be relieved by rainfall.
3. Lower rate of crop growth and incomplete LI. The population density is too low and the crop growth rate is limited by incomplete canopy coverage. Crops with incomplete LI may be limited by excessive tb, and a resulting low CGR and reduced BM by the end of the season.
Vegetative periods are shortened when soybean is sown late or when early maturing cultivars are used. In addition, stresses such as water, fertility, and pests may further limit crop growth. These factors dictate a continuum of plant populations necessary to exploit light and maximize yield for that system. A primary consideration in optimizing yield potential in short-season crops is to lessen the time needed to achieve LI values >90%. This ensures full light interception as early as possible. Therefore, a higher population density is required for a short-season crop than is necessary for a full season crop, so that maximum LI will continue as the weather and maturity rate of the crop permit. The greater the time constraint or stress pressure, the higher the population density and the narrower the row spacing required to maximize LI.SAS Institute 1989
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
Received for publication August 23, 1999.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. D. Lee, D. B. Egli, and D. M. TeKrony Soybean Response to Plant Population at Early and Late Planting Dates in the Mid-South Agron. J., June 16, 2008; 100(4): 971 - 976. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Kyei-Boahen and L. Zhang Early-Maturing Soybean in a Wheat-Soybean Double-Crop System: Yield and Net Returns Agron. J., February 7, 2006; 98(2): 295 - 301. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. S. Poag, M. Popp, J. Rupe, B. Dixon, C. Rothrock, and C. Boger Economic Evaluation of Soybean Fungicide Seed Treatments Agron. J., November 17, 2005; 97(6): 1647 - 1657. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. T. Edwards and L. C. Purcell Soybean Yield and Biomass Responses to Increasing Plant Population Among Diverse Maturity Groups: I. Agronomic Characteristics Crop Sci., August 1, 2005; 45(5): 1770 - 1777. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. T. Edwards, L. C. Purcell, and D. E. Karcher Soybean Yield and Biomass Responses to Increasing Plant Population among Diverse Maturity Groups: II. Light Interception and Utilization Crop Sci., August 1, 2005; 45(5): 1778 - 1785. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J. Heitholt, J. B. Farr, and R. Eason Planting Configuration x Cultivar Effects on Soybean Production in Low-Yield Environments Crop Sci., August 1, 2005; 45(5): 1800 - 1808. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Popp, J. Edwards, L. Purcell, and P. Manning Early-Maturity Soybean in a Late-Maturity Environment: Economic Considerations Agron. J., November 1, 2004; 96(6): 1711 - 1718. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. E. Board, M. S. Kang, and M. L. Bodrero Yield Components as Indirect Selection Criteria for Late-Planted Soybean Cultivars Agron. J., March 1, 2003; 95(2): 420 - 429. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. P. Jones, D. L. Holshouser, M. M. Alley, J. K.F. Roygard, and C. M. Anderson-Cook Double-Crop Soybean Leaf Area and Yield Responses to Mid-Atlantic Soils and Cropping Systems Agron. J., March 1, 2003; 95(2): 436 - 445. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. K. Norsworthy and J. R. Frederick Reduced Seeding Rate for Glyphosate-Resistant, Drilled Soybean on the Southeastern Coastal Plain Agron. J., November 1, 2002; 94(6): 1282 - 1288. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. E. Board A Regression Model to Predict Soybean Cultivar Yield Performance at Late Planting Dates Agron. J., May 1, 2002; 94(3): 483 - 492. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. C. Purcell, R. A. Ball, J. D. Reaper III, and E. D. Vories Radiation Use Efficiency and Biomass Production in Soybean at Different Plant Population Densities Crop Sci., January 1, 2002; 42(1): 172 - 177. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. A. Ball, R. W. McNew, E. D. Vories, T. C. Keisling, and L. C. Purcell Path Analyses of Population Density Effects on Short-Season Soybean Yield Agron. J., January 1, 2001; 93(1): 187 - 195. [Abstract] [Full Text]
|
|
|
|
|
|
R. A. Ball, L. C. Purcell, and E. D. Vories Short-Season Soybean Yield Compensation in Response to Population and Water Regime Crop Sci., July 1, 2000; 40(4): 1070 - 1078. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
