Published online 1 November 1999
Published in Agron J 91:928-933 (1999)
© 1999 American Society of Agronomy
677 S. Segoe Rd., Madison, WI 53711 USA
Agronomy Journal 91:928-933 (1999)
© 1999 American Society of Agronomy
INTEGRATED SOIL AND CROP MANAGEMENT
Growth Analysis of Soybean under No-Tillage and Conventional Tillage Systems
Raji I. Yusufa,
John C. Siemensa and
Donald G. Bullocka
a Dep. of Agricultural Engineering, Univ. of Illinois, Urbana, IL 61801-4798 USA
dbullock{at}uiuc.edu
 |
ABSTRACT
|
|---|
Soybean [Glycine max (L.) Merrill] plants grown with no-tillage (NT) often appear smaller than those grown with conventional tillage (CT), yet they produce similar grain yield. Our objective was to test the hypothesis that the early-season growth depression is offset by compensatory growth and changes in plant development. A 2-yr field study was conducted at Urbana, IL, on a long-term tillage experiment. Grain yield, moisture, protein, and oil content were similar for CT and NT treatments. Total plant, stem, leaf, and pod dry biomass were all initially about 15 to 20% greater under CT, but the difference declined until about R5 or R6; thus, compensatory growth did occur. At the initiation of sampling (V2) crop growth rate was about 20% greater under CT, but the difference declined until about R2. The advantage shifted to NT until about R6. Leaf area index (LAI) was greater for CT until about R4. Net assimilation rate was greater for NT until about R5. Increases in early-season crop growth rate for CT was due to increased LAI. Greater crop growth rate for NT late in the season was due to increased net assimilation rate. Leaf weight ratio was larger for the CT crop until about R6. Specific leaf area was less in CT than NT. This work supports our hypothesis that compensatory growth and alterations in plant development occur when soybean is grown in NT systems and helps to explain why grain yield does not decrease with NT even though early-season growth is affected.
Abbreviations: CGR, crop growth rate • CT, conventional tillage • L, leaf dry biomass • LAI, green leaf area index • LWR, leaf weight ratio • NAR, net assimilation rate • NT, no-tillage • P, pod dry biomass • S, stem dry biomass • SLA, specific leaf area • W, total plant dry biomass
|
INTRODUCTION
|
|---|
NO-TILLAGE PRODUCTION
results in changes in soil physical properties, including increases soil organic matter content (Douglas and Goss, 1982), aggregate stability (Heard et al., 1988), and macroporosity (Lal et al., 1990; Blackwell and Blackwell, 1989). Collectively and individually, these changes influence plant growth (Ehlers et al., 1983; Voorhees and Lindstrom, 1984; Dao, 1993). The changes can be detrimental, neutral, or beneficial for crop growth and yield, depending on soil texture and structure (Dick and VanDoren, (1985), climatic factors such as rainfall (Boyer, 1970), and weed control (Kapusta, 1979; Miguel and Matt, 1987).
In general, NT systems have greater positive effects on crop growth and yield when used on soils characterized by low organic matter levels and poor structure, rather than on well-structured soils high in organic matter (Kladivko et al., 1986). Many of the soils in the central Corn Belt have relatively high organic matter levels and good structure. On such soils, NT is detrimental to early-season plant growth but usually does not substantially decrease soybean grain yield (Kladivko et al., 1986). This lack of yield decrease is an enigma, since a decrease in early-season vegetative mass can decrease soybean grain yield due to premature flowering and insufficient leaf area index (Egli and Leggett, 1973; Board and Hall, 1984).
Growth analysis methods are useful tools for describing plant response to environmental variations (Radford, 1967; Hunt, 1982). The objective of this study was to test our hypothesis that a combination of compensatory growth and changes in plant development are responsible for stable grain yield in the face of reduced early-season plant growth for soybean on a high organic matter, well-structured soil in the central Corn Belt.
|
Materials and methods
|
|---|
Field Procedures
Field studies were conducted during the summers of 1993 and 1994 at the Agricultural Engineering Research Farm of the University of Illinois at Urbana–Champaign. This study was superimposed on a long-term tillage experiment in a corn [(Zea mays (L.)]–soybean rotation that was established in 1986 on a Thorp silt loam field (fine-silty, mixed, mesic Argiaquic Argialbolls) with 3.1% organic matter and a pH of 6.7. From the six tillage systems included in the long-term tillage experiment, we selected two for this study: conventional tillage and no-tillage. The CT consists of fall moldboard plowing (20–25 cm deep) of corn residue followed by spring disking (7.5–10 cm deep) and field cultivating in preparation for soybean. The NT, by definition, has no tillage or cultivation. The experiment design is a complete randomized block with four replications, and the experimental units consist of plots 85.3 m long by 6.1 m wide.
The indeterminate soybean cultivar Williams-82 was planted on 17 May 1993 and 18 May 1994 in 76-cm rows at a seeding rate of 33 seed m-1. Alachlor [2-chloro-N-(2,6-diethylphenyl)-N-(methoxymethyl)acetamide] and metribuzin [4-amino-6-(1,1-dimethyl)-3-(methylthio)-1,2,4-triazin-5(4H)-one] were applied preemergence for weed control at rates 3.4 kg ha-1 and 0.63 kg ha-1, respectively. Prior to planting, NT plots were sprayed with glyphosate [N-(phosphonomethyl)glycine] at 0.84 kg ha-1 and 2,4-D [(2,4-dichlorophenoxy)acetic acid] at 0.47 kg ha-1 tank-mixed with X-77 at 0.5% (v/v) in 187 L ha-1.
Sampling Procedures
Experimental units were hand-thinned to a stand of 20 plants m-1 between VC and V2 (Ritchie et al., 1983). A 1-m2 sampling area was selected randomly from each experimental unit at weekly intervals from V2 to R7 and all plants in the sampling area were clipped at the ground. Primary data consisted of green leaf area (A); dry biomass of leaf (L), stem (S), and pods (P); and total plant biomass (W), which was calculated as the sum of the component dry biomass values (L + S + P). Sampling was not conducted in the center two rows of each experimental unit, as these were reserved for final grain yield estimation.
Leaf area was estimated by measuring the green leaf area of all leaves with a leaf area meter (Model LI-3100, LI-COR, Lincoln, NE). The dry weight of the plant materials was measured after drying for 6 d in a forced-air dryer at 60°C. Growth stages (vegetative and reproductive) were recorded as the mean from three plants, selected randomly, from each sample. A 1.52- by 82-m strip at the center of each experimental unit was harvested for grain yield using a plot combine. The grain yield was adjusted to 130 g kg-1 moisture.
Estimation of Growth Analysis Components from Fitted Growth Curves
The analytical methods used in this study are similar to those of Bullock et al., (1998). Readers interested in a complete discussion of plant growth analysis are directed to Hunt (1982).
Initial analysis indicated that plant growth was similar between years. That, in combination with the factor Year being a random component in the model, led us to pool the primary data for each sampling period over years. The means of the primary data were transformed to natural logarithms to obtain homogeneity of errors (Steel and Torrie, 1980) and then were subjected to smooth curve fitting to describe the relationships between the primary measures and time. The logistic function a/[1 + b x exp(-ct)] with estimated parameters a, b, and c and time (t) (Hunt, 1982) was used to describe W and P, while cubic polynomial functions were used for L, S, and A. The logistic and polynomial curves were fit with the PROC NLIN and REG procedures, respectively, SAS (1996).
The relationship between primary biomass data [ln(g m-2)] and time (t, in weeks) may be written as Eq. [1–4] and between primary leaf area data [ln(m2 leaf area m-2 land area)] and t as Eq. [5]. Note that the units reflect the transformation of the primary data, and also that the sample area was consistently 1 m2 and thus land area designation is not required in the following equations.
 | (1) |
 | (2) |
 | (3) |
 | (4) |
 | (5) |
Crop growth rate (CGR; g m-2 land area per week) is the rate of change of the total plant dry biomass over time and was calculated as the first derivative of Eq. [1]:
 | (6) |
Net assimilation rate (NAR; g m-2 leaf area per week) is a measure of the change in total plant dry biomass per unit leaf area per unit time and was calculated as
 | (7) |
The units for CGR and NAR do not reflect the transformation of the primary data, since derivatives were taken. Leaf weight ratio (LWR; ln g/ln g) is the ratio of leaf dry biomass to total plant dry biomass and thus a measure of the proportion of the plant dry biomass residing in the leaf material. Leaf weight ratio was calculated as
 | (8) |
Specific leaf area (SLA; ln m2 leaf area/ln g leaf dry biomass) is the ratio of leaf area to leaf plant dry biomass and thus a measure of leaf thickness. Specific leaf area was calculated as
 | (9) |
Equations [1–9] allowed the calculation of predicted values at each sampling time ti (where i = 1 to 14) for the primary data and the crop growth analysis variates CGR, NAR, LWR, and SLA. Analysis of variance was performed on the predicted values at each sampling time. The F-test of tillage treatment tested the hypothesis of no difference between NT and CT. The advantage of the method is that it utilizes smaller standard errors of predicted values.
Grain Protein and Oil Content
Several hundred seeds were selected randomly from the harvested grain of each plot and dried in a forced-air oven at 60°C for 24 h. A 12-seed subsample from each experimental unit was then ground to pass through a 1-mm screen. The total N of the seed was determined using a micro-Kjeldahl digest procedure (Nelson and Sommers, 1980) and grain protein was estimated as 6.25 x N. The oil content of the grain from each tillage system was determined using a Soxhlet extraction method (AOAC, 1984). Analysis of variance was performed on the grain yield, moisture, oil content, and protein content via the PROC GLM procedure of SAS (1996).
|
Results and discussion
|
|---|
Weather
Long-term monthly precipitation and air temperature means at Urbana, IL, and deviation from those means for 1993 and 1994 are presented in Fig. 1 and 2
respectively. In general, 1993 was wetter and cooler than the long-term mean. In contrast, with the exception of April, 1994 was drier than the mean during the growing season and slightly warmer than the mean, particularly in June and September.
View larger version (28K):
[in this window]
[in a new window]
|
Fig. 1 Long-term (1961–1990) monthly total precipitation means and deviation from those means during 1993 and 1994 at Urbana, IL
|
|
View larger version (27K):
[in this window]
[in a new window]
|
Fig. 2 Long-term (1961) monthly air temperature means and deviations from those means during 1993 and 1994 at Urbana, IL
|
|
Grain Yield and Moisture
The cool and wet conditions experienced in 1993 were expected to have exacerbated any negative effects of NT on soybean yield, but that was not the case. Analysis of variance indicated that soybean yield was not affected by either the main effects of tillage and year or by the year x tillage interaction (Table 1) . The lack of a mean yield difference between tillage treatments is in agreement with the majority of the literature in which tillage systems are compared on high organic matter, well-structured soils (Colvin and Erbach, 1982; Erbach, 1982; Richey et al., 1977).
View this table:
[in this window]
[in a new window]
|
Table 1 Abbreviated ANOVA and means for grain yield, grain moisture, oil content, and protein content soybean grown with conventional (CT) and no-tillage (NT) at the University of Illinois Agricultural Engineering Research Farm, Urbana, IL, in 1993 and 1994
|
|
Grain moisture was affected only by the main effect of year (Table 1). This was due to a slightly later harvest and thus more field drying in 1993 than 1994. Grain moisture was affected by neither the main effect of tillage nor the year x tillage interaction.
Grain Oil and Protein
Soybean oil and protein content were significantly affected only by the main effect of year (Table 1). Soybean grain contained more oil in 1994 than in 1993 (22 g kg-1 vs. 19 g kg-1), but contained less protein in 1994 than in 1993 (37 g kg-1 vs. 42 g kg-1). This inverse relationship between grain oil and grain protein is well known (Scott and Aldrich, 1983). Differences in grain oil and protein response between the years was probably due to temperature. Cool conditions during grain fill generally result in a decrease in grain oil and an increase in grain protein (Calvin, 1965). Air temperature during the grain-filling period (mid to late August and early September) was cooler in 1993 than 1994 (Fig. 2).
Dry Biomass
Note again that due to heterogeneous variances the primary data were transformed Eq. [1–5] and thus the units of some of the predicted values are affected by that transformation as shown in Eq. [1–9] and as labeled on Fig. 3 to 11
. To avoid cumbersome language the transformed units will no longer be designated as such in the discussion. For example the total plant dry biomass will not be referred to as the transformed total plant dry biomass or ln plant dry biomass, but rather simply as the total plant dry biomass. This is an accepted practice in the growth analysis literature (Hunt, 1982) and is appropriate, since the transformation is done only for statistical reasons and does not alter the appropriate interpretation of the data.
The estimated coefficients and their respective standard errors for the nonlinear fits of the total plant and pod dry biomass accumulation with time to the logistics function are presented in Table 2
. Likewise, coefficients and standard errors for fits of the cubic polynomial function to the green leaf area index and leaf and stem dry biomass accumulation with time are shown in Table 3
.
View this table:
[in this window]
[in a new window]
|
Table 2 Estimated parameter coefficients and their standard errors (SE) of the nonlinear regression equations of the logistics function fitted to the total plant dry biomass and total pod biomass of soybean grown with conventional (CT) and no-tillage (NT) at the University of Illinois Agricultural Engineering Research Farm, Urbana IL over a 2-yr period (1993 and 1994)
|
|
View this table:
[in this window]
[in a new window]
|
Table 3 Estimated parameter coefficients and their standard errors of cubic polynomial regression equations for leaf area index, leaf dry biomass, and stem dry biomass of soybean grown with conventional (CT) and no-tillage (NT) at the University of Illinois Agricultural Engineering Research Farm, Urbana, IL, over a 2-yr period (1993 and 1994)
|
|
While yield was not affected by tillage system in this study it was visually obvious early in the season of both years that soybean plants grown with NT conditions were smaller and less vigorous than plants grown with CT. The total plant dry biomass accumulation is presented in Fig. 3. Treatment did not affect the time of emergence, but did affect subsequent growth and development. Soybean plants on CT plots developed rapidly and were visually larger for much of the vegetative period. Two weeks after emergence when sampling commenced, soybean growing on the CT plots had produced a substantially larger total plant dry biomass than those on NT plots, and the difference was maintained until R6 (Fig. 3). The total plant dry biomass did not differ between tillage treatments at the end of the season. This observation confirms the first part of our hypothesis that, soybean does demonstrate compensatory growth which contributes to the similar grain yield in the CT and NT systems.
The leaf (Fig. 4), stem (Fig. 5), and pod (Fig. 6) dry biomass were all initially greater with CT than NT; however, the magnitude of the difference varied. For leaf dry biomass the difference lasted until about R6 (Fig. 4), while the difference in stem dry biomass was observed until R7, although the mean difference declined with time (Fig. 5). The difference in pod dry biomass for the CT and NT treatments was also largest during the early season and vanished by R6. Thus, for all measures of dry biomass, soybean biomass accumulation was greater with CT than NT until late into the grain-filling period, but the magnitude of these differences declined with time. The differences were substantial only prior to R5. After R5, dry biomass measures were essentially the same.
Crop Growth Rate
Crop growth rate over the entire sampling period is presented in Fig. 7. Soybean growing on CT plots had an initial higher CGR than those on NT plots. The difference was present from the start of sampling and continued until late R2, after which the soybean growing in NT possessed a greater CGR (Fig. 7). Note also that CGR for plants with CT peaked earlier than the CGR for plants with NT. This again demonstrates that plants growing with NT conditions were delayed relative to the CT treatment, but compensated for the early delay by rapid growth prior to and during grain fill. This later increase in CGR is responsible for the similar total plant dry biomass at the end of the season and presumably contributed to NT and CT soybean producing similar yields.
Leaf Area Index
Plants ultimately depend on green leaf surface, measured here by the LAI, for the vast majority of the photosynthetic output that is deposited into the developing grain and other plant parts. The LAI under CT was larger than that under NT during the early season prior to R5 (Fig. 8). From R5 through the remainder of the season, the difference in CT and NT LAI vanished. There was no difference in LAI between CT and NT for the majority of the grain-filling period (R4–R7). Thus, the potential photosynthetic capacity of the plants as indicated by green leaf surface did not differ in the two different tillage systems when it mattered the most (i.e., during grain fill).
Net Assimilation Rate
At the commencement of sampling, NAR of soybean under NT was greater than that under CT (Fig. 9). This difference was maintained until late R6; however, the difference decreased with time and was largest prior to R4. The decrease in NAR in both CT and NT is attributed to the shading effect of lower leaves by the upper canopy and a general decline in photosynthetic efficiency with leaf age.
By definition, CGR is the product of LAI and NAR (Hunt, 1982). The early-season CT advantage (prior to R2) in CGR (Fig. 7) was due entirely to a greater LAI (Fig. 8), which was able to compensate for the greater NAR under NT (Fig. 9). As the season progressed, the difference in LAI for the CT and NT soybean decreased to R4 and was similar from R5 to R7 (Fig. 8). The larger NAR of the NT soybean crop was responsible for significantly greater CGR in NT. Thus, while both CT and NT soybean had periods in which their CGR was significantly greater, the basis for the advantage differed. In the case of CT soybean, the CGR advantage was due to a greater LAI; for the NT soybean, the CGR advantage was due to a greater NAR. In general, an increase in CGR is more commonly due to an increase in LAI rather than an increase in NAR (Clawson et al., 1986).
Specific Leaf Area
We have shown that the CT treatment resulted in greater LAI until R4 (Fig. 8) and greater leaf dry biomass accumulation until R5 (Fig. 4). The SLA (defined as the ratio of leaf area to leaf weight), however, was less with the CT treatment than the NT treatment (Fig. 10) except when there was a reversal at R5 to R6. It is not clear why such an event should have occurred, and it may be more an artifact of the curve fitting process than an indication of a real event. The increase in SLA indicates that the leaves of soybean plants growing with CT conditions were thicker than the leaves of soybean plants growing with NT conditions for most of the season. These data also indicate that the increased LAI with CT noted previously was indeed due to an increased leaf dry biomass which, in fact, compensated for and overcame a decreased SLA in CT. The increased LAI in turn was responsible for the larger CGR noticed for the CT crops during the early season.
Leaf Weight Ratio
Leaf weight ratio is a measure of the proportion of the total plant dry biomass devoted to leaf material. The LWR for the CT was significantly greater than for the NT soybean through mid R6 and did not differ for the remainder of the season (Fig. 11). Thus, prior to R6, soybean grown under CT conditions preferentially distributed photosynthates to the production leaves as compared to soybean grown under NT.
|
Conclusion
|
|---|
The changes in LWR and SLA are fundamental changes in plant growth and development. This work supports our initial hypothesis that a combination of compensatory growth and plant form are responsible for stable grain yield in the face of reduced early-season plant growth for soybean on a high organic matter, well-structured soil in the central Corn Belt.SAS Institute 1996
Received for publication August 24, 1998.
|
REFERENCES
|
|---|
- AOAC. 1984. Fat (crude) or ether extract in animal feed. In S. Williams (ed.) Official methods of analysis of the Association of Official Analytical Chemists. AOAC, Washington, DC.
- Blackwell P.S., Blackwell J. The introduction of earthworms to an ameliorated, irrigated duplex soil in south-eastern Australia and the influence on macro pores. Aust. J. Soil Res. 1989;27:807-814.
- Board J.E., Hall W. Premature flowering in soybean yield reductions at non-optimal planting dates as influenced by temperature and photoperiod. Agron. J. 1984;76:700-704.[Abstract/Free Full Text]
- Boyer J.S. Leaf enlargement and metabolic rates in corn, soybean, and sunflower at various leaf water potentials. Plant Physiol. 1970;46:233-235.[Abstract/Free Full Text]
- Bullock D.G., Khan S., Rayburn A. Early soybean [Glycine max (L.) Merr.] growth is affected by row spacing. Crop Sci. 1998;38:1011-1016.[Abstract/Free Full Text]
- Calvin D.T. The effect of temperature on the oil content of fatty acid composition of the oils from several oil seed crops. Can. J. Bot. 1965;43:63-69.
- Clawson K.L., Specht J.E., Blad B.L. Growth analysis of soybean isolines differing in pubescence density. Agron. J. 1986;78:64-172.
- Colvin T.S., Erbach D.C. Soybean response to tillage and planting methods. Trans. ASAE 1982;25:1533-1535,1539.
- Dao T.H. Tillage and winter wheat residue management effects on water infiltration and storage. Soil Sci. 1993;57:1586-1595.
- Dick W.A., VanDoren D.M., Jr. Continuous tillage and rotation combinations effects on corn, soybean, and oat yields. Agron. J. 1985;77:459-465.[Abstract/Free Full Text]
- Douglas J.T., Goss M.J. Stability and organic matter content of surface soil aggregates under different methods of cultivation and in grassland. Soil Tillage Res. 1982;2:155-175.
- Egli D.B., Leggett J.E. Dry matter accumulation patterns in determinate and indeterminate soybeans. Crop Sci. 1973;13:220-222.[Abstract/Free Full Text]
- Ehlers W., Kopke U., Hesse F., Bohm W. Penetration resistance and root growth for oats in tilled and untilled loess soil. Soil Tillage Res. 1983;3:261-375.
- Erbach D.C. Tillage for continuous corn and corn–soybean rotation. Trans. ASAE 1982;25:906-911,918.
- Heard J.R., Kladivko E.J., Mannering J.R. Soil macro-porosity, hydraulic conductivity, and air permeability of silty soils under long-term conservation tillage in Indiana. Soil tillage Res. 1988;11:1-18.
- Hunt R. Plant growth curves: The functional approach to plant growth analysis. Baltimore, MD: Arnold, London, and Univ. Park Press, 1982.
- Kapusta G. Seedbed tillage and herbicide influence on soybean (Glycine max) weed control and yield. Weed Sci. 1979;27:520-526.
- Kladivko E.J., Griffith D.R., Mannering J.V. Conservation tillage effects on soil properties and yield of corn and soybean in Indiana. Soil Tillage Res. 1986;8:277-287.
- Lal R., de Vleeschauwer D., Ngaje R.M. Changes in properties of a newly cleared tropical Alfisol as affected by mulching. Soil Sci. Soc. Am. J. 1990;44:823-827.
- Miguel A.A., Matt L. Weed management in ecological approaches. Boca Raton, FL: CRC Press, 1987.
- Nelson D.W., Sommers L.E. Total nitrogen analysis of soil and plant tissues. J. Assoc. Off. Anal. Chem. 1980;63:770-778.
- Radford P.J. Growth analysis formulae: Their use and abuse. Crop Sci. 1967;7:171-175.
- Richey C.B., Griffith D.R., Parsons S.D. Yields and cultural energy requirements for corn and soybeans with various tillage–planting systems. Adv. Agron. 1977;29:141-182.
- Ritchie, S.W., J.J. Hanway, H.E. Thompson, and G.O. Benson. 1983. How a soybean plant develops. Rev. ed. Spec. Rep. 53. Iowa State Univ. Coop. Ext. Serv., Ames.
- SAS Institute. SAS user's guide, 3rd ed Cary, NC: SAS Inst, 1996.
- Scott W.O., Aldrich S.R. Modern soybean production, 2nd ed Champaign, IL: S & A Publications, 1983.
- Steel R.G.D., Torrie J.H. Principles and procedure of statistics: A biometrical approach, 2nd ed New York: McGraw-Hill, 1980.
- Voorhees W.B., Lindstrom M.J. Long-term effects of tillage method on soil tilth independent of wheel traffic compaction. Soil Sci. Soc. Am. J. 1984;48:152-156.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
J. W. Singer, S. D. Logsdon, and D. W. Meek
Soybean Growth and Seed Yield Response to Tillage and Compost
Agron. J.,
June 16, 2008;
100(4):
1039 - 1046.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
P. Pedersen and J. G. Lauer
Soybean Growth and Development Response to Rotation Sequence and Tillage System
Agron. J.,
July 1, 2004;
96(4):
1005 - 1012.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
P. Pedersen and J. G. Lauer
Soybean Growth and Development in Various Management Systems and Planting Dates
Crop Sci.,
March 1, 2004;
44(2):
508 - 515.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
= 0.01, 0.05, 0.10 and lack of significance at 
