Crop Science 41:1162-1168 (2001)
© 2001 Crop Science Society of America
CROP ECOLOGY, MANAGEMENT & QUALITY
Cotton Growth and Development under Different Tillage Systems
Charles W. Kennedy*,a and
Robert L. Hutchinsonb
a 104 Sturgis Hall, Dep. of Agronomy, Louisiana State Univ. Agric. Center, Baton Rouge, LA 70803
b Northeast Research Station, Louisiana State Univ. Agric. Center, St. Joseph, LA 71366
* Corresponding author (ckennedy{at}agctr.lsu.edu)
 |
ABSTRACT
|
|---|
Adoption of reduced tillage systems is a means to lower production costs and improve soil productivity, but yield of cotton (Gossypium hirsutum L.) can be variable. Our objective was to determine when and how different tillage systems affected growth quantities and yield. The effect of conventional tillage (CT), ridge tillage (RT), and no tillage (NT) on crop growth rate (CGR), leaf area index (LAI), net assimilation rate (NAR), fruiting form numbers and weights, plant population, and crop yield was determined on a Gigger silt loam (fine-silty, mixed, thermic, Typic Fragiudalf). In 1991, the NT system produced a maximum prebloom CGR of 11 g m-2 d-1 compared with 7 and 7.5 g m-2 d-1 for the CT and RT systems, respectively. In 1992, a year with more adverse growing conditions, prebloom CGR values were lower for all systems but NT was greatest at 7 g m-2 d-1. Differences in prebloom CGR among tillage systems in 1994 were similar to the1992 results although values were similar to those of 1991. Prior to blooming, the RT system produced a consistently lower CGR than one or both of the other systems. Differences in LAI corresponded to CGR differences, but there were no differences among the three tillage systems in NAR. Greater early CGR led to faster ontogenic development and eventually to greater lint yields. Lint yields averaged 1057 kg ha-1 for NT, 1007 kg ha-1 for CT, and 890 kg ha-1 for RT. The NT system, which in this study had the greatest prebloom CGR and lint yield, was the conservation tillage system with the greatest production potential in this soil type. Early-season differences in soil characteristics among tillage systems at 0 to 0.15 m was considered a factor.
Abbreviations: CT, conventional tillage • CGR, crop growth rate • LAI, leaf area index • NAR, net assimilation rate • NT, no tillage • RT, ridge tillage
|
INTRODUCTION
|
|---|
INTEREST IN CONSERVATION TILLAGE SYSTEMS often with the use of cover crops has increased with the need to reduce production costs and improve soil productivity. Cotton response to conservation tillage has been variable (Keeling et al., 1989; Stevens et al., 1992; Boquet et al., 1997). This situation exacerbates the overall objective of conservation tillage, which is to maintain crop productivity while providing additional soil benefits attributed to these types of tillage systems (Touchton and Reeves, 1988). Stand establishment problems in these systems have been implicated as causing lower productivity in some studies (Grisso et al., 1984; Morrison et al., 1985), but lower population did not always result in lower yields (Touchton and Reeves, 1988). Because of the wide plant population range acceptable for cotton production in the Mid-South (Bridge et al., 1973), yield effects may not necessarily be attributed to any population differences. Analysis of crop growth and development can provide insight into differences between various treatment inputs that affect yield (Ashley et al., 1974). Such analyses would lead to a better understanding of how conservation tillage–cover crop systems improve or impair cotton productivity. Our objectives were to quantify cotton crop growth and development under different tillage–cover crop systems throughout the growing season and relate these growth parameters to yield.
|
MATERIALS AND METHODS
|
|---|
The cotton cultivar Stoneville 453 was seeded on 14 May 1991, 4 May 1992, and 7 May 1994 into a Gigger silt loam (fine-silty, mixed, thermic, Typic Fragiudalf). Tillage systems consisted of conventional tillage (CT), ridge tillage (RT), and no tillage (NT). The CT system consisted of disking four times during the month of April each year, followed by disk hipping and, finally, bedding with a reel and harrow. The RT system consisted of using a sweep (Buffalo row cleaner, Fleischer Manufacturing, Inc., Columbus, NE) to clear residue and soil from a portion of the raised seed bed that was prepared during the previous season. The portion cleared was about 0.5 m wide and 25 mm deep. The NT system consisted of using a fluted coulter in front of double-disk openers on the planter on beds originally developed in 1987. The same planter was used for all tillage systems. Each tillage system was factorially arranged with four different cover crops: native vegetation, winter wheat (Triticum aestivum L.), crimson clover (Trifolium incarnatum L.), and hairy vetch (Vicia villosa L.). The experiment was a randomized complete block factorial design with four replications. Plots consisted of eight 1-m-wide rows that were 15.2 m long. Tillage and cover crop management, seeding, fertilizer, herbicide, and harvesting methods in this study have been previously described by Boquet et al. (1997). These treatment plots were maintained continuously since 1987.
Data Collection
Plant populations were determined about 20 d after planting (DAP) on a row adjacent to a border row. Plant samples for growth analyses were taken at approximately biweekly intervals or less. All plants were removed from a 0.6-m section of row for each sampling time. The number of plants in this section had to correspond proportionately to the average plant population determined in that plot. Leaf area was determined with a Li-Cor 3000 area meter (LI-COR Inc., Lincolin, NE), measuring all leaves in each sample early in the season. As plants became larger, leaf area was taken on one-half to one-third of plants in the sample. The total leaf area for these samples was determined by the specific leaf area method (Wells and Meredith, 1986). Total leaf area was divided by the land area below the sampled plants to determine leaf area index (LAI). Total fruiting structures from plant samples were counted, and bolls were grouped according to location on the plant. Groupings were node position on a sympodia (1, 2, and 3 and beyond) and sympodial location on the mainstem (Mainstem Nodes 5–8, 9–12, 13–16, and 17 and above). All bolls produced on monopodia were pooled separately. All plant tissues were dried at 70°C for a minimum of 48 h before weighing. Individual boll weights were determined by averaging within each boll grouping. Crop growth rate (CGR) and net assimilation rate (NAR) were determined by the method of Hunt and Parsons (1981) using total dry matter and LAI values from each sampling date. Yield data were collected from the four center rows of each plot.
Weather data were collected from two nearby locations. Maximum and minimum daily temperatures were collected from a weather station about 10 km from the experiment site, and rainfall data were collected less than 1 km from the site.
Statistical evaluation used general linear models analysis of variance (SAS, 1985) followed by LSD mean separation when F-tests were significant. Analysis of CGR and NAR was based on the stepwise regression method of Hunt and Parsons (1981). Standard errors generated by the regression program were used to determine significant differences using t-tests. In our data, means ± standard error that did not overlap reflected the same statistically significant difference as those found with the t-test. Thus standard errors were used to show significant differences in the figures for CGR and NAR.
|
RESULTS AND DISCUSSION
|
|---|
Weather conditions were best for cotton growth and yield during 1991, followed closely by conditions in 1994. Temperatures during the growing season in those years were near to slightly above normal. Rainfall was fairly well distributed and of amounts considered to be effective (Fig. 1). In 1992, temperatures were below normal, especially early in the growing season, and rainfall was not as well distributed. Cotton was initially exposed to a cool, wet period followed by a warm, dry period (Fig. 1). These climatic differences had a large effect on growth parameters among years, but were not considered to have a differential influence for these parameters among treatments within years.
View larger version (38K):
[in this window]
[in a new window]
|
Fig. 1. Rainfall from 50 d prior to 140 d after planting and maximum and minimum daily temperatures from 10 d prior to 140 d after planting. Rainfall data were taken less than 1 km from the experiment, and temperature data were taken about 10 km away.
|
|
There were generally no interactions between tillage system and cover crop for the parameters analyzed. The occasional interactions that did occur were inconsistent over time and not considered a major influence on the results. Therefore, for the purposes of these results, the data were pooled across cover crop for each tillage system.
Differences in CGR between tillage systems began very early in the development of the crop (Fig. 2). Generally, NT reached the exponential phase of growth sooner and this phase was greater than that for the RT system and, in some cases, for the CT system. Maximum CGR values in 1991 approached rates previously published for cotton (Mauney, 1986) and C3 crops in general (Larcher, 1983). The NT system usually did not have the highest maximum CGR, but did have the greatest rate of exponential growth. The components upon which CGR is based, LAI and NAR, varied in response to tillage system. As with CGR, development of LAI was generally faster for NT than for the other systems, but maximum LAI was similar for all (Fig. 3). Maximum LAI ranged from about 2 in 1992 to 4 in 1994 (Fig. 3). Baker and Meyer (1966) determined that a canopy LAI of 3 to 4 produced maximum photosynthesis. This level was achieved in 1991 and 1994. While LAI was affected by tillage system, NAR generally was not (Fig. 4). This indicated that photosynthetic efficiency per se was not altered by tillage system. Factors other than this were apparently involved in tillage effects on growth and development.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 2. Tillage effects on crop growth rate (CGR) over the growing season for cotton cv Stoneville 453 on a Gigger silt loam in 1991, 1992, and 1994. Tillage treatments had been in place since 1987. Error bars represent SE. Differences in means ±SE reflect significant t-test differences.
|
|
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 3. Tillage effects on leaf area index (LAI) over the growing season for cotton cv Stoneville 453 on a Gigger silt loam in 1991, 1992, and 1994. Tillage treatments had been in place since 1987. Error bars represent LSD 0.05 unless subtended by  (LSD 0.10). Differences between points at a given time not having an error bar are NS.
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 4. Tillage effects on net assimilation rate (NAR) over the growing season for cotton cv Stoneville 453 on a Gigger silt loam in 1991, 1992, and 1994. Tillage treatments had been in place since 1987. Error bars represent SE. Differences in means ±SE reflect significant t-test differences.
|
|
The importance of early development has been noted by many studies (Ashley et al., 1965, 1974; Muramoto et al., 1965; Potter and Jones, 1977; Watson, 1952; Wells and Meredith, 1986). The faster exponential growth exhibited by NT and, in some cases, CT early in the season (compared with RT) resulted in more rapid ontogenic development both vegetatively (LAI) and reproductively. Evidence for the relationship of early crop growth rates and LAI with faster ontogeny is shown by the correlation coefficients in Table 1. Crop growth rate early in the season prior to visible flower buds was highly related to subsequent LAI, flower bud development prior to blooming, early boll set, and individual boll weight after cut-out. In turn, these parameters were each correlated moderately to highly with lint yield of the crop. As seen in Fig. 5, NT and, depending on year, CT produced more early flower buds than RT leading to earlier boll set. This usually led to faster boll development by the NT and/or CT system(s), compared with RT. An earlier and somewhat faster bollset was coupled with a numerically to significantly higher weight per boll in an area of the plant that produced more than 50% of the yield, i.e., Position 1 on sympodia off Mainstem Nodes 5 to 12 (Fig. 6). These effects led to a numerically and/or significantly higher lint yield for the NT and CT systems, compared with RT (Fig. 7). In 1991 when plant response to CT was similar to that to RT and lower than to NT, yield responses reflected these differences. Plant population did not contribute greatly to these differences in any year. Ridge till and CT plant populations were comparable and NT slightly lower (Fig. 8) but still within the range of 70 000 to 120 000 plants ha-1 found by Bridge et al. (1973) to maximize yields for Mid-South conditions. Moreover, lint yield had a larger correlation coefficient with prebloom CGR than with plant population (Table 1).
View this table:
[in this window]
[in a new window]
|
Table 1. Correlation coefficients of lint, plant population, growth parameters, and yield components across years and tillage systems for cv. Stoneville 453. N = 36.
|
|
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 5. Tillage effects on flower bud production and boll set over the season for cotton cv Stoneville 453 on a Gigger silt loam in 1991, 1992, and 1994. Tillage treatments had been in place since 1987. Error bars represent LSD 0.05 unless subtended by (LSD 0.10). Differences between points at a given time not having an error bar are NS.
|
|
View larger version (29K):
[in this window]
[in a new window]
|
Fig. 6. Tillage effects on dry matter accumulation of first position bolls located on sympodia off Mainstem Nodes (MSN) 5 to 8 and 9 to 12 for cotton cv Stoneville 453 grown on a Gigger silt loam in 1991, 1992, and 1994. Tillage treatments had been in place since 1987. Error bars represent LSD 0.05 unless subtended by (LSD 0.10). Differences between points at a given time not having an error bar are NS.
|
|
View larger version (41K):
[in this window]
[in a new window]
|
Fig. 7. Tillage effects on lint yield of cotton cv Stoneville 453 grown on a Gigger silt loam in 1991, 1992, and 1994. Tillage treatments had been in place since 1987. Differences between bars subtended by the same letter are NS (P < 0.05) for a given year.
|
|
View larger version (49K):
[in this window]
[in a new window]
|
Fig. 8. Tillage effects on plant population of cotton cv Stoneville 453 on a Gigger silt loam in 1991, 1992, and 1994. Tillage treatments had been in place since 1987. Differences between bars subtended by the same letter are NS (P < 0.05) for a given year.
|
|
As previously indicated, the reason for the lower CGR in the RT system was not caused by differences in NAR (Fig. 4), but was instead related to LAI (Fig. 3). This effect may be attributable to root growth and development and the relationship with shoot growth. Previous studies indicated that root restriction can result in a reduction in vegetative growth (Ben-Porath and Baker, 1990; Carmi and Shalhevet, 1983). Research has shown that root elongation rate is a function of soil resistance to penetration as well as internal factors (Greacen and Oh, 1972; Taylor and Gardner, 1963; Whisler et al., 1986). Any changes in soil strength will alter root elongation rates and could ultimately affect shoot growth. We did not collect soil impedance data during the years of this study. However, these data were collected in 1993 on these plots. Measured 53 DAP, soil strength from 0 to 0.15 m was significantly lower for NT (0.37 MPa) compared with CT (0.5 MPa) and RT (0.56 MPa). These data do not indicate severe limitations to root growth, but show that differences between tillage systems did exist in this long-term study. These differences, although perceived as slight, may have had some influence on growth especially in early stages of ontogeny. Alterations in shoot growth could occur either as a result of reduced nutrient uptake (Klepper, 1990) or possibly by some as yet unknown feed forward process as suggested by the results of Young et al. (1997) for monocot species.
The NT system had the lowest soil impedance, indicating improved soil structure for that conservation tillage system. Improvements more than likely occurred sooner than 1993 and had approached equilibrium, especially in the case of NT, by the time this experiment was initiated. In the plots of this study Boquet et al. (1997) found a greater amount of organic matter in the top 0.15 m of soil with the NT system. A higher organic matter content in NT would lead to improved soil structure and less soil impedance to root growth. Triplett et al. (1996) found that a NT system improved yields above a CT system over time and attributed this to the maintenance of soil macropores resulting in long-term improvement of soil structure. In our case, the CT system probably provided adequate soil aggregation for minimal early season impedance to root growth but, according to our data, the CT effects varied from year to year, depending on conditions present during and after tillage. The RT system employed a sweep to clear a planting strip on beds established the previous year. This process removed the top 25 mm of surface soil, including organic residue and related soil aggregation, and may also have caused compaction near the germinating seed. Stephens and Johnson (1993) found planter closing units could increase soil strength in the seed zone considerably under susceptible soil conditions. This susceptibility may have been greatest for the RT system. In our study, the RT system apparently did not improve soil structure as did NT or provide a short-term improvement in soil aggregation prior to planting as CT practices generally do.
|
CONCLUSIONS
|
|---|
Factors that affect crop growth early in ontogeny often produce modifications that extend through the season and may be manifest in altered economic yield. Our results reflected this phenomenon. Higher lint yields were related to faster early-season CGR and LAI development through the effect these parameters had on subsequent ontogenic development. These growth parameters may have been modulated by soil conditions (soil strength and organic matter) that developed from the type of tillage systems used in our study. The NT system was a superior conservation tillage system for cotton than the RT system. The NT system produced cotton with a consistently greater prebloom CGR and final economic yield than the RT system. Cotton in the CT and NT systems was usually, but not always, similar in performance. We did not conclusively determine why prebloom CGR was consistently lower for RT, but soil compaction that reduced or slowed seedling root growth may have been involved to some extent.
|
NOTES
|
|---|
Approved for publication by the Director of the Louisiana Agric. Exp. Stn. as Manuscript No. 00-09-0180.
Received for publication April 27, 2000.
|
REFERENCES
|
|---|
- Ashley, D.A., J.E. Elsner, O.L. Brooks, and C.E. Perry. 1974. Factors affecting early growth of cotton and subsequent effects on plant development. Agron. J. 66:20–23.[Abstract/Free Full Text]
- Ashley, D.A., B.D. Doss, and O.L. Bennett. 1965. Relation of cotton leaf area index to plant growth and fruiting. Agron. J. 57:61–64.[Abstract/Free Full Text]
- Baker, D.N., and R.E. Meyer. 1966. Influence of stand geometry on light interception and net photosynthesis in cotton. Crop Sci. 6:15–19.
- Ben-Porath, A., and D.N. Baker. 1990. Taproot restriction effects on growth, earliness, and dry weight partitioning of cotton. Crop Sci. 30:809–814.[Abstract/Free Full Text]
- Boquet, D.J., R.L. Hutchinson, W.J. Thomas, and R.E.A. Brown. 1997. Tillage and cover crop effects on cotton growth, yield and soil organic matter. p. 639–641. In P. Duggar and D. Richter (ed.) Proc. Beltwide Cotton Prod. Res. Conf., New Orleans, LA. 3–7 Jan. 1997. National Cotton Council, Memphis, TN.
- Bridge, R.R., W.R. Meredith, Jr., and J.F. Chism. 1973. Influence of planting method and plant population on cotton (Gossypium hirsutum L.). Agron. J. 65:104–109.[Abstract/Free Full Text]
- Carmi, A., and J. Shalhevet. 1983. Root effects on cotton growth and yield. Crop Sci. 23:875–878.[Abstract/Free Full Text]
- Greacen, E.L., and J.S. Oh. 1972. Physics of root growth. Nature (London) New Biol. 235:24–25.
- Grisso, R., C. Johnson, and W. Dumas. 1984. Experience from planting cotton in various cover crops. p. 58–61. In J.T. Touchton and R.E. Stevenson (ed.) Proc. of the 7th Ann. Southeast No-Tillage Systems Conf. Alabama Agric. Exp. Stn., Auburn, AL.
- Hunt, R., and E.T. Parsons. 1981. Plant growth analysis. In Users instructions for the stepwise and spline programs. Unit of Comparative Ecology, Univ. of Sheffield, Sheffield, UK.
- Keeling, W., E. Segarra, and J.R. Abernathy. 1989. Evaluation of conservation tillage cropping systems for cotton in the Texas Southern High Plains. J. Prod. Agric. 2:269–273.
- Klepper, B. 1990. Root growth and water uptake. p. 281–322. In B.A. Stewart and D.R. Nielson (ed.) Irrigation of agricultural crops. ASA, Madison, WI.
- Larcher, W. 1983. Physiological Plant Ecology. 303 pp. Springer-Verlag, Berlin, Germany.
- Mauney, J.R. 1986. Carbohydrate production and partitioning in the canopy. p. 183–188. In J.R. Mauney and J.McD. Stewart (ed.) Cotton physiology. The Cotton Foundation, Memphis, TN.
- Morrison, J.E., T.J. Gerik, and F.W. Chichester. 1985. No-tillage systems for high clay soils. p. 1055–1069. In Traction and Transport as Related to Cropping Systems. Proc. of the International Conference on Soil Dynamics. National Soil Dynamics Lab., Auburn, AL.
- Muramoto, H., J. Hesketh, and M. El-Sharkaway. 1965. Relationships among rate of leaf area development, photo synthetic rate, and dry matter production among American cultivated cottons and other species. Crop Sci. 5:163–166.[Free Full Text]
- Potter, J.R., and J.W. Jones. 1977. Leaf area partitioning as an important factor in growth. Plant Physiol. 59:10–14.[Abstract/Free Full Text]
- SAS Institute, Inc. 1985. SAS user's guide: Statistics. SAS Institute, Cary, NC.
- Stephens, L.E., and R.R. Johnson. 1993. Soil strength in the seed zone of several planting systems. Soil Sci. Soc. Am. J. 57:481–489.[Abstract/Free Full Text]
- Stevens, W.E., J.R. Johnson, J.J. Varco, and J. Parkman. 1992. Tillage and winter cover crop management effects on fruiting and yield of cotton. J. Prod. Agric. 5:570–575.
- Taylor, H.M., and H.R. Gardner. 1963. Penetration of cotton seedling taproots as influenced by bulk density, moisture content, and strength of soil. Soil Sci. 96:153–156.
- Touchton, J.T., and D.W. Reeves. 1988. A beltwide look at conservation tillage for cotton. p. 36–41. In J. Brown (ed.) Proc. Beltwide Cotton Prod. Conf., New Orleans, LA. 3–7 Jan. 1988. National Cotton Council of America, Memphis, TN.
- Triplett, G.B., Jr., S.M. Dabney, and J.H. Siefker. 1996. Tillage systems for cotton in silty upland soils. Agron. J. 88:507–512.[Abstract/Free Full Text]
- Watson, D.J. 1952. The physiological basis of variation in yield. Adv. Agron. 4:101–145.
- Whisler, F.D., B. Acock, D.N. Baker, R.E. Hye, H.F. Hodges, J.R. Lambert, H.E. Lemmon, J.E. McKinion, and V.R. Reddy. 1986. Crop simulation models in agronomic systems. Adv. Agron. 40:141–208.
- Wells, R. and W.R. Meredith, Jr. 1986. Heterosis in upland cotton: I. Growth and leaf area partitioning. Crop. Sci. 26:1119–1123.[Abstract/Free Full Text]
- Young, I.M., K. Montagu, J. Conroy, and A.G. Bengough. 1997. Mechanical impedance of root growth directly reduces leaf elongation rates of cereals. New Phytol. 135:613–619.
This article has been cited by other articles:
|
|
|
|
 
J. J. Marois, D. L. Wright, P. J. Wiatrak, and M. A. Vargas
Effect of Row Width and Nitrogen on Cotton Morphology and Canopy Microclimate
Crop Sci.,
May 1, 2004;
44(3):
870 - 877.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
F. B. Fritschi, B. A. Roberts, R. L. Travis, D. W. Rains, and R. B. Hutmacher
Response of Irrigated Acala and Pima Cotton to Nitrogen Fertilization: Growth, Dry Matter Partitioning, and Yield
Agron. J.,
January 1, 2003;
95(1):
133 - 146.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
