Crop Science
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 22 October 2009
Published in Crop Sci 49:2233-2245 (2009)
© 2009 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
Open Access Article
This Article
Free via Open Access
Right arrow OA Abstract
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Google Scholar
Right arrow Articles by Katsvairo, T. W.
Right arrow Articles by Wiatrak, P. P.
PubMed
Right arrow Articles by Katsvairo, T. W.
Right arrow Articles by Wiatrak, P. P.
Agricola
Right arrow Articles by Katsvairo, T. W.
Right arrow Articles by Wiatrak, P. P.
Related Collections
Right arrow Crop Rotation Systems
Right arrow Crop Ecology

Comparative Plant Growth and Development in two Cotton Rotations under Irrigated and Non-Irrigated Conditions

Tawainga W. Katsvairo*, David L. Wright, James J. Marois, Jimmy R. Rich and Pawel P. Wiatrak

Univ. of Florida, NFREC, 155 Research Rd., Quincy, FL 32351. Mention of a trademark, proprietary product, or vendor does not constitute a guarantee of warranty for the product, and does not imply its approval to the exclusion of other products or vendors that may be suitable

* Corresponding author (katsvair{at}ufl.edu).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 
Incorporating perennial grasses such as bahiagrass (Paspalum notatum Fluegge) to diversify the conventional two-crop rotation of peanut (Arachis hypogaea L.) and cotton (Gossypium hirsutum L.) prevalent in the U.S. Southeast (SE) is advocated. However, little is reported on growth and development for cotton grown in rotation with perennial grasses. Our objectives were to compare plant characteristics including height, leaf area index (LAI), chlorophyll meter readings (chlorophyll index), N uptake, weed densities, and residual soil nutrients in a conventional rotation of peanut-cotton-cotton vs. a bahiagrass-bahiagrass-peanut-cotton rotation. Field studies were conducted in Quincy, FL from 2000 to 2006. Plant height, LAI and N, P, and K uptake were generally greater for cotton in the bahiagrass rotation compared to the conventional cotton/peanut rotation. Weed densities were reduced for cotton in the bahiagrass rotation. Residual nutrients at the end of the season including N, P, and K and soil organic matter (SOM) showed no differences between the rotations. In spite of the improvements in plant growth characteristics, rotating cotton with bahiagrass overall did not improve yield above the conventional rotation. Potential exists for greater cotton yield in the bahiagrass rotation once effective management practices have been identified.

Abbreviations: CT, conservation tillage • DAP, days after planting • LAI, leaf area index • SE, U.S. Southeast • SOM, soil organic matter


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 
PROPOSALS HAVE BEEN made to integrate perennial grasses such as bahiagrass and Bermuda grass into the conventional two-crop rotation of peanut and cotton prevalent in the U.S. Southeast (SE). While the numerous advantages of diverse cropping systems are described across the nation (Katsvairo et al., 2006a; Sulc and Tracy, 2007; Russelle and Franzluebbers, 2007; Russelle et al., 2007; Allen et al., 2007; Franzluebbers, 2007), very few articles report on plant growth characteristics when cotton is grown in rotation with perennial grasses.

Cotton physiological responses to micro-environment changes can be complex to manage, furthermore, diverse cropping systems come with new challenges. Because soil physiochemical changes such as moisture regimes and soil nutrient status can affect cotton growth, development, and subsequently yield, the introduction of perennial grasses in cotton cropping systems can be expected to pose challenges in cotton crop management. Excess soil N can result in cotton developing more vegetative growth which can delay maturity and reduce overall yield (Howard et al., 2001; Gaylor et al., 1983). The perennial grass rotation alters soil fertility conditions by recycling nutrients from deep soil profiles and making them available for crop uptake. Furthermore, the perennial grass rotation can increase soil organic matter. Soil organic matter accumulation results in a slow release of soil nutrients which can provide nutrients above levels normally found in conventional systems. Among the earliest research on perennial grass-based cotton rotations, Elkins et al. (1977) reported greater nutrient cycling for cotton in a wheat grass sod rotation. Nutrient recycling, however, resulted in excessive nutrient availability which resulted in excess growth that reduced cotton yield.

A zone of soil compaction spans most of the soils at the 15 to 30 cm soil depth in the SE (Kashirad et al., 1967; Campbell et al., 1974). Bahiagrass develops a deep and extensive root system that can penetrate through this zone of compaction (Elkins et al., 1977). Cotton following bahiagrass reuses the root channels as paths of least resistance and develops a more extensive root system (Elkins et al., 1977; Katsvairo et al., 2007a). Improved soil conditions resulting in larger rooting systems would be expected to result in more aboveground biomass.

Irrigation water has become less available, and the cost of irrigation continues to rise. Biomass production for renewable energy is likely to expand into marginal lands that will require irrigation. This will call for even more competition for irrigation water. Profitability of cotton has generally been marginal and new crop rotations that reduce irrigation will be beneficial to farmers. We hypothesize that improved soil conditions after bahiagrass (Katsvairo et al., 2007a) will reduce the need for irrigation.

Although mostly produced under conservation tillage (CT) (National Crop Residue Management Survey, 2002), cotton in the SE is mainly grown on soils prone to erosion, nutrient leaching, and general loss of pesticides to the environment. Cropping systems which improve nutrient uptake would be expected to reduce nutrient losses to ground water. No article, to the best of our knowledge, reports on the status of residual nutrients at the end of the growing season for cotton in the sod rotation compared to cotton in the conventional rotation. There is usually a four-week window between cotton harvest in late fall and the establishment of a rye or wheat cover crop which protects the soil against nutrient leaching and soil erosion.

The effect of perennial grasses on pest dynamics, especially on plant-parasitic nematodes and fungal pathogens, is well documented and one of the strong points of the rotation system (Kabana et al., 1988; Brenneman et al., 2003; Katsvairo et al., 2006b); however, little has been reported on weed control in perennial grass–cotton based rotations. While diverse cropping systems can support a wider spectrum of weed species (Derksen et al., 1995; Froud-Williams, 1988), diverse speciation can reduce the predominance of any one weed type (Liebman and Dyck, 1993; Anderson, 1998; Derksen et al., 1995). Equally important, rotation systems space crops temporally and allow growers to rotate herbicides with different modes of action, and thus delay and possibly prevent the onset of herbicide-resistance in weeds (York and Culpepper, 2004).

Plant growth characteristics including N, P, K uptake, leaf area index (LAI), height, node to height ratio, and weed management affect yield and can be influenced by crop rotation. Also, nutrient recycling can reduce losses to ground water. Our objectives were to (i) measure plant growth parameters including height, LAI, and N uptake under a perennial grass-based rotation vs. the conventional peanut/cotton rotation and under irrigated vs. non-irrigated conditions; (ii) determine residual soil nutrients in a perennial grass-based cotton rotation vs. the conventional peanut/cotton rotation; and (iii) determine weed dynamics in perennial grass-based cropping systems.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 
A six-year irrigation x rotation study was initiated in 2000 at the University of Florida's North Florida Research and Education Center in Quincy, FL (84°33' W, 30°36' N). The 1.75 ha experimental site was planted to cotton in the summer of 1999 and was fallow during the following fall and winter. Before 1999, the field had been in a CT/winter cover cropping sequence for several years. The soil type at the site is a Dothan sandy loam (fine, loamy siliceous, thermic Plinthic Kandiudults). Soil tests in the fall of 1999 indicated a pH of 5.4, 36 mg kg–1 P, 117 mg kg–1 K, and 116 mg kg–1 Ca, based on the Mehlich-1 extraction procedure (Mehlich, 1978).

Treatments were arranged in a strip plot (also called split-block) experimental design with three replications (Little and Hills, 1978). The use of a lateral move irrigation unit imposed the necessity for the strip-plot treatment layout. Three strips, each 128 m long x 45.7 m wide, were utilized and each of the strips consisted of alternating irrigated and non-irrigated treatments (Fig. 1 ). The irrigation unit stayed in the same area all six years of the study. Irrigation was applied when tensiometer (Irrometer Co., Riverside, CA) readings at the 30-cm soil depth indicated 40 kPa in irrigated plots. Crop rotation subplots were 45.7 m long x 18.3 m wide (20 rows wide) and subplots were aligned perpendicular to the irrigated and non-irrigated strips. Crop rotations were a peanut cotton-cotton-peanut rotation, which is the conventional rotation used by growers in the region, and a bahia-bahia-peanut-cotton rotation. All phases of the rotations were present in all years (Table 1 ). In the conventional rotation, the first and second year of cotton were also compared in 2004, 2005, and 2006. Bahiagrass sod was grown for two years before planting peanut to ensure good ground coverage and vigorous growth of the bahiagrass.


Figure 1
View larger version (13K):
[in this window]
[in a new window]

  		Figure 1. Treatment layout for two irrigation systems and three rotations in FL. {dagger}Irrigation constitutes the main strip. {ddagger}Rotation constitutes the sub-strips.

 

View this table:
[in this window]
[in a new window]

  		Table 1. Crop sequence for bahiagrass (bahia) peanut, and cotton, 2000–2006 in FL.

 
Crop Management
In April of each year, two to three weeks before cotton planting, plot rows were strip-tilled using a Brown Ro-till implement (Brown Manufacturing Co., Ozark, AL). In 2000, cotton cultivar Paymaster PM 1500 BG/RR was used, while in 2001 to 2006, cultivar Deltapine DP 458 BR was planted. All plantings were made from late April to early May with a Monosem pneumatic planter (ATI Inc., Lenexa, KS) at 144,000 seeds ha–1. Starter fertilizer (5–10–15) at a rate of 560 kg ha–1 was band applied adjacent to each row at planting. Cotton was side-dressed with 200 kg ha–1 34–0–0 at square initiation from 2000 through to 2004. In 2005 the rotation plots were split and two N rates of 67 kg ha–1 and 33 kg ha–1 were used. Because no yield differences were found between the two N rates in 2005, we used rates of 67 N kg ha–1 and 0 N kg ha–1 in 2006. Early season cotton plant densities were determined when cotton was 10 cm high by counting the number of plants in an 18.3 m length of the two harvest rows in each plot. Bahiagrass management and other cotton crop cultural management including insect control and herbicide use were done according to standard University of Florida recommendations (Ferrell et al., 2006) and also reported in an earlier publication (Katsvairo et al., 2007b).

Plant height was determined throughout the season by estimating the height of 10 representative plants in each plot. After reducing N rate to 0 kg ha–1, plant height was monitored more frequently in 2006 than other years. Chlorophyll index was measured on the uppermost fully expanded leaf of thirty plants per plot with Minolta's SPAD meter (Spectrum Technologies, Plainfield, IL). Weed counts and identification were determined by counting all weeds 5 cm and above in the entire net plot area. Weed counts were conducted only in 2003 and 2004, the years with high weed pressure.

At cotton maturity, whole plants in a 1-m2 net area for the rotations and N rates were harvested for NO3–N determination. The plants were dried at 60°C in a forced-air oven to constant moisture, and dry weight was determined. The samples were then ground in a Wiley Mill (Thomas Scientific, Swedesboro, NJ), and plant N concentrations were determined by Kjeldahl procedures. Total N uptake was estimated as the product of dry weight x whole plant N concentration. N use efficiency was calculated as the lint yield divided by the N rate.

The plots were harvested in October or November of each year. A 1-kg subsample of seed cotton from each plot was ginned to determine percentage lint content and seed yield. Harvest index was calculated as the lint yield divided by the total aboveground biomass.

Immediately after harvest, soil samples were taken at the 30-cm depth in each of the rotation plots. A total of 10 soil cores were taken for each rotation and in each plot. The samples were submitted to a laboratory for soil nutrient analysis. The Mehlich-1 extraction procedure was used to determine the soil nutrient concentrations (Mehlich, 1978).

Statistical Analysis
All data were analyzed using SAS general linear models procedures (SAS Institute, 2002). In this paper only cotton data are reported. Irrigation main effects are confounded with block effects and as a result, they cannot be discussed from a statistical significance standpoint, but can be discussed using means and standard errors. However irrigation x rotation interactions are appropriate for discussion. A detailed discussion on irrigation is provided in a companion paper. Although the study started in 2000, the rotations made the first complete cycle in 2003, so only data from 2003 to 2006 are presented. The year-to-year variances for all measured parameters in this experiment were not homogenous, therefore separate analysis for each year are presented. Mean separation for main effects and interactions were obtained by Fisher's protected LSD, as described by Little and Hills (1978). Effects were considered significant in all statistical calculations if P ≤ 0.05.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSION
 REFERENCES
 
Precipitation
Precipitation differed among the four growing seasons. The years 2003 and 2006 were generally drier, with 2006 having below 30 yr precipitation average for the site (Table 2 ). The years 2004 and 2005 were generally wetter; however, rainfall was not always uniformly distributed within each season. Total irrigation amounts of 110, 125, 207, 240 mm were applied to the cotton and bahiagrass in the irrigated treatment in 2003, 2004, 2005, and 2006, respectively. The month of July 2004 was dry and had rainfall below the 30-yr average. July and/or August are important months in cotton production in the SE because that is when cotton blooms and fruit set occurs. The year 2004 was also marked by three hurricanes, which were mostly felt as strong winds in Quincy, FL, resulting in lodged plants.


View this table:
[in this window]
[in a new window]

  		Table 2. Monthly precipitation and irrigation in FL, 2003–2006 growing seasons.

 
Plant Height
Cotton showed irrigation x rotation interactions (Table 3 ). At the beginning of the season, cotton in both rotations were of similar heights under the two rotation systems; as the season progressed cotton in the bahiagrass rotation attained plateau heights quicker than cotton in the conventional rotation. Toward the end of the season, plant height was similar for both rotations


View this table:
[in this window]
[in a new window]

  		Table 3. Significant levels for cotton height in FL in 2006.

 
Cotton plant height varied between years. The plants were tallest in 2003 where they reached a peak height of 1.44 m compared to a peak of less than 1.25 m for the other years (Table 3). Greater rainfall and late application of the herbicide Pix (Tenkoz Inc., Alpharetta, GA) could have resulted in the much taller plants in 2003. A different study from the same site reported similar results of plant height varying between years (Wiatrak et al., 2005). Plant height was generally greater in the perennial grass rotation compared to the conventional rotation in 2003 and 2006 (Fig. 2 ). There were no differences in plant height between the rotations in 2004 and 2005. Results from the same study showed greater root biomass, root length, and total root area for cotton in the perennial grass rotation compared to the conventional system (Katsvairo et al., 2007b). Similarly, Elkins et al. (1977) reported greater root length and total root area in cotton after bahiagrass. Improved soil health conditions including reduced soil mechanical resistance, greater organic matter, increased earthworm activity, and greater infiltration rates have also been reported for cotton in rotation with perennial grasses (Franzluebbers and Triplet, 2006; Katsvairo et al., 2007b). A combination of improved soil conditions coupled with greater root development could have contributed to increased plant height. In most cotton growing regions, moisture is a major limiting factor and the benefits of the sod rotation on plant height were more pronounced in the drier years 2003 and 2006. The year 2006 was the driest of the four years of the study period. Although irrigation effects cannot be discussed from a significance standpoint because they are confounded with blocks, in 2006 cotton under irrigation was mostly taller than cotton under non-irrigated conditions (Fig. 3 ).


Figure 2
View larger version (28K):
[in this window]
[in a new window]

  		Figure 2. Cotton height for three rotations in FL. Rotations are as follows: B-B-P-C, bahiagrass-bahiagrass-peanut-cotton; P-C-C, peanut-cotton-cotton; P-C-C, peanut-cotton-cotton. Underline indicates the phase of the crop in rotation. For each day after planting, treatment means followed by the same letter are not significantly different at P < 0.05.

 

Figure 3
View larger version (22K):
[in this window]
[in a new window]

  		Figure 3. Cotton height under irrigated and non-irrigated conditions in 2006 in FL. Statistical analysis was not done for irrigation because irrigation effects were confounded with block.

 
In 2005, a decision was made to compare two N rates (67 kg ha–1 and 33 kg ha–1) because no yield differences had been observed between the rotations over the previous two years. There were no differences in plant height between the two N rates in 2005 (Fig. 4 ). It is possible that the residual soil N was still high from the previous fertilizer applications. Soils at this site generally contain more clay and silt than most soils of Florida. The finer textured soils hold N better than sandy soils. Fritschi et al. (2003) alludes to the problem of greater residual soil nutrients obscuring treatment effects in croplands. It's also possible that the overall improved conditions in soils after the perennial grass would have enabled cotton to grow more than it would normally.


Figure 4
View larger version (26K):
[in this window]
[in a new window]

  		Figure 4. Cotton height for two N rates in FL. For each day after planting, treatment means followed by the same letter are not significantly different at P < 0.05.

 
In 2006 the two rates of side-dress N were changed to 67 kg ha–1 and 0 kg ha–1. Plant height showed rotation x irrigation, and rotation x N rate, and rotation x irrigation x N rate interactions throughout the growing season (Table 3). Plant height was similar between the N rates until 102 DAP at which point cotton plants were consistently shorter under 0 N compared to the 67 kg ha–1 N (Fig. 4). In 2006 cotton was also shortest in the plots which went from 33 to 0 kg ha–1 N compared to the plots which went from 67 kg ha–1 to 0 kg ha–1 (data not shown). Two years of reduced N may have exhausted the residual soil N pool to reduce overall plant growth. N rates of first-year cotton in the conventional rotation and cotton in the bahiagrass rotation would have changed from 67 kg ha–1 to 0 kg ha–1. While N rates of second-year cotton would have changed from 33 kg ha–1 to 0 kg ha–1. When these two N rates were analyzed separately, cotton was shortest in the plots which changed from 33 to 0 kg ha–1 N compared to the plots which changed from 67 kg ha–1 to 0 kg ha–1.

Nodes and Node/Height Ratio
A tendency for greater numbers of nodes for cotton in the bahiagrass rotation compared to the conventional rotation was observed (Fig. 5 ). Greater number of nodes is indicative of greater potential for fruiting bodies. Similarly, there was also a tendency for greater node-to-height ratio for cotton in the bahiagrass rotation compared to the conventional rotation (Fig. 6 ). However, not all the fruiting bodies matured into viable cotton bolls as will be seen in the yield section (data not shown). The excessive cotton bolls in the bahiagrass rotation likely aborted as the plants would have been unable to support (nutrition-wise) large numbers of bolls. There were more nodes and greater node-to-height ratio under irrigated compared to non-irrigated conditions. Both nodes and node-to-height ratio showed interactions with irrigation, rotations, and N rate in 2006 (Table 4 ).


Figure 5
View larger version (23K):
[in this window]
[in a new window]

  		Figure 5. Cotton average node counts for three rotations in Fl. For each day after planting, treatment means followed by the same letter are not significantly different at P < 0.05.

 

Figure 6
View larger version (27K):
[in this window]
[in a new window]

  		Figure 6. Plant height/node ratio for three cotton rotations in FL. Rotations are as follows: B-B-P-C, bahiagrass-bahiagrass-peanut-cotton; P-C-C, peanut-cotton-cotton; P-C-C, peanut-cotton-cotton. Underline indicates the phase of the crop in rotation. For each day after planting, treatment means followed by the same letter are not significantly different at P < 0.05.

 

View this table:
[in this window]
[in a new window]

  		Table 4. Significant levels for cotton nodes under different rotations and rates in 2006 in FL.

 
Leaf Area Index
The greatest LAI was observed in 2004 compared to the other years (Fig. 7 ). At its peak, LAI was above 7 in 2004 compared to peak LAI values less than 3.5 in 2006. The greater rainfall in 2004 would be expected to result in dense cotton canopies. LAI was generally greater for either cotton in the bahiagrass rotation and/or first-year cotton in the conventional rotation compared to second-year cotton in the conventional rotation in 2003, 2004, and 2006 (Fig. 7). LAI showed no rotation effect in 2005. The high LAI under the bahiagrass and first-year cotton could have been a result of improved soil properties in that rotation which enabled the cotton to utilize resources such as light, moisture, and nutrients more efficiently. LAI showed an irrigation x rotation interaction 88 DAP in 2006 while bahiagrass under non-irrigated conditions showed greater LAI than second-year cotton in the conventional rotation. The lower LAI values toward the end of each season are an indication of senescence as the season progresses. Visual observation showed that cotton in the conventional rotations, especially second-year cotton reached physiological maturity approximately two weeks earlier than cotton in the bahiagrass rotation. Cotton maturity (cut out) is determined by nodes above white flower and time to defoliate is determined when 60% of the bolls are open. However LAI values were not measured for much longer after physiological maturity. LAI are an essential indicator to the interaction between crop growth and environment.


Figure 7
View larger version (20K):
[in this window]
[in a new window]

  		Figure 7. Leaf area index for cotton rotations in FL. Rotations are as follows: B-B-P-C, bahiagrass-bahiagrass-peanut-cotton; P-C-C, peanut-cotton-cotton; P-C-C, peanut-cotton-cotton. Underline indicates the phase of the crop in rotation. For each day after planting, treatment means followed by the same letter are not significantly different at P < 0.05.

 
Weeds
Weeds were generally well-managed in the rotations, and weed counts were only done in years with moderate weed pressure including 2003, 2005, and 2006. Weed populations and biomass were generally lower for cotton in the bahiagrass rotation compared to the conventional rotations which were also the rotations with the high LAI (Table 5 ). Cotton in the bahiagrass rotation grew more vigorously and quickly developed a canopy. The more developed plant canopy was able to effectively shade the weeds rendering them less competitive to the cotton. Weeds reduce plant growth and yield by competing with crops for essential resources such as moisture, nutrients, and light. Weeds also make harvesting more difficult and can stain the cotton lint, thus reducing both yield and quality. The reduced weed pressure in the bahiagrass-rotated cotton may mean less herbicide application, less herbicide cost for the growers and a lower potential for pesticide contamination of the environment. Growing cotton in rotation with bahiagrass would be a cost-effective way to control weeds. We did not observe any evidence for shift in weed spectrum species. This may have been due to the limited time-span of the experiment and also the plot sizes are small and could allow easy dispersion of weeds. The major weeds were morning glory (Ipomoea spp.) and quack grass (Elymus spp.).


View this table:
[in this window]
[in a new window]

  		Table 5. Weed densities for three rotations in FL.

 
Chlorophyll
There were generally no differences in chlorophyll index between the rotations in 2005. This probably indicates sufficient levels of N in the two rotations. In 2006, chlorophyll index was generally greater for either cotton in the perennial grass rotation or first-year cotton in the conventional rotation compared to second-year cotton (Fig. 8 ). Greater soil N from the decaying residues in the bahiagrass rotation and residual N from the peanuts in the first-year cotton all could have resulted in improved plant growth and nutrient uptake in the two rotations and this could have resulted in the greater chlorophyll levels. Chlorophyll index levels ranged from 39.3 to 53.7 across the rotations. High chlorophyll index level in plants is an indication of good N availability (Everitt et al., 1985; Longstreth and Nobel, 1980) Although not discussed from a perspective of statistically significant, chlorophyll index was generally greater under irrigation compared to non-irrigated conditions in 2006 (Data not shown). Chlorophyll index was not determined for the different N rates in 2005. However, chlorophyll index was consistently greater for the 67 kg ha–1 N rate compared to the 0 N rate after side-dressing in 2006 (Fig. 9 ). This is consistent with Bronson et al. (2001) who reported correlation between N availability and total chlorophyll index concentration in cotton leaves. There was no irrigation x rotation interaction for chlorophyll index.


Figure 8
View larger version (34K):
[in this window]
[in a new window]

  		Figure 8. Cotton chlorophyll levels for three rotations in FL. Rotations are as follows: B-B-P-C, bahiagrass-bahiagrass-peanut-cotton; P-C-C, peanut-cotton-cotton; P-C-C, peanut-cotton-cotton. Underline indicates the phase of the crop in rotation. For each day after planting, treatment means followed by the same letter are not significantly different at P < 0.05.

 

Figure 9
View larger version (28K):
[in this window]
[in a new window]

  		Figure 9. Cotton chlorophyll levels at 0 kg vs. 67 kg ha–1 N in FL in 2006. For each day after planting, treatment means followed by the same letter are not significantly different at P < 0.05.

 
Biomass and N uptake
Total biomass varied among the years and was on average greater in 2003 and 2004 compared to 2005 and 2006 (Table 6 ). There was an almost threefold difference in biomass between 2003 and 2006. Greater moisture in 2003 and 2004 could have resulted in the bigger plants. Also, growth regulators were applied earlier in the season in 2005 and 2006 compared to 2004 and 2003. This could have reduced the overall growth for cotton. Highest total plant biomass was observed in either the bahiagrass rotation or cotton in first year of the conventional rotation in 2003, 2005, and 2006 and averaged up to 10 Mg ha–1 greater for the bahiagrass rotation compared to the conventional rotation (Table 6). N concentration was generally the same between the rotations in all years. With the exception of 2004, N uptake was greater in either the bahiagrass rotation or cotton in the first year of the conventional peanut/cotton rotation compared to cotton in the second year in the conventional rotation. Cropping systems which increase N uptake efficiency are an asset because they reduce excess N leaching to groundwater.


View this table:
[in this window]
[in a new window]

  		Table 6. Cotton dry weight, N, P, K concentration and uptake at physiological maturity for conventional and sod rotation in FL.

 
Total biomass and P uptake showed an irrigation x N rate interaction. Both biomass and P uptake were greater for the 0 N rate under non-irrigated conditions compared to the 67 kg ha–1 N rate under irrigated conditions. Potassium uptake showed a three-way interaction of irrigation x rotation x rate. There generally were no differences in plant P concentration between the rotations in all years except for 2004 where second-year cotton in the conventional system had greater concentration than the other two rotations (1.4 vs. 1.6 g kg–1) (Table 6). However, P uptake was generally greater for cotton in the bahiagrass rotation compared to first-year cotton in the conventional rotation over the four years. Plant K concentrations were more erratically distributed across the rotations (Table 6). It was greater for cotton in either the bahiagrass rotation or cotton in the second year of the conventional rotation in two of the four years compared to cotton in the conventional rotation. However, K uptake was greater overall for cotton in the bahiagrass rotation. As was the case with plant height and LAI, the improved soil health conditions and the resultant larger rooting system (Katsvairo et al., 2007a) in the sod rotation could have resulted in the greater plant biomass and nutrient uptake for the sod rotation.

There was no difference for all measured parameters (N,P, K concentration and uptake, and plant biomass) between the 67 kg ha–1 vs. 0 kg ha–1 N side-dress rates in 2005 (Table 7 ). However in 2006, differences in favor of the higher N rate were observed for total biomass (9500 vs. 7891 Kg ha–1), N concentration (16.1 vs. 12.7 g kg–1), N uptake (154 vs. 101 kg ha–1), P concentration (2.5 vs. 2.0 g kg–1), and K uptake (260 vs. 206 kg ha–1) for the 67 kg ha–1 N side-dress rate compared to the 0 kg ha–1 side-dress rate (Table 7).


View this table:
[in this window]
[in a new window]

  		Table 7. Cotton dry weight, N, P, and K concentration, and uptake at physiological maturity for 2 N rates in FL.

 
Yield
Yield varied across the years with the greatest yield obtained in 2005 and 2006, which were also the years with the lowest plant height, LAI, and biomass, indicating a negative relationship between excessive growth and yield (Table 8 ). The negative relationship between excessive growth and yield can partially explain the lack of rotation response. There was an almost twofold difference in yield between the years 2003 and 2004 vs. 2005 and 2006. A related study which evaluated spatial yield variation in cotton fields in Florida reported lower yields in areas with the tallest cotton plants (Katsvairo et al., 2007c). Averaged over the four years, the perennial grass-based rotation did not significantly improve yield over the conventional cropping system (Table 8). In this article we defined harvest index as the lint yield divided by the total aboveground biomass. Harvest indices were greater in 2005 and 2006 compared to 2003 and 2004 as a result of the higher yields obtained in those years. Lower harvest indexes were observed in cotton in the bahiagrass rotation compared to cotton in the conventional rotation (Table 9 ). The lack of yield differences for cotton in the bahiagrass rotation compared to cotton in the conventional rotation may have been due to several factors. The greater biomass produced in the bahiagrass rotation could have occurred at the expense of fruiting bodies resulting in reduced yield. This may imply using more Pix to control growth in the sod rotation. The standard N application rate recommended for the conventional cropping system was used in the sod rotation; however, the standard N application rate caused excessive vegetative growth for the bahiagrass rotated cotton. Soil N levels at the beginning of the experiment may have been high and masked the effects of the rotations. High levels of soil baseline nutrients in fields masking treatments effects were reported by Fritschi et al. (2003).


View this table:
[in this window]
[in a new window]

  		Table 8. Cotton yield under conventional and bahiagrass rotation in FL.

 

View this table:
[in this window]
[in a new window]

  		Table 9. Harvest indices for cotton yield under conventional and bahiagrass rotation in FL. Harvest indices were calculated as the ratio of lint yield divided by the total aboveground biomass

 
Reducing N rates from 67 to 33 kg ha–1 had no effect on yield in 2005 across the rotations, further suggesting high carryover N in the soil. However, in 2006 the cotton plots, which went from 33 kg ha–1 in 2005 to 0 kg ha–1 in 2006, had reduced yield compared to cotton which went from 67 kg ha–1 N in 2005 to 0 kg ha–1 N in 2006 (1511 vs. 1447 kg ha–1) (Table 10 ). Fertilizer use efficiency was 27 at an N rate of 67 kg ha–1 compared to 57 at an N rate of 33 kg ha–1 in 2005 (Table 10). In 2006, the fertilizer use efficiency was only 25% at an N rate of 67 kg ha–1 and the value approaches infinity with 0 N applied. At the same site, Wiatrak et al. (2005) reported a linear increase in cotton yield with increase in fertilizer rates and no yield response above 67 kg ha–1. It appears that N rates may need to be reduced in some fields and situations in both conventional and perennial grass based cropping systems in the SE.


View this table:
[in this window]
[in a new window]

  		Table 10. Cotton lint yield for two N rates in FL.

 
In general, we observed that cotton in the conventional rotation reached physiological maturity at least two weeks earlier than cotton in the bahiagrass rotation, indicating better soil conditions. It has been almost 30 years since Elkins et al. (1977) reported challenges in obtaining consistently greater cotton yields when grown in rotation with perennial grasses. To date that problem has not been resolved. There is a need to learn to better manage cotton in the perennial grass rotation system. Cotton in the bahiagrass systems at this site showed improved root growth and aboveground growth (Katsvairo et al., 2007b), yet this did not translate into yield improvement.

Residual Soil Nutrients
It was hypothesized that the cotton in the perennial grass rotation could develop a deeper and extensive rooting system, which can take up more nutrients and result in reduced residual nutrients For example, Long and Elkins (1983) reported reduced residual K, Ca, and NO3–N for cotton in bahiagrass rotation compared to continuous cotton. In our study, the effects of crop rotations on soil residual nutrients over the four-year period were surprisingly inconsistent and lead to no definite conclusions (data not shown). Furthermore in both 2005 and 2006, no differences in residual soil N between the N rates which averaged 11 mg kg–1 and 20 mg kg–1 between the two years, respectively, were observed. It is possible that nutrient leaching occurs rapidly at this site, and most nutrients could have leached out of the soil profile by the time soil tests were determined. This further confirms the concern of nutrient leaching between crop harvest and cover crop establishment. It should also be pointed out that soil nutrient levels are particularly difficult to measure in SE soils, especially residual N. It is feasible that the methodology used to measure soil nutrients in this study could have failed to show differences in the residual soil nutrients among the rotations. P and K values averaged 68 and 215 mg kg–1 respectively over the four-year period and did not differ between the rotations.

Soil organic matter averaged 1.56% over the four-year period. And similarly, while differences in SOM were expected between the rotations, none were observed during the study. Numerous possible reasons for this exist: The lack of difference may have been due to the short duration of this study. The top 1 to 2 cm is where the initial build-up of SOM occurs. Soil samples were taken to a depth of 15 cm, and it is plausible that sampling to a depth of 15 cm could have obscured the top layer and diminished the differences between the treatments. The build-up of SOM levels occurs over a long time (Martin, 2003) and longer studies may be required to observe differences in SOM between the rotations. It seems likely that for these soils, CT may have been the dominant cultural practice affecting SOM more than crop rotations.


   CONCLUSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
REFERENCES
 
Cotton grown after bahiagrass showed greater biomass, plant height, and LAI. The greater plant biomass system and higher nutrient uptake in the bahiagrass rotation is indicative of better resource utilization including soil moisture, soil nutrients, and light. The greater aboveground biomass for the cotton in the sod rotation could have been a result of larger and deeper root systems in the cotton in the bahiagrass rotation (Katsvairo et al., 2007b). The vigorously growing cotton in the bahiagrass rotation was in turn able to outcompete weeds. Our inability to obtain yield improvement from bahiagrass rotation in spite of better plant growth characteristics, improvements in soil conditions, and reduced pest problems strongly suggests a need for development of perennial grass-based crop management guidelines which are different from the current conventional cropping system recommendations. Data from this study suggests that more research is needed on cotton fertilizer application rates when grown in rotation with perennial grasses. Pest management strategies may also need to be reviewed since the greater cotton biomass produced in the bahiagrass rotation may reduce pesticide penetration into the canopy.


   ACKNOWLEDGMENTS
 
This study was supported in part by Cotton Incorporated, USDA Special Research Grants, and Northwest Florida Water Management District. Brian Kidd, Iwona Jarczykowska, Cynthia Davis-Holloway, Lauren Wyckoff, Hunter Pittman and Jennifer Morrow helped in conducting the field, laboratory work and in compiling the data.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

Received for publication June 22, 2008.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 





This Article
Free via Open Access
Right arrow OA Abstract
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Google Scholar
Right arrow Articles by Katsvairo, T. W.
Right arrow Articles by Wiatrak, P. P.
PubMed
Right arrow Articles by Katsvairo, T. W.
Right arrow Articles by Wiatrak, P. P.
Agricola
Right arrow Articles by Katsvairo, T. W.
Right arrow Articles by Wiatrak, P. P.
Related Collections
Right arrow Crop Rotation Systems
Right arrow Crop Ecology

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