Winter Annual Cover Crops in a Virginia No-till Cotton Production System: II. Cover Crop and Tillage Effects on Soil Moisture, Cotton Yield, and Cotton Quality

Winter annual cover crops could help control soil erosion problems on sloping Piedmont soils in a no-till cotton (Gossypium hirsutum L.) production system. Field experiments were conducted from 1995 to 1997 to monitor the effects of winter annual cover crops in a no-till cotton production system on soil moisture, cotton yield, and cotton quality using the variety Deltapine 50. Cover crops, crimson clover (Trifolium incarnatum L.), hairy vetch (Vicia vilosa L.), hairy vetch and rye (Secale cereale L.), rye, wheat (Triticum aestivum L. em. Thell.), and white lupin (Lupinus albus L.), and two tillage systems (conventional and no-till) were arranged in a split-block design with four replications. Volumetric soil moisture was measured at 15, 30, 61, and 92 cm depths every 7 to 10 days during the 1996 and 1997 cotton growing seasons. Cotton was hand picked, weighed, and ginned for lint yield determination. Sub-samples of the ginned cotton from each plot were analyzed for quality (length, uniformity, strength, and micronaire). Soil moisture results indicated that no-till plots had higher soil moisture compared with conventional tillage during periods of drought in 1997. The no-till rye treatment conserved more soil moisture than any other cover crop treatment from pinhead square through the first three weeks of cotton flowering at the 15 cm depth. Cotton yield and quality were not affected by tillage system. However, the hairy vetch + rye cover crop treatment had higher cotton lint yields during 1995, compared with the wheat cover crop treatment, probably due to N immobilization by the wheat residue. Although differences occurred between cover crop treatments for the different quality parameters during 1995 and 1996, the market value of lint was only affected by micronaire in the 1995 growing season. High micronaire measurement for cover crop treatments in 1995 resulted from unseasonable heat unit accumulation in October and over maturity of the cotton fiber. Using winter annual cover crops in a no-till cotton production system provides greater soil moisture conservation during periods of drought, while producing lint fiber of similar yield and quality compared with a conventional tillage system. A cotton is taprooted and considered drought tolerant, an adequate supply of soil water is still critical for producing high yielding, good quality cotton. Stored soil moisture often limits yield of dryland cotton (Azevedo et al., 1996). Drought stress on cotton has been linked to yield decreases due to reduced size and production of sympodial leaves, in addition to reduced photosynthetic rates. Drought stress limits the amount of photosynthetic assimilate available for plant growth (Krieg, 1997). By using a no-till system, surface residue can be managed to better conserve soil water for greater use efficiency by the cotton plant (Hill and Blevins, 1973). Water withdrawal is different under a no-till system, compared with a conventional tillage system (Blevins et al., 1971). Naderman (1991) reported that surface residue potentially increases infiltration of water into the soil by 25 to 50% under no-till compared with a conventional tillage system. In addition, cover crop surface residue decreases the effect of wind and temperature on soil water evaporation and increases water storage in the soil profile (Brun et al., 1986; Smart and Bradford, 1996). Unger and Parker (1976) concluded that wheat straw was four times better than cotton residue for decreasing evaporation from the soil. The different physical characteristics of the wheat straw including specific gravity, thickness, and surface coverage were given as the reasons for the differences in the water conserving ability of the residues (Unger and Parker, 1976). Burnett and Fisher (1954) reported that moisture is needed in the top 30 cm of soil for crop establishment, but cotton yields are more directly correlated with moisture stored between 30 and 90 cm below the soil surface. Water availability between pin head square and first flower (FF) influences the maximum boll load capacity of the cotton crop (Lawlor et al., 1992). With conventional tillage, the soil surface is unprotected and vulnerable to moisture evaporation from the beginning of the growing season until the cotton canopy closes the rows. By using cover crops to maximize ground cover, the ratio of soil water evaporation to crop transpiration decreases. With increased soil moisture reserve, cotton may endure short-term, low-rainfall conditions without detrimental effects (Blevins et al., 1971). Previous 86 JOURNAL OF COTTON SCIENCE, Volume 3, Issue 3, 1999 research concluded that use of cover crops in a reduced tillage production system increased available soil water and led to higher lint yields compared with conventional tillage systems (Lawlor et al., 1992). Regardless of moisture conservation other researchers have continually found that the combined effect of cover crops and conservation tillage maintains or increases cotton yields compared with conventional tillage (Bloodworth and Johnson, 1995; Boquet et al., 1994). In addition, Bauer and Busscher (1996) found that cotton lint quality was not affected by tillage system or winter cover, but a 0.1 decrease in micronaire was observed in cotton following rye compared with legumes. The objectives of this study were to determine the effects of cover crops and tillage systems on soil moisture, cotton yield, and lint quality under central Virginia piedmont soil and climate conditions. MATERIALS AND METHODS A field study was conducted during the 1995, 1996, and 1997 growing seasons at the Southern Piedmont Agricultural Research and Extension Center, in Blackstone, VA. We had emergence problems in 1997, and an adequate cotton stand was obtained only after the third planting. This delayed maturity to the point where the cotton bolls failed to open, reducing yield. Thus, data for yield and quality will be reported only for 1995 and 1996, while soil moisture data will be reported for the 1996 and 1997 growing seasons. The soil type at the site was a Mayodan sandy loam (fine, mixed, semiactive, thermic Typic Hapludults) for 1995 and 1996, and a Dothan–Norfolk, sandy loam (fine-loamy, kaolinitic, thermic Plinthic and Typic Kandiudults) for 1997. The experiment design used was a split block with four replications. Cover crops were randomly assigned to strips within each block. Tillage practices (conventional and no-till) were randomly assigned to strips perpendicular to cover crop strips. Plots were 4.27 m wide and 7.63 m long with 4 rows and 1.1 m between the rows. The cover crop treatments were crimson clover, hairy vetch, hairy vetch and rye, rye, wheat, and white lupin. About 3 wk prior to the estimated cotton planting date the conventional tillage plots were mowed and disked while the no-till plots were desiccated with 2.24 kg ha-1 glyphosate. The no-till plots received an additional burndown herbicide application when it was needed (Daniel et al., 1999). The cotton cultivar Deltapine 50 was planted 1 wk after the second burndown application, at the rate of 16.4 seeds m-1 of row. Cotton was planted using a 4-row no-till planter equipped with fluted coulters to cut through surface residue followed by double disk openers to make a furrow for the seed, and press wheels to firmly cover the seed. At planting, aldicarb (granular insecticide) and metalaxyl (granular fungicide) were applied infurrow at 5.6 and 11.2 kg ha-1, respectively. Fertilizer N, P, K, and B according to soil test recommendations was broadcast on no-till plots and disked into conventional tillage plots. Standard production management practices were conducted throughout the cotton growing season each year. Volumetric soil moisture was measured in each plot of the first two experimental replications in 1996 and 1997, to monitor differences in soil moisture under the two tillage systems and the cover crop treatments. The measurements were taken with the Troxler Sentry 200-AP soil moisture probe, operated from a permanent access tube (Troxler Electronic Laboratories Inc., 1991). Access tubes were constructed from Schedule 40 PVC pipe that had an inside diameter of 5.22 cm and an outside diameter of 6.03 cm. Access tubes were cut 1.22 m long, and sealed on the bottom with an inside-diameter plastic plug and PVC glue. The access tubes were inserted tightly into a newly augured hole, with 15 cm of the tube extending from the soil surface. Once all the access tubes had been installed, volumetric soil moisture measurements were taken at the 15, 30, 61, and 92 cm depths every 7 to 10 d from the day of cotton planting until the middle of the cotton flowering period. For reliable soil moisture readings the Sentry 200-AP moisture probe was calibrated to the specific soils in the experiment area using a procedure was based the technique of Khosla and Persaud (1997). The probe was calibrated to within 2.56% accuracy for 1996 and 0.37 % accuracy for 1997. Reproducibility of the instrument was high based on repeated measurements at the same depth that showed little or no drift in the measurement. Data will be presented for all the measurements and three moisture measurement dates chosen to match critical periods in the growth stage of the 87 DANIEL ET AL. :WINTER COVER CROPS IN NO-TILL COTTON: II. SOIL MOISTURE Fig. 1. Monthly rainfall totals for May through September during the 1996 and 1997 growing seasons, and the 30yr average. Fig. 2. Soil moisture percent under conventional tillage and no-tillage at the 15, 30, and 61 cm depth 2 d before planting (27 May) during the 1997 growing season. Means for bars within depths followed by the same letter are not significantly different P = 0.05 (Duncan’s multiple range test). cotton plant when moisture availability could affect stand establishment, fruit development, maturity, and overall cotton yield. The dates analyzed in 1996 were 20 May, 1 July, 5 August. In 1997 those