TDOC Z TA245.7 B873 NO.1585 B-1585 Furrow Diking and SUbSOmIIg Studies in the Rolling Plains THE TEXAS AGRICULTURAL EXPERIMENT STATION/ Nevill P ' . Cl k , D - - _ e ar e lrector/ The Texas A&M Umversny System/College Station, Texas Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Methods and Materials . . . . . . . . . . . . . . . . . . . . . . . 1 Sorghum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Cotton . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . 4 Sorghum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Cotton . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Sorghum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Cotton . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Literature Cited . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 ‘\ Furrow Diking and Subsoiling Studies in the Rolling Plainsl C. J. Gerardz Introduction Research has demonstrated that water is the dominant factor for yields in the Rolling Plains of Texas (5). This area is plagued by periods of drought during the growing season. The soils in the Roll- ing Plains are low in organic matter and have a weak or fragile structure that contributes to significant loss of water and soil fol- lowing even moderate rainfall. Furrow diking or basin tillage is the practice of putting small dams in the furrow to obstruct water flow, reduce runoff, and increase moisture storage for crop production. The idea of furrow diking or basin tillage is not recent. Luebs ( 12) reported on research with furrow dikes at Hays, Kansas, between 1937 and 1953. He stated that holding rain with dikes on nearly level land appeared to have merit for increasing yields, particu- larly for continuous sorghum. In the High Plains of Texas, Bilbro and Hudspeth (1), Hudspeth (10), and Lyle and Dixon (13) used basin tillage or diking to re- duce runoff and increase cotton yields. Similarly, Jones and Clark (1 1) reported that diking conserved water and increased sorghum yields. Gerard et al. (7) found that diking every 5O ft in the furrow increased sorghum yields by 11 percent to 17 percent in 1979 in the Rolling Plains of Texas. Furrow diking increased cotton yields by about 35 percent in the Rolling Plains in 1981, according to Clark (4) and Gerard et al. (6). Diking alternate rows increased cot- ton yields from 10.5 percent to 15 percent, according to Bor- dovsky (2) and Gerard et al. (8). Gerard et al. (8) summarized studies on sorghum and cotton in 1981 and 1982. They noted that furrow dikes seem to be effective in reducing runoff and increasing moisture storage on gently slop- ing land, and that position on the slope has an important effect on yields. Hanna et al. (9) found that in Nebraska, landscape position on the slope can influence the amount of water retained by soils for dryland crop production. To evaluate the effectiveness of dik- ing in capturing rainfall or reducing runoff, according to Gerard et al. (8), it is necessary that response be measured on a substantial part of the slope. For example, in these studies, diking increased sorghum yields an average of 302 percent, 140 percent, and 42 percent more than the check treatments on the upper, middle,and lower parts of the slope, respectively, and diked cotton produced 57 percent, 37 percent, and 12 percent more than the check on the upper, middle, and lower parts of the slope, respectively. This publication summarizes furrow diking and subsoiling studies on sorghum during 1979-85, and cotton during 1980-82 and 1984-85 in the Rolling Plains. Methods and Materials Sorghum The effects of subsoiling and diking on yields of sorghum were evaluated from 1979 to 1985 at the Texas A&M Research Center at Chillicothe-Vemon. Rainfall data during 1979-1985 are re- ported in Table 1. This experiment, conducted on an Abilene loam soil, was a randomized block design with three replications. Mois- ture retention curve for the Abilene soil is shown in Figure 1. Selected properties of this soil are reported in Table 2. All treat- ments described in Table 3 were 12 rows 200 ft long. The locations of all treatments were the same throughout the experiment. The slope of the field down the rows rangedfrom 0.1 percent on the lower half to 0.4 percent in the upper half. The lower 50 ft of the field had a low place that tended during significant rainfall to ac- cumulate water from the upper side of the field. The entire field , except the lower part was bordered to intercept runoff water from adjacent fields and keep it out of the experimental area. To test the feasibility of diking, dikes were installed by hand 50 ft apart in 1979 (Table 3). In 1980, dikes were installed with dikers manufactured in Lockney, Texas. These were mounted behind lis- \ter sweeps and adjusted to establish dikes about 4- to 6-in. high on 4- or 8-ft intervals in the furrows. The tripping mechanism that determined intervals between dikes was a 3-ft-diameter wheel. In 1981, dikers were modified to trip by means of a hydraulic motor- ‘Contribution otTexas Agricultural Experiment Station, Texas A&M University Re- search and Extension Center, Vernon, Texas 76384. zProfessor, Texas Agricultural Experiment Station, Vernon, Texas 76384. driven mechanism described by Lyle and Dixon (13). This re- placed the 3-ft-diameter wheel used in 1980 and provided more uniform diking intervals. Removing the wheel shortened the di- kers by about 3 ft, which allowed easier handling of less weight. As noted in Table 3, a diking interval of 6 ft in the furrow was used from 1981 to 1985. Treatment 2 was a check treatment in 1979 (Table 3). ln 1980, this treatment was changed to evaluate diking interval (8 ft apart), and from 1981 to 1985 it was changed to evaluate diking alternate middles (half-diked). Typical cultural operations for all treatments including the check or conventional-tilled sorghum consisted of the following: (1 ) bed- ding and applying 1.2 lbs/A of propazine in early spring; (2) rod weeding if needed in April and/or early June; (3) fertilizing land with 40 to 60 lbs of N/A and 20 to 25 lbs of P/A; (4) planting and applying 1.2 lbs/A of propazine in late June or early July, and (5) cultivating for reshaping beds and controlling weeds in late July. Pioneer 8501 sorghum hybrid was planted on dates given in Table 4 at a rate of about 3 seeds/ft of row (3 lbs/A). As reported in Table 4, diked and subsoiling operations for specific treatments in Table 3 were done on indicated dates and years. Dikes were installed, sometimes in conjunction with bedding or cultivation on treat- ments 2, 4, and 5, at the time of bedding and herbicide applica- tions in early spring and again after stand establishment in July or August (Table 4). Subsoiling in treatments 3 and 5 was usually done when land was bedded. After the sorghum was harvested, stubble remained standing until land was bedded and diked in early spring. During this fallow period dikes, even though reduced in size, were still functional on appropriate treatments. TABLE 1. RAINFALL DATA FROM 1979 TO 1985 AT CHILLICOTHE, TEXAS Months 1979 1980 1981 1982 1983 1984 1985 ‘R inches ‘J January 1.49 2.14 0.04 1.24 2.47 0.09 0.33 February 0.43 0.71 0.80 0.34 0.22 0.75 1.77 March 2.37 0.42 1.55 2.17 2.25 1.94 2.71 April 2.10 0.72 3.19 1.53 1.56 0.88 1.93 May 6.23 7.43 5.20 7.82 4.49 0.91 g 2.51 June 4.16 1.17 3.87 4.20 3.51 1.15 8.31 July 9.09 0.00 0.09 2.09 1.4a 1.5a 2024/ August 5.31 0.44 0.83 3.11 0.23 1.96 2.56 September 0.01 2.42 0.74 1.99 0.59 0.23 2.73 October 1.97 0.78 2.28 0.11 13.74 6.29 5.35 November 1.36 0.60 0.66 1.87 1.76 2.23 1.36 December 1.21 1.18 0.28 1.12 0.61 4.30 1.95 Total 29.67 18.01 20.13 28.16 32.86 22.31 33.53 TABLE 2. THE pH, PERCENT ORGANIC MATTER, AND PARTICLE SIZE DISTRIBUTION OF ABILENE LOAM AND MILES FINE SANDY LOAM Soils Abilene loam Miles fine sandy loam Soil depth organic organic (inches) pH matter sand silt clay pH matter sand silt clay % % 0-6 7.1 0.88 47 36 17 6.1 0.40 70 21 9 6-12 7.2 0.87 47 36 17 6.1 0.41 72 18 10 TABLE 3. DESCRIPTION OF TILLAGE TREATMENTS ON SOR- t‘; GHUM IN 1979, 1980, AND 1981-85. Treatment‘ Description of treatments 1979 1. Check Conventional 25 2. Check Conventional ‘ 3. Subsoiled“ Land was subsoiled 16 in. below beds O MILES FINE SAND Y LOAM and furrows 4. Diked Rows were diked 50 ft apart 2 0 ‘ A B I I- E N E I- O A M 5. Subsoiled and diked Rows were subsoiled and diked as described under treatments 3 and 4 8 1990 ' 1s 9 4 1. Check Conventional g 2. Diked Rows were diked 8 ft apart 3 3. Subsoiled“ Land was subsoiled 16 in. below beds l“; and furrows — 1Q ‘ 4. Diked Rows were diked 4 ft apart g w‘ ‘ 5. Subsoiled and diked Rows were subsoiled and diked as ‘ described under treatments 3 and 4 0g 5 O 1991-1995 \#\Q‘___,.___-o 1. Check Conventional _ 2. Half-diked Diked every other row about 6 ft apart J 3. Subsoiled“ Subsoiled 16 in. below beds and o fl furrows 0 O 4. Diked Rows were diked 6 ft apart 0'5 1' 5 10 15 5. Subsoiled and diked Rows were subsoiled and diked as described under treatments 3 and 4 ‘Treatments were 12 rows wide and about 200 ft long and replicated three times. “Time of subsoiling operation is given in Table 4. BARS \ figure 1. Moisture retention curves for Miles fine sandy loam Q and Abilene loam soils. I Soil moisture of the top foot was determined gravimetrically; soil moisture at 1 to 4 or 5 ft at 6-in. increments was determined using neutron scattering techniques. Moisture in top foot and neutron readings were converted to inches. In 1979, moisture use was determined about 75 ft from the upper part of the slope. From 1980 to 1985, moisture use was measured on the upper, middle, and lower parts of slope. Distances of neutron access pipes with respect to slope for monitoring soil moisture were about 30, 90, and 150 ft down the slope for upper, middle, and lower parts of slope, respectively. Yield determinations were made by combine-harvesting three center rows. Fifty feet were harvested from each section of the slope in 1979, and 6O ft were harvested in remaining years. Grain yields were corrected to 13 percent moisture. Yield data were sub- jected to appropriate analysis of variance as explained in a previ- ous publication (8). Soil particle size distribution shown in Table 2 was determined by the hydrometer method (3). Organic matter was determined according to the method described by Walkley (14). Strength of soil as affected by subsoiling furrows and beds was evaluated using a recording soil penetrometer. The penetrometer, which is about 0.5-in. diameter with a 60° cone-shaped tip, is mounted on a 300- or 500-lb force transducer with an output of i 5V. The shaft behind the lip is about 0.4 in. in diameter. The output from the transducer was a linear function of mechanical resistance of the soil. Strength measurements were made at 2-in. depth incre- ments to a depth of 20-22 in., and data were recorded with a Om- nidata polycorder. The output was converted to mechanical im- pedance in bars (0.99 atmosphere or 14.5 lbs/in.2). Runoff from the check, half-diked, and diked treatments was measured in 1985 using level stage recorders. Cotton The effects of subsoiling and furrow diking on cotton yields were evaluated in 1980-82, 1984, and 1985. Cotton failed to emerge to satisfactory stands in 1983. Because it was too late to replant cotton, mungbeans were planted on the site but failed to produce marketable yields. Chemical and physical properties of this Miles fine sandy loam are given in Table 2. Water retention characteristic of this soil is shown in Figure 1. This deep and uniformly coarse-textured soil frequently develops compacted zones that severely restrict water infiltration and root growth. The test site is in a flat area between two parallel terraces. The slope down the row is 0.1 percent to 0.2 percent. Tillage treatments that remained at the same location during 1980-85 are described in Table 5. Plot included eight 40-in. rows, 650-750 ft long. Treat- ments were randomized in three complete blocks. Typical cultural operations for all treatments including the check or conventional-tilled cotton were as follows: (1 ) shredding stalks, bedding, and applying 11/2 pt/A of Treflan in early spring; (2) fer- tilizing with 20 to 40 lbs/A of N, cultivating, and planting in late May or early June; (3) cultivating or reshaping beds after stand es- tablishment, and (4) if needed, cultivating for weed control in July. . Diking and subsoiling operations for specific treatments were I done on indicated dates and years (Table 6). Dikes were installed, sometimes in conjunction with bedding or cultivation on treat- ments 2, 3, and 4, at time of bedding in early spring and again after stand establishment in June or July (Table 6). Subsoiling of land in treatment 2 was done at time of early bedding. Dates of main cultural operations are given in Table 6. Bedding and diking as reported in Table 6 were performed before spring rains. Rainfall at the site is reported in Table 7. In 1980, Tamcot SP-37 was planted at a rate of about 4 seeds/ft (12 lbs/A), and from 1981 to 1985 a similar variety, G&P3774, was planted in these experiments. Cotton was harvested with a I two-row brush stripper. The effects of slope and treatments were measured by harvesting 100-ft sections of two rows in the upper, TABLE 4. DATES OF CULTURAL OPERATIONS FOR SORGHUM Years Diking Subsoiling Planting 1979 Mar. 9*-July 1" Mar. 8*“ June 21 1980 Apr. 8*-July 18*’ — June 18 1981 Jan. 28*-Aug. 3*’ — July 6 1982 Jan. 16*-July 26“ Nov. 15, 1981'“ June 28 1983 Apr. 18*-July 25“ — June 27 1984 Feb. 1*-Aug. 7*‘ Feb. 1**** July 5 1985 May 7*-July 3*’ — June 25 *Appropriate treatments were diked before planting and undiked at planting. “Appropriate treatments were rediked after stand establishment. “Beds and furrows were deep-plowed 14-16 in. ****Furrows were deep-plowed 14-16 in. TABLE 5. DESCRIPTION OF TILLAGE TREATMENTS FOR COT- TON ON MILES FINE SANDY LOAM IN 1980, 1981, 1982, 1984, AND 1985 Treatment Description of treatments 1. Check Conventional 2. Half-diked Diked every other row about 6 ft apart 3. Diked Diked every row about 6 ft apart 4. Subsoiled and diked Diked every row as in treatment 3 and subsoiled furrows 14 in. deep TABLE 6. DATES OF MAJOR CULTURAL OPERATIONS FOR COTTON Years Diking Subsoiling Planting 1980 Apr. 15"-July 14" Apr. 15*“ June 5 1981 Mar. 24*-July 22“ Mar. 23'“ June 24 1982 Jan. 25*-July 27“ Jan. 25"“ June 16 1984 Feb. 6*-June 14*’ Feb. 6*" May 25 1985 Apr. 10"-June 17“ — May 30 *Appropriate treatments were diked before planting and undiked at planting. "Appropriate treatments were rediked after stand establishment. '**Furrows were deep-plowed 14—1 6 in. ****Beds and furrows were deep-plowed 14-16 in. TABLE 7. RAINFALL DURING 1980-82 AND 1984-85 ATVERNON Months 1980 1981 1982 1984 1985 inches January —* 0.05 1.33 0.10 0.28 February —* 0.83 0.25 0.88 2.10 March —* 1.34 1.66 2.33 6.82 April 1.00 2.45 1.70 0.95 4.05 May 3.52 4.97 8.94 1.55 1.54 June 1.70 4.73 4.18 0.43 5.57 July 0.00 1.73 2.46 0.74 0.05 August 1.15 3.03 1.47 1.95 2.54 September 2.06 1.40 0.77 0.52 3.11 October 0.28 3.21 0.14 3.25 5.80 November 0.76 0.44 2.31 3.01 0.81 December 1.35 —* 1.04 4.72 0.07 Total 1 1.82 24.18 26.25 20.43 32.74 ‘Rainfall not recorded. middle, and lower parts of the slope. Fiber quality analyses for each harvested plot were determined in 1981, 1982, and 1984. Analyses of variance of yield data by location on slope and total yields were performed as outlined in a previous publication (8). Runoff on undiked and half-diked treatments was measured in 1985 with level stage recorders. The effect of subsoiling was measured using a recording soil penetrometer as previously described. Average soil strength was calculated at intervals of 2 in. to a depth of 22 in. The effects of time and moisture on changes in strength were evaluated. Implica- tions of these results on rootgrowth and moisture use are discussed. I“ Results and Discussion Sorghum Yields of sorghum grown under different fillage practices in 1979 are reported in Table 8. High rainfall from May through August (Table 1) was responsible for high yields in 1979 that ranged from about 2,900 to almost 5,400 lbs/A. The average increases due to diking, subsoiling, and diking plus subsoiling ranged from 11 percent to 17 percent. Average yields for diked or subsoiled treatments were about 1,215 to 1,960 lbs/A higher than the check on the upper side of the slope, but there was no difference among the treatments at the lower position of the slope. Even though the handmade dikes were 50 ft apart, their effects were highly significant. Plants close to the dikes on the upper part of the slope (5) were 36 percent taller and appeared more productive than plants 25 to 50 ft from the dikes. These results indicated the need for shorter in- tervals between dikes and the need for determining plant re- sponse and moisture use with respect to location down the slope. Because of the low yields, sorghum was not harvested with respect to the slope in 1980. Sorghum yields were very low and ranged from 550 to almost 800 lbs/A (Table 9). Low rain- fall (Table 1) and high evaporative conditions contributed to the lower grain yields. However, as shown in Table 9, diked sorghum produced about 250 lbs/A or about 35 percent more than the sorghum on the check treatments. subsoiled treatments and about 1,450 to 1,620 lbs/A on diked a and diked-subsoiled treatments, respectively. Yield increases caused by diking treatments in 1984 and 1985 as shown in Table 10 were small compared with previ- ous years. The greatest effects of diking in 1984 and 1985 were found on the upper part of the slope. Average yields for treatments and position on the slope for 1981-1985 are reported in Table 11 and shown in Figure 2. The data presented in Tables 8, 9, and 10 illustrate that diking significantly affected yields every year during 1979-1985. Yields of diked sorghum shown in Table 11 were 88 percent, 48 per- cent, and 30 percent higher than the yields of the check treat- ments on the upper, middle, and lower parts of the slope, re- spectively. Diking increased the average yield of sorghum 50 percent over the check for 1981-1985. Half-diking increased sorghum yields an average of about 62 percent, 21 percent, and 17 percent on the upper, middle, and lower parts of the slope, respectively, and an overall average of 29 percent higher than the average check yield. The yields of the check in the upper and middle parts of the slope reported in Table 10 and Figure 2 are probably indica- tive of the expected yields of sorghum on undiked large fields. This assumption is based on the fact that most of the produc- five land area on large, gently sloping fields would be com- posed of the upper and middle parts of the slope. The lower part of the slope often includes a drainage ditch or an area Yields as affected by treatments, position on the slope, and that conducts runoff to drainage waterways or to low areas T" years 1981-1985 are reported in Table 10. As shown in Table where water accumulates. These areas are often not produc- 10, yield increases due to diking were very high in 1981 and tive because excessive runoff prevents planting or waterlog- 1982. In 1981, yields of the diked treatments were four and ging conditions prevent plant growth. However, the sorghum six times higher than yields of the check and subsoiled treat- yields from undiked land would probably vary with factors ments on the upper part of the slope (Table 10) and slightly such as soils, location, rainfall, and degree of slope. more than two times the average yields of the check and sub- The effects of subsoiling on yields varied with years. Yield soiled treatments. Yields of the diked treatments in 1982 were increases from subsoiling were greatest the same year the til- more than four times the yields of the check on the upper part lage operation was performed. Data presented in Table 8 show of the slope and averaged more than two times the average a marked increase in yields from subsoiling on the upper part yield of the check treatment. of the slope in 1979. Subsoiling significantly increased yields Results in 1983 indicated diking increased yields an average in 1982, but residual effects of subsoiling in prior years, which of 43 percent over the check treatment. Average yields in were measured in 1980, 1981, and 1983, were small or insig- 1983 were slightly more than 1,000 lbs/A on the check and nifant. The only possible exception was the response of sor- TABLE 8. EFFECT OF SUBSOILING AND DIKING TREATMENTS OF SORGHUM HYBRID (PIONEER 8501) IN 1979 Distance down slope in feet 0-50 50-100 1 00-150 1 50-200 Average c/o Of °/o Of ‘V: Of °/o Of °/o Of Treatments lbs/A diked lbs/A diked lbs/A diked lbs/A diked lbs/A‘ diked 1,2 Conventional (check) 29394 69 4184a 86 4973a 100 5298a 101 4349a 89 - 3 Subsoiled‘ 4153 95 5237a 108 5244a 106 5111a 97 4936ab 102 w 4 Dikedz 4384 100 4855a 100 4947a 100 5256a 100 4861 ab 100 5 Subsoiled and diked 4898 112 4999a 103 5255a 106 5372a 102 5131b 106 Average?’ 3863c 4692b 5078ab 5267a ‘Land was subsoiled 16 in. below the beds and furrows. zDikes were 50 ft apart (put in manually). 3Average values for distance down slope followed by same letter are not significantly different at 5% probability level. v ‘Average yields of treatments 3, 4, and 5 were significantly higher yielding than check at 5% probability level. 4 k TABLE 9. GRAIN YIELDS OF GRAIN SORGHUM HYBRID (PIONEER 8501) UNDER DIFFERENT CULTURAL TREATMENTS OF SUBSOILING AND DIKING IN 1980 Treatments lbs/A % of Diked Conventional (check) 547a‘ 73 Subsoiledz 580ab 77 Dikeda \47ab 99 Diked‘ 751 ab 1 0O Subsoiled and diked‘ 791 b 105 ‘Average values for treatment followed by same letter are not signifi- cantly different at 5% probability level. Yields from diking were signifi- cantly higher yielding than the check and subsoiling treatments at 5% level of probability. 2Land was subsoiled 16 in. below furrows and beds in 1979. °Land was machine-diked 8 ft apart. ‘Land was machine-diked 4 ft apart. ghum to subsoiling and diking in 1983, which averaged 17 percent more than the yield of diked sorghum (Table 10). As shown in Table 11, average yields from subsoiling were 5 per- cent, 4 percent, and 22 percent higher than the check on the upper, middle, and lower parts of the slope, respectively. These results suggest that runoff water from the upper and middle parts of the slope was captured on the lower part of the slope. Subsoiling may prove to be most effective on gently sloping land with long rows. Surface sealing during and after rainfall probably reduces effectiveness of subsoiling in captur- ing rainfall for increased yields. Regression equations for sorghum yield on rainfall for the check and diked treatments shown in Table 12 indicate the increase in efficiency of rainfall achieved by diking. Total rainfall from April through September accounted for only 18 percent of the variabil- ity in yield of the check on the upper part of the slope and 4O per- cent of the variability of average yield. In contrast, rainfall during the same period accounted for 51 percent and 58 percent of the variability of yield for the diked treatments. Rains during August accounted for 57 percent and 83 percent of the variability of yield of the check on the upper part of the slope and of the average yield. These percentages compare to 89 percent and 91 percent of the variability of yield for the diked treatment. For both peri- ods of rainfall, diking substantially increased the effect of rainfall on yield. The relative low contribution of rainfall to variability of yield on the upper part of the slope for the check can be attributed to high runoff. Diking was particularly effective in increasing the yield response to rainfall on the upper part of the slope as evi- denced by the slopes of the regression lines. The contribution of rainfall to variability of yield on diked treatments was similar for yield on the upper part of the slope and the average yield. Water use for years with no runoff and where runoff was mea- sured in 1985 are reported in Table 13. Previous studies showed that water was the dominant factor for yields in the Rolling Plains (5) (7). High temperatures and lack of timely and adequate rainfall are often responsible for severe stress on growing plants in the Rolling Plains. The severity of this stress in any year determines the amount of water required to produce acceptable yields. For example, according to regression analyses, 4.5 to 5 in. of water were required before any grain was produced in the dry years of 1977 (5) and 1981 (8). Previous results and data pre- sented in Table 13 indicate that under typical years the diked sor- ghum produced about 300 lbs of grain/inch of water. The produc- tion/inch of water in 1980, an extremely dry year, was only about 100 lbs/in. of water. Measured runoff in inches from check, half-diked, and diked treatments during the 1985 crop year was 1.64, 0.75, and O in., respectively. Factors such as residue cover, soils, antecedent mois- ture, and tillage affect the surface permeability of soils and runoff. TABLE 10. SORGHUM GRAIN YIELDS UNDER DIFFERENT TREATMENTS AND LOCATIONS ON THE SLOPE IN 1981-1985 Position on slope Treatments Upper Middle Lower Average 1981 lbs/A Check 478a* 949a 1 681 a 1 036a Subsoiled 312a 527a 2499a 1 1 13a Half-diked 1439b 1 778b 2355a 1 857b Diked 1 806b 2298bc 2559a 2221 b Subsoiled and diked 1939b 2559c 2234a 2243b Average 1 1 95 1 622 2266 1982 lbs/A Check 655a 1 509a 281 7a 1661 a Subsoiled 1 282b 2279a 3385a 231 5b Half-diked 2036c 2147a 2916a 2366b Diked 2747d 3596b 3853a 3399c Subsoiled and diked 3410a 3715b 3237a 3454c Average 2026 2649 3242 1983 lbs/A Check 760a 1 1 12a 1 195a 1022a Subsoiled 587a 900a 1 602a 1 030a Half-diked 1 002ab 1 556a 1 847a 1 468b Diked 1 1 03ab 1 481 a 1 778a 1 454b Subsoiled and diked 1347b 1585a 1936a 1623b Average 960 1327 1672 1984 lbs/A Check 1 549a 2266a 2056a 1 957a Subsoiled 1 457a 2500a 2323a 2093a Half-diked 2423b 2649a 2568a 2547b Diked 2363b 2682a 2408a 2484b Subsoiled and diked 2408b 2675a 2380a 2488b Average 2040 2554 2347 1985 lbs/A Check 2661 "* 3342 3202 3068*“ Subsoiled 2756 3344 3578 3226 Half-diked 2991 2934 3079 3001 Diked 3479 3555 3646 3560 Subsoiled and diked 3183 3147 3638 3323 Average 301 4 3264 3429 ‘Values within each location on slope or averages followed by same letter are not significantly different at 5% level according to Duncan’s multiple range test. "Half-diked, diked, and subsoiled-diked treatments significantly higher yielding than check and subsoiled treatments at 0.08 level of probability. ***Diked and subsoiled-diked treatments produced significantly more grain than othertreatments (sign = 0.05). The land in this study has been in sorghum since 1979. The re- sidue from preceding sorghum crops probably reduced the amount of runoff on the check and half-diked treatments. Penetrometer resistance measurements of the soil profile in 1985 below beds and furrows are shown in Figures 3 and 4, re- spectively. The beds were deep-plowed in November 1981; how- ever, the effects of deep tillage are still apparent in 1985. There was a greater difference in strength of soil between the check and the subsoiled-diked treatment in the furrows than in the beds, par- ticularly in the zone of 8- to 12-in. depth. This difference likely can be attributed to the furrows having been subsoiled in February 1984 compared to the beds having been subsoiled in November 1981. Cotton Cotton yields from various tillage treatments and slope posi- tions are given in Table 14. Yields of cotton, like sorghum, were very low in 1980, which was a year with extremely high summer temperatures and low rainfall. Cotton in the lower parts of the field produced higher yields than cotton on the upper and middle parts of the slope, but yields were not significantly affected by tillage treatments. Diking significantly increased yields in 1981 and 1982. The diked treatments averaged from 87 to 125 lbs/A of cotton more than the check treatments. Position on the slope significantly af- fected yields, especially the yields of the check treatments. For example, in 1981 and 1982 average cotton yields of the check treatments on the upper and middle parts of the slope averaged 7O percent and 88 percent, respectively, of the yields of the check on the lower part of the slope. Yields of half-diked cotton were higher in comparison to the check in 1982 than in 1981. Since the tillage treatments were in the same location throughout the exper- iment,diking could have caused residual effects on soil moisture from one year to the next. Soil moisture profiles of conventional beds are often only partially replenished by late fall and early spring rains. Cotton continues to extract water from the soil profile until the first killing frost, and valuable water also is lost to runoff. Treatments and locations on the slope did not affect cotton yields in 1984. Yields in 1985 were lower than in 1981, 1982, and 1984, and treatment effects were small but significant. As shown in Table 15, cotton on diked treatments produced about 25 lbs/A more lint than the cotton on the check treatments. III?! 3000 2000 1000 YIELD - POUNDS/ACRE CHECK SUBSOILED HALF- DIKED A summary of diking studies on cotton (Table 15 and Figure 5) shows half-diking increased yield about 28 lbs/A and full-diking in- creased yields almost 50 lbs/A. Diking treatments increased yields by an average of 22 percent, 21 percent, and 11 percent on the upper, middle, and lower parts of the slope, respectively, and an average of 18 percent during the 5 years of investigation. Half- diking increased yields over the check by 10 percent from 198C to 1985. Factors such as years, percent slope, and soils probably af- fect the response of cotton to diking. However, diking of the soil at this location with slopes of 0.1 percent to 0.2 percent increased soil moisture storage and yields in 3 of the 5 years. The benefits of dik- ing cotton would probably be greater on soils with steeper slopes. However, the slope of some fields may be such that furrow dikes alone could not effectively prevent runoff and severe erosion. The effects of treatments on fiber length and micronaire for 1981, 1982, and 1984 are reported in Table 16. The main effect of diking is reduction of drought stress. In dry years such as 1981, diking significantly increased fiber length and micronaire. In 1981, cotton from the diked and diked-subsoiled treatments had a higher grade and was estimated to be worth 4¢/lb more than the cotton from the check (6). Half-diked cotton in 1981 was worth about 2¢/lb more than cotton from the check. As shown in Table 16, the half-diked and diked cotton was about 1/32 in. longer than cotton from the check treatment. These data indicate diking could increase fiber length and micronaire of certain cotton varieties. Es- timates of runoff using yield data on this site were made in 1981. The average runoff on the check, half-diked, and diked treatments were estimated at 35 percent, 18 percent, and 0 percent, respec- tively (6). Water-use estimates for years or treatments free of UPPER PART OF SLOPE MIDDLE PART OF LOWER PART OF SLOPE SLOPE E 5‘, llllllllllll D DIKED 8t SUBSOILED Figure 2. Average sorghum yields under different tillage treatments and parts of slope from 1981 to 1985. 6 TABLE 11. AVERAGE SORGHUM GRAIN YIELD UNDER DIFFER- ENT TREATMENTS AND LOCATIONS ON THE SLOPE FOR 1981 THROUGH 1985 Position on slope Treatments Upper Middle Lower Average I —-—-—-——— lbs/A Check 1 221 1 836 21 90 1 749 Subsoiled 1279 191 0 2677 1 955 Half-diked 1 978 221 3 2553 2248 Diked 2300 2722 2849 2624 Subsoiled and diked 2457 2736 2685 2626 Average 1 847 2283 2591 2240 TABLE 12. REGRESSION OF SORGHUM YIELD ON RAINFALL FOR CHECK AND DIKED TREATMENTS DURING 1979-85 Treatment Time of rainfall Slope position Equation’ R2 Check April-September upper 9 = 331 .6 + 72.4x 0.18 Check April-September average“ 9 = — 276.4 + 142.2x 0.40 Check August upper 9 = 464.1 + 439.1 x 0.57 Check August average“ 9 = 550.7 + 678.0x 0.83 Diked April-September upper 9 = — 44.5 + 154.6x 0.51 Diked April-September average“ 9 = — 87.1 + 176.5x 0.58 Diked August upper 9 = 980.6 + 676.5x 0.89 Diked August average“ 9 = 1162.0 + 734.1 x 0.91 '9 is estimated yields in lbs/A; x refers to rainfall in inches. “Average yield refers to an average for upper, middle, and lower parts of slope. TABLE 13. AVERAGE WATER USE AND GRAIN YIELD/INCH OF WATER FROM 1979-1985* 1 980 Grain 1 983 Grain water lbs/m. water lbs/in. Treatment use of H20 Treatment use of H20 Check 6.5 84 Check 3.8 269 Subsoiled 7.3 80 Subsoiled 4.6 224 Diked 7.3 1 03 Half-diked 4.4 336 Subsoiled and diked 7.2 110 Diked 5.0 297 Subsoiled and diked 4.9 299 1981 Grain 1984 Grain water lbs/in. water lbsfin. Treatment use of H20 Treatment use of H2O Check 6.1 170 Half-diked 8.1 314 Subsoiled 6.5 171 Diked 8.5 291 Half-diked 6.5 285 Diked 7.3 306 Subsoiled and diked 7.2 312 1982 Grain 1985 Grain A water lbs/in. water lbs/in. Treatment use of H20 Treatment use of H20 Diked 11.0 310 Check 8.7 354 Subsoiled and diked 11.0 311 Half-diked 9.9 305 Diked 10.2 350 ‘Water-use data reported for treatments and years with no runoff experience during 1979-1984. In 1985, runoff was measured and subtracted from check and half-diked treatments. Runoff did not occur from diked treatment during crop year in 1985. 7 TABLE 14. LINT COTTON YIELDS FROM DIFFERENT TILLAGE TREATMENTS AND SLOPE LOCATIONS IN 1980-82 AND 1984-85 Position on slope 1980 Treatments Upper Middle Lower Average lbs lint/A Check 82 71 116 90 Half-diked 64 76 120 87 Diked 69 77 115 87 Subsoiled and diked 7O 86 150 102 Average 71 78 1 25 92 1981* p lbs lint/A Check 270a 330a 376a 325a Half-diked 334b 361 b 429a 375b Diked 399c 4590 429a 4290 Subsoiled and diked 3970 480c 475a 451c Average 350 408 427 395 1982* lbs lint/A Check 260a 339a 386a 328a Half-diked 388b 403ab 435a 409b Diked 437c 459b 41 7a 438b Subsoiled and diked 426c 430b 390a 415b Average 378 408 407 398 1984" lbs lint/A Check 41 1 380 350 380 Half-diked 371 390 351 371 Diked 378 383 389 383 Subsoiled and diked 347 400 396 381 Average 377 388 372 379 1985 lbs lint/A Check 265 243 255 254"" Half-diked 301 266 274 280 Diked 289 276 296 287 Subsoiled and diked 281 257 295 278 Average 284 261 280 275 ‘Values within each location on slope or averages followed by same letter are not significantly different at 5% level according to Duncans multiple range test. “Yields not significantly affected by treatments in 1984. “Average yields of diked treatments were significantly higher than the check in 1985. 0 o CHECK ‘ a omen - SUBSOILED -4 XA \\ Q _ a l z \ _ on ~ \\ I " 12 ‘ é ,1‘- -16 . \. \ l A o - 2o \¢ l \\. - 24 t o 5 1o 15 2o STRENGTH - BARS figure 3. Average penetnometer resistance of beds of wet soil — for diked and subsoiled, and check treatments in sor- ghum in 1 985. Beds for diked and subsoiled treatment were deep-plowed in November 1 981. TABLE 15. AVERAGE COTTON YIELDS ON DIFFERENT PART OF THE SLOPE FOR VARIOUS DIKING PRACTICES FROM 1980 TO 1985 Position of slope Treatments Upper Middle Lower Average Lint cotton—Ibs/A Check 258 273 297 276 Half-diked 292 299 322 304 Diked 31 4 331 329 325 Subsoiled and diked 304 331 341 325 Average 292 309 322 \ vi ‘U U J 0 o CHECK , a DIKED - SUBSOILED ‘ '\\ ~12 DEPTH - IN 5/ / -16 8O TABLE 17. WATER USE AND PRODUCTION OF LINT COTTON/ i \ INCH OF WATER* A - 2 0 I Water use lbs/in. \ \ Treatment (inches) of water “ 1980 Check 8.4 11 - 2 4 Half-diked 8.5 1O Q 5 1 Q 1 5 2 Q Diked 8.4 10 Subsoiled and diked 8.7 11 STRENGTH — BARS 1981 Diked 1 3.5 31 .8 Figure 4. Penetrometer resistance of furrows when soil was wet Subsmled and diked 13-2 34-1 \ for diked and subsoiled, and check treatments in sor- 1932 ghum in 1985. Furrows for diked and subsoiled treat- Diked 120 367 men! were deep-plowed in Febwaw 1984. Subsoiled and diked 10.8 38.5 1 984 Diked 9.6 39.9 Subsoiled and diked 9.6 39.8 ‘Water-use data reported for treatments and years with no runoff ex- perienced during 1980-1985. TABLE 16. FIBER LENGTH AND MICRONAIRE AS AFFECTED BY TILLAGE TREATMENTS IN 1981, 1982, AND 1984 Treatment 1981 1982 1984 Average Fiber length—inches Check 0.99a* 1.04 1.01 1.01 Half-diked 1.01b 1.05 1.04 1.03 Diked 1.03bc 1.06 1.00 1.03 Subsoiled and diked 1.040 1.05 1.02 1.04 Micronaire” ii "‘ Check 3.6 4.3 4.2 4.0 Half-diked 1: 3.6 4.4 3.9 4.0 Diked 3.5 4.4 4.1 4.0 Subsoiled and diked 3.9“ 4.5 4.2 4.2 ‘Average values for treatment followed by same letter are not signifi- cantly different at 5% probability level. \ “Micronaire refers to fineness. In 1981 micronaire of diked and sub- soiled treatment was significantly greater than other treatments at 5% probability level. UPPER PART OF SLOPE MIDDLE PART OF SLOPE IIIIZ LOWER PART OF SLOPE g 400- O g 0A In g - _ _ -4 o n4 n U = " “ /? 2 I I I u - — : : % g I _ _ z -8 II >- 3 I I " Al 9,10 6 10o —- - '- ' F‘ . ° n - " " I / 9/11 l; I : I E -12 i \A ‘ 3 o _ : I a u ‘ I 9/12 * I CHECK HALF- DIKED omen a g x 9 9'14 omen SUBSOILED w -16 Figure 5. Average cotton yields under different tillage treatments 4L4 and parts of slope from 1 980 to 1 985. llL - 2O IO // |\ B I U - 24 o 0 1O 2O 3O 40 STRENGTH - BARS J 1\ HQUTQ 7. Penetrometer resistance of a Miles soil after subsoiling 0 I I as a function of depth and moisture changes in cotton // in September 1985. Soil moisture was at about 0.1 bar I suction (field capacity) on September l0, 1985. -12 $0". DEPTH - IN 0% 9/10 _ 1 6 O K/ A 9/ 1 1 runoff were determined for 4 of the 5 years (Table 1 7). Production u of lint per inch of water for 1980 1981 1982 and 1984 ranged / . 9’ 12 from about 1O to 4O lbs/in. of wafer. Previous studies (5) indicate _ 2 o X n 9 / 14 that climatic conditions and varieties affect the production/inch of water. Under the severe stress conditions of the Rolling Plains, cot- - n u ton produced only 1O to 11 lbs/in. of water in 1980. In 1976 and - 24 iglilsli)’lliéilefiifivillifirllfflilffiiilfliliilrfftfil a o 1 o 2 o 3 o 4 O low as 4O lbs/in of water l a STRENGTH - BARS Soil strength affects root penetration and is negatively related to infiltration rate of the soil. Soil strength of a Miles soil in cotton was measured with a penetrometer at near field capacity and on Figure 6. Penetrometer resistance of a Miles soil as a function of three subsequent days. Figure 6 shows soil strength at various depth and moisture changes in cotton in September depths for conventionally tilled cotton, and Figure 7 shows soil » 1985. Soil moisture was atabout0.1 bars suction (field strength at comparable depths for subsoiled tilled cotton. Soil capacity) on September 10, 1 985. strength (penetrometer resistance) increased rapidly at depths of 1O -. "iii. 1\.\.\. -. 1.11 1 i U. \../ 11/1 t; \_| 53-16 IIXx -24 O 5 1O 15 96 MOISTURE Figure 8. Moisture changes with time and depth in cotton from September 1O through September 14, 1 985. from 4 to 14 in. in the conventionally tilled soil. Over the 4 days, strength increased from 2O to 35 bars at the 4-in. depth. Changes were similar to the 14-in. depth. ln contrast, there was little change in soil strength below the 4-in. depth in the subsoiled treatment, and soil strength was from 1O to 2O bars less in the subsoiled treat- ment than in the conventionally filled treatment. The substantial changes in soil strength of the conventionally tilled soil occurred over a 4-day periodwith very small changes in moisture suction. Figures 8 and 9 show the moisture suction and percent moisture of these soils over the 4-day period. The moisture suction changed from less than 0.1 bar to about 0.30 bar, and percentage moisture changed from 12 percent to 7.5 percent. These data emphasize the dynamic physical characteristics of this sandy loam soil. Cot- ton growing in soils such as the Miles often show deformed roots when soil strength is greater than 2O bar. 11 O -4 0A I II E —8 | i5 5, '12 “' u 09/10 a J A 9/11 Q _16 n 9112 9/14 0A! fl n -20 l -24i 0 0.1 0.2 0.3 0.4 SUCTION — BARS Figure 9. Soil moisture suction changes at four depths in cotton from September 1O through September 14, 1 985. Summary Sorghum Studies on sorghum production with diking and subsoiling til- lage practices were conducted from 1979 to 1985. Diking signifi- cantly increased yields in every year of the study. Average yields of diked treatments from 1981 to 1985 were 2,625 lbs/A, which was about 900 lbs/A more than the check treatments. During this period, yields of diked sorghum were 88 percent, 48 percent, and 3O percent higher than the check on the upper, middle, and lower parts of the slope, respectively. Half-diking during the period from 1981 to 1985 increased yields 62 percent, 21 percent, and 17 percent higher than the check on the upper, middle, and lower parts of the slope, respectively. The increase in yield from diking in some years was restricted primarily to the upper side of the slope. However, data from the upper part of the slope in these ex- periments may be the best estimate of potenfial yields for large, undiked fields. The effect of subsoiling on yields varied with years. Subsoiling had the greatest effect on yield during the year the subsoiling was performed. The average increases in yields by subsoiling from 1981 to 1985 percent were 5 percent, 4 percent, and 22 percent higher than check on the upper, middle, and lower parts of the slope, respectively. These results suggest that runoff water from the upper and middle parts of the slope is cap- tured on the lower part of the slope when land is deep-plowed or subsoiled. Yields of sorghum in the subsoiling plus diking treat- ment were usually not significantly different from the yields of sor- ghum in the diking treatment alone. These results indicate that subsoiling did not stop the movement of water from the upper and middle parts of the slope to the lower part of the slope on this soil. Regression analyses showed that rainfall, especially August rainfall, had a highly significant effect on yields of diked treat- ments. Finally, yields varied with years and rainfall, but these in- 1 vestigations demonstrated the effectiveness of diking in reducing runoff and increasing sorghum yields. These results demonstrate that with diking in combination with recommended varieties, planting date, and cultural practices, dryland production of grain sorghum is feasible in the Rolling Plains of Texas. High-residue crops such as sorghum can play an important role in maintaining desirable physical characteristics of poorly structured soils in the Rolling Plains. Cotton Studies on cotton production with diking and subsoiling til- lage were conducted from 1980 to 1985. Years and treatments significantly affected yields. Significant lint yield increases from til- lage treatments were observed in 3 of the 5 years. Diking in- creased the average yield about 32 percent above conventional tillage for 2 of the 5 years. For the period 1981 to 1985, diking in- creased the yield 22 percent, 21 percent, and 11 percent above the check on upper, middle, and lower parts of the slope, respec- tively. Average yields for the check and diked treatments from 1980 to 1985 were 275 and 325 lbs/A, respectively. Half-diking increased yields 13 percent, 10 percent, and 8 percent above the check on the upper, middle, and lower parts of the slope, respec- tively. Diking plus subsoiling did not significantly increase yields more than the diking by itself. Diking was effective in reducing . runoff and drought stress. The water captured by dikes signifi- cantly increased fiber quality in 1981 and tended to increase fiber length by about 1/32 in. during this period of investigation. Dik- ing may increase yields from 5O to 100 lbs/A. Literature Cited 1. Bilbro, C. J., and E. B. Hudspeth. 1977. Furrow diking to prevent runoff and increase yields of cotton. Texas Agric. Exp. Sta. PR-3436. 2. Bordovsky, D. G. 1983. Effect of furrow diking on cotton in the Rolling Plains. Texas Agric. Exp. Sta. PR-41 76. 3. Bouyoucos, G. J. 1927. The hydrometer as a new method for mechanical analysis of soils. Soil Sci. 23:343-353. 4. Clark, L. E. 1983. Response of cotton to cultural practices. Texas Agric. Exp. Sta. PR-4175. 5. Gerard, C. J., D. G. Bordovsky, and L. E. Clark. 1980. Water management studies in the Rolling Plains. Texas Agric. Exp. Sta., B-1321. 6. Gerard, C. J., P. D. Sexton, and D. M. Matus. 1983. Furrow diking for cotton production in the Rolling Plains. Texas Agric. Exp. Sta. PR-4174. 7. Gerard, C. J., P. D. Sexton, L. E. Clark, and E. C. Gilmore Jr. 1983. Sorghum for grain: Production strategies in the Rolling Plains. Texas Agric. Exp. Sta. B-1428. 8. Gerard, C. J., P. D. Sexton, and D. M. Conover. 1984. Ef- fect of furrow diking, subsoiling, and slope position on crop yields. Agron. J. 76:945-950. 12 9. Hanna, A. Y., P. W. Harlan, and D. T. Lewis. 1982. Soil- available water as influenced by landscape position and as- pect. Agron. J. 74:999-1004. Hudspeth, E. B. 1978. Basin tillage for water conservation and maximum dryland cotton production. The Cotton Gin and Oil Mill Press, p.18. Jones, O. R., and R. N. Clark. 1982. Effect of furrow dikes on water conservation and dryland crop yields. Agron. Abstr., p. 250. Luebs, R. E. 1962. Investigations of cropping systems, til- lage methods, and cultural practices for dryland farming. Kansas Agric. Exp. Sta. B-449. Lyle, W. M., and M. R. Dixon. 1977. Basin tillage for rainfall retention. ASAE Trans. 2011013-1017, 1031. Walkley, A. 1947. A critical examination of a rapid method for determining organic matter in soils—effect of variations in digestion conditions and of inorganic soil constituents. Soil Sci. 63:251-264. 10. 11. 12. 13. 14. ,..,.§-5§1,i;, . ,3; __, . » \_- v. w ‘w. [Blank Page in Original Bulletin] ‘ Mention of a trademark or a proprietary product does not constitute a guarantee or a warranty of the product by The Texas Agricultural ‘ Experiment Station and does not imply its approval to the exclusion of other products that also may be suitable. "i All programs and information of The Texas Agricultural Experiment Station are available to everyone without regard to race, color, religion, sex, age, handicap, or national origin. 2.5M—12—87