Soybean Field Losses as Influenced by Harvest Delays

Conflicts for time and machinery can postpone harvests beyond the initial time when optimum conditions exist. This study was conduced to determine the effects of delaying soybean [Glycine max (L.) Merr.] harvests on grain losses in the field. Field studies were conducted each year from 1983 to 1986 at Arlington, WI. Two cultivars from each of maturity groups (MG) 0, I, and II, one more susceptible to lodging than the other, were used. Initial harvest for each maturity group began 3 to 7 d beyond stage, R8. Three additional harvests were made for each maturity group at 14, 28 and 42 d beyond their initial harvest. Average soybean field losses were 10% of the potential yield, but ranged from 5.5% in 1983 to 12.7% in 1984. Loss of potential yield increased linearly at a rate of 0.2% d from an average of 6.1% at the initial harvest to 13.9% 42 d later. In 1984 and 1986 net yields were reduced 14 and 18 kg ha d respectively. Harvest delays of 42 d resulted in plant deterioration and, in turn, lodging increased 20%, and preharvest, shatter, and stem losses increased 62, 95, and 70 kg ha , respectively. Shatter losses were influenced by moisture conditions at harvest, but plant deterioration also increased shattering beyond that accounted for by moisture. On the average the MG I cultivars Hardin and Northrup King S1346 lost only 7.3 and 8.3% of their potential yield versus 10.4 and 11.4% for the MG O cultivars Ozzie and Evans, and 10.8 and 9.2% for MG II cultivars Wells II and Corsoy 79. Both of the MG I cultivars exhibited slower rates of harvest loss increases. The proportion of potential yield lost was inversely related to potential yield, indicating that harvest efficiency was improved with higher yields. Harvest delays can ultimately result in plant deterioration, increased grain losses, increased harvesting difficulties, and reductions in net yield of 11 kg had ARVEST delays for a portion of the soybean crop are inevitable each year. Unsuitable weather can postpone soybean harvest when the grower is otherwise prepared to harvest. Availability of labor and equipment can delay harvesting of soybean for several weeks after their harvest maturity. Nave et al. (1973) reported little progress in Illinois since the mid-1920s in reducing harvesting losses in soybean, from an average total loss of 11.7% of the potential yield in 1927 to 9.2% in 1968. Schnug and Beuerlein (1987) report that average soybean harvest losses remain greater than 10% of the harvestable seeds remaining on the plants at harvest, but with proper machine operation and adjustment, losses can be reduced to 1 to 3%. Burnside et al. (1969) attributed high harvest losses from the combine gathering unit to harvest delays while waiting for weeds to desiccate. 1 Dep of Agronomy, 1575 Linden Dr., Univ. of Wisconsin-Madison, WI 53706. Contribution from the Dep. of Agronomy, Univ. of Wisconsin-Madison. Supported by Hatch Project 1890 and the Wisconsin Crop Improvement Assoc. Received 22 April 1988. *Corresponding author. Published in Agron. J. 81:251-258 (1989). H Soybean and Small Grains Page 2 © 2004 Board of Regents of the University of Wisconsin System, doing business as the Division of Cooperative Extension of the University of Wisconsin-Extension. In an Ohio study soybean yield losses due to pre-harvest shattering were negligible prior to the crop reaching 100 g kg grain moisture, but increased up to 1% d when the crop remained in the field with grain moisture below 100 g kg (Lamp et al. 1962). Shatter losses during harvesting of soybeans also increased as grain and pod moistures decreased. The gathering unit is the source of greatest loss during soybean harvest, and grain shattering makes up the largest proportion (80%) of the total gathering unit losses (Schnug and Beuerlein, 1987). Lodging in two separate studies was responsible for 1.1 and 1.3% out of 9.7 and 8.0% total field losses, respectively (Park and Webb, 1959; Weber and Fehr, 1966). Other factors reported to affect harvest losses include plant population, time of day, row width, crop condition, weed infestations, and weather patterns (Nave and Cooper, 1974; Nave et al., 1973; Nave and Wax, 1971; Burnside et al., 1969; Weber and Fehr, 1966; Lamp et al., 1962). Schnug and Beuerlein (1987) recommended that soybean harvest begin when the crop reaches 170 to 190 g kg grain moisture, with most efficient harvest occurring between 130 to 160 g kg grain moisture. Only brief attention has been given to the specific effects of harvest delays on harvest losses. Lamp et al. (1962) devoted 2 yr of a 5-yr study to the examination of harvest date effects on soybean with average potential yields of only 1685 kg ha. Our study was initiated to examine the effects of delaying soybean harvest on field and harvest losses, and associated yield reductions. Our objective was to compare the influence of harvest delays on field and harvest losses among cultivars of differing lodging susceptibilities and maturities. MATERIALS AND METHODS Field studies were conducted each year from 1983 to 1986 at Arlington, WI (43°20’N, 89°25’W) on a Plano silt loam (fine-silty, mixed, mesic Typic Argiudolls) soil with a pH of 6.5, 3.4% organic matter, and an average of 464 kg of K and 122 kg of P ha. In all years maize (Zea Mays L.) was the preceding crop. Plots, 7.6 m long, were planted using a specially designed plot planter (Oplinger et al., 1983) and consisted of 11 rows spaced 0.18 m. Planting was on 26 May 1983, 31 May 1984, 7 May 1985, and 6 May 1986 at 12 seeds m. Plots were end trimmed to 6.4 m between the V1 and V3 growth stages, and the center seven rows were harvested with an ALMACO (Allen Machine Co., Nevada, IA 50201) SPC Model 20 plot combine. The experiment was designed as a randomized complete block with a split plot arrangement of treatments and four replicates. Main plots consisted of six cultivars, two similarly maturing cultivars from each of the three maturity groups 0 (Ozzie and Evans), I (Northrup King S 1346 [NK1346] and Hardin), and II (Wells II and Corsoy 79). Cultivars were selected to have low (Ozzie, NK1346, and Wells II) versus high (Evans, Hardin, and Corsoy 79) lodging. Determination of these characteristics were based on multiple-year cultivar evaluation results in southern Wisconsin (Oplinger et al., 1982). Subplots were four scheduled harvest dates at 0, 14, 28 and 42 d after harvest maturity. Harvest maturity occurred 3 to 7 d after growth stage R8 when grain moisture first neared 160 g kg. Harvests were made as scheduled or at the first opportunity after the scheduled harvest that weather permitted. The combine was adjusted and operated consistently, utilizing a cutting height of 0.076 m at each harvest date in order to minimize variation due to harvest mechanics. Plant population and plant height were determined in each plot just prior to harvest. Lodging was evaluated just prior to harvest using a rating from 1, all plants erect, to 5, all plants prostrate. Grain moisture and seed weight were determined using harvested grain samples. Soybean and Small Grains Page 3 © 2004 Board of Regents of the University of Wisconsin System, doing business as the Division of Cooperative Extension of the University of Wisconsin-Extension. Soybean losses were determined using modifications of the methods most recently described by Schnug and Beuerlein (1987). Losses were determined inside of a 0.93 m rectangular frame, which extended across all harvested rows (1.24 m) and a 0.75 m length of row. One end of the frame was removable so the frame could be slid into the plot from the side, thus facilitating measurement of preharvest losses without damaging the standing plants. Each soybean seed on the ground was counted whether it was free from the pod, in a detached pod, or in a pod that was attached to a detached portion of stem. In 1983 as seeds comprising preharvest and harvest losses were counted they were removed from the measurement area. The same specific portion of the plot was used for determining all other loss categories. A portion of the plot was harvested and the combine was backed up in order to determine seeds lost due to the action of the gathering unit prior to the trash being dropped in that area. The combine then completed the plot, and soybean seeds which came out of the back of the combine were counted. In 1984 to 1986 soybean from the sampled area was not removed after counting. Instead when the front of the combine had progressed through the plot, the machine was stopped, the sieves were swept clean, and then the combine continued through the plot. Seeds lost from the action of the gathering unit could then be determined from the portion of the plot last harvested and those seeds forced out of the back of the combine were counted from the portion first harvested. This procedure was more efficient in that harvest loss data could be collected later without occupying the combine or operator’s time. Harvest loss categories were then determined by subtraction of overlapping categories (seed number from the first harvested plot portion – seed number from the last harvested plot portion) – seed number prior to harvest = seeds lost during threshing). The following preharvest and harvest loss categories were measured: 1) Preharvest losses: All seeds detached from standing soybean plants prior to harvest. 2) Gathering Unit losses: All seeds lost due to the action of the combine header during harvest. a) Shatter losses: All seeds free from pods or in detached pods. b) Stem losses: All seeds attached to stems that were broken or cut free from the harvested plants. c) Stubble losses: All seeds on the remaining stem portion below the point were the plants were cut off during harvesting or remaining on uncut lodged plants. 3) Threshing: All seeds which came out of the back of the combine by contact with the cylinder or sieves, with the trash, or during winnowing. 4) Combine losses: All the soybean seeds lost that were attributed to the harvest machinery which was the total of the gathering unit and threshing losses. 5) Total losses: All soybean seeds lost in all categories (preharvest and combine). Losses on an area basis were calculated using counted seed numbers and seed weight measurements. Two types of grain yield were determined and were defined as: 1) Net yield: Determined from the total weight of soybean actually harvested by the combine and adjusted to 130 g kg moisture. 2) Potential yield: Yield that would have been obtained from a plot if it had been harvested at the initial opportunity with a machine that had no losses (net yield + [total losses – initial harvest preharvest losses]). Soybean and Small Grains Page 4 © 2004 Board of Regents of the University of Wisconsin System, doing business as the Division of Cooperative Extension of the University of Wisconsin-Extension. All data were subjected to analysis of variance. Some comparisons between treatment means were made using Fischer’s protected LSD test. Sums of squares for single degree of freedom comparisons between treatment means were made using Fisher’s protected LSD test. Sums of squares for single degree of freedom comparisons and orthogonal polynomials were partitioned using treatment totals. RESULTS AND DISCUSSION