Assessment of Irrigation Scheduling Approaches in a Soybean Field Located in a Humid Region.

To determine better irrigation scheduling method for soils in the Blackland Prairie of humid southeastern USA, two commonly used irrigation scheduling methods, soil moisture via in-situ sensors (SM) and evapotranspiration (ET), were evaluated at Brooksville Experiment Stations of Mississippi State University. Three treatments with three replicates, SM, ET and rainfed (RF) were completely randomized to three blocks on a Brooksville silty clay soil in a three-acre, furrow-irrigated field. A group IV soybean cultivar, Asgrow 4632, was planted at 120,000 seeds per acre in a 1-m row spacing on May 9, 2016. Soil water content and potential in the layers of 0-15 cm, 15-30 cm, 30-45 cm and 45-60 cm were measured using TDR and Watermark sensors in each plot (6 rows × 100 m). Soil was sampled at these same layers were taken before and after irrigation to measure soil gravimetric water content. In the SM treatment, we compared daily soil water storage in rooting zone with the plant available water (PAW) value and once storage was equal or less than 50% of PAW, irrigation was applied to recharge soil water profile to its field capacity (FC). The ET method compared soil water storage estimated through calculated daily ET and an initial soil water content and applied sufficient irrigation water to replenish the soil FC when the soil water was equal or less than 50% of PAW. Rainfed treatment was not irrigated in the entire season. Water balance was estimated using a pen lysimeter (Soil Moisture, Corp) in each plot and microflume runoff collectors located at the tail end (north side) of the plots. The effective rainfall and canopy interception are estimated using a rain gage above the canopy and a complementary gage below the canopy of different plots. A weather station with an automatic evaporation pan was installed near the field to measure potential ET and evaporation. Crop growth stage, height and canopy cover, rooting depth, leaf area index, and dry biomass were determined weekly. Reference ETo was 665 mm and ETc was 504 mm at Brooksville station during soybean season in 2016. It is a dry year, soybean received 284 mm of total rainfall during its entire growing season, total effective rain water deficit was 220 mm. In vegetative growth stages, rainfall (82 mm) fully satisfied soybean ETc requirement (59 mm). All irrigation was required in R stages this year. The SM treatment required six times of irrigation and received 228 mm, while the ET treatment required five times of irrigation and received 183 mm. Both SM and ET  treatments were triggered at the same time on June 28 (R2), July 5 (R2), July 15 (R3), July 27 (R4), and August 4 (R5), 2016. SM treatment was triggered earlier and one more time on June 22 (R1), 2016. Soybean consumed 511 mm water in the entire growing season based on detail calculation of soil water balance. Soybean used 66 mm more water in SM treatment than ET treatment. During reproductive (R) growth stages, all treatments consumed 75% of total water consumption in the entire growing season. Soybean irrigated six times on June 22 (R1), June 28 (R2), July 5 (R2), July 15 (R3), July 27 (R4), and August 4 (R5), 2016 stage had mean grain yield of 3078 kg ha^-1 (based on combine harvest) and a harvest index of 32%. While soybean irrigated five times at R2, R3, R4 and R5 had mean grain yield of 2984 kg ha^-1 and a harvest index of 32% too. Irrigation six times and five times significantly increased grain yield by 41% and 40%, aboveground biomass by 46% and 38%, and harvest index both by 10%, respectively. Pod filling was not affected, and thousand grain weight averaged 146 g and 160 g for irrigated and rainfed treatments, respectively. Irrigation increased the number of pods, irrigated treatments had twice pod numbers than rainfed treatment. Irrigation and rainfed treatments had grain water use efficiency 6.4 and 5.3 kg ha^-1 mm^-1. Though sensor-based method triggered one more time of irrigation, it only increased grain yield by 3%. It appears that both sensor-based method and ET-based method can be used for scheduling soybean irrigation for the relatively shallow and heavy clay soil in the Blackland Prairie humid region.