Analysis of spatial yield variability and economics of prescriptions for precision agriculture: A crop modeling approach.

摘要

Non-uniformity of soil properties, soil moisture and rooting depth, and other factors such pest and disease pressures can cause significant soybean and corn yield variability within a field. In this study, two crop growth models were used to characterize factors that cause spatial yield variability in soybeans and corn, and to evaluate economic consequences of variable rate management prescriptions. Analysis of yield data from 224 grids within a 16-hectare field in Boone, Iowa focused on water stress effects using CROPGRO-Soybean and CERES-Maize models for soybean and corn, respectively. Water stress explained 69% of the variability in soybean, and population and water stress explained 57% of corn yield variability. Grid-level optimum nitrogen fertilizer rate prescriptions for corn were also developed. Distribution of optimum nitrogen fertilizer prescription was highly spatially varied. Optimum nitrogen rates were found to range from 141 to 160 kg ha−1 in 64 of 224 grids (28.6%) which are typical fertilizer rates farmers apply for corn in Iowa. Based on model predictions, grid-level nitrogen fertilizer management used lower amounts of nitrate, produced higher yields and was more profitable than either transect- or field-level (single rate) fertilizer application. In another study, four factors affecting soybean yield variability namely, water stress, soybean cyst nematode (SCN), soil pH, and weeds, were examined in each of 100 grids within a 20-hectare field in Perry, Iowa using the CROPGRO-Soybean model. Average estimated yield loss due to the combined effects of water stress, SCN, pH, and weeds in each 0.2-hectare grid was 842 kg ha−1. Water stress had the biggest impact on soybean yield with an average yield reduction of 626 kg ha−1. Yield impact and economic consequences of three strategies namely, variable plant population density (PPD), soybean cyst nematode (SCN) resistant and susceptible varieties, and irrigation management schemes, were evaluated using 34 years of weather data. Implementing the best PPD for each year produced higher grid-level soybean yield and net return compared to using the 34-year average optimum rate. Achieving maximum net return may not be possible on a yearly basis due to uncertainties in weather condition. Using a SCN-resistant variety resulted in significant yield increase over that of a susceptible variety. Several grids had a significant increase (>350 kg ha−1 ) in average yield with some grids having as much as 995 kg ha −1 (17 bu ac−1) yield increase when a SCN-resistant variety was used. Irrigating when available soil moisture reached a value of 40% and 50% significantly increased average field-level soybean yields by 1585 and 1619 kg ha−1, respectively. Excluding the cost of equipment, irrigation would significantly increase net return.