Using Grain Yield History To Predict Future Yield Of Corn and Soybean.

To maximize field productivity and profitability, growers are increasingly using site-specific management rather than whole field management practices. Our objective is to describe spatial and temporal yield variability to predict grain yield of specific land cells. The goal is to determine if yield maps accurately delineate management zones for prescription applications. Grain yield from twenty-six years (experiment 1) of continuous corn (CC), continuous soybean (SS), and corn-soybean rotations (CS) in no-tillage (NT) and conventional tillage (CT) systems and twelve years (experiment 2) of CC and CS rotations with CT and NT tillage treatments were used in the analysis. Across all tillage-rotation systems in both experiments, spatial variability within corn systems was +11 to +27 bu/A, while temporal variability was +30 to +44 bu/A. Within soybean systems, spatial variability was +4 to +5 bu/A and temporal variability was +8 to +13 bu/A. Spatial variability was 5.5 to 15% and temporal variability was 17 to 27% of the average grain yield. Temporal variability was 1.1 to 3.9 times greater than spatial variability. CS-NT soybean had the lowest temporal variability, and CC-NT exhibited the highest temporal variability. Spatial analysis found cells to be significantly different for grain yield and may be ranked within a tillage x rotation combination. CC required the least amount of time to rank cells, and NT required more time than CT. High corn yields did not always predict high soybean yields and vice-versa. Since spatial variation is lower than temporal variation prescription predictions remain challenging.