Nitrogen Rate‐Yield‐Precipitation Relationships and N Rate Forecasting for Corn Crops1

Publisher Website
Google Scholar
Abstract
The purpose of this study was to determine relationships between precipitation, N rate, and yield, and to develop a criterion that would permit to forecast in the spring the most probable optimum N rate. The study was carried out on a chernozem soil, in a continental temperate climate, characterized by an average annual precipitation of about 500 mm that vanes greatly between years and between seasons. Regression analysis of 14 years of data has shown that optimum N rate was more highly correlated with November to February inclusive; accumulated precipitation (r=0.77**) than with May to August precipitation (r=0.41). The multiple correlation coefficient of optimum N rate and the two precipitation periods was highly significant (r=0.94**). In years with roughly normal summer precipitation, November to February precipitation was more important (I=0.86**). In years with less than 100‐120 mm in November to February, the applied N was less efficient even when the summer precipitation was greater than normal. The May to August precipitation had a higher direct effect on the N rate only in wet winter years, (with precipitation of over 200 to 220 mm in November to February); here, the simple correlation of the N rate with the winter precipitation decreased (r=0.50) and that with May to August precipitation increased (r=0.84**; 0.01 level), but the interaction effect of winter and summer precipitation was still significant (R=0.89**). This suggests that in all cases, winter precipitation can be used to predict the optinum rate of N fertilizer. To calculate the optimum N rate in March,the following equation is proposed: N rate = 19.53 + 0.2854X1 + 0.0066X1X2 (R=0.90) If the November to December precipitation exceeds 220 mm and the spring continues to be wetter than normal, the N rate can be calculated again in June, using the following equation: N rate = 28.37 + 0.0021X1X4 + 0.0032X3X4 ‐ 0.2538X4 (R=0.97) where: X1, X2, and X4 are the November to January, February and May‐ July precipitation (mm), respectively. Estimating the N rates in March and reestimating in June with the above two regressions gives an average yield increase of 69% and a profit 200% greater than with annual application of the long‐term average N rate. The yield increase was also more highly correlated with November to Februay (r=0.63; 0.05 level) than with May to August precipitation (r=0.59*; 0.05 level), and the interaction effect of these two variables was high (r=0.92**; 0.01 level). Yields can be predicted in summer using both winter and summer precipitation.
Keywords

This publication has 0 references indexed in Scilit: