LARS Tech Report Number
Knowledge of when critical crop stages occur and how the environment affects them should provide useful information for crop management decisions and crop production models. This research evaluated two sources of data for predicting dates of silking and physiological maturity of corn (Zea mays L.). Initial evaluations were conducted using data of an adapted corn hybrid grown on a Typic Agriaquoll at the Purdue University Agronomy Farm from 1979 to 1981. The second phase extended the analyses to large areas using data acquired by the Statistical Reporting Service of USDA for crop reporting districts (CRD) in Indiana and Iowa from 1969 to 1980. Several thermal models were compared to calendar days for predicting dates of silking and physiological maturity. Mixed models which used a combination of thermal units to predict silking and days after silking to predict physiological maturity were also evaluated. At the Agronomy Farm the models were calibrated and tested on the same data. For each CRD the models were calibrated using 4 or 5 years of data and tested using 7 different years of data.
The thermal models were significantly less biased and more accurate than calendar days for predicting dates of silking. Differences among the thermal models were small. Significant improvements in both bias and accuracy were observed when the mixed models were used to predict dates of physiological maturity. The results indicate that statistical data for CRD can be used to evaluate models developed at agricultural experiment stations.
Date of this Version