Weather conditions significantly affect corn yields. While weather remains as the major uncontrolled variable in crop production, an understanding of the influence of weather on yields can aid in early and accurate assessment of the impact of weather and climate on crop yields and allow for timely agricultural extension advisories to help reduce farm management costs and improve marketing decisions. Based on data for four representative counties in Indiana from 1960 to 1984 (excluding 1970 because of the disastrous southern corn leaf blight), a model was developed to estimate corn (Zea mays L.) yields as a function of several composite soil-crop-weather variables and a technology-trend marker, applied nitrogen fertilizer (N). The model was tested by predicting corn yields for 15 other counties. A daily energy-crop growth (ECG) variable in which different weights were used for the three crop-weather variables which make up the daily ECG-solar radiation intercepted by the canopy, a temperature function, and the ratio of actual to potential evapotranspiration-performed better than when the ECG components were weighted equally. The summation of the weighted daily ECG over a relatively short period (36 days spanning silk) was found to provide the best index for predicting county average corn yield. Numerical estimation results indicate that the ratio of actual to potential evapotranspiration (ET/PET) is much more important than the other two ECG factors in estimating county average corn yield in Indiana.

One reason for the slow implementation of conservation tillage methods on the nation's farms is the perception that results from plot-level research at the agricultural experiment stations-presently the primary information source for interpreting the relative crop yield potential of alternative tillage systems-may not be representative. In this study, data for corn (Zea mays L.) production from 78 sites in Wabash County, Indiana, were obtained from cooperating farmers as part of their normal farming operations during 1983. The data were analyzed with simulation models, regression techniques, and partial budgeting. The average farm corn yields, and hence net returns, observed from different tillage systems appeared to be significantly different, corn on ridge-till sites yielding substantially more than corn on conventional and no-till sites. After differences among sites in precipitation and soil characteristics-expressed as a single moisture stress variable-and applied N and P fertilizers were taken into account, however, there were few yield differences among tillage systems, reconciling the farm findings with those from universities. The moisture stress variable (S) was the sum of 90 daily ratios of actual-to-potential evapo-transpiration, from 39 d before corn silking to 50 d after. The study also provided empirical evidence that, as moisture stress increased, net returns to those corn producers using no-till systems increased over those obtained for other tillage systems. Although the extensive data gathering was done in only 1 yr, that year (1983) provided as much range in the moisture stress variable (S = 15 to 45) and final corn yields (7 to 160 bu/acre among sites) as might be expected from a single soil plot over 10 or more years in Indiana. Therefore, the results are believed to provide a representative view of the yield-moisture stress relationships, especially at the high moisture stress end.

Public agencies are likely to enact policies aimed at reducing agricultural contributions of nitrates to groundwater. These policies are likely to be aimed at N application rates and methods. Current application methods and rates relative to crop needs would be a primary determinant of how farmers respond to particular policies and the costs of those responses. A budget model is used to estimate both aggregate and farm level N needs, sources, and deficiencies or excesses. The model is applied to the six-county karst area of southeastern Minnesota (SE MN), because of its intensive crop and livestock production and demonstrated susceptibility to groundwater contamination. The model also is estimated for Fillmore County, within the six county area, and four individual farms in Fillmore County. The four individual farms include a mixed dairy and crop operation, two livestock-crop operations, and a continuous-corn farm. Secondary data and direct interviews were the primary data sources in the aggregate and farm level applications, respectively. Nitrogen sources exceeded crop needs in Fillmore County and SE MN by ≥ 50 lb/corn acre. Individual farm estimates indicate highly variable N applications across farms. The continuous-corn farm exhibited a close balance between crop needs and N applications, the non-dairy livestock operators had excesses of 20 to 60 lb/acre, and the dairy operator's sources exceeded crop needs by about 130 lb. Due to the indicated concentration of “excess” N applications on farms with livestock, policies aimed at reducing N applications and otherwise improving N management should consider manure and legume sources, as well as commercial fertilizer.