Development of models for specific crop calendar events

In order to test growth prediction and yield prediction equations on historical weather data for areas where planting dates were not published, or for areas where dates of planting will not be as readily accessible as weather data, there is a need for a model to predict the date of planting. Such a model was developed by regressing reported percentages of the wheat crop planted on various weather variables and their transformations. The model for predicting percentage of wheat acreage planted in North Dakota was tested on data not used in formulating the coefficients of the model. Although the predictions may have errors of several days the model is considered to be an improvement over a system utilizing a constant year-to-year average planting date.The research was supported by funds supplied by the National Aeronautics and Space Administration under contract No. NAS 9-14006.Presented at the Seventh International Biometeorological Congress, 17–23 August 1975, College Park, Maryland, USA.