Explaining rice milling quality variation using historical weather data analysis

Rice quality, specifically head rice yield (HRY), can vary inexplicably from one lot to another, and from year to year. In an effort to correlate air temperatures during various growth stages to HRY, growth staging data expressed in degree day units was used to predict the occurrence of sequential growth stages within a set of 17-year historical data, which included HRY and 50% heading dates for two long-grain rice cultivars, (Otyza sativa L) Newbonnet and Lemont, and area weather data. HRY was most strongly affected by the average daily low temperature (or nighttime temperature) during the R8 developmental stage. Lower HRY were associated with high nighttime air temperatures during this stage for both Newbonnet and Lemont. When used as a single variable in a regression model, the nighttime temperature during the R8 developmental stage explained over 25% of the variation in HRY.