A degree-day model for regional prediction of first occurrence of frit flies in oats in Sweden

Abstract

A degree-day model for regional prediction of first occurrence of frit flies, Oscinella frit (L.), in oat fields in Sweden was developed. The model parameters were calibrated using water trap and temperature data from one province. Another set of data from the same and other provinces was used to validate the predictions of the model. Different methods of degree-day calculation, based on either air or soil minimum temperatures, were compared and the least variation of temperature sums associated with date of first occurrence was found using actual degree-days calculated from air temperature data. In the validation of the mode, median date of first occurrence could be predicted with an error of less than five days in 12 of 15 province-year combinations. The mean predictive errors differed between years and provinces and also, in one case, due to choice of weather station within region. Based on these results we conclude that the degree-day model predicts regional occurrence of frit flies with reasonably small errors and that it will be an important part in the development of a decision support system for control of frit flies.