Predicting phenological development for Australian wheats

Dates of ear initation and anthesis were recorded for 16 wheat cultivars at a wide range of sowing dates in four field experiments conducted over four years.In general for the majority of cultivars number of days from sowing to ear initiation increased as sowing was delayed through May and then declined with sowings after June. The effects of sowing date and cultivar on anthesis were similar to those observed for ear initiation. Maximum time to anthesis was observed from sowings in early May.A linear regression model relating rate of development to mean temperature and photoperiod accounted for 47-98% of the variation in rate of development from sowing to ear initiation and from 68 to 98% of the variation from ear initiation to anthesis. A five-parameter non-linear model was also tested but was not superior. Observations in a single year were sufficient to characterize a cultivar provided the range of mean temperature and photoperiod was large.Comparison with data from other field sites of ear initiation and anthesis showed that the regression equations gave a good fit to the occurrence of these events when used in the incremental sense, that is, by summing increments of development rate calculated from daily temperature and photoperiod.The prediction model is discussed in relation to its application in simulation models of crop growth, analysis of cultivar adaptation to environments and in day-to-day crop management.