METEOROLOGICAL AND NUTRITIONAL DATA USED TO PREDICT LOW TEMPERATURE BREAKDOWN IN STORED BRAMLEY'S SEEDLING APPLES.

A multiple regression model using three meteorological variables as explanatory variables (mean daily maximum temperature in June, the difference between the mean daily maximum temperature in August and September, and total rainfall in August and September) accounted for 44% of the variation in Low Temperature Breakdown (LTB) in stored Bramleys Seedling apples from up to 24 orchards over a 5-year period (1981–85). The inclusion of two other explanatory variables (leaf N and fruit P concentration) in the model increased the percentage variance accounted for to 56% and generally improved correlations between observed and estimated values within each year. Although the latter model accurately predicted the average LTB severity in the 1986 Bramley crop, there was little agreement between the observed and predicted values for individual orchards. Possible mechanisms by which the variables used in the models might affect LTB development are discussed.