Development and Testing of a Model to Rapidly Predict Apple Thinner Response

Abstract

Several thinning experiments were initiated in 2003 to test the hypothesis that monitoring fruit growth is an appropriate and accurate method to predict final fruit set early enough to apply supplemental thinners if appropriate. A total of eight thinning treatments were applied in Massachusetts and New York. On the day of thinner application 70 to 100 spurs were tagged on 4-8 trees (replications). All fruit within a spur were individually identified and fruit were measured. At 2 to 3 day intervals fruit diameter was measured at a designated point on the fruit. Growth rate of the fastest growing 20 fruit on the untreated trees was used as the criteria to determine growth rate of fruit that would persist to harvest. A fruit on a treated tree was predicted to abscise if growth rate slowed to 50% or less of the growth rate of the 20 fastest growing fruit on untreated trees. Cold weather in 2003 following thinner application slowed the response time to thinners. Thinning treatments were applied to Delicious, Golden Delicious, McIntosh, and Gala at 7-9-mm stage. BA, carbaryl, and combinations of NAA and carbaryl were used. In Massachusetts accuracy of prediction of final fruit set at 7-11 days after application ranged from 87% to 100% with and average of 95% accuracy compared to final observed drop at the end of June drop in July. In Geneva, N.Y., the temperature was so unseasonably cold following application that prediction of final set at 7 to 11 days after application was between 68% and 79% with an average of 74% accuracy. We conclude that prediction of final fruit set following growth rate of individual fruit shows promise as an accurate predictor of final fruit set early enough to apply supplemental thinners if appropriate.