An Apple Fruit Set Prediction Model From Distributions of Fruitlet Mass Accurately Estimates Abscission in Four Disparate Regions of the United States

Abstract

Crop load management of apple is one of the most important management practices for ensuring consistent annual yields of high-quality fruit. A previously developed fruit set prediction model, coined the FGR (Fruit Growth Rate) was shown to predict apple fruit set within 7 to 8 days of chemical thinning. Although this could aid repeat thinning decisions, the FGR model had not been widely adopted due to the time demand for implementation. The objective of the current research was to develop a user-friendly alternative using the conceptual framework of the FGR model to generate fruit set predictions. Because fruitlet weight was highly correlated to fruitlet diameter, fruitlets were assessed by mass using an automated balance with a data export function to an Excel program. Without the need for repeated measures of fruit growth and manual data entry, the method was tested alongside the FGR model using two commercially important cultivars, Gala and Honeycrisp, in four distinct locations of the United States: Michigan, New York, North Carolina, and Washington. At all sites, post-bloom chemical thinning applications were made over a range of developmental timings according to fruitlet diameter and weather conditions. A given model’s capacity to generate a prediction was considered successful if the accuracy of the fruit set prediction was within ± 10% of the absolute fruit set observed at “June drop.” Over the course of 12 separate experiments, the FSD model (fruitlet size distribution) successfully predicted fruit set in 83% of trials; this was markedly higher than the FGR (67% success rate). However, the FGR model, when accurate, returned a prediction by 5.5 days (±0.7 day), which was 2.8 days earlier than the FSD. When thinners were applied before 6-mm fruitlet diameter, both models returned a prediction when fruitlets would still be considered susceptible to select thinning chemistries. When thinner applications were delayed until later fruitlet development stages, a 2.8-day delay in the prediction could limit repeat applications given increasing fruitlet resistance to thinners with time. Model validation under disparate regions, cultivars, and years produced consistent results and may improve labor-use-efficiency and precision management of apple fruit set, especially in combination with bloom and early fruitlet (∼≤6 mm diameter) thinning.