Apple fruitlet drop prediction using spectral Vis/NIR and growth diameter times-series
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
In apple production different thinning strategies are applied to produce optimal yields of marketable fruit and to reduce the risk of alternate bearing. However, the effectiveness of a chemical thinning treatment is highly variable being influ- enced by the tree’s physiological state and environmental conditions both before and after application. Optical sensing with a visible/near-infrared spectropho- tometer is tested as a prediction tool to decide if a further thinning treatment is required and precisely thin trees in the orchards. In 2023, in Southwest Ger- many and in Northern Italy, the apple cultivars ‘Gala’ and ‘Braeburn’ were thinned between the 8-15 mm fruit diameter growth stage. King and second lateral fruitlets, which were the target of the thinning, were measured with dig- ital calipers and scanned before and up to 13 days after thinning treatments. Thinning treatments effectively reduced fruit set by one half. Receiver operating characteristic curves were tested separately on spectral indices and fruit diame- ter growth. Spectral monitoring recorded similar information as measuring fruit diameter growth. The obtained area under the curve was > 75% on 5-8 days after treatment for both cultivars. Persisting fruitlets showed higher light absorption in the chlorophyll and water related wavelength regions and higher light reflectance from the near-infrared region (699 to 939 nm).