New citizen science initiative enhances flowering onset predictions for fruit trees in Great Britain

Abstract

Accurately predicting f lowering phenology in fruit tree orchards is crucial for timely pest and pathogen treatments and for the introduction of managed pollinators. Making predictions requires large datasets of f lowering dates, which are often limited to single locations. Consequently, the resulting phenology predictions are not representative across larger geographic areas. Citizen science may offer a solution to this data gap, with millions of biological records across a wide range of taxa recorded annually. Here, a new citizen science platform called ‘FruitWatch’ is introduced, monitoring the f lowering dates of fruit trees in Great Britain. The objectives of this study are to assess the suitability of FruitWatch submissions to (i) detect latitudinal variation in f lowering onset dates, (ii) parameterize existing phenology modelling frameworks, and (iii) make predictions of f lowering onset dates across Great Britain for a single year. Using data for four cultivars from 2022, linear models reveal significant latitudinal delays in f lowering onset of as much as 1.49 ± 0.63 days per degree latitude further north (Pear ‘Conference’), with significant delays also seen in Cherry ‘Stella’ (1.39 ± 0.48 days) and Plum ‘Victoria’ (1.22 ± 0.18 days). FruitWatch informed phenology modelling frameworks performed well for predicting f lowering onset, with root mean square error values of predictions from validation datasets ranging between 4.6 (‘Victoria’) and 8.0 (‘Conference’) days. The parameterized models also provided realistic f lowering onset predictions across Great Britain in 2022, with earlier f lowering dates predicted in warmer areas. These findings demonstrate the potential of citizen science data to offer growers cultivar- and location-specific phenology predictions to help inform orchard management.