Satellite-based detection of apple proliferation and pear decline

Apple proliferation (AP) and pear decline (PD) are important diseases of fruit trees in Europe that are difficult to control. Timely uprooting of infected trees is a successful measure to limit the spread of the diseases and is therefore mandatory in some production areas. Large-scale monitoring strategies are required. AP and PD induce in infected trees an early leaf reddening in late summer or early autumn that is suitable to be monitored by remote sensing. Very high-resolution satellite images covering the region of reference orchards were analysed with machine learning methods to identify apple and pear orchards and AP-and PD-infected trees. High accuracy of apple and pear orchard identification was possible with values from 0.98 to 0.99. Symptomatic trees were identified in multispectral data of the WorldView satellites with a spatial resolution of 15 cm x 15 cm. Vegetation indices which combine information on the biomass and pigment proved to be suitable for detecting both AP and PD at tree level. Vitality maps and diagnostic maps are two targeted products out of the current study. While the diagnostic map reveals trees with phytoplasma-induced symptoms such as red leaves at a specific time point, the vitality map also causally records other vitality parameters of the trees in a time series. © 2025, Technology Society of Basic and Applied Sciences.