Detecting olive quick decline syndrome: A satellite-based dataset for a case study in Apulia Region.

Abstract

The bacterium Xylella fastidiosa (Xf) is a plant pathogen first identified in Europe in 2013, specifically in olive groves in the Apulia region (south-eastern Italy). It is now spreading across the Mediterranean basin and poses a serious threat to the local economy by causing branch desiccation and the rapid death of olive trees, a condition known as olive quick decline syndrome (OQDS). Several studies have investigated the potential of remote sensing (RS) technology to monitor OQDS over time and space; however, accurate and reliable data on OQDS occurrence remain scarce. To enhance the distribution data of Xf-infected trees in the Apulia region, we investigated an infection hotspot of 25 km² area in the province of Brindisi, where records of infections were documented in 2019 and 2020. Three very high resolution, commercial WorldView-2 images were acquired and segmented, resulting in a dataset of 76637 olive trees. Through visual interpretation, 2340 trees were identified most likely as either infected or removed due to OQDS. This dataset provides a valuable resource for developing or validating RS techniques for early detection of OQDS. Furthermore, it could support studies aimed to evaluate spectral bands or indices most correlated with infection presence. Finally, the dataset can be integrated with other Xf-infection presence data to support species distribution model studies.

Plant phenotyping relevance

衛星画像のセグメンテーションと感染・枯死オリーブ樹のラベル化による、植物病害状態の検出・検証用データセットが研究の中心である。

Code and data availability

The paper is a Data in Brief article describing a public Figshare dataset (OQDS-Insight) containing WorldView-2 satellite raster imagery (RGB and NDVI GeoTIFFs) and a shapefile of 76,637 olive tree points with OQDS infection labels — directly the paper's phenotyping measurements.