Italy Orchard Map at 12 m Resolution — Wall-to-Wall Tree Crop Classification

Italy Orchard Map at 12 m Resolution — Wall-to-Wall Tree Crop Classification Overview This dataset is a companion product of the peer-reviewed article: Lodato, F., Jayne, R., Santonico, M., Pollino, M., Bellino, F., De Gara, L., & Basso, B. (2025). High-resolution mapping of orchard distribution across Italy. Science of Remote Sensing, 11, 100237. https://doi.org/10.1016/j.srs.2025.100237 It provides a wall-to-wall thematic refinement of the agricultural areas of CORINE Land Cover, mapping the distribution of orchards and permanent tree crops across the entire Italian territory at a spatial resolution of 12 m. The map refers to the end of 2023 (satellite time series 2021–2023). Updated versions covering more recent years are planned for future releases. For all methodological details, please refer to the article above. The raster is provided as int8 (NoData = 0) and has been sieve-filtered with a 10-pixel minimum mapping unit. Classification scheme Value Class Description 0 NoData / Background Non-orchard areas 1 Actinidia Kiwifruit plantations. 2 Citrus Oranges, lemons, bergamot, mandarins and other citrus species. 3 Stone fruit Mainly peach, apricot, plum and cherry. Minor crops such as pomegranate and persimmon may occasionally be included in this class. 4 Apple From intensive production systems of Trentino-Alto Adige to traditional plantings (e.g. Annurca in Campania) and scattered orchards nationwide. 5 Olive The most extensive class, encompassing highly heterogeneous contexts: from intensively managed groves (e.g. Tuscany) to fragmented, semi-managed and diffuse "patchwork" systems (e.g. Liguria, Sicily). A small but non-negligible false-positive rate should be expected for this class. 6 Pear Pear orchards. 7 Vineyard Vineyards; as with olive, this class spans highly heterogeneous management systems. Relation to the published article This raster constitutes a second, revised edition of the map presented in the article above. The classification model was retrained with an improved pipeline using the High Performance Computing Center (HPCC) of Michigan State University. As a consequence, per-class total area figures may differ from those reported in the original publication; the achieved classification accuracies nonetheless make the dataset reliable for regional- to national-scale analyses. Intended use and limitations The dataset is suitable for regional- and national-scale analyses of permanent crop distribution, agricultural statistics support, land-use planning, and environmental modelling. Because one of the objectives of this work was to improve the detection of the small and heterogeneous agricultural parcels that are common in many parts of Italy, the mapping approach prioritises sensitivity. As a result, some crop classes, particularly olive groves and vineyards, may show a slight overestimation due to an increased number of false positives. Conversely, permanent crops temporarily covered by protective nets or plastic films may occasionally be omitted from the map. The dataset is intended for research and informational purposes. Field-scale applications should be supported by additional local verification. The authors make no warranty regarding fitness for a particular purpose. Users are responsible for verifying the information before applying it to operational, regulatory, or commercial decisions. Acknowledgments The authors gratefully acknowledge Timac Agro Italia for funding the PhD scholarship that supported this research, and the Michigan State University High Performance Computing Center (HPCC) for computational resources. License and citation This dataset is released under a Creative Commons Attribution 4.0 International (CC-BY 4.0) license. If you use this dataset, please cite both the dataset (DOI of this Zenodo record) and the original article: Lodato, F., Jayne, R., Santonico, M., Pollino, M., Bellino, F., De Gara, L., & Basso, B. (2025). High-resolution mapping of orchard distribution across Italy. Science of Remote Sensing, 11, 100237. https://doi.org/10.1016/j.srs.2025.100237