A national extent map of cropland and grassland for Switzerland based on Sentinel-2 data

[摘要] Agricultural landscapes support multiple functions andare of great importance for biodiversity. Heterogeneous agricultural mosaicsof cropland and grassland commonly result from variable land use practicesand ecosystem service demands. Switzerland's agricultural land use isconsiderably spatially heterogeneous due to strong variability inconditions, especially topography and climate, thus presenting challenges toautomated agricultural mapping. Nationwide knowledge of the location ofcropland and grassland is necessary for effective conservation and land useplanning. We mapped the distribution of cropland and permanent grasslandacross Switzerland. We used several indices largely derived from Sentinel-2satellite imagery captured over multiple growing seasons and parcel-basedtraining data derived from landholder reporting. The mapping was conductedwithin Google Earth Engine using a random forest classifier. Theresulting map has high accuracy in lowlands as well as in mountainous areas.The map will act as a base agricultural land cover dataset for researchersand practitioners working in agricultural areas of Switzerland andinterested in land cover and landscape structure. The map as well as thetraining data and calculation algorithms (using Google EarthEngine) are freely available for download on the EnviDat platform https://doi.org/10.16904/envidat.205 (Pazúr et al., 2021).

[学科分类] 眼科学