food-LIFE

This dataset contains spatially aggregated "Land-cover Impacts on Future Extinctions (LIFE)" values at administrative levels 0-2, along with a spatial raster of LIFE scores at 5 arcminute resolution. The units of both the aggregated and spatial data are "mean change in extinction risk across 30,000 vertebrates per km2 agricultural land-use". food-LIFE builds on the original LIFE metholodogy described in Eyres et al. 2025 (https://royalsocietypublishing.org/rstb/article/109696) and uses species range and habitat preference data from the IUCN Redlist (iucnredlist.org) along with spatial agricultural data to estimate the extinction risk impacts associated with agricultural land-use. This work builds on work originally published in Nature Food (https://www.nature.com/articles/s43016-025-01224-w). Subsequent work involved an improvement on the original methodology and a switch to MapSPAM (https://www.mapspam.info/) as opposed to GAEZ (https://gaez.fao.org/) as the input source for underlying crop distribution data. At the time this was due to more recent data availability in MapSPAM, as well as alignment with https://commodityfootprints.earth/; the primary motivation behind this work was the inclusion of LIFE as a metric in commodityfootprints and subsequently a UK biodiversity indicators toolkit. This is the most recent version of the work, which uses the latest version of MapSPAM as well as several pipeline refinements. The code underpinning this analysis is publicly available at https://github.com/thomasball42/food_life_mapspam/tree/working_prehybrid. Aggregations were performed using administrative boundaries from https://gadm.org/ and calculated values are the weighted mean LIFE score per km2 within those boundaries. Downstream applications of this data can be found here: https://github.com/louis-de-neve/FLMP. Please contact Thomas Ball at [e-mail] with any questions about the development of this dataset or appropriate usage.