CRACLE: Inferring Climate from Vegetation

Correlations between plant distributions and the climate and environmental conditions to which species have adapted are well known, and models can predict distributions accurately based on occurrences and climate. This study describes a novel approach for estimating climate from vegetation. Researchers created climate tolerance profiles for more than 4,300 species by pairing 3.75 million GBIF-mediated occurrences with climate data, then using the presence of the same plants in 165 survey sites to predict the local climate. The CRACLE method—‘climate reconstruction analysis using coexistence likelihood estimation’—returned results with a strong linear correlation with actual climate and, when compared to other methods, yields lower error rates in estimating variables such as temperature and precipitation.