I've updated the predictions of the world biomes https://t.co/5EKyOPtYQo. Now using 40% more covariates + Ensemble ML (RF, xgboost, deepnet and glmnet). PNV methodology explained in https://t.co/schUfJD3UT. Can be used to derive soil carbon sequestration potential and similar. https://t.co/e7VW7wmw3f
@SCaldararu Even the first part is not novel as such. Potential natural vegetation has been mapped already using (partly) the same data and algorithm. https://t.co/cW5HnQYmAU
@restoreforward We tested some Machine Learning algorithms for predicting forest tree species in Europe https://t.co/schUfJD3UT and the results were promising. Hopefully soon we will be able to provide state-of-the-art species distribution data globally and support regreening of our
@tom_hengl on a role - also our work 'Global mapping of potential natural vegetation: an assessment of machine learning algorithms for estimating land potential' was a top 5 most viewed #PlantScience and #SpatialScience article in @thePeerJ in 2018
Global mapping of potential natural vegetation: an assessment of machine learning algorithms for estimating land potential
https://t.co/JS71Q1RRVu via @thePeerJ
Global mapping of potential natural vegetation: an assessment of machine learning algorithms for estimating land potential https://t.co/SojCDXcXhg https://t.co/rlfEPjnN58