Markus Drüke

TL;DR: In this paper , a generative adversarial network is proposed to improve local distributions and spatial structure simultaneously by enforcing a physical constraint to preserve global precipitation sums, which can generalize to future climate scenarios unseen during training.

TL;DR: Using a machine learning approach, it is found that remaining errors in simulated forest carbon turnover can be explained with bioclimatic variables, demonstrating the need to improve model formulations for climate effects on vegetation turnover and mortality despite the apparent successful constraint of simulated vegetation dynamics with multiple satellite observations.

TL;DR: In this paper, the authors focus on improving the control of climate and hence fuel moisture content on fire danger in the LPJmL4-SPITFIRE DGVM in South America, especially for the Brazilian fire-prone biomes of Caatinga and Cerrado.

TL;DR: In this paper, the authors couple the well-established and comprehensively validated dynamic global vegetation model LPJmL5 (Lund-Potsdam-Jena managed Land) to the coupled climate model CM2Mc, the latter is based on the atmosphere model AM2 and the ocean model MOM5 (Modular Ocean Model 5).