Modelling the long-term carbon storage of a mixed forest under climate change: a meta-modelling approach

This study presents an up-scaling method that builds a statistical emulator, or meta-model, of the NOTG (Not Only Tree/Grass) forest model to overcome computational constraints of detailed physiological and biogeochemical simulations at large scales. Applied to the mixed-species Font-Blanche forest in south-eastern France, the meta-model enables century-scale projections under future Mediterranean climate scenarios. Results indicate that despite declining spring and summer precipitation, climate change may enhance forest productivity and long-term carbon storage through species interactions. The approach demonstrates feasible, accurate ecosystem predictions over extended periods.