E2.5 The evaluation and global offline simulations using ELM v1

Abstract

Nitrogen-enabled models in CMIP5indicated that the inclusion of nitrogen (N) cycle reduces CO₂fertilization effect and warming-induced carbon (C) loss from land ecosystems. None of the CMIP5 models, however, has considered phosphorus (P) as a limiting nutrient, which is generally considered as the most limiting nutrient in productive lowland tropical forests. One of the major developments in ELM v1 is the introduction of a prognostic P cycle and C-N-P cycle interactions. Here we demonstrate that the P-enabled ELM v1 is able to better capture the productivity and biomass gradient in Amazon tropical forests. We also provide results on the performance of ELM v1 using the ILAMB benchmarking system. Particularly we show that the inclusion of P dynamics and C-N-P interactions improved simulated historical land carbon sink.  We have performed a set of offline global-scale simulations using ELM v1 constrained by realistic maps of phosphorus distribution to examine the influence of including phosphorus cycle dynamics and C-N-P interactions on C-climate feedbacks.