Wu, HH; Fu, CS; Wu, HW; Zhang, LL (2020). Plant Hydraulic Stress Strategy Improves Model Predictions of the Response of Gross Primary Productivity to Drought Across China. JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES, 125(24), e2020JD033476.

The response of the terrestrial carbon cycle to droughts, which have recently increased in incidence and severity, is a hot topic in research, but the role played by vegetation water use strategies remains unclear. Both the soil moisture stress (SMS) and plant hydraulic stress (PHS) strategies are used in land surface models, but their performance in simulating the terrestrial carbon cycle and its response to droughts in various climatic zones remains uncertain. In this study, we used Community Land Model Version 5 to simulate gross primary productivity (GPP) across four different climatic subregions of China during 2000-2015 to investigate the roles played by PHS and SMS in the response of GPP to dry events and dry or rainy years in various climate zones. Simulated GPP with the two strategies were overestimated by 2.0-2.3% for the entire China compared to Moderate Resolution Imaging Spectroradiometer (MODIS) data. GPP simulated using PHS configuration fit eddy covariance and MODIS data better than that obtained using SMS across China, except for the Qinghai-Tibet Plateau. PHS increased simulated GPP by 0.10-0.63 g C m(-2) day(-1) during dry events across different climatic regions because of root hydraulic lift but decreased GPP by 0.37-0.45 g C m(-2) day(-1) during growing season in the humid southeastern region because of high atmospheric vapor pressure deficit and root hydraulic descent. Simulations forced by data collected locally in China performed better than those forced by CRUNCEP7 global data at the site but worse at the regional scale, in particular for the Qinghai-Tibet Plateau.