GRASSLAND PRODUCTIVITY IN THE COMMUNITY LAND MODEL: BIAS AND CLIMATE SENSTIVITY ACROSS NATIONAL ECOLOGICAL OBSERVATORY NETWORK SITES

Abstract

Grasslands are responsible for a large portion of terrestrial carbon fixation; therefore understanding how grasslands will respond to future global change is essential for predictions about the global carbon cycle. This study uses the Community Land Model (CLM) paired with the NCAR-NEON system at National Ecological Observatory Network sites to evaluate model bias and assess grassland productivity and plant functional type (PFT) responses to future environmental change scenarios including modified CO2, drought, and warming. We found that CLM overestimates gross primary productivity (GPP) at sites classified as C4 and C3 arctic PFTs and slightly underestimates GPP at sites classified as the C3 PFT. Across all PFTs grassland GPP significantly decreased in response to low CO2 and showed a nonsignificant increase in response to high CO2. We imposed two drought treatments of different duration, a year-long drought and a growing season drought and found that both treatments significantly decreased GPP across all PFTs, but the response to year-long drought was stronger. Simulated 2°C warming caused the most variable and smallest GPP response, with a significant decrease for the C3 PFT, a significant increase for the C3 arctic PFT, and no significant response for the C4 PFT. These results highlight both the utility of the NCAR-NEON system for model evaluation and areas where CLM’s representation of PFT-level grassland physiology could be improved.