Dynamic crop growth improves regional climate simulation in continental United States

The land surface exerts control on climate through surface energy fluxes, which are largely determined by the surface parameters, such as leaf area index. The prescribed leaf area index in many climate models ignores the environmental controls on vegetation growth and resulting interannual variability in leaf area, which could introduce biases into climate simulations. For example, previous simulations using a newly coupled regional climate model (WRF3.3-CLM4) showed the strongest warm bias in the Midwest due in part to the underestimated crop leaf area index. In this work, we incorporated the dynamic crop growth module in CLM4 and created a new version of a coupled regional climate model (WRF3.3-CLM4crop), where the crop leaf area index, stem area index, and canopy height are updated every time step. We set up a 5-year (2002-2006) simulation using WRF3.3-CLM4crop and compared to the previous simulation that used prescribed LAI, the leaf area index increased by 90% in annual average and the site level validation showed the crop growth module better captured the monthly variation of LAI and reduced the LAI bias by 30%. The Large warm bias in the Midwest was reduced by 1-2K in annual average as a result of increased latent heat flux that generated by the higher LAI. The increased latent heat flux also played an active role in modifying the Midwest water cycle. The precipitation and soil moisture increased by 9% and 5% respectively than the simulation with prescribed LAI. Our results suggested that incorporate dynamic crop growth scheme into climate model could improve not only the surface energy fluxes but also the water cycle simulations.