This study explores the effects of vegetation and groundwater dynamics on the precipitation during the North American warm season through sensitivity studies by using a coupled land-atmosphere model, the Weather Research and Forecasting (WRF) model with the Noah land surface model (LSM). A dynamic vegetation model and a simple groundwater model (SIMGM) are implemented into the unified Noah LSM. Three sensitivity experiments are conducted with time-varying sea surface temperatures: 1) the vegetation fraction is prescribed according to the value on the initial time of the model integration ; 2) vegetation dynamics as represented by the dynamic vegetation model in the Noah LSM; and 3) based on 2) but with groundwater dynamics as represented by the SIMGM.

Results show that the inclusion of the dynamic vegetation improves the July precipitation in the Northern North American Monsoon System (NNAMS) region. The simulation is further improved with the SIMGM in the Noah LSM and more peak precipitation in the NNAMS and Southern Great Plains are captured. The default WRF/ Noah LSM produces much less precipitation in the NNAMS and the Central U.S. The increased precipitation simulated by the experiments 1) and 2) agrees well with the increase of vegetation fraction. In addition, the implementation of the SIMGM induces higher latent heat flux in the Central U.S. and the NNAMS, resulting in more precipitation. These sensitivity experiments demonstrate the need for augmented representations of vegetation and groundwater dynamics in the current Noah LSM.