Is Soil Moisture a Major Factor That Affects Short-Range Near-Surface Weather Forecasts?

Near-surface weather prediction presents a great challenge in numerical weather prediction. This is mainly due to complicated interactions between land and atmospheric boundary layer and uncertainties in coupling the land and atmosphere in the numerical models. Previous studies have demonstrated that the land surface data assimilation (e.g., soil moisture data assimilation in the most cases) influences numerical weather prediction.

In this study, we conduct comprehensive studies to examine the linkage between soil moisture and near-surface atmospheric variables (e.g., 2-m temperature and humidly and 10-m winds). First, we use in-situ data over 8-year to investigate the correlations between the soil moisture and near surface variables. Second, a single column model was used to evaluate the influences of the changes in soil moisture on numerical forecasts of these near-surface variables. Finally, soil moisture data assimilation was performed with Noah land-surface model to examine the potential influence of soil moisture data assimilation on short range weather forecasts using the mesoscale community Weather Research and Forecasting (WRF) model. Different coupling methods are examined. The preliminary efforts with developing efficient land-atmosphere coupling for the NCEP Next Generation Global Prediction System (NGGPS) are also discussed.