Reduction of the summer surface warm/dry bias in the FV3GFS by improving the evaporation from the bare soil (47)

Helin Wei, IMSG at NOAA/NCEP/EMC, College Park, Maryland

Weizhong Zheng, IMSG at NOAA/NCEP/EMC






The proper partition of the surface incoming energy into latent heat and sensible heat fluxes (Bowen ratio) is critical to the forecast of surface temperature and humidity. The summer surface warm/dry bias is one of the outstanding issues in the current NCEP operational Global Forecast System (GFS). It was found that too weak surface evapotranspiration cooling could be the major cause. Canopy resistance of some types of vegetation was reduced to increase the plant transpiration in the last GFS implementation. The short-term impact is very positive. The warm/dry bias has been reduced significantly. However, the negative consequence is found after one-year evolution. The increasing transpiration pulled too much moisture from the deep soil layer and the deep soil became too dry which prevented further evapotranspiration, thus reintroducing warm/dry bias.

The latent heat flux includes not only the transpiration from the canopy but also the evaporation from the bare soil. Previous study found the latter was being too low while the former was overestimated in the current Noah land surface model. In this study the formula calculating the bare soil evaporation is revisited. It is found that the key parameter (soil moisture residual) is relatively higher and thus prevents further evaporation. The value of the parameter in the EC model is adapted. The results show the warm and dry biases can be reduced by increasing latent heat flux while the deep soil moisture can keep intact. The revision will be included in the FV3GFS Beta implementation during Q3FY18.