Impact of Tropical Initial Water Vapor from MT-SAPHIR Observations on Medium-Range Forecasts Using the KMA Operational Model

Abstract

The proper representation of water vapor helps set accurate initial conditions in numerical weather prediction (NWP) models. Water vapor levels have been observed over the tropics by the Sounder for Atmospheric Profiling of Humidity in the Inter-tropics by Radiometry (SAPHIR) on the Megha-Tropiques (MT) satellite. To determine how tropical convection, which is relevant to water vapor levels, can affect medium-range forecasts in both the tropics and extratropics, we compare the differences between simulations with and without MT-SAPHIR data using the operational NWP model from the Korea Meteorological Administration (KMA), which includes a hybrid four-dimensional variational process in data assimilation. More observed water vapor in the lower tropical troposphere leads to stronger tropical convection, which is well represented in the 300 hPa velocity potential, and strengthens mean meridional circulation. Root-mean-square errors (RMSEs) for relative humidity, temperature, and wind fields in the tropics are reduced by up to 7%. This improvement enhances the forecast field in the middle latitudes. In particular, RMSEs are reduced by up to 3.5% in the humidity field in the extratropics. Overall, the correct simulation of tropical water vapor via MT-SAPHIR data improves medium-range forecasts in the extratropics.