Moistening and heating effects of convection simulated in the JMA global NWP model

A convection scheme in the global numerical weather prediction (NWP) model plays important roles not only to produce good forecast fields but also to make good first guess fields for a data assimilation system. Some global NWP models have a so-called spindown problem associated with a data assimilation cycle, which is caused by a rapid decrease of humidity (thereby, column integrated precipitable water vapor) accompanied by a strong precipitation. Because the problem seems to be closely related with convective precipitation, tackling the spindown problem in the light of a data assimilation system is an important approach to evaluate a convection scheme in a global NWP model. On the other hand, moistening and heating effect of shallow convection often interacts with a deep convection and contributes to precipitation in the diurnal cycle and the large-scale atmospheric mechanisms. The balance of occurrence between shallow and deep convections, therefore, is also one of the important check points in developing a better global NWP model. In this study, the role of convection is focused on investigating the forecast performance of the global NWP model of the Japan Meteorological Agency (JMA) under the KAKUSHIN project (funded by the Japanese Ministry of Education, Culture, Sports, Science and Technology), and the results will be discussed.