A study on the impact of dynamic open water surfaces by river inundation in a global climate model

Abstract:

Current climate models do not take into account the role of open water surfaces on the global water cycle. In spite of the small area coverage of rivers and floodplains, the impact of these water surfaces to various hydrologic processes should not be overlooked. Therefore, in this study, we focused on implementing river inundation processes from the state-of-the-art river model, CaMa-Flood, into the climate model, MIROC5. The two models were coupled using Multi-Program Multi-Data(MPMD), thereby enabling independent runs and reduced computational time. In the coupled model, the surface area of rivers and daily variation of floodplain areas were considered as a fraction of the land surface in each grid.

The implementation of these increased water surfaces resulted in large changes in variables such as latent heat flux, precipitation and temperature. Particularly, precipitation and latent heat flux tend to have a positive correlation with water surfaces, while surface air temperature had a negative correlation. The surface air temperature showed a decrease in high latitudinal areas during the summer months of June to August when floodplains increase, thereby weakening a common known bias of high temperature in atmospheric models. The results from this study suggest the necessity of appropriate consideration of open surface waters in a climate model and the need for a more detailed high-resolution analysis of the impacts mentioned above.