On the impact of arctic wetlands on the climate system: Model sensitivity simulations with MIROC5 AGCM and the simplified wetland scheme

Abstract:

There are many wetlands and lakes in the Arctic. They influence the surface water and energy budget, surface hydrology, and hence climate. In the present study, we examine a simplified wetland scheme that stores part of snowmelt with MATSIRO land surface model in a climate model, MIROC5. We conduct AGCM experiments using MIROC5 with climatological monthly SST and sea ice boundary conditions from 1979 to 1996 with and without the simplified wetland scheme. First, we compare simulated climatology of soil moisture in JJA with satellite based observation data. The result shows that eastern Eurasia has wet bias, and western Eurasia and North America except Alaska have dry bias; this might be caused by precipitation biases. The simulated soil moisture with the wetland scheme increases in high latitudes, about 0.2 for the most affected regions. Then, we evaluate simulated precipitation and surface air temperature using JRA55 reanalysis dataset. In the experiment without the wetland scheme, warm bias exists in most of the land area. By introducing the wetland scheme, the warm bias decreases because the latent heat flux increases and the sensible heat flux decreases. Simulated precipitation without the wetland scheme shows similar bias in soil moisture. It decreases in western Eurasia with the wetland scheme. Further, we evaluate the land-atmosphere coupling strength using rank correlation coefficient of soil moisture and lifting condensation level. The result shows that the effects of wetlands reduce the coupling strength.