Influence of projected snow and sea-ice changes on future climate in heavy snowfall region

Snow/ice albedo and cloud feedbacks are critical for climate change projection in cryosphere regions. However, future snow and sea-ice distributions are significantly different in each GCM. Thus, surface albedo in cryosphere regions is one of the causes of the uncertainty for climate change projection. Northern Japan is one of the heaviest snowfall regions in the world. In particular, Hokkaido is bounded on the north by the Okhotsk Sea, where is the southernmost ocean in the Northern Hemisphere that is covered with sea ice during winter. Wintertime climate around Hokkaido is highly sensitive to fluctuations in snow and sea-ice. The purpose of this study is to evaluate the influence of global warming on future climate around Hokkaido, using the Pseudo-Global-Warming method (PGW) by a regional climate model. The boundary conditions of the PGW run were obtained by adding the difference between the future (2090s) and past (1990s) climates simulated by coupled general circulation model (MIROC3.2 medres), which is from the CMIP3 multi-model dataset, into the 6-hourly NCEP reanalysis (R-2) and daily OISST data in the past climate (CTL) run. The PGW experiments show that snow depth significantly decreases over mountainous areas and snow cover mainly decreases over plain areas, contributing to higher surface warming due to the decreased snow albedo. Despite the snow reductions, precipitation mainly increases over the mountainous areas because of enhanced water vapor content. However, precipitation decreases over the Japan Sea and the coastal areas, indicating the weakening of a convergent cloud band, which is formed by convergence between cold northwesteries from the Eurasian continent and anticyclonic circulation over the Okhotsk Sea. These results suggest that Okhotsk sea-ice decline may change the atmospheric circulation and the resulting effect on cloud formation, resulting in changes in winter snow or precipitation. We will also examine another CMIP3 model (MRI-CGCM2.3.2), which sensitivity of surface albedo to surface air temperature is the lowest in the CMIP3 models.