Northeasterly winds blowing from high latitudes of the North Pacific in boreal summer, including the Sea of Okhotsk and the Bering Sea, are called the “Yamase” in Japanese. The Yamase brings unusually cold and cloudy summers over northeastern Japan, in contrast to climatological southeasterly winds, having great impact on agriculture and life in the region. Therefore, future changes of the Yamase, which may be caused by global warming, are a major concern.
This study is the first attempt to investigate future changes of Yamase frequency, which are defined using 10day mean surface winds, by analyzing eighteen coupled atmosphere-ocean general circulation model (AOGCM) experiments for May to August in the CMIP3 archives. We first assess present-day climate experiments (1981-2000, 20C3M) and then examine the changes in future climate (2081-2100, SRES A1B scenario). In the present-day climate, an eighteen multi-model ensemble (MME18) mean modestly reproduces seasonal variation of the Yamase frequency, although each model generally underestimates the Yamase frequency compared to the reanalysis data and large differences are seen among the models. In the future climate, most models project increases of the Yamase frequency in August in contrast to decreases of the frequency in May, whereas projected frequency changes in June, July and May to August (MJJA) are inconsistent among the MME18. Inter-model comparison suggests that weakening of mean tropical circulation, including the Walker circulation, may contribute to the increased Yamase frequency in August. A projection employing only nine of the models with higher skill (MME9hi), based on a defined metric, is also tried. Negative anomalies in June over the Sea of Okhotsk and eastern Siberia in the mean sea level pressure field are contrasted with positive anomalies in July, which are unclear in the MME18 projections. In August, almost all the MME9hi project increase of the Yamase frequency, consistent with the MME18 projections.