Future change in Southern Hemisphere atmospheric blocking simulated using a 20-km-mesh AGCM

In this study, we focus on Australia-New Zealand (AU; summer and winter) and Andes (AN; winter) blockings simulated using a 20-km-mesh AGCM (TL959L60) for present-day (1979-2003), near-future (2015-2039) and future (2075-2099) climates. For the present-day simulation, the model was integrated with observed historical SST (HadISST). For the future simulation, SST was made by adding change in SST projected by the CMIP3 multi-model ensemble (MME) mean to annual variation of the HadISST. IPCC SRES A1B scenario was assumed for future emission of green house gases. To estimate uncertainty in the future projection of blocking frequency, three-member initial value ensemble simulations were conducted using a 60-km-mesh AGCM (TL319L60). Blocking is detected by meridional gradients for 500hPa geopotential height. Blocking is detected by meridional gradients for 500hPa geopotential height.

In the present-day climate, 20-km-mesh AGCM is in good agreement with the Japanese reanalysis (JRA25) in terms of the AU and AN blocking frequencies. The higher the horizontal resolution of the AGCM, the more accurate the simulated blocking frequency. 20-km-mesh AGCM can simulate the wintertime long-lived AU blocking well.

In the near future climate, no significant decreases in the frequencies of the AU and AN blockings are predicted during both summer and winter. In the future climate, a significant decrease in the AU blocking frequency is predicted during both summer and winter, whereas no significant decreases in the AN blocking frequency is predicted. The decrease in the AU blocking frequency during winter is more remarkable than that during summer. In the future winter, it is possible that long-lived (13>days) AU blocking will disappear altogether.