No abstract

Abstract. The rainfall over Bangladesh during the 1995 summer monsoon season has been investigated in terms of the intraseasonal variation of monsoon activities. The rainfall over Bangladesh is basically dominated by the north-south oscillation of the monsoon trough. The rainfall increases when the monsoon trough is located at the foot of the Himalayas, because synoptic-scale convective activity is much more vigorous to the south of the monsoon trough axis than to the north of it. In addition, the strong southwesterly wind to the south of the monsoon trough intensifies local convective activity owing to the effects of the orography to the north and east of the country. It is also found that the monsoon rainfall over Bangladesh in 1995 varies with a periodicity of -20 days, and this rainfall variation is closely associated with synoptic-scale monsoon activities spreading over South and Southeast Asia. The active/break cycle of the rainfall variation during the 1995 summer monsoon season can be mostly explained by the northward propagation of what is called the 10-20 day variation of monsoon activities.

Wind resource assessment for coastal areas requires accurate wind speed simulation using a mesoscale model. Our previous study found that the annual mean wind speed simulated by the advanced research Weather Research and Forecasting (WRF-ARW) model has a remarkable positive bias in the lower part of the planetary boundary layer (PBL). This result was obtained from a comparison with wind profiler measurements at Mihama, which is an observation station of the WInd profiler Network and Data Acquisition System (WINDAS) operated by the Japan Meteorological Agency (JMA). In this study, we examine whether such a positive bias can be seen at other WINDAS stations from a comparison of the WRF-simulated wind speed profile using the Mellor-YamadaJanjic (MYJ) PBL scheme with wind profiler measurements at ten WINDAS stations. The results show that the positive bias is found at all stations, and, moreover, that the positive bias is unlikely to be caused by either error in wind profiler measurement or the objective analysis data input into WRF. In addition, we compare the wind speed profiles simulated by WRF with seven different PBL schemes for a month. The result shows that the positive bias cannot be simply reduced by using other PBL schemes.