Simulation of Surface Temperature in Southeast Asia during the Southeast Asian Southwest Monsoon Using RegCM4

摘要

The performance of RegCM4 in simulating the surface temperature in Southeast Asia (SEA) during the Southeast Asian southwest monsoon using six different cumulus parameterization schemes (CPSs) was assessed in this study using the model output from the Southeast Asia Regional Climate Downscaling/ Coordinated Regional Climate Downscaling Experiment Southeast Asia (SEACLID/CORDEX-SEA) project. The simulated surface temperatures were evaluated against the surface temperature from the CRU observation dataset from year 1990 to 2008. The assessment shows that all the simulations produced an overall cold bias in SEA. The cold bias is the strongest in the mountain region of west Papua, followed by that over the northeastern region of Thailand. Generally, the cold bias over the maritime continent is found to be much weaker compared to that of Indo-China mainland. As the bias is not unique to any simulations, the bias simulated is not very much CPSs dependence. It is also noticed that the model has the tendency to underestimate the maximum surface temperature (Tmax) explaining the systematic cold bias produced by the model. Also, the model tends to overestimate the minimum surface temperature (Tmin). The overestimation of Tmin is slightly higher over the maritime continent, explaining the weaker cold bias in surface temperature observed over that area. Through assessing the simulations bias and the simulated spatial pattern agreement, we noticed that the reasonable choice of CPS to be used has to have Emanuel set on land when the simulation of southwest monsoon surface temperature is concerned. Nevertheless, the ensemble standard deviation calculated across all the simulations reveals that the surface temperature over SEA is less sensitive to the choice of CPS used, especially at the maritime continent. Comparatively, the surface temperature over Indo-China mainland is more sensitive to the choice of CPS used.