STUDY THE APPLICABILITY OF GLOBAL MODEL IN SEASONAL FORECASTING FOR VIET NAM

Abstract

This article presents some experimental results of application of Conformal-Cubic Atmospheric Model (CCAM) in seasonal forecasting for the global and regional Viet Nam. The results of forecast in period 2012-2014 with different boundary conditions (sea surface temperature - SST) were analyzed and assessed, source SST forecast from global climate model CFS of the National Centers for Environmental Prediction (CCAM_CFS), and SST from the International Research Institute for Climate and Society (CCAM_IRI). The results show that model has well predicted large-scale circulation all over the global compared to CFS analysis data and CFS forecast for both cases. Over Viet Nam, the forecast of temperature tends to be lower than observed for most of the months, the error in winter months is greater than in summer months, the error in the north is greater than in the south. The average error of lead time 3 months of both cases is about 2.5^oC, the error is quite stable and systematic. For rainfall, the forecast is more variability and tends to be lower in the rainy season. Relative error in the rainy season is about 35-70% in the northern climatic regions, the southern part is smaller. The predictive skill of temperature of CCAM_IRI gives better CCAM_CFS and at medium lead times are better than short and long. But with rainfall, the skill in short lead times are better than others and skill of CCAM_CFS is significantly better.