Abstract: Based on the daily observational data of fog, minimum temperature, relative humidity and mean wind speed at 591 national meteorological observation stations of China during the period of 1971-2005, the correlation between regional annual fog days and the three climate factors were analyzed for each of the nine fog regions of China. Furthermore, the annual fog days for each region in the first 50 years of the 21st century were projected by using the model output data under the SRES-A2, -A1B, and -B1 emission scenarios provided by IPCC AR4. Results show that the projection results of regression equations of annual fog days for 2001-2005 are so good that these equations can be used for long-term projection. The projected results indicate that fog days in the first half of the 21st century will overally and obviously decrease by 16.2%, 13.4% and 12.9% under the SERS-A1B, -A2, and -B1 emission scenarios, respectively. As far as the spatial distribution is concerned, fog days under the three scenarios will all decline in most areas of China except a few individual areas; the most obvious decline will occur in Yunnan-Guizhou Plateau, South China, North China and Sichuan Basin etc. under the SRES-A1B scenario, while the most obvious increase will occur in the Tianshan Mts. and the Longdong-Shanxi region in Northwest China under the SRES-B1 scenario.

Key words: fog days, regression analysis, projection, emission scenarios, trend analysis, climate models