Simulation And Projection For Regional Climate In China By Multiple Global Climate Models

Through comparing the history data of the5global models of CMIP5:BCC-CSM1-1, CanESM2, CNRM-CM5, INMCM4, NorESMl-M with grid data of observation dealt by the interpolation of an objective analysis method, the article test the ability of5models to simulate the mean regional climate and extreme weather events in the period1961-2005, and then estimate the changes of China’s regional climate mean state and extreme weather events within the next100years under RCP4.5and RCP8.5circumstances. The main conclusions are as follows:(1) the research suggests a good ability of the models to simulate the regional average temperature in China because of reflecting the characters of spatial distribution with high special and time correlation, inter-annual variability, warming trend of mean temperature, greater linear trend of the minimum temperature than the maximum temperature, greater rise rate of the northern regions than the southern regions. But the model temperature values is much more9times lower than the observations, which may be attributed to the about0℃(0.73°C) value of regional mean min temperature and the ability of the models to simulate the inter-annual variation of the min temperature is better than that of the highest temperature, and better in north than in south with the lowest time correlation in southwest in China mainly in Yunnan-Guizhou Plateau, Sichuan Basin and southeastern Tibetan Plateau.And can’t script the break of the maxmam temperature in1994.(2) The CanESM2has the max value of linear rate instead of the deviation, INMCM4has the lowest ability to script the temperature inter-annual changes, and CNRM-CM5has the max deviation among all the models.The Ensemble has a good abality of simulation.(3) Because of a good capability of temperature simulation, further estimate for average temperature change of the2006-2099is be made by model Ensemble. The result is as follows:in the case both RCP8.5and RCP4.5emissions, the linear trend of the average temperature, max temperature and min temperatures in2006-2099were0.54℃/10a0.25℃/10a;0.52℃/10a,0.24℃/10a;0.57℃/10a,0.25℃/10a, under the higher emission scenarios, there is a0.3℃/10a higher warming trend, and the same time the linear of minimum temperature is slightly larger than that of maximum temperature, but each mode is very different from each other, so that it is necessary to make further research to find out the rate of contribution of max and min to the warming of whole region; under the circumstances of both RCP4.5and RCP8.5emissions, China’s regional average temperature, max temperature, min temperature will respectively have an increase by50%,120%;20%,45%;150%,325%in2099, and under the RCP8.5case, the heating rate in Tibetan Plateau is greater than in the eastern region with same latitude and in the western region than the eastern region.(4) It also has been tested that the models have a good ability to reflect the feature of extreme temperature events with the high correlation of space and time of the Ensemble in the three extreme events, such as heat waves, warm nights, cold days, but the severe vibration in anomalies of extreme temperature doesn’t be reflected. The ensembling has the better inter-annual variability of extreme temperature events in the northern regions than in the southern area of China of35N and a lower the simulation capability in the southwest. Further results of estimates for extreme temperature events is as follows:in2006-2099under both RCP4.5and RCP8.5circumstances, linear trend of the heat waves, warm night and the cold days were2.57days/10and9.99days/10a,1.77%/10a and4.28%/10a,-0.83%/10a and-1.18%/10a, and the rate in the high emission was significantly higher than the moderate emission. Compared to the1986-2005climate mean state, heat waves, respectively, will increase by10%,20%; warm nights by4%and7%; cold days reduce by60%and80%, and the rise of the warm events is much larger than the decline of the cold events. In both cases, the higher trend of the heat wave events lie in the north part of the Northeast, Xinjiang and the south of Tibet; warm night mainly in the South; cold days in the Northeast and the Tibetan Plateau.(5) A certain capability to simulate regional precipitation has been testified with a greater spatial correlation than0.67.The relative error of less than55%and a slight increase of regional precipitation trends. The ensemble can correctly script the first two modes of precipitation in eastern China and the change of the time series. However, the ability of simulating inter-annual variation of precipitation have much differences. The estimated results of the2006-2099of the trend of annual rainfall are in the both emissions are [0.01,0.04] mm/day·10a and [0.02,0.07] mm/day·10a, with a difference of [0.01,0.03] mm/day·10a between them. The linear trend of CanESM2is highest. In the two cases, the higher estimated linear trend lies in northeast of China, Neimeng, the South of Tibetan Plateau and the regions of Yellow river. In the higher emission, the trend of the precipitation rises more rapidly in Tibetan Plateau than the other regions.(6)The ability to simulate extreme precipitation events has been testified with higher special correlation than0.63; relative error of CWD is about200%and those of the rest five indexes are less than40%. The two values of time correlation of CWD and R10are negative and the rest are positive, and those of CDD and R95T are larger.It’s estimated that in the RCP4.5and RCP8.5cases, the average value of the two decades in2080-2099will change relative to that in1986-2005:the CDD decrease by-8%,-11%; CWD increase by8%,5%; the R10increase of14%,20%; R95T increase by18%,38%; R5D increase by15%,27%; SDII an increase by8%,13%. In both cases, the most influenced areas is located in southwest of China.