中文摘要：

针对珠江流域,分析了在全球气候模式（BCC_CSM1.1）驱动下,区域气候模式RegCM4进行的中国区域气候变化模拟中,珠江流域在RCP4.5和RCP8.5温室气体排放情景下,未来2010—2099年的气候变化。结果表明,RegCM4对珠江流域气候特征具有很强的模拟能力。未来RCPS情景下珠江流域气温将持续增大。与参照时段（1980—1999年）相比,RCP4.5和RCP8.5情景下的年平均温度在2020s分别增加0.7℃和0.8℃,2050s分别增加1.0℃和1.6℃,2080s分别增加1.6℃和2.9℃。而未来年降水并未表现出显著的变化趋势,但不同情景、不同地区预估的降水呈现不同的变化趋势。RCP4.5情景下,流域降水2020s将减少4.3%,2050s和2080s将分别增加0.7%和0.1%;RCP8.5情景下,未来不同时段流域降水均呈减少趋势,2020s、2050s和2080s分别减少1.7%、2.9%和0.2%,表明降水预估具有更大的不确定性。两种排放情景下未来降水在东南沿海增加、西北部减少,变化率为±8%。此外,两种排放情景下未来珠江流域的日平均温度统计特征发生改变,揭示未来高温事件可能增加,同时,大雨级别以上的降水发生频率增加,可能导致洪涝事件增加。

英文摘要：

The changing tends of temperature and precipitation during 2010—2099 in the Pearl River basin under RCPs scenarios were analyzed. It is based on projections over China simulated by RegCM4（a high resolution regional climate model）, which is driven by the GCM（BCC_CSM1.1）. The results show that RegCM4 can simulate local climate distribution characteristics over the Pearl River Basin quite well, judging from comparisons of the simulation with in situ observation. For the future, the projected temperature shows a persistently increasing trend in the Pearl River Basin under RCPS scenarios. Compared to the reference period（1980—1999）, the average temperature in 2020s will increase 0.7 ℃ and 0.8 ℃ under RCP4.5 and RCP8.5 Scenario, and increase 1 ℃ and 1.6 ℃ in 2050s, and 1.6 ℃ and 2.9 ℃ in 2080s. There is no significant change trend in the annual precipitation, but the projected precipitation has different change trends for different scenarios and areas. Under the RCP4.5 scenario, the precipitation will reduce by 4.3% in 2020s while increasing by 0.7% and 0.1% in 2050s and 2080s, respectively. Under the RCP8.5 scenario, precipitation shows a decreasing trend, with 1.7%, 2.9%, and 0.2% in 2020s, 2050s and 2080s, respectively. It indicates that the precipitation has greater uncertainty. Under the two emission scenarios, there would be an increasing trend of precipitation in the southeast coast while a decreasing trend in the northwest in the next 90 years. Meanwhile, the frequency distribution of daily temperature indicates that the potential possibility of high temperature event might rise. And a substantial increase of the extreme precipitation events is also projected and may lead to increased frequency of flood events.