中文摘要：

利用1958—2005年西沙永兴岛气象观测站逐月气温资料、美国国家海洋大气部（NOAA）提供的1866—2014年逐月海表温度（SST）资料、逐月ENSO指数（MEI）、南方涛动指数（SOI）分析了厄尔尼诺-南方涛动（ENSO）事件与南海西沙气温、海温异常的关系。1958—2005年西沙群岛永兴岛观测站气温与MEI指数比较分析显示：西沙群岛气温正距平百分率积累值阈值能较好地响应75.00%的中等及其以上强度厄尔尼诺事件;西沙气温负距平百分率积累值阈值能较好地响应100%的中等及其以上强度拉尼娜事件。1866—2014年西沙海域SST异常与同期SOI指数对比分析表明：西沙SST正距平百分率积累值阈值能对76.47%的中等及其以上强度的厄尔尼诺有较好的响应;西沙SST负距平百分率积累值阈值能对79.41%的中等及其以上强度的拉尼娜事件有较好的响应。该研究结果表明西沙气温、SST距平百分率积累值阈值法能够得到中等及其以上强度ENSO事件的良好记录,这为利用南海高分辨率珊瑚、砗磲重建所得到的古温度记录定量重建过去ENSO事件提供了可靠的科学依据。

英文摘要：

Background, aim, and scope El Nino-Southern Oscillation （ENSO） is the most important mode of interannual changes in global climate, and understanding past ENSO variability is essential for us to understand the ENSO mechanism and predict its future trend. The paleo-climate records from the South China Sea are ideal for the past ENSO reconstructions because of the significant impacts of ENSO on regional inter-annual climate variations. Indeed, the investigations of ENSO periodicity have been involved in some high resolution paleoclimate records derived from the δ^18O and Sr/Ca of biogenic carbonate. However, these studies only discussed the periodicity of ENSO and few of them involved a more confident quantitative reconstructions. The essential difficulty is how to infer the ENSO signal from the South China Sea climate records which convolve the impacts from various components and aspects the highly dynamical climate system, such as the Asian monsoon, ENSO, Pacific Decadal Oscillation, and so on. In order to provide a detailed method for the quantitative ENSO reconstruction from the paleo proxy records, the modern observational data were first used to establish the relationship between ENSO events and regional temperature around the Xisha Islands, South China Sea, in this study. We analyzed the accumulated positive （negative） percentage of monthly temperature anomalies of Xisha to determine a suitable ＂ENSO threshold＂. Materials and methods We used monthly air temperature from weather station, Yongxing Island, South China Sea and gridded sea surface temperature （SST） data sets of National Oceanic and Atmospheric Administration （NOAA） （http：//www.esrl.noaa.gov/psd/data/gridded/data.noaa.ersst.html. The resolution of SST is 2°×2°, and we just chose the areas of 111°-113°W, 15°-17°N）. We used Multivariate ENSO index （MEI） sets of National Oceanic and Atmospheric Administration （NOAA） （https：//www.esrl.noaa.gov/psd/enso/mei/） and Southern Oscillation Index （SOI） sets