气候分类在揭示各地不同状况下的区域气候特征，帮助各地及各行业有针对性地利用气候资源、趋利避害及适应未来气候变化等都具有重要意义。历史上由于缺乏长期观测数据，在应用传统柯本分类时，研究者都尽可能使用已有资料对长期气候平均状况进行分类。过去二三十年的研究表明，在不同时间尺度下，地区的降水与温度会发生明显变化。由此可以推论，在不同时间尺度下，气候分类也会发生相应变化。ENSO是气候波动中最显著的信号，它的发生及在频率与强度等方面的变化对全球大部分地区的社会经济产生巨大影响。因此对ENSO背景下开展区域气候分类研究，有助于进一步了解ENSO影响，为制定相应对策提供必要的科学依据，也为应对ENSO年气候异常、防御灾害以及保证经济安全提供科学参考，具有一定的应用价值。 本文利用1916-2015年全球及中国格点逐月插值资料，采用柯本气候分类法与合成分析法，分析过去近百年全球及中国在不同数据产品、ENSO类型以及ENSO强度下的柯本气候分类，并与1916-2015年柯本气候分类的结果对比。结果表明： （1）基于不同数据源，全球气候带与气候型的差异分布都较为吻合。UD（美国特拉华大学地理海洋环境学院）数据得出的结果在不同气候情景下波动较大，CRU（东英格利亚大学气候研究中心）数据得出的结果相对稳定。中国气候分类通过三种数据计算得出的结果也基本较吻合，中国格点数据得出的结果在不同气候情景下数值的波动最大，CRU数据得出的结果相对稳定，数据波动最小。 （2）基于不同ENSO类型，从全球和中国气候分类来看，东部型和中部型ENSO气候带与50年平均态差异分布基本吻合。但气候型分布差异较大，在部分气候敏感区域出现明显差异。 （3）基于不同ENSO强度，从全球和中国气候分类来看，不同类型ENSO气候带与50年平均态差异分布基本吻合。但气候型分布差异也较大，在部分气候敏感区域出现明显差异，并且相较于其他ENSO类型，超强ENSO背景下的气候分类与50年平均态的差异最明显。 （4）在不同气候情景下，全球及中国均存在厄尔尼诺/拉尼娜敏感区、厄尔尼诺敏感区及拉尼娜敏感区。 本研究同时基于区域灾害风险理论，结合自然及社会经济数据针对中国和南非沿海地区的潜在干旱风险尝试进行案例对比研究，结果表明： 从总体上来看，中国沿海各省在不同气候情景下的旱灾风险区划，总体差异不大。中国高风险区主要集中在广西省、海南省、福建省、辽宁省及河北省，中等风险区包括天津市和浙江省以及上海市，剩余山东省、江苏省与广东省风险相对较低。南非沿海各省的旱灾风险区划在不同气候情景下，总体差异较大。南非沿海高风险区主要集中在东开普敦省西部地区、北开普敦省东部地区，中等风险区包括北开普敦省西部地区，剩余西开普敦省及东部地区的夸祖鲁-纳塔尔省风险相对较低。

Climate classification is of great significance in revealing regional climate characteristics under different conditions in different regions and helping various regions and industries to make targeted use of climate resources, seek advantages，avoid disadvantages and adapt to future climate change. In the past, due to the lack of long-term observation data, when using traditional K?ppen climate classification, researchers use the existing data to classify the long-term average climate status as far as possible. Studies over the past two or three decades have shown that precipitation and temperature in a region vary significantly on different time scales. It can be inferred that climate classification changes correspondingly in different time scales. Therefore, the study on regional climate classification under the background of ENSO is helpful to further understand the impact of ENSO, and provide necessary scientific basis for formulating corresponding countermeasures and scientific reference for coping with ENSO annual climate anomalies, disaster prevention and ensuring economic security, which has certain application value. In view of this situation, this paper uses the monthly interpolation data of global and Chinese grid points from 1916 to 2015, and adopts the K?ppen climate classification and synthetic analysis method. The K?ppen climate classification of the world and China in the past 100 years under different datasets, ENSO types and intensities was analyzed, and compared with the results of the K?ppen climate classification from 1916 to 2015. The results show that: (1) Based on different data sources, the difference distributions of global climate zones and types are relatively consistent. The results of UD (School of Geography and Marine Environment, University of Delaware) data fluctuate greatly in different climate scenarios, and the results of CRU (Climate Research Unit, University of East Anglia) data are relatively stable. The results obtained from the calculation of three types of data products for climate classification in China are basically consistent with each other. The results obtained from Chinese grid data show the largest fluctuations in different climate scenarios, while the results of CRU data products are relatively stable with the smallest fluctuations. (2) Based on different types of ENSO, from the perspective of global and Chinese climate classification, the distribution of eastern and central type of ENSO climate zones is basically less different from the 50-year average. But the difference of climate types are more different, and there are significant differences in some climate-sensitive regions. (3) Based on different intensities of ENSO, from the perspective of global distribution and Chinese climate classification, the distribution of different types of ENSO climate zones is basically less different from the 50-year average. But the climate types are more different, and there are also significant differences in some climate-sensitive regions. (4) Under different climatic scenarios, there are El Ni?o/ La Ni?a sensitive areas, El Ni?o sensitive areas and la Ni?a sensitive areas all over the world and in China. Based on the theory of regional disaster risk, combined with nature and socio-economic data, the potential drought risk in typical areas with different climate types was analyzed as a case study. The results show that: Generally speaking, there is little difference between coastal provinces in China in different climate scenarios. The higher coastal risk areas of China include the provinces of Guangxi, Hainan, Fujian, Liaoning and Hebei, and the middle risk areas include the provinces of Tianjin, Zhejiang, as well as Shanghai. The remaining provinces of Shandong, Jiangsu and Guangdong have relatively low risks. The coastal provinces of South Africa vary greatly in different climate scenarios. The higher coastal risk areas of South Africa include the western part of the province of East Cape and the eastern part of the province of North Cape, the middle risk areas include the western part of the province of North Cape, and the remaining the provinces Western Cape and the eastern part of Kwazulu-natal have relatively low risks.