四川省高低气温时空变化特征分析

Analysis on the Characteristics of Spatio-Temporal Variation of High and Low Temperature in Sichuan Province

摘要: 本文利用四川省148个主要气象站气温逐日观测资料，主要对年平均气温、平均最高气温及平均最低气温的长期变化趋势、时空特征和突变特征进行分析。通过对气温的分析得到四川地区气温变化的特征如下：1961~2014年四川年平均气温、平均最高气温以及平均最低气温呈上升趋势，但是不同地区具体变化情况也不尽相同。在川西高原地区以及四川西南地区气温上升较快，四川东北地区以及四川盆地甚至出现降温中心。降温中心主要站点有汉源县和位于四川东北部的安岳县、盐亭县、苍溪县、渠县、南江县等。对年平均气温、平均最高气温、平均最低气温的EOF分析中，第一模态的特征值均为正值。同时也能说明四川的气温呈现上升趋势。且对应的时间系数表明在1990年以后气温上升速率较快，2000年以后气温仍然上升但是速度有所减缓。通过突变分析发现平均最高气温发生突变的年份大致在2000年前后，与年平均温度的突变时间基本一致。同时发现年平均最低气温发生突变的年份比平均气温和平均最高气温要早10年左右。

Abstract: Based on the daily temperature observation data of 148 major meteorological stations in Sichuan Province, this paper mainly analyzes the long-term variation trend, spatio-temporal characteristics and abrupt changes of the annual average temperature, the average maximum temperature and the average minimum temperature. From 1961 to 2014, the annual average temperature, the average maximum temperature and the average minimum temperature of Sichuan showed an upward trend, but the temperature changes were different in different regions. In the western Sichuan plateau and southwestern Sichuan, the temperature rose fast, and there were even cooling centers in northeastern Sichuan and Sichuan Basin. The main sites of the cooling center occurred in Hanyuan County and some County in the northeast of Sichuan such as Anyue, Yanting, Cangxi, Quxian and Nanjiang. Through the EOF analysis of the annual average temperature, the average maximum temperature and the average minimum temperature, the eigenvalues of the first mode all shown to be positive. This also proves that the temperature in Sichuan presents an ascend trend. And the corresponding time coefficient shows that the rate of temperature rise is faster after 1990. It still rises after 2000, but the rate of growth slows down. According to the analysis of abrupt change, the year in which the average maximum temperature abruptly changes is around 2000, which is basically consistent with the time of abrupt change of the average temperature of the year. At the same time, it is found that the annual mean minimum temperature mutation year was 10 years earlier than the average temperature and the average maximum temperature mutation year.