我国西南地区气象要素变化特征及预测

Variation Characteristics of Meteorological Elements and Forecasting in Southwest China

基于我国西南地区112个测站1961~2012年的逐月降水和气温观测资料，采用趋势系数法、线性倾向估计法、Mann Kendall突变检验法、相关分析及逐步回归周期分析法等分析了我国西南地区近52年降水量和气温的时间变化特征，并在深入分析影响该地区年平均降水量的大气环流因子的基础上，建立了年平均降水量的多因子逐步回归周期分析预报模型，对当前及未来5年西南地区年平均降水量进行了预测。结果表明，西南地区年均降水量和秋季平均降水量呈显著下降趋势，年均降水量在2003、2009年出现突变，秋季平均降水量在1990年存在突变点；年均及四季平均气温都呈显著上升趋势且存在明显的突变点；副高、欧亚经向环流和极涡强度指数对我国西南地区年平均降水量影响较为显著；年平均降水量的多因子逐步回归周期分析预报模型的拟合预报合格率为100%，预测结果可信。

Based on the monthly observational precipitation and temperature data of 112 survey stations in southwest China during1961~2012, time change characteristics of precipitation and temperature in southwest China during 52 years were analyzed by using trend coefficient method, linear trend analysis method, Mann Kendall test method, correlation analysis and stepwise regression cycle analysis. On the basis of analysis the influence of circulation indexes on annual average precipitation in southwest China, multi factor stepwise regression cycle analysis forecast model of annual average precipitation is established. And then the annual average precipitation in current and next five years are predicted. The results show that the annual average precipitation and autumn average precipitation appear downward trend; annual average precipitation has abrupt drier condition in 2003 and 2009, and autumn average precipitation has an abrupt drier condition in 1990; annual and four seasons average temperatures have upward trend and have remarkable abrupt drier condition; subtropical high, Eurasian meridional circulation and Polar vortex intensity index have greater impact on annual average precipitation in southwest China; the qualified rate of fitting and forecasting with the multi factor stepwise regression cycle analysis model of annual average precipitation reaches 100%, and prediction result is reliable.