科尔沁沙地及其周围地区降水与气温变化的多时间尺度分析

MUITIPLE TEMPORAL SCALES ANALYSIS OF RAINFALL AND TEMPERATURE CHANGE IN KERQIN DESERT AND ITS SURROUNDING AREAS

根据科尔沁沙地及其周围地区5个国家基本台站1950年～2000年的降水和气温数据，本文运用小波分析方法研究了该地区降水与气温变化的多时间尺度结构。研究结果表明，研究地区5个台站降水和气温变化的多时间尺度结构特征虽然存在一定的差异，但是，在16a尺度上，各台站降水和气温具有比较一致的明显变化趋势。其中，降水方面主要表现为3个相对少雨期和2个相对多雨期；气温方面主要表现为4个相对冷期和3个相对暖期。此外，根据小波变换系数的正负中心变化推断，2000年～2010年研究地区可能降水减少、气温下降，呈现冷干化的趋势。最后，文章指出与常规的统计方法相比，尽管小波分析在研究降水与气温时间序列的多时间尺度特征方面具有不可比拟的优势，但是它也存在自身的局限性。因此，只有将小波分析与多种其它方法相结合，才有可能深入理解气候复杂系统的变化规律。

Based on precipitation and temperature time series data （1950 -2000） collected from 5 basic national meteorological observatory stations located in Kerqin Desert and its surrounding area, multiple temporal scales＇characteristics of climate change was investigated by Mexican Hat - form wavelet transformation. The results indicated that obvious common trends of precipitation and temperature changes existed at 16 - year scale among these 5 stations during the past 50 years, though each contained small variations in multiple temporal characteristics. There were 3 drier periods, 2 rainy periods, and 4 colder periods and 3 warmer periods in the study area from 1950 to 2000. In addition, according to the conversion of wavelet transformation coefficients between positive and negative, it was predicted that it would be drier and colder from 2000 to 2010. Finally, it was pointed out that wavelet transform also has its own limitations in studying climatic changes, though it has significant advantages compared with traditional statistical methods. Thus it should be used together with other related methods in order to understand complex climatic systems.