摘要
频谱分析法能分离和提取存在于降水过程中的周期成分,并将分离出的周期分量进一步用于降水量的中长期预报。通过分析1971～2009年郑州市的降水量资料,应用频谱分析法构建了郑州市降水量的预报模型,并应用该方法对郑州市未来3 a的降水量进行了预报。结果表明,该区年降水量存在两个主要周期(2 a和5 a左右),反映了该地区的气候变化规律。为了验证预报结果的准确性,还与Monte-Carlo方法的预报结果进行比较,证明了频谱分析法具有更高的预报精度。
The periodic components existing in the precipitation process could be separated and extracted by spectrum analysis method, and the separated periodic component can be used for forecast of long - term rainfall. After analyzing precipitation data from 1971 to 2009 in Zhengzhou City, the precipitation forecast model was established using spectrum analysis method and applied in the forecast of precipitation in the future three years in Zhengzhou City. The results show that two dominant cycles existed in this area, namely two years and five years, which reflects the regularity of climate change in the area. In order to verify the accuracy of the prediction results, the analysis results by Monte - Carlo method has been used for comparison. It shows that the frequency spectrum analysis has higher precision.
出处
《人民长江》
北大核心
2012年第10期50-52,共3页
Yangtze River
基金
2010年华北水利水电学院高层次人才科研启动项目(001316)
