摘要
采用惩罚最大F检验、惩罚最大T检验以及二相回归方法,并结合台站迁移元数据,对我国西北五省的142个气象台站1961—2010年月平均、最高、最低气温序列进行了均一化的检验。结果表明,惩罚最大F检验和惩罚最大T检验适合对月气温序列的均一性检验,台站迁移是造成西北五省月气温序列非均一的最重要原因。针对检验出的存在非均一的气候序列采用差值方法和QM方法进行了均一性的订正,并且对订正前后区域气温的趋势变化和突变性特征进行了对比分析。结果表明,经过均一化的订正,在保持区域整体趋势不变的基础之上,区域中由于非均一台站而造成的虚假中心得以滤去,个别季节的突变性在订正前后也发生了明显的变化,区域整体的均一性状况到了明显的改善。
Based on the penalized F test,the penalized T test,the two-phase regression test and the detailed metadata of the station migration,the homogeneity of monthly average,maximum,minimum temperature data from 1961 to 2010 of the 142 meteorological observation stations of the five provinces in the northwest China were detected and adjusted.The results showed that,the penalized F test and the penalized T test work well for the homogeneity test of the monthly temperature series of the five provinces in northwest China,and the change of location was the main causes for the no-homogeneity.Using the difference method and the QM method,the non-homogeneity series were adjusting and the trend and the mutation characteristic of the before and after the adjusting in this area were compared and analyzed.The results showed that,the false centers had been filtered,the mutation of some seasons had been changed and the homogeneity of the area had been well improved after the homogeneity adjusted,but the trend of the whole area kept no change.
出处
《农业灾害研究》
2013年第2期27-32,共6页
Journal of Agricultural Catastrophology
基金
国家科技支撑计划(2012BAC22B04)
陕西省气象局科技创新基金计划项目(2012M-50)
关键词
均一性
气温
西北地区
Homogeneity
Air temperature
Northwest China