中国西北地区参考作物蒸散量的估算与变化特征

摘要

基于中国西北地区42个气象站1956~2011年逐日气象观测资料,采用Penman-Monteith公式,估算该地区的参考作物蒸散量(ET0),分析西北地区ET0的时空变化特征,并利用Mann-Kendall检验、小波分析和多元线性回归分析等方法分别对ET0进行突变检验、周期变化和主要影响因子分析研究.结果表明:(1)西北地区ET0存在明显的月和季节变化,ET0自夏季、春季、秋季和冬季依次减少;(2)近56 a来,西北地区年均ET0呈明显减少趋势,且存在一定的季节性差异,春、夏、秋季ET0均呈显著减小趋势,且夏季减少率最大,而冬季则呈平缓的增加趋势;(3)ET0减小幅度最大的地区位于哈密—和田的东北—西南向一带;(4)1956~2011年,春、夏、秋、冬季及全年平均ET0分别在1984、1986、1981、1995与1980年前后发生了一次减小的突变,且年均ET0存在2~3 a显著震荡周期和准6 a的震荡周期;(5)ET0与2 m风速、日照时数呈显著正相关,而与相对湿度和平均气温呈负相关,其中与平均气温的相关性较弱,可见平均气温、相对湿度的升高与日照时数、2m风速的下降导致西北地区ET0的减小,同时考虑日照时数、平均气温、风速和相对湿度4个气象因子的多元线性回归方程的均方根误差最小,表明西北地区ET0的变化是上述多种气象因子综合作用的结果.%Based on the daily meteorological data at 42 weather stations in Northwest China from 1956 to 2011,the reference crop evap-otranspiration (ET0 )was estimated by using Penman-Monteith (P-M)equation,and the temporal and spatial distributions of ET0 in Northwest China were analyzed.The mutation and period of ET0 during 1956-2011 were studied by using Mann-Kendall (M-K) test and Morlet wavelet analysis methods,and the correlation between ET0 and main meteorological factors was done by using the multi-variable linear regression method.The results are as follows:(1 )The monthly and seasonal changes of ET0 in Northwest China were obvious,the value of ET0 in July was the maximum,while that in December was the minimum,and that in summer,spring,autumn and winter decreased in turn.(2)In the past 56 years,the annual average ET0 in Northwest China reduced significantly,and the sea-sonal difference was significant,the trends of average ET0 in spring,summer and autumn decreased,and the decreasing rate in summer was the biggest,while the average ET0 in winter increased slowly.(3)The maximum ET0 was located in Kumul to Hotan area.(4) The decreasing mutation of ET0 in spring,summer,autumn and the whole year appeared in the early of the 1980s,and that occurred respectively in 1984,1986,1981,1980,while that in winter mutated in 1995.The annual average ET0 in Northwest China during 1956 -2011 had 2-3 years obvious periods and 6 years quasi-periods.(5)The ET0 were positive correlated with wind speed at 2 m and sunshine duration,while that was negative correlated with relative humidity and mean surface air temperature,and the correlation between the ET0 and air temperature was weak.Therefore,the rising of air temperature and relative humidity,and the decreasing of sunshine duration and wind speed caused the ET0 reducing in Northwest China during 1956-2011.The root mean square error consid-ering the above four climatic factors was the minimum in all multitiple linear regression models.In general,the decreasing of ET0 was caused by the variation of four meteorological factors.