Adjustment of Urbanization Bias in Surface Air Temperature over the Mainland of China
张爱英,周雅清,任玉玉,任国玉,李娇,温康民,孙秀宝
DOI: https://doi.org/10.18306/dlkxjz.2019.04.012
2019-01-01
Abstract:This study developed a method for correcting urbanization bias of station monthly mean surface air temperature data. By using this method and the data of 143 reference stations obtained from a previous research,we corrected urbanization bias of annual and monthly mean temperature data from 1961 to 2015 for 685 national reference climate and basic weather stations (national stations). We explained the rationality and evaluated the effect of the adjustment in the monitoring and analysis of surface air temperature change in the mainland of China. This study adopted a method of iterative correction that corrects longitudinally from east to west, and a revised target station can also serve as a reference station. First, the reference stations of a target station was set as within the range of 300 km, and the correlation coefficients of the detrended annual mean temperature between the target and candidate reference stations were taken as the criteria for selecting the reference stations. Second, by using the correlation coefficients of the reference stations as weights, the weighted average of the annual and monthly mean temperature of all the reference stations around a target station was calculated,obtaining annual and monthly mean temperature reference series for each of the target stations. Third,urbanization biases of the target stations were adjusted by using the linear temperature trend differences of the target station and the reference series as the total correction amounts. The areas with large adjustments are located in North China, part of Central China, northern Northeast China, part of Southwest China, and western China, ranging from 0.1 to 0.3 ℃/10 a, and negative urbanization bias exists in some areas of northwestern China, western and southern Tibet, southern Northeast China, coastal South China, and a few stations in East and Central China. For the whole of China's mainland, the relative urbanization bias is 19.6％. As representative metropolis observational sites in North China, Central China, Northwest and South China, respectively, Beijing,Wuhan, Yinchuan, and Shenzhen stations are found to have larger relative adjustments of 67.0％, 75.4％, 32.7％,and 50.3％ respectively in the past 55 years. The adjustments are in line with the results of previous studies on assessment of impacts of urbanization for the stations. The results show that the adjustment is reasonable. The corrected temperature data largely have eliminated the uncertainties caused by urbanization biases.