不同时间分辨率对气象要素月平均值统计的影响

EFFECT ON THE STATISTICAL MONTHLY MEAN VALUES OF METEOROLOGICAL ELEMENTS RESULTING FROM DIFFERENT TIME LEVEL OF OBSERVATIONS

通过我国12个国家基准气候站建站至2000年的气压、气温、相对湿度、风速每日4次与24次定时观测资料月平均值的差值、差值平均值、标准差的对比分析，研究不同时间分辨率统计的资料序列的均一性；在此基础上，对全国所有基准站建站至2000年气温历年月平均值进行统计，绘制了全国基准站建站至2000年气温4次与24次定时累年月平均值差值的平均值分布图，并做了初步分析。结果表明，采用两种不同时间分辨率统计气象要素月平均值会出现统计误差，气压、相对湿度、风速4次与24次定时观测资料月平均值偏差在观测精度之内，气温的差值超过了观测精度。我国绝大部分地区的气温4次定时观测资料要比24次观测资料统计的月平均值偏小0．1～0．3℃。在气候分析和气候变化研究中，必须考虑因不同时间分辨率统计气温月平均值引起的统计误差，排除非气候原因造成的影响。

The differences, the mean differences and the standard deviations of monthly mean values were analyzed from 4 times fixed time observations and 24 times fixed time observations each day. The observations on air pressure, air temperature, air relative humidity and wind speed were from 12 base stations in China from the stations building to 2000. The homogeneities of data sequences from different time level of observations were studied. The distributing chart of mean difference of air temperature in years from the stations building to 2000 are given, which is between 4 times fixed time observations and 24 times fixed time observations each day in each base station in China. It indicates that the deviation of monthly mean air pressure between above two time level of observations lies in observation precision, so do that of monthly mean relative humidity and monthly mean wind speed, but that of monthly mean air temperature is in excess of its observation precision. The monthly mean air temperature from 4 times fixed time observations is about 0.1-0.3℃ less than that from 24 times fixed time observations. The statistical error of monthly mean air temperature from different time level of observations should be considered in climate research to exclude the non-climate effect.