新疆阿勒泰站人工观测与自动观测气温差异分析

Differences Between Manual Observation and Automatic Observation of Temperature in the Altay Weather Stations

对新疆阿勒泰气象站2004年1月-2011年12月，人工与CAWS600SE—H型自动气象站，气温的平行观测资料，统计分析了差异原因和变化的一般规律。结果表明：两者差值较大，有日、季节变化，最低、最高气温系统性误差，人工观测低于自动观测0．05～0．20℃；误差具有正态分布特征。气温整点观测，人工比自动观测提前，造成系统性误差，气温急剧升降时误差更大，通过时差订正，绝对误差一般收窄0．1～0．2℃，均方差减少0．142。1和10月，4、24次日平均模式，差值变化幅度≥1℃的频率达15％。平均差值，除7月外，4次平均值低于24次平均0．1℃，10月低于0．5℃以上。平均模式对日值的影响最大，变化幅度3．7oC；对季的影响较小，变化幅度只有0-4℃。自动观测仪器的系统性偏差、测量元件的精度及对气象要素变化响应的灵敏度、观测时间的差异和人为因素的影响是造成观测误差的原因。

The temperature data from the CAWS60OSE-H Automatic and manual weather station in Altay from January 2004 to December 2011, parallel observations difference was analyzed. The results show that： the difference was large between the two stations; there are daily, seasonal variation and for the minimum and maximum temperatures, the systematic errors of manual observation was 0.05 -0.20 ℃ below the auto-observation with normal distribution characteristics. The temperatures at the right points of the observation hours, manual observation was earlier than the automatic observation, resulting in the systematic errors, the error was greater when rapid change of temperature occurred, while after revised with the time difference method, the absolute error was generally reduced by 0.1-0.2 ℃ and both variance reduction was 0.142. For January and October, 4 and 24 number of times in day average mode, the frequency of the magnitude above 1 ℃ of the error change difference reached to 15%. Except in July, the average difference of four times was 0.1 ℃ less than that of 24 times, and in October less than 0.5 ℃ or more. The impact of the average value model on day values was the biggest reaching to 3.7 ℃ but less on the quarterly data reaching to only 0.4℃. Systematic bias of automatic observation equipment, precision of measuring devices, sensitivity and differences in observation time and the cause of human factors caused the impact of observational errors.