摘要
针对气象资料难以满足高斯分布的情况,引进了具有耐抗性的三组耐抗线方法,分析了满足不同信度情况下最小二乘回归和三组耐抗线的差异,并用西藏近30 a气温资料进行了检验。结果表明三组耐抗线方法比最小二乘法研究气候变化趋势有明显优势;采用最小二乘法得到的西藏气候变暖估计偏高。
In the light of the question that meteorological surface data are unfit to Gaussian distribution, the variation tendencies computed with Tibet surface air temperature data by Three Group Resistance Line and Least Square Regression Method were compared in this paper. Results show that Three Group Resistance Line is not significantly influenced by oufliers and is useful for tendency analysis of meteorological records sometimes without sufficient number of samples. The tendencies from Three Group Resistance Line were less than those from Least Square Regression method in Tibet during 1971 -2000.
出处
《干旱气象》
2007年第1期79-82,共4页
Journal of Arid Meteorology
基金
国家自然科学基金项目(40375003)资助
关键词
三组耐抗线
最小二乘法
气温变化
Three Group Resistance Line
Least Square Regression Method
temperature variation