Effect of change of radiation error correction methods on China radiosonde temperature series data

Radiation error correction is one of the most important error corrections in a radiosonde observation process. The addition and revision of radiation error corrections are the main reasons for the inhomogeneity of radiosonde temperature in China. Using penalized maximum F test( PMFT) methods developed by the Center of Environment in Canada and the detailed metadata of each radiosonde station in China,the inhomogeneity test and corrections were carried out based on the monthly temperature data at mandatory levels at 123 radiosonde stations in China. The results showthat the most of temperature data series have a discontinuous problem caused by the change of radiation error correction methods. The breakpoints caused by the change of radiation error correction methods reach 46. 17% of total breakpoints. The adjustment value of temperature at 00: 00 UTC increases with height and it is negative,and it is larger at high mandatory level than in lowmandatory level.