二十世纪全球再分析数据集

使用了比较简单的通化方案，仅同化了表层气压资料，以Hadley中心的月平均海温资料和海冰分部资料（HadISST1.1）作为边界条件，集合Kalman滤波资料同化技术，6小时分析一次。本数据集首次提供了跨越20世纪的全球对流层变率的估计、随时间变化的数据集质量的估计。与探空观测和台站温度资料的比较结果表明，本再分析数据集质量大体接近数值预报的三天预报结果，应该对模式评估和气候诊断有用。

A relatively simple data assimilation scheme was adopted, which only assimilated surface pressure data. Monthly mean sea surface temperature and sea ice concentration data (HadISST1.1) from the Hadley Centre were used as boundary conditions, and the ensemble Kalman filter data assimilation technique was employed, with analyses updated every 6 hours. This dataset provides, for the first time, estimates of global tropospheric variability spanning the 20th century, as well as time-varying assessments of dataset quality. Comparisons with radiosonde observations and surface temperature station data show that the quality of this reanalysis dataset is generally comparable to 3-day numerical weather prediction forecast results, and it should be useful for model evaluation and climate diagnostics.

搜集汇总

背景与挑战

背景概述

二十世纪全球再分析数据集是由美国NOAA发布的全球对流层变率估计数据集，采用集合Kalman滤波技术同化表层气压资料，时间分辨率为6小时至月，空间分辨率为2度或T-62高斯格点，覆盖全球范围，数据格式为NetCDF，适用于模式评估和气候诊断。