澳大利亚区域大气加权平均温度建模

大气加权平均温度T_m是计算水汽转换因子和大气可降水量的重要参数。利用2007—2017年全球大地观测系统(global geodetic observing system, GGOS) Atmosphere T_m格网数据和欧洲中尺度天气预报中心(European centre for medium-range weather forecasts, ECMWF) 2 m温度数据，建立一种适合澳大利亚区域、顾及T_m残差季节性和日周期变化的T_m模型——qTm。此外，采用2018年的GGOS Atmosphere T_m格网数据和探空资料对该模型进行评估。结果表明，qTm模型在澳大利亚区域具有较高的精度和适用性，与GGOS Atmosphere T_m相比，qTm模型的年均偏差(Bias)和均方根误差(root mean square error, RMSE)分别为－0.31 K和1.97 K，相对于GPT2w-1和GPT2w-5模型，RMSE分别提高21.8%和25.9%；qTm模型值与探空积分值更符合，模型的年均Bias和RMSE分别为－0.44 K和2.45 K，相比GPT2w-1和GPT2w-5模型分别提高10.2% 和11.8%。qTm模型可为澳大利亚区域提供精确的T_m值，为该区域大气水汽分析和厄尔尼诺现象研究提供基础。

Atmospheric Weighted Mean Temperature Modeling for Australia

  Objectives  Atmospheric weighted average temperature T_m is essential for calculating water vapour conversion factor and atmospheric precipitation.  Methods  Based on the global geodetic observing system(GGOS) Atmosphere T_m grid data and European centre for medium-range weather forecasts (ECMWF) 2 m temperature data from 2007 to 2017, a T_m model (qTm) suitable for Australia was established and the seasonal and daily variation of T_m residuals was taken into account. In addition, GGOS Atmosphere T_m grid data and radiosonde data in 2018 were selected to evaluate the model.  Results  The results reveal that the qTm model is more accurate and applicable in the Australian region. Compared with GGOS Atmosphere T_m, the annual average Bias and RMSE(root mean square error) of the qTm model are －0.31 K and 1.97 K, respectively. While compared to the GPT2w(global pressure and temperature 2 wet)-1 and GPT2w-5, RMSE of qTm model increased by 21.8% and 25.9%, respectively. qTm model values are more consistent with sounding integral values, and the model's annual average Bias and RMSE are －0.44 K and 2.45 K, respectively. Compared with GPT2w-1 and GPT2w-5, they increased by 10.2% and 11.8%, respectively.  Conclusions  The qTm model can provide accurate T_m values for the Australian region and is a basis for the region's atmospheric moisture analysis and El Ni?o studies.