基于2020-2021年的中国气象局(CMA)陆面数据同化系统(CLDAS)逐小时地面气温(T2m)产品,融合CMA上海快速更新循环数值预报(CMA-SH3)的T2m预报数据,构建深度学习语义分割模型(MT-Cunet),实现逐小时滚动更新的24 h T2m网格预报,并对2022年...基于2020-2021年的中国气象局(CMA)陆面数据同化系统(CLDAS)逐小时地面气温(T2m)产品,融合CMA上海快速更新循环数值预报(CMA-SH3)的T2m预报数据,构建深度学习语义分割模型(MT-Cunet),实现逐小时滚动更新的24 h T2m网格预报,并对2022年预报结果进行了检验评估。结果表明,在研究范围内,MT-Cunet在3~9 h时效订正效果最好,平均MAE和平均RMSE分别降低42.4%、40.89%;10~24 h时效的订正效果较好,平均MAE和平均RMSE分别下降26.7%、26.3%。低温(≤0℃)和高温(≥35℃)事件检验评估表明,MT-Cunet在高温预报整体表现为正偏差,而低温整体为负偏差,但误差幅度远低于CMA-SH3;空间尺度上,MT-Cunet能较大幅度减少复杂地形下的T2m预报误差,降低CMA-SH3的MAE离散度,使预报误差分布较为稳定。通过对2022年2月和3月的区域性增温、寒潮过程分别进行检验评估发现,MT-Cunet能较好预报出增(降)温转折时间和增(降)温幅度。在增温和寒潮过程中,MT-Cunet的MAE比CMA-SH3分别降低28.9%和33.8%,表明MT-Cunet模型在转折性天气过程中同样具有较好的预报能力。因此,利用可以快速增加预报样本数量的快速更新循环数值预报,经过语义分割深度学习模型客观方法订正,就能较大幅度降低数值模式预报误差,解决常规数值预报由于数据量太少,深度学习训练效果较差的问题,这对充分利用国产模式资源,更广泛地开展国产模式后处理和应用提出了一个新的思路。展开更多
In real-time operations,the minutely/hourly updated high-resolution rapid refresh(HRRR)system is one of the most expensive numerical weather prediction(NWP)models.Based on a twenty-member HRRR-time-lagged-ensemble(HRR...In real-time operations,the minutely/hourly updated high-resolution rapid refresh(HRRR)system is one of the most expensive numerical weather prediction(NWP)models.Based on a twenty-member HRRR-time-lagged-ensemble(HRRR-TLE)system developed from two real-time convection-permitting HRRR models,CMA-GD(R3)and CMA-SH3,from the China Meteorological Administration(CMA),this study proposes an optimized probability-matching(OPM)technique to improve 0−12 h quantitative precipitation forecasts(QPFs)based on the correlation and error relationships between ensemble forecasts and observations during the training window.Then,a series of sensitivity experiments using different cost functions and optimized ratios was conducted to further improve OPM predictions.The results indicate that:(1)In the HRRR-TLE system,there is no always optimal member in both weak rain and severe rain forecasts,as measured by the equitable threat score(ETS)and bias extent(BE)at four thresholds(1+,5+,10+,and 20+mm h^(-1);e.g.,“1+”means≥1).(2)Compared with the HRRR-TLE system,the QPFs generated by the traditional PM technique showed a notable increase in ETS and a decrease in BE at all of the above thresholds.Compared with the traditional probability-matching method(PM),OPM can generate more skillful forecasts on both spatial representations and rain rates by using the sliding-weight method and optimized ensembles,respectively.(3)In particular,in the 20+mm h^(-1)forecasts,which are often difficult to predict,the ETS of the optimal OPM test,with a 20%optimization ratio and symmetric mean absolute percentage error cost function,increased by 64.6%,and the BE decreased by 5.7%,relative to PM.Moreover,OPM shows good stability in both daytime and nighttime periods.展开更多
文摘基于2020-2021年的中国气象局(CMA)陆面数据同化系统(CLDAS)逐小时地面气温(T2m)产品,融合CMA上海快速更新循环数值预报(CMA-SH3)的T2m预报数据,构建深度学习语义分割模型(MT-Cunet),实现逐小时滚动更新的24 h T2m网格预报,并对2022年预报结果进行了检验评估。结果表明,在研究范围内,MT-Cunet在3~9 h时效订正效果最好,平均MAE和平均RMSE分别降低42.4%、40.89%;10~24 h时效的订正效果较好,平均MAE和平均RMSE分别下降26.7%、26.3%。低温(≤0℃)和高温(≥35℃)事件检验评估表明,MT-Cunet在高温预报整体表现为正偏差,而低温整体为负偏差,但误差幅度远低于CMA-SH3;空间尺度上,MT-Cunet能较大幅度减少复杂地形下的T2m预报误差,降低CMA-SH3的MAE离散度,使预报误差分布较为稳定。通过对2022年2月和3月的区域性增温、寒潮过程分别进行检验评估发现,MT-Cunet能较好预报出增(降)温转折时间和增(降)温幅度。在增温和寒潮过程中,MT-Cunet的MAE比CMA-SH3分别降低28.9%和33.8%,表明MT-Cunet模型在转折性天气过程中同样具有较好的预报能力。因此,利用可以快速增加预报样本数量的快速更新循环数值预报,经过语义分割深度学习模型客观方法订正,就能较大幅度降低数值模式预报误差,解决常规数值预报由于数据量太少,深度学习训练效果较差的问题,这对充分利用国产模式资源,更广泛地开展国产模式后处理和应用提出了一个新的思路。
基金Natural Science Foundation of Hunan Province(2021JC0009)National Natural Science Foundation of China(42375002,42105146)Key Projects of Hunan Meteorological Bureau(XQKJ22A004)。
文摘In real-time operations,the minutely/hourly updated high-resolution rapid refresh(HRRR)system is one of the most expensive numerical weather prediction(NWP)models.Based on a twenty-member HRRR-time-lagged-ensemble(HRRR-TLE)system developed from two real-time convection-permitting HRRR models,CMA-GD(R3)and CMA-SH3,from the China Meteorological Administration(CMA),this study proposes an optimized probability-matching(OPM)technique to improve 0−12 h quantitative precipitation forecasts(QPFs)based on the correlation and error relationships between ensemble forecasts and observations during the training window.Then,a series of sensitivity experiments using different cost functions and optimized ratios was conducted to further improve OPM predictions.The results indicate that:(1)In the HRRR-TLE system,there is no always optimal member in both weak rain and severe rain forecasts,as measured by the equitable threat score(ETS)and bias extent(BE)at four thresholds(1+,5+,10+,and 20+mm h^(-1);e.g.,“1+”means≥1).(2)Compared with the HRRR-TLE system,the QPFs generated by the traditional PM technique showed a notable increase in ETS and a decrease in BE at all of the above thresholds.Compared with the traditional probability-matching method(PM),OPM can generate more skillful forecasts on both spatial representations and rain rates by using the sliding-weight method and optimized ensembles,respectively.(3)In particular,in the 20+mm h^(-1)forecasts,which are often difficult to predict,the ETS of the optimal OPM test,with a 20%optimization ratio and symmetric mean absolute percentage error cost function,increased by 64.6%,and the BE decreased by 5.7%,relative to PM.Moreover,OPM shows good stability in both daytime and nighttime periods.
