1.中山大学地理科学与规划学院,广东 广州 510006
2.广东省粤北岩溶区碳水耦合野外科学观测研究站,广东 广州 510006
庄胜杰(2000年生),男;研究方向:水文气象预测;E-mail:zhuangshj3@mail2.sysu.edu.cn
王大刚(1975年生),男;研究方向:极端气候变化、水文气象预测;E-mail:wangdag@mail.sysu.edu.cn
纸质出版日期:2024-07-25,
网络出版日期:2024-04-24,
收稿日期:2023-11-15,
录用日期:2024-03-13
移动端阅览
庄胜杰,王大刚,林泳恩等.基于CFSv2产品和机器学习的东江流域月降水预报[J].中山大学学报(自然科学版)(中英文),2024,63(04):9-18.
ZHUANG Shengjie,WANG Dagang,LIN Yongen,et al.Monthly precipitation forecast in the Dongjiang Basin based on CFSv2 products and machine learning[J].Acta Scientiarum Naturalium Universitatis Sunyatseni,2024,63(04):9-18.
庄胜杰,王大刚,林泳恩等.基于CFSv2产品和机器学习的东江流域月降水预报[J].中山大学学报(自然科学版)(中英文),2024,63(04):9-18. DOI: 10.13471/j.cnki.acta.snus.ZR20230016.
ZHUANG Shengjie,WANG Dagang,LIN Yongen,et al.Monthly precipitation forecast in the Dongjiang Basin based on CFSv2 products and machine learning[J].Acta Scientiarum Naturalium Universitatis Sunyatseni,2024,63(04):9-18. DOI: 10.13471/j.cnki.acta.snus.ZR20230016.
中长期降水预报一直以来是研究水文气象变化的热点,其精度与可靠性不高的问题亟待解决。以东江流域为研究对象,通过距平相关系数ACC、标准化均方根误差NRMSE、平均绝对误差MAE和多模型稳定性指数MSI评估CFSv2模式产品在月尺度的预测精度与稳定性,采用CFSv2模式降水预报、CFSv2模式预报因子结合机器学习模型预报2种方法预测未来降水。结果表明,不同预见期下,CFSv2模式降水预测与实测降水量具有较高的相关性,对于枯水期的预测效果好于汛期,但随着起报时间发生改变,降水预测的差异性较大,模型稳定性较差;CFSv2模式预报因子结合机器学习模型提高了预测的稳定性,相较于CFSv2模式降水预测,MSI从0.45降低到0.25,在很大程度上减小了由于起报时间改变产生的预报随机性。研究成果可为中长期降水预测提供一种新的思路,并为中长期水文预报和水资源管理提供决策依据。
Mid to long-term precipitation forecasting has always been a hot topic in hydro-meteorological research,with the issue of low accuracy and reliability needing urgent solutions. This study focuses on the Dongjiang Basin and evaluates the prediction accuracy and stability of CFSv2 model products at the monthly scale using the anomaly coefficient of correlation (ACC),normalized root mean square error (NRMSE),mean absolute error (MAE),and the multi-model stability index (MSI). Two methods,namely the CFSv2 model precipitation forecast and the machine learning model forecast combined with CFSv2 model predictors,are employed to predict future precipitation. The results show that under different lead times,the CFSv2 model precipitation forecast exhibits a high correlation with observed precipitation,performing better during the dry season compared to the flood season. However,there is significant variability in precipitation forecasts and poor model stability with changes in the initial time. Combining CFSv2 model predictors with machine learning models improves the forecast stability,reducing the MSI from 0.45 to 0.25 and effectively reducing the randomness in forecasts caused by changes in the initial time. The findings contribute to providing a new approach for mid to long-term precipitation forecasting and offer decision-making support for mid to long-term hydrological forecasting and water resource management.
CFSv2中长期预报机器学习产品评估
CFSv2mid to long-term forecastmachine learningproduct evaluation
陈柯兵,郭生练,王俊,等,2020. 长江上游ECMWF降水和径流预报产品评估[J]. 人民长江,51(3):73-80.
冯志州,2017. 基于统计方法的东江流域季节降水预报研究[D]. 广州:中山大学.
董满宇, 王炳钦, 廖剑宇, 等, 2013. 近50年东江流域极端降水事件变化特征[J]. 资源科学, 35(3): 521-529.
黄超,李巧萍,谢益军,等,2022. 机器学习方法在湖南夏季降水预测中的应用[J]. 大气科学学报,45(2):191-202.
黄赛男,李文韬,段青云,2022. GEFSv12降水再预报数据在淮河流域的适用性评估[J]. 南水北调与水利科技(中英文),20(5):925-934.
刘金凤,田兆伟,沈雪娇,2018. 近60年东江流域降雨径流特性分析[J]. 广东水利水电,(12):31-36+74.
刘雅云,2018. 基于机器学习的降水数据分析算法的研究[D]. 南京:南京信息工程大学.
马浩,于翡,葛敬文,等,2022. CFSv2对2018年浙江梅雨降水多尺度预测的性能评估[J]. 气象科学,42(5):690-702.
唐兆康,2021. 地基垂直观测网数据对数值预报的影响评估研究[D]. 南京:南京信息工程大学.
吴裕珍,2018. 基于CFSv2的东江流域中长期降雨预报[D]. 广州:中山大学.
肖颖,庞轶舒,马振峰,等,2023. NCEP CFSv2模式对川渝夏季降水次季节预测技巧评估及预报偏差分析[J]. 高原气象,42(6):1576-1588.
谢舜,孙效功,张苏平,等,2022. 基于SVD与机器学习的华南降水预报订正方法[J]. 应用气象学报,33(3):293-304.
张东方,成青燕,何慧根,等,2021. CFSv2模式资料在成都市延伸期降水预测中的应用评估[J]. 气象研究与应用,42(2):1-6.
赵俊虎,封国林,2015. 全球变暖背景下中国东部夏季三类雨型预测概念模型新建[J]. 中国科学:地球科学,45(4):414-426.
甄亿位,郝敏,陆宝宏,等,2015. 基于随机森林的中长期降水量预测模型研究[J]. 水电能源科学,33(6):6-10.
BREIMAN L,2001. Random forests[J]. Mach Learn,45(1):5-32.
GUO Y,NIE H W,2020. Summertime daily precipitation statistics over East China in CFSv2[J]. Phys Chem Earth,115:89-95.
LANG Y,LUO L F,YE A Z,et al,2020. Do CFSv2 seasonal forecasts help improve the forecast of meteorological drought over mainland China[J/OL]. Water,12(7):2010. https://doi.org/10.3390/w12072010https://doi.org/10.3390/w12072010.
LEE S S,LEE J Y,HA K J,et al,2011. Deficiencies and possibilities for long-lead coupled climate prediction of the western North Pacific-East Asian summer monsoon[J]. Clim Dyn,36(5):1173-1188.
LI C Z,YANG S,MO W Q,et al,2022. Seasonal prediction for may rainfall over Southern China based on the NCEP CFSv2[J]. J Trop Meteorol,28(1):29-44.
LUO L F,TANG W,LIN Z H,et al,2013. Evaluation of summer temperature and precipitation predictions from NCEP CFSv2 retrospective forecast over China[J]. Clim Dyn,41:2213-2230.
PENG T,ZHI X,JI Y,et al,2020. Prediction skill of extended range 2-m maximum air temperature probabilistic forecasts using machine learning post-processing methods[J]. Atmosphere,11(8):823.
SAHA S, MOORTHI S, WU X,et al,2014. The NCEP climate forecast system version 2[J]. J Clim,27:2185-2207.
VALVERDE RAMÍREZ M C, de CAMPOS VELHO H F, FERREIRA N J,2005. Artificial neural network technique for rainfall forecasting applied to the São Paulo region[J]. J Hydrol,301(1/2/3/4):146-162.
WANG W Q,XIE P P,YOO S H,et al,2011. An assessment of the surface climate in the NCEP climate forecast system reanalysis[J]. Clim Dyn,37:1601-1620.
0
浏览量
93
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构