摘要
水文预报不可避免地存在着输入、水文模型参数和结构等不确定性,导致预报结果也具有不确定性。因此,定量估计水文预报的不确定性,实现概率水文预报,不仅可得到比确定性预报更高的精度,而且还能为决策者提供更丰富的预报信息。本文根据不确定性来源的不同,从输入资料、模型结构、模型参数和综合不确定性等方面,详细综述了贝叶斯水文概率预报的研究进展,归纳了精度评定指标和效果检验方法,并展望了贝叶斯概率水文预报未来的研究重点和方向:(1)科学有效地解释、沟通和传播水文预报不确定性信息和概率水文预报产品;(2)建立水文集合概率预报框架,估计并降低水文预报的总不确定性;(3)开展考虑预报变量时空相关性的贝叶斯概率水文预报研究;(4)深入推动概率水文预报信息在风险决策中的应用。
The uncertainty of input,hydrological model structure and parameters inevitably exists in hydrological forecasting,which leads to the uncertainty of forecast results. Therefore,quantitative estimation of hydrological forecasting uncertainty and realization of probabilistic hydrological forecasting are not only more scientific and reasonable in theory than deterministic forecasting,but also can provide more abundant forecasting information for decision-making. According to the different sources of uncertainty considered, a state art review of Bayesian probabilistic hydrological forecasting from four aspects are systematically summarized in this paper,including input data,model structure and parameters,total uncertainties. Then,the accuracy evaluation index and effect verifying method of probabilistic hydrological forecasting are reviewed and presented. Finally, the research emphasis and directions of Bayesian probabilistic hydrological forecasting are suggested as follows:(1) explaining,communicating and disseminating hydrological forecasting uncertainty information and probabilistic hydrological forecasting products scientifically and effectively;(2) setting up ensemble probabilistic hydrological forecasting framework to estimate and reduce the total uncertainty of hydrological forecasting;(3) carrying out the research on Bayesian probabilistic hydrological forecasting considering the spatial-temporal correlation of forecasting variables;(4) to further promote the application of probabilistic hydrological forecasting information in water related risk decision-making activity.
作者
刘章君
郭生练
许新发
成静清
钟逸轩
巴欢欢
LIU Zhangjun;GUO Shenglian;XU Xinfa;CHENG Jingqing;ZHONG Yixuan;BA Huanhuan(Jiangxi Provincial Institute of Water Sciences,Nanchang 330029,China;State Key Laboratory of Water Resources and Hydropower Engineering Science,Wuhan University,Wuhan 430072,China)
出处
《水利学报》
EI
CSCD
北大核心
2019年第12期1467-1478,共12页
Journal of Hydraulic Engineering
基金
国家自然科学基金项目(51909112)
江西省重点研发计划项目(20181ACG70018)
国家重点研发计划项目(2016YFC0402206)
关键词
水文预报
定量降水预报
不确定性
概率预报
集合预报
贝叶斯理论
hydrological forecasting
quantitative precipitation forecast
uncertainty
probabilistic forecast
ensemble forecast
Bayesian theory
