Integration of a Kalman filter in the HBV hydrological model for runoff forecasting

被引:9
作者
Ouachani, Rim
Bargaoui, Zoubeida
Ouarda, Taha
机构
[1] Ecole Natl Ingn Tunis, Lab Modelisat Hydraul & Environm, Tunis 1002, Tunisia
[2] Inst Natl Rech Sci Eau Terre & Environm, Quebec City, PQ G1K 9A9, Canada
来源
HYDROLOGICAL SCIENCES JOURNAL-JOURNAL DES SCIENCES HYDROLOGIQUES | 2007年 / 52卷 / 02期
关键词
data assimilation; Kalman filter; HBV model; updating; rainfall-runoff modelling; real time;
D O I
10.1623/hysj.52.2.318
中图分类号
TV21 [水资源调查与水利规划];
学科分类号
081501 ;
摘要
A standard Kalman filter was developed with the purpose of updating certain state variables of the lower zone function and the output of the HBV model. The global version of the HBV model, functioning with an hourly time step, was calibrated by trial and error to the Tunisian basins of the Barbara and Mellila Wadis. The daily time step was considered for evaluating the quality of the data reconstitution. The model was then coupled with the Kalman filter by reformulating the conceptual model as a system. of dynamic state equations and implementing the Kalman filter procedure. The hourly discharge hydrograph constitutes the introduced measurement to correct state variables. Despite a literature-driven selection of the errors on measurements and system states, this filter clearly allowed reduction of the uncertainty on discharge produced by the rainfall-runoff process and the precipitation input.
引用
收藏
页码:318 / 337
页数:20
相关论文
共 38 条