Improved parameter inference in catchment models: 2. Combining different kinds of hydrologic data and testing their compatibility

摘要: Often some of the parameters catchment models fitted to runoff data are poorly determined thereby making task developing useful regionalization relationships more difficult. The Bayesian methodology developed in part 1 is extended utilize several kinds hydrologic parameter inference, goal being improve precision parameters. concept compatibility using statistical hypothesis tests. Different said be compatible if differences between their not statistically significant. pooling incompatible may undermine model's ability predict and also induce bias A hierarchy three levels information introduced enable systematic checking for compatibility. Finally, a case study presented. Using on runoff, soil moisture, interception, it shown that substantial reductions uncertainty can realized; importance testing demonstrated.