Résumé :
|
The combined use of radar hydrology with distributed hydrological modelling creates powerful numerical tools for forecasting storms, floods and related natural disasters. Such tools, when used with historical observations, can also contribute rich information for water resources assessment. Considerable attention has been focused on these developments. However, fundamental issues of model development associated with the existence of uncertainties or errors in radar rainfall estimation and prediction, and in distributed model parameters, still remain unresolved. Clarification as to how hydrological prediction accuracy will be improved is essential. Scale issues also need further study. Suitable model resolution and required forcing resolution might be a function of catchment scale, but the relationship between them is still unknown. In association with the scale issues, assessing in what environment a distributed model shows better performance than a lumped model, is also an important research topic.
|