New Approaches to Real Time Flood Forecasting: Weather Radar and Fluvial Geomorphology in Hydrology

美国有两项重要的技术进步，一是数字化地形图对全国的有效覆盖，分辨率达到30米× 30米，现在可以用来开发实时洪水预报方法。该项目将开发这样一个模型，其目标是通过使用尽可能多的数据获得模型的简约参数化。降水预报将基于降雨发展在时间和空间上的连续公式。降雨和雷达数据将通过分布式滤波算法共同用于更新模型状态。提出的降雨径流模型是基于地貌和水文响应的分布式知识，并将利用实验室范围内的雷达设施和河流流域在实地测试这一概念。

Two important technological advances in the United States, one, the effective coverage of the nation by the digitized topographic maps to resolutions of 30 meter by 30 meter pixels can be used now to develop real time flood forecasting methods. The project will develop such a model with the goal of obtaining parsimonious parametrization of models by using as much data as possible. Precipitation prediction will be based on a continuous formulation, in time and space, of the development of rainfall. Raingage and radar data will be used conjunctively to update model states via a distributed filtering algorithm. The rainfall-runoff model proposed is based on the distributed knowledge of the geomorphology and the hydrologic response will be suggested and this concept will be tested in the field by making use of radar facilities and river basins within the laboratory's range.