An Evaluation of Small Catchment Flood Estimation Techniques

ABSTRACT Hydraulic control structures are frequently designed based on limited data and pragmatic modeling approaches. The success of the design depends largely on the experience and judgement of the engineer. An evaluation was made of the suitability of using the Rational Method, a unit hydrograph procedure, a time-area method, and a kinematic model to predict flood flows on several watersheds in Southern Africa. In order to test the flood estimation techniques for a range of conditions, watersheds were selected based on size, geographic location and land use. Watershed sizes ranged from a few hectares to nearly 100 km^, and land uses ranged from 100% urban to 100% forest. In general, each method worked best for the particular land-use condition it was designed to model. However, the performance of all the models was very disappointing, particularly when used in an uncalibrated mode. The study illustrated deficiencies in: (1) user documentation for these hydrologic models; (2) documented model input parameters; (3) field and laboratory methods to determine model input parameters; and (4) the physically-based model algorithms.