Testing of Several Runoff Models on an Urban Watershed.
Abstract
Six models, plus two variants of one and a variant of another, were tested with the objective of making a preliminary evaluation of their relative capabilities, accuracies, and ease of application. For four of the models, plus two variants of one of them, the primary performance criterion was the degree to which simulated values matched observed daily and monthly runoff volumes for the 5.5 square mile Castro Valley Watershed near Oakland, California. In addition, tests were performed for several individual runoff events for all six models. The results showed that each model could be calibrated on a single set of data and verified with acceptable accuracy on a different data set. The ease of application was decidedly different for all models, due to the differing level of detail in input data required. Going from the simplest to most difficult to apply, the continuous models rank as follows: STORM, HEC-1C, SSARR, and HSP. Similar ranking of the single-event models is: HEC-1, SWMM and MITCAT. Also, a recent capability added to the STORM model (i.e., SCS procedures for computing runoff and routing) produced more accurate results than the coefficient method of computing quantity of runoff incorporated in the original version of STORM. These limited tests were not intended to serve as a basis for comparison of the accuracy of the various models. However, they did show that the more complex models did not produce better results than the simple models for the Castro Valley Watershed data. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 1978
- Accession Number
- ADA106258
Entities
People
- Jess Abbott