Predicting Storm Runoff on Small Experimental Watersheds

Careful estimates of rates and. amounts of runoff are required for design of flood retarding and channel stabilizing structures, and bridges and culverts in small upstream watersheds. The period of record of many of the small experimental watersheds in the United States is too short to permit direct analysis for reliable estimates of the magnitude and frequency of storm runoff. A method is presented for extending the period of runoff records based on analysis of existing short term records of rainfall and runoff for the watershed, and a longer record of the rainfall alone. The method involves: (1) estimating storm runoff volumes from the rainfall pattern and antecedent rainfall, and (2) distributing this runoff through an adaptation of the unit hydrograph principle. A method is also presented for developing synthetic unit hydrographs for ungaged areas.