G. Padmanabhan, professor of civil engineering, and Brent Johnson, graduate student in civil engineering, have written an article titled “Regression Estimates of Design Flows for Ungaged Sites Using Bankfull Geometry and Flashiness” that will appear in the international journal Catena. Padmanabhan is Johnson's adviser.
In their study, Johnson and Padmanabhan used bankfull geometry data at 22 U. S. Geological Survey stream flow-gaging sites in the Red River Basin of Minnesota and North Dakota for developing regression estimates of flows of different return periods. Regression equations were developed first by using only basin characteristics and then by using only bankfull geometry data as independent variables. Flashiness was included later as an independent variable in both cases. Stream flashiness was found to be an important explanatory variable in both bankfull geometry and basin characteristics regressions.
According to the authors, the bankfull geometry data produced slightly better regression equations than the basin characteristics data. The bankfull geometry regressions will be useful to predict design flows and to compare with flow estimates developed by traditional methods for ungaged sites in the Red River Basin.