TR-20 Segmental and Lag Time Approaches for Peak Flow: Comparison to Stream Gage Flow Frequency

One way most roadway design sections of state Department of Transportation predict peak flow for culvert design is using NRCS TR-20 methodology. Even though the approach is more than 50 years old, there is no clear guideline for how large a single watershed drainage area may be while remaining appropriate for predicting peak discharge with this method. We developed flow frequency estimates for 130 small-area stream gages in rural Nebraska and compared the estimated return period discharges with those from TR-20 using both the segmental and lag equation approaches for calculating the time of concentration. There is no significant difference in root mean square errors between peak discharges estimated using the lag and segmental methods. Overprediction of low return period discharges was common using both times of concentration. Results suggest that TR-20 error becomes larger after 15 square miles. Despite having its lowest errors for watershed areas between 1 and 15 square miles, using an uncalibrated TR-20 model is discouraged due to its likelihood to overpredict peak flows.