Record Details
Watershed and climate influences on flood frequency distributions in the Willamette River basin
ScholarsArchive at Oregon State University
| Field | Value |
|---|---|
| Title | Watershed and climate influences on flood frequency distributions in the Willamette River basin |
| Names |
Tolle, Timothy Victor
(creator) Rosenfeld, Charles (advisor) |
| Date Issued | 1978-05-18 (iso8601) |
| Note | Graduation date: 1979 |
| Abstract | Problem areas identified are: The needs a) to develop curve fitting procedures to estimate flood flow distributions; b) to incorporate precipitation data into frequency analysis procedures; and c) to better define flood potentials for ungauged streams. Closely related to the above, it is noted that variables commonly used to estimate flood magnitudes usually do not serve as optimum predictors. Hydrophysical variables are shown to increase the ability to estimate flood magnitudes of all frequencies within the Willamette River basin of western Oregon. The major hypotheses tested show that the traditional "regionalization" procedure is not hydrologically sound. The objective herein is to regionalize flood distributions by basin parameters. These parameters include climatic variables that are hydrologically meaningful and easy to measure or estimate. Regression equations are developed to estimate the descriptive statistics of each stream's flood distribution. The mean of the logarithms of the annual peak flows describes numerically the position of a flood frequency curve. The standard deviation of the logarithms of the peak flows describes numerically the slope of the flood frequency curve assuming a log-normal or log Pearson type III distribution. Watershed relief is the strongest estimator of flood distributions for all but the low elevation streams (R² = 56%). Relief influences within-basin climatic variability. Large flood climatology differences yield desynchronized runoff from the various source areas. Only rarely do the different elevation areas synchronize and then they yield exceptionally large peak flows. Antecedent conditions are shown to be related to flood distributions. Those basins with large year-to-year differences in antecedent conditions have steeper flood frequency curves (R² = 76%). Precipitation zones, soil permeability, and forest cover all influence antecedent conditions. Drier basins with permeable soils and forest cover have steeper flood frequency curves. Lakes and ponds act as sinks on all peak flows, and streams which flow through them have less steep flood frequency curves. In combination with watershed relief and mean annual precipitation, these three variables explain 86% of the stream-to-stream differences in the slopes of flood frequency curves in the Willamette River basin. Drainage area, terrain roughness, forest cover, and precipitation explain over 95% of the stream-to-stream differences in average flood magnitudes. Steeper slopes yield larger flood peaks as does lesser forest cover. As elevation increases, average flood magnitudes decrease, but extreme peak flows are larger. Collectively these estimation equations reduce standard errors of 50-year floods from over 40% to less than 14%. They also reduce standard errors of ten-year and 25-year floods from 46% to 15%. |
| Genre | Thesis/Dissertation |
| Topic | Floods -- Willamette River Basin |
| Identifier | http://hdl.handle.net/1957/42607 |
