Prediction equations were developed for estimating the flow
regime at ungaged stream locations in the Oregon coastal range.
Principal components analysis was used to screen the initial data set
of physiographic and meteorological parameters. The final regression
equations for predicting mean monthly flow had standard errors of
estimate ranging from 3 to 42 percent, with an average standard error
of 13.5 percent. A linear prediction equation was found to give the
best results for drainage basins larger than 150 square miles, while
a logarithmic equation gave best results for basins of less than 150
square miles in area. A simple linear relationship was also established
between mean monthly flow and the standard deviation of
monthly flow. A test on an independent sample indicated that the
monthly estimates of standard deviation made using the simple linear
relations were comparable to those reported by others using equations containing physiographic and meteorological parameters.
Equations were also developed to forecast monthly streamflow for Oregon coastal streams. When observed rainfall for the current month was used, the average standard error of the forecast equations was 18 percent. The use of the National Weather Service's 30-day precipitation outlooks in forecasting monthly streamflow was also investigated. The results showed that the forecasts based upon the 30-day outlook precipitation were worse than those based upon median historical precipitation. It was suggested that the monthly streamflow forecast equations could best be applied on a probability basis. / Graduation date: 1974
| Identifer | oai:union.ndltd.org:ORGSU/oai:ir.library.oregonstate.edu:1957/35287 |
| Date | 13 July 1973 |
| Creators | Orwig, Charles Edwin |
| Contributors | Klingeman, Peter C. |
| Source Sets | Oregon State University |
| Language | en_US |
| Detected Language | English |
| Type | Thesis/Dissertation |
| Relation | Lower Columbia Explorer, Explorer Site :: Oregon Explorer |
Page generated in 0.0023 seconds