Abstract:
Summer low-flow behavior in the semi-arid areas of the Pacific Northwest was studied. Long-term data from thirty-eight streams/rivers and thirty-six precipitation stations was exploited. The study area was divided into five zones based on annual average precipitation. Trends and patterns
in summer low-flows and precipitation were identified. Different zones showed different trends and patterns in precipitation over the period of record but significant
similarities within each zone. Most of the summer low-flow and precipitation records showed that the 1930's and the late 1980's to early 1990's experienced major droughts. A
possible return interval of 50-60 years appeared reasonable for similar major droughts in the study area. Spring, summer and fall precipitation, on average, were
found insignificantly related to the summer low-flows, except for northern Idaho and southeastern Oregon where summer precipitation was significant in explaining the summer low-flow trends. Winter and annual precipitation were found significantly related to summer low-flows. But the zonal equations constructed to predict summer low-flows using precipitation alone were considered unsuitable for
practical use. Zonal and regional recession models to forecast summer stream flows with significant accuracies were constructed
successfully. Extreme summer low-flows were not significantly related to different watershed cover types in eastern Oregon. However, percentage of rangelands appeared
to be more related to the extreme summer low-flow than other cover types. An extreme summer low-flow prediction model was
constructed using several watershed and precipitation variables. Many of these variables were found to be significantly related to extreme summer low-flow. Watershed
average width and annual minimum precipitation explained 71% of the variations in the extreme summer low-flow. The model finally selected, with the inclusion of watershed end point elevation, was able to explain 79% of the variability in the extreme summer low-flow. Stream and precipitation gauges need to be carefully maintained during dry periods. Also,
generalization of climatic trends based on a few observations in a large region can be misleading.
