USGS Washington Water Science Center
|Project Home | Publications and Products | Project Summaries | Partnerships ||
WA370 - Magnitude and Frequency of Water Available for Runoff During Rain-on-Snow Events at Selected Sites in Washington State - Completed FY1996
Problem - Runoff during rain-on-snow events has been associated with mass-wasting of hillslopes, damage to riparian zones, downstream flooding and associated damage, and loss of life. Forest managers, among others, in Washington State are concerned about the effects that timber harvest may have on runoff during rain-on-snow events. Watershed studies in the Pacific Northwest suggest that timber investigations have shown how a forest canopy can affect the amount of water available for runoff (WAR) during rain-on-snow events. Our current state of knowledge regarding timber-harvest effects on rain-on-snow events is a fair understanding of the physical processes involved and how these processes may be affected by both timber-harvest and topography. There is a need to look at the significance of rain-on-snow events with a region-wide, probabilistic perspective. To do this, one needs to investigate the frequency of occurrence of rain-on-snow events in different climatic regions, and to estimate the magnitude of WAR that could be expected from sites with different amounts of forest cover, and different slopes and aspects. 10/01/93 (Update for FY 1994) Runoff during rain-on-snow (ROS) events has been associated with mass-wasting of hillslopes, damage to riparian zones, downstream flooding, and associated damage and loss of life. Forest managers, amoung others, in Washington State are concerned about the effects that timber harvest may have on runoff during ROS events, because based on available data and theoretical considerations, runoff during ROS events is greater in open than forested areas. There is a need to look at the effect different rain-storm and antecedent snowpack conditions have on the water-available-for-runoff (WAR) in open areas during ROS events, and how frequently specific storm conditions are expected in the ROS zone of western Oregon and Washington.
Objectives - The objectives of this study are to: (1) calibrate and test a numerical model for estimating WAR during rain-on-snow events. The model will be calibrated and tested using detailed climate and WAR data that have been collected at 22 sites in the Pacific Northwest; (2) to use the above model for investigating the effects of forest-canopy density, land slope, and aspect on the magnitude and frequency of WAR during rain-on-snow events. These events will be investigated at 38 National Weather Service sites in Washington State; and (3) to compare the magnitude and frequency of WAR from rain-on-snow events to the magnitude and frequency of WAR from rainfall-only events. This will be done at the same 38 sites. 10/01/93 (Update for FY 1994) The objectives of this study are to: (1) verify a numerical model for estimating WAR from open land during ROS events; (2) use the verified numerical model to investigate the effects of different rain-storm and antecedent snowpack conditions on the magnitude of WAR during ROS events; and (3) determine the frequency of occurrence of different rain-storm conditions in the ROS zone of western Oregon and Washington that may lead to ROS events.
Approach - The interception and snow-accumulation and melt routines of the USGS' Precipitation Runoff Modeling System (PRMS) numerical model will be used to synthesize the WAR data from existing climate data. The PRMS model has not had much use in this climate region, so some refinement of the model algorithms will be done. The PRMS model will be calibrated and tested using existing data sets from the Pacific Northwest. The data sets, or subsets thereof, will be split into two groups--one group will be used for model calibration and the other will be used for model validation. The tested model will be used to synthesize long-term time-series data of WAR at the 38 weather-station sites. The synthesized WAR records will then be statistically analyzed to determine magnitude-frequency relationships, one relationship for each cover-slope-aspect condition at each weather station site. The magnitude-frequency relations for each cover-slope-aspect condition at a given site will be compared with one another for significant differences. The final task will be to calculate magnitude-frequency relations for the precipitation data alone at the 38 weather stations, and to compare these relations to those previously determined from the WAR data.
10/01/03 (Update for FY 1994)
The point-scale, energy-balance snow-accumulation-and-melt model by Marks (Marks, Danny, 1988, Climate, energy exchange, and snowmelt in Emerald Lake Watershed, Sierra Nevada: Santa Barbara, University of California, Ph.D. dissertation, 149 p.) will be used to simulate accumulation and melt at clear-cut data collection sites in the western Cascades of Oregon and Washington. Following verification of the model's performance, the sensitivity of snowmelt generation to different combinations of atmospheric and antecedent snowpack conditions will be computed. The relative occurrence of rain-storm conditions that may lead to ROS and for which snowmelt sensitivity was calculated will be found from a synthetic, historical, climate time-series, representative of the ROS zone of the western Cascade Mountains of Oregon and Washington. The latter is created by combining historical, hourly, climate data from Stampede Pass and SeaTac National Weather Service (NWS) stations in Washington State.