WHERE IS THE RAIN-ON-SNOW ZONE IN THE WEST-CENTRAL WASHINGTON CASCADES? MONTE CARLO SIMULATION OF LARGE STORMS IN THE PACIFIC NORTHWEST
Recognition of ROS has grown during big PNW events, as investigations based on theory, weather records and field studies clarified many aspects of the phenomena. ROS seems more important in middle elevations; land use can affect processes; climatic shifts are changing its geography and magnitude. Questions remain: how much does ROS affect the long-term frequency of water input? can we define a peak ROS zone, in some specific area in the Northwest?
Sporadic occurrence hinders study of ROS, but some issues can be addressed by modeling. A computer program using probabilistic and deterministic elements simulates large “storms” over “1000 yr”, generating realizations of weather and snowpack. In an event precipitation falls, snow accumulates/melts, and water is tracked to the ground. Frequency distributions are based on data from Washington’s Cascades, some combined into functions of elevation and date, so the model can be applied to specific places or general elevations. Validation focuses on Stampede Pass, site of a NWS station, snow course and SNOTEL on the Cascade crest. The model is evaluated by comparing statistics and frequency relations of the instrumental record against model realizations, for precipitation and water to runoff.
The model is used first to evaluate the significance of ROS varying with elevation. Over “1000 yr”, snow in some events usually reduces the water reaching the ground during model storms, more often where common and deep, but sometimes also at moderate elevations. Several metrics suggest the greatest ROS effect in this region occurs at ~800 m, where rain+melt enhance water delivery to the ground during ~20% of major storms and increases the magnitude of the rarer events.