Precipitation Phase and Runoff Characteristics in High Relief Topography

Mountainous watersheds are characterized by high relief and complex meteorological conditions. Because temperature decreases with elevation, high relief landscapes experience strong differences in the dominant precipitation phase (i.e. rain or snow) within a given elevation zone. High elevation portions of mountainous watershed are often snow dominated, whereas mid and low elevations receive varying proportions of rain and snow. The dominate phase of precipitation can vary extensively over small spatial scales and short temporal extents making it difficult to characterize the dominant hydrologic regime within different elevation zones of mountainous landscapes (Figure 1 & 2).

The idea that meteorological and hydrologic process can be quite variable and complex in mountainous topography is not a new realization; however, there is a lack of research documenting these complex relationships in large mountainous watersheds. Meteorological observation networks tend to be sparse in mountainous terrain and lack the ability to characterize the spatial variability in precipitation phase and storage of precipitation as snowpack. Furthermore, streamflow measurement networks tend to monitor mainstem channels of large-drainage-area, high-relief watersheds; this type of monitoring approach mixes the meteorological inputs of low-elevation rain-dominated terrain and high-elevation snow-dominated terrain into a composite streamflow signal. Thus, direct quantification of the influence of rain or snow on hydrograph form using traditional streamflow gaging networks is difficult. This ambiguity of the influence of precipitation phase on hydrograph form limits our ability to accurately predict how changes in precipitation phase associated with climate warming will affect water resources and hydrologic processes in mountainous landscapes.

In the Salmon River watershed, central Idaho, United States, complex interactions between the atmosphere and mountain topography produce spatial and temporal variability in local climate and weather patterns. For example, a day flight over the Salmon River watershed reveals the significant variability in the extent of snow cover in sub-catchments throughout the watershed (Figure 2). Snow cover is patchy and thin in low elevations, more extensive with minor patchiness in mid elevations and completely covers the landscape at high elevations. Temporal variability occurs in changes in the elevation of the snow line with individual storm events and seasonal and annual fluctuations. Meteorological analysis reveals how often a given portion of the landscape receives rain or snow and can be used to classify tributary watersheds as snow-dominated, rain-dominated or influenced by a balance of rain and snow (Figure 3).

Runoff characteristics of tributary watersheds in the Salmon basin are influenced by complex spatial and temporal precipitation patterns. The phase and intensity of precipitation affects the timing, magnitude, and rates of change in water flux through channels. In the rain-dominated portions of mountainous watersheds precipitation is readily transported through channels; thus there is little to no storage of water on the landscape. The only impediment to the process is watershed mechanisms that act as a filter (soil, vegetation, ground water storage, etc.). In snow-dominated watersheds precipitation is accumulated until temperatures warm enough to melt the snowpack and release the water for transport through channels.

In the study presented here, we designed a streamflow monitoring network to isolate the influence of precipitation phase on hydrograph form. This was accomplished by measuring streamflow in tributary catchments that are bound within elevations that correspond to a dominant precipitation phase. Four tributaries were selected within low, mid, and high elevations and are referred to as rain-dominated (RD), mixed rain and snow (MRS) and snow-dominated (SD).

Meteorological analysis using meteorological stations, remote sensing, and physical-driven precipitation models confirmed that elevation is a good proxy for precipitation phase. For the period from 2003-2012, low elevation watersheds received precipitation as 80% rain and 20% snow, mid elevations 50% rain and 50% snow, and high elevations 30% rain and 70% snow. Differences in precipitation phase generated contrasts in the timing of water movement through the landscape, the magnitude and duration of flood events, the frequency of hydrologic events, the probability of channel-drying during summer months, and the overall form of the hydrograph (Figure 4). RD channels were characterized by active hydrographs during winter months when precipitation events caused rapid departures from and returns to baseflow. The timing and dominance of rain during winter months increased the likelihood of channel-drying during dry summer months for RD watersheds. MRS watersheds have characteristics of both RD and SD watersheds. When snow accumulation was significant, the overall form of the hydrograph was snow-dominated (i.e. long rising and falling limbs during spring snowmelt); however, a high proportion of rain caused spikier snowmelt events (reflecting rain on snow events) and earlier movement of water through channels. The SD watersheds have the latest seasonal runoff and tend to have smooth rising and falling limbs. Runoff from the SD watersheds was of higher magnitude than runoff from RD and MRS watersheds and tended to be longer in duration.

The results from this study are important because they demonstrate the hydrologic variability that exists within high-relief topography and highlight the need to look beyond traditional streamflow gaging techniques to better understand how changes in climate will affect hydrologic processes and water resources.

• National Operational Hydrologic Remote Sensing Center. 2004. Snow Data Assimilation System (SNODAS) Data Products at NSIDC, [2003 - 2010]. Boulder, Colorado USA: National Snow and Ice Data Center. Digital media. http://bcal.geology.isu.edu/statplanet/