Quantifying and Modeling the Influence of Forest on the Magnitude and Duration of Mountain Snow Storage in the Pacific Northwest, USA

Thesis (Ph.D.)--University of Washington, 2016-08

Forests strongly influence the amount and duration mountain snow storage because forest cover modifies both snow accumulation and ablation processes. Quantifying and predicting forest effects on snow processes and snow storage is critical for understanding the effects of forest change on snow storage, and subsequent impacts on downstream water resources. However, both the magnitude and direction of forest modifications of individual snow processes vary with climate, topography, and forest characteristics. Accurate prediction of the net effects of forest change on mountain snow storage, particularly in a warming climate, depends on accurately representing the spatiotemporal variability of forest-snow interactions. With a goal to better understand forest-snow processes in the maritime snow zone, we collected snow observations over four winters within diverse forest types in western Washington, USA. We utilize these new observations to quantify forest effects on snow duration, as well as to assess the robustness of remote methods to observe snow-covered area within a forest. We find that mean snow duration is 8 days longer in forest gaps than in forested plots, but that snow duration in thinned forest and dense forest are indistinguishable at the 1600 m2 plot-scale. We additionally show that time-lapse cameras and spatially distributed ground temperature sensors are both robust methods for observing snow duration, and make suggestions about the optimal spatial density of snow observations within forests. The entire four-year dataset and related metadata are extensively described, and are now publicly available for potential use in numerous modeling applications. To expand our focus on forest-snow interactions to the Pacific Northwest, USA, regional-scale, we collaborate with other research institutions and engage citizen scientists. Regional synthesis and analysis of snow depth and duration at 12 out of 14 paired open-forest locations show that differential snow duration ranges from synchronous, to snow lasting up to 13 weeks longer in the open. The differences in snow duration are attributed to forest effects on snow accumulation, with larger differences between snow accumulation rates than between ablation rates in the open and forested sites through the duration of the forest snowpack. In 2 out of the 14 locations, differential snow duration is 2-5 weeks longer in the forest. These 2 sites are subject to hourly average wind speeds ranging up to 8 and 17 m s-1. Therefore, longer snow duration in the forest likely results from a combination of enhanced deposition of snow and reduced snow loss from canopy interception in the forested sites. These findings suggest that a regional framework to understand forest effects on snow storage in the maritime to maritime-continental transitional climate across the Pacific Northwest must account for high interception efficiencies in warmer climates as well a high winds due to topographic exposure and climate. Lastly, we assess the influence of forest structural characteristics on snow storage in western Washington by linking lidar-derived forest canopy metrics to snow depth and snow duration. By using a matrix decomposition method to collapse the variance of spatially distributed observations of snow depth onto a few dominant modes, we show that the top two modes represent forest effects on snow accumulation and ablation, respectively. Furthermore, gridded metrics of canopy cover and height that quantify the canopy directly overhead, rather than to the south, correlate equally strongly (r2 of up to 0.74) with the spatial coefficients that scale both of these modes. This finding suggests that the role of forests in shading the snowpack from sunlight is diminished at this site. Furthermore, multivariate analysis of physiographic predictors of snow duration across a range of elevations and years quantifies the important role of canopy characteristics in controlling snow duration. At the study site in western Washington, the binary simplification of considering forested versus open locations is supported by evidence for a stepped response, in which snow duration shifts from longer to shorter around values of 60-70% canopy cover. Collectively, the findings demonstrate that forest effects on snow accumulation dominate the overall influence of forest on snow storage in the Pacific Northwest, USA, resulting in larger magnitude and longer duration snow storage in canopy gaps, except in locations subject to high wind speeds.