[1] The results of a 3 year field study to observe the processes controlling snow interception by forest canopies and under canopy snow accumulation and ablation in mountain maritime climates are reported. The field study was further intended to provide data to develop and test models of forest canopy effects on beneath-canopy snowpack accumulation and melt and the plot and stand scales. Weighing lysimeters, cut-tree experiments, and manual snow surveys were deployed at a site in the Umpqua National Forest, Oregon (elevation 1200 m). A unique design for a weighing lysimeter was employed that allowed continuous measurements of snowpack evolution beneath a forest canopy to be taken at a scale unaffected by variability in canopy throughfall. Continuous observations of snowpack evolution in large clearings were made coincidentally with the canopy measurements. Large differences in snow accumulation and ablation were observed at sites beneath the forest canopy and in large clearings. These differences were not well described by simple relationships between the sites. Over the study period, approximately 60% of snowfall was intercepted by the canopy (up to a maximum of about 40 mm water equivalent). Instantaneous sublimation rates exceeded 0.5 mm per hour for short periods. However, apparent average sublimation from the intercepted snow was less than 1 mm per day and totaled approximately 100 mm per winter season. Approximately 72 and 28% of the remaining intercepted snow was removed as meltwater drip and large snow masses, respectively. Observed differences in snow interception rate and maximum snow interception capacity between Douglas fir (Pseudotsuga menziesii), white fir (Abies concolor), ponderosa pine (Pinus ponderosa), and lodgepole pine (Pinus contorta) were minimal.

No abstract

We present an approach to diagnosing salmonid habitat degradation and restoring habitat‐forming processes that is focused on causes of habitat degradation rather than on effects of degradation. The approach is based on the understanding that salmonid stocks are adapted to local freshwater conditions and that their environments are naturally temporally dynamic. In this context, we define a goal of restoring the natural rates and magnitudes of habitat‐forming processes, and we allow for locally defined restoration priorities. The goal requires that historical reconstruction focus on diagnosing disruptions to processes rather than conditions. Historical reconstruction defines the suite of restoration tasks, which then may be prioritized based on local biological objectives. We illustrate the use of this approach for two habitat‐forming processes: sediment supply and stream shading. We also briefly contrast this approach to several others that may be used as components of a restoration strategy.

We modeled large woody debris (LWD) recruitment and pool formation in northwestern Washington streams after simulated stand‐clearing disturbance using two computer models: Forest Vegetation Simulator for stand development and Riparian‐in‐a‐Box for LWD recruitment, depletion, and pool formation. We evaluated differences in LWD recruitment and pool formation among different combinations of channel size, successional pathway, and stand management scenario. The models predict that time to first recruitment of pool‐forming LWD is about 50% shorter for red alder Alnus rubra than for Douglas‐fir Pseudotsuga menziesii at all channel widths. Total LWD abundance increases faster in red alder stands than in Douglas‐fir stands but declines rapidly after 70 years as the stand dies and pieces decompose. Initial recovery is slower for Douglas‐fir stands, but LWD recruitment is sustained longer. Total LWD abundance increases faster with decreasing channel size, and pool abundance increases faster with decreasing channel width and increasing channel slope. The models predict that thinning of the riparian forest does not increase recruitment of pool‐forming LWD where the trees are already large enough to form pools in the adjacent channel and that thinning reduces the availability of adequately sized wood. Thinning increases LWD recruitment where trees are too small to form pools and, because of reduced competition, trees more rapidly attain pool‐forming size. On channels less than 20 m wide, thinning of red alder and underplanting shade‐tolerant conifers will reduce near‐term alder recruitment and increase long‐term conifer recruitment. However, the same treatment on channels more than 20 m wide may increase both near‐term and long‐term recruitment. Compared with the natural fire regime, timber harvest rotations of 40–80 years during the past century have reduced the percentage of riparian stands that can provide LWD of pool‐forming size to streams, especially in channels at least 20 m wide.

Despite increasing attention to management of headwater streams as sources of water, sediment, and wood to downstream rivers, the extent of headwater channels and perennial flow remain poorly known and inaccurately depicted on topographic maps and in digital hydrographic data. This study reports field mapping of channel head and perennial flow initiation locations in forested landscapes underlain by sandstone and basalt lithologies in Washington State, USA. Contributing source areas were delineated for each feature using a digital elevation model (DEM) as well as a Global Positioning System device in the field. Systematic source area-slope relationships described in other landscapes were not evident for channel heads in either lithology. In addition, substantial variability in DEM-derived source area sizes relative to field-delineated source areas indicates that in this area, identification of an area-slope relationship, should one even exist, would be difficult. However, channel heads and stream heads, here defined as the start of perennial flow, appear to be co-located within both of the lithologies, which together with lateral expansion and contraction of surface water around channel heads on a seasonal cycle in the basalt lithology, suggest a controlling influence of bedrock springs for that location. While management strategies for determining locations of channel heads and perennial flow initiation in comparable areas could assign standard source area sizes based on limited field data collection within that landscape, field-mapped source areas that support perennial flow are much smaller than recognized by current Washington State regulations.