No abstract

Modern land-use planning and conservation strategies at landscape to country scales worldwide require complete and accurate digital representations of river networks, encompassing all channels including the smallest headwaters. The digital river networks, integrated with widely available digital elevation models, also need to have analytical capabilities to support resource management and conservation, including attributing river segments with key stream and watershed data, characterizing topography to identify landforms, discretizing land uses at scales necessary to identify human-environment interactions, and connecting channels downstream and upstream, and to terrestrial environments. We investigate the completeness and analytical capabilities of national to regional scale digital river networks that are available in five countries: Canada, China, Russia, Spain, and United States using actual resource management and conservation projects involving 12 university, agency, and NGO organizations. In addition, we review one pan-European and one global digital river network. Based on our analysis, we conclude that the majority of the regional, national, and global scale digital river networks in our sample lack in network completeness, analytical capabilities or both. To address this limitation, we outline a general framework to build as complete as possible digital river networks and to integrate them with available digital elevation models to create robust analytical capabilities (e.g., virtual watersheds). We believe this presents a global opportunity for in-country agencies, or international players, to support creation of virtual watersheds to increase environmental problem solving, broaden access to the watershed sciences, and strengthen resource management and conservation in countries worldwide.

We developed an automated procedure for modeling spatial distribution of fish occurrence using logistic regression models and geographic information system (GIS) tools. Predictors were measured from a digital elevation model (DEM) and stream layers. We evaluated the accuracy of GIS measures of reach slope through a comparison with field measures. Resource selection function models were used to explain presence-absence of bull trout (Salvelinus confluentus), rainbow trout, (Oncorhynchus mykiss), nonnative brook trout (Salvelinus fontinalis), and all fishes. Our models were extrapolated based on low, medium, and high levels of probability to produce reach-scale maps across 12 000 km2. We attempted to improve models by adding land-use variables; however, the terrain best suited to road building and harvest also contained the habitat selected by rainbow trout, whereas bull trout generally selected terrain too steep for land use. These confounding factors emphasize the need for process-based investigations in addition to correlative approaches to identify habitat requirements. This automated method provides a rapid evaluation of fish habitat across remote areas useful for salmonid conservation and research planning.

Paired-catchment studies conducted on small (< 10 km2) rain-dominated catchments revealed that forest harvesting resulted in a period of increased warm-season low flows ranging from less than five years to more than two decades, consistent with the results of stand-level studies and process considerations. Of the five paired-catchment studies in snow-dominated regions, none revealed a statistically significant change in warm-season low flows in the first decade following harvest, although two exhibited non-significant higher flows in August and September and one had lower flows. Two studies, one of rain-dominated catchments and one of snow-dominated catchments, found that summer low flows became more severe (i.e., lower) about two decades or so following harvest. These longer-term results indicate that indices such as equivalent clearcut area, as currently calculated using monotonic recovery curves, may not accurately reflect the nature of post-harvest changes in low flows. Studies focussed on medium to large catchments (tens to thousands of km2 in area) found either no statistically significant relations between warm-season low flows and forest disturbance, or inconsistent responses. Attempts to synthesize existing studies are hampered by the lack of a common low-flow metric among studies, as well as detailed information on post-harvest vegetation changes. Further fieldresearch and process-based modelling is required to help elucidate the underlying processes leading to the results from these paired-catchment studies and to enhance the ability to predict streamflow responses to forest harvesting, especially in the context of a changing climate. KEYWORDS: streamflow; forestry; low flows; fish habitat; hydrologic recovery

[1] Lithologic transitions and glaciations create complex longitudinal profiles that control contemporary erosion and deposition processes. In areas with these characteristics, traditional morphometric approaches for predicting process domains, such as area-slope plots, can be augmented by considering other predictors measured from high resolution lidar-derived digital elevation models (DEMs). Ordinal logistic regression was used to model the distribution of hillslope, swale, colluvial channel, and fluvial channel domains, as identified during field surveys. The study area was a glaciated region of the Rocky Mountain foothills with a complex lithostructural setting. Relationships between domains and a suite of geographic information system-derived descriptors were explored. Predictors included profile anomalies measured at the reach and basin scale using a normalized stream length-gradient (SL/k) index. Drainage area was the dominant factor controlling domains. A model with area as the only predictor was 82% accurate. Reach slope relations were not consistent. A model that also included lithology and basin-scale SL/k index variation was 87% accurate. Domain transitions had larger area thresholds in basins with resistant conglomerate versus sandstone or shale formations and where SL/k index was more variable along a profile. In a restricted model of hillslope, swale, and colluvial channel domains, profile curvature measured over 100 m was also related to domain occurrence. A model for regional-scale mapping applications with six additional predictors was 95% accurate. The results showed that ordinal logistic regression can be used to predict and map process domains in regions with complex physiography using descriptors measured from highresolution DEMs.