H53B-1227
Hierachical Landscape Controls on Channel Morphology in Mountainous River Basins in Scotland.
Scottish upland rivers exhibit some of the greatest diversity of process and form in the UK reflecting the diversity of landscape evolution history, climate and landuse. This contribution outlines key findings from the mountainous River Dee drainage basin (2300 km2), north-east Scotland. Preliminary findings ascertained using a combined field and GIS based approach indicate that the glacial legacy seems to form the dominant control on the hierarchy of landscape controls that influence the morphology and sensitivity of channel morphology at three different scales: catchment, valley segment and reach. At the catchment scale, differences in the distribution of channel type between catchments are broadly related to the varying intensity of glacial denudation and deposition. Within individual catchments at the valley segment scale, spatial differences in glacial erosion and deposition control the broad spatial distribution of channel morphology types by affecting valley topography, drift cover and base level positioning. At a finer scale, local differences in valley topography, drift cover and landuse affect the fluvial sediment supply and transport capacity regimes which in turn create channel morphology variability. By better characterizing the hierarchy of controls within the area of study that influence channel morphology, improvement of the prediction of upland channel morphology and sensitivity is possible which in turn could potentially explain the distribution of certain aquatic species and aid riverine habitat management.
H53B-1228
Characterizing Controls of Riparian Width for Mountain Streams in the Colorado Front Range
High variability of mountain streams causes riparian width to vary greatly from changes in drainage, valley and channel characteristics. GIS- based models for predicting flood-prone width, valley bottoms, or riparian zones may not accurately reflect processes at the reach scale, therefore field verification and reach-specific studies are needed. Management of riparian areas often designates a generalized width, which may under- or over-estimate the true riparian width. This study examines correlations between potential control variables and riparian zone width in the Colorado Front Range. Results from this study will be used to predict the riparian zone as a proxy for flood-prone width in the semi-arid Colorado Front Range. We hypothesize that local controls interact with large- scale controls to determine floodplain processes. Large-scale controls identified are elevation, which reflects hydroclimatology and glacial history, gradient and drainage area. Local controls are entrenchment, the ratio of the valley width to channel width, connectedness, defined as the distance from the channel to valley edge, presence of colluvium, and vegetation type, affecting roughness during flooding and bank stability. We chose twenty reaches based on elevation, connectedness, gradient and drainage area using a GIS base map in anthropogenically undisturbed areas of the Colorado Front Range, which included the Cache la Poudre and North St. Vrain drainages. Riparian width was defined using a three-tiered approach: evidence of fluvial processes and presence of riparian vegetation, compared with the Q100 stage. A longitudinal and two valley and channel cross-section surveys were completed at each stream reach to determine valley and channel geometry and bed gradient. Preliminary results show significant positive correlations between drainage area, entrenchment, and connectedness and riparian width, and negative correlations between gradient and riparian width, supporting the hypothesis that controls at various scales influence floodplain processes. Statistical analyses are used to determine the relative importance of each control variable using multiple regressions and best subsets regressions using Adjusted R2 and Mallow's Cp model selection criteria. Results from these data will allow for more accurate delineations of the riparian zone for mountain streams, in addition to an understanding of the reach-scale significance of interactions between floodplain and local hillslope versus drainage basin processes.
H53B-1229
Wood Loading in Tropical Forested Headwater Streams
The influence of wood on forested headwater streams has been explored in numerous studies conducted in temperate climate zones. As of yet few studies reported in the literature have expanded the research into other climate zones, and it is unclear which, if any, generalizations about wood-channel interactions will hold beyond the temperate zone. In this study we start to address this knowledge gap by characterizing wood loads in forested headwater streams located in the steep eastern foothills of the Central Volcanic Cordillera of Costa Rica. In surveys of 30 reaches, each 50 m long, we found the mean wood load to be 188 m3/ha (or 12.5 m3/100 m), with loads ranging from 41 to 612 m3/ha (3.0 to 34.7 m3/100 m) among the 30 study reaches. In each of the 30 study reaches we measured contributing drainage area, stream gradient, bedform roughness as quantified by the mean square error of the thalweg elevation, intermediate diameter of the 84th percentile bed material, grain sorting measured with the inclusive graphic standard deviation, average bankfull width, width to depth ratio, average valley slope, maximum valley slope, sinuosity, a surrogate for stream power, and a surrogate for unit stream power. Multiple regression analysis indicates that although no single variable adequately explains wood loading, more adequate models involving five significant variables can be constructed using best subset model selection with Mallows' Cp as the selection criterion. Variable subsets used to model wood volume per 100 m of channel and wood volume per hectare of channel differ slightly, but both have R- square values near 0.5 and adjusted R-square values of approximately 0.4. Additionally, both utilize drainage area, slope, and stream power, indicating that transport of wood through the channel network is likely to be an active and influential process in controlling wood load. Drainage area is positively correlated with wood load, suggesting that channels with large contributing networks are more likely to have large wood loads. Slope, stream power, and unit stream power are negatively correlated with wood load, suggesting that channels with an increased capacity to transport wood have lower wood loads. Channel width is a significant component of the model of wood load per hectare, but not of the model of wood load per 100 m of channel. Wider channels correlate with lower wood volumes per hectare, but with slightly higher wood loads per 100 m of channel.
H53B-1230
Geomorphic and Ecologic Interactions of Large Wood and Pacific Salmonid Redds Across Habitat Units on a Regulated California River
Large wood pieces (LW, >1 m length, >10 cm diameter) are important components of geomorphic and ecologic dynamics within river systems. Physical presence of LW within a bankful channel can influence flow, sediment deposition and scour patterns, and storage of organic matter, whereas ecologic elements of LW include hydraulic variability, habitat, and nutrient sources for aquatic species. In regulated rivers hydrologic connectivity has been lost and ecosystem dynamics disrupted, yet lower reaches continue to serve as habitat, and now as headwaters, for a myriad of species including anadromous salmonids returning to spawn and complete their life-cycles. Regardless of condition, lower reaches of regulated rivers must serve as ecosystem hotspots in response to anthropogenic manipulations of the watershed. In this research, interactions between large wood, Pacific salmonid redds, and aquatic habitat units (i.e. riffle, run, glide, and pool as defined by depth and velocity) were explored in a regulated, mid-sized (i.e. channel width is greater than most tree heights), Mediterranean-climate (i.e. smaller, softer-wood trees dominate the landscape) river draining a portion of the Sierra Nevada of California. Because watershed connectivity has been severed, riparian zones highly altered, and LW removal remains common, LW levels are thought to be very low in regulated ecosystems. The study hypothesis was that a dynamic and healthy ecosystem might have areas of low, optimal, and overabundances of wood, which would correlate to low, optimal, and low redd abundances, respectively. On the other hand, in an ecosystem where connectivity is diminished, an increase in the amount of LW may potentially convert otherwise unsuitable spawning habitat to highly preferred spawning habitat. In exploring the dynamics of wood and redds at the habitat unit scale, characteristics of 530 LW pieces, locations of 650 redds, and habitat units along a 7.5 km reach directly below a dam were mapped during a spawning season. Findings suggest that fall-run Chinook salmon preferentially spawned in riffles often where LW was present, but also spawned in glides and occasionally runs, and across all these habitats both with and without LW. No spawning was observed in pools. LW was present in all habitat types, but not in every habitat unit occurrence. There were no areas in the study reach with an overabundance (i.e. complete coverage of channel bed) of LW, nor did LW presence appear to prohibit redd activity in spawning habitats. Further research is needed to determine what quantities of LW might be considered optimal in below-dam ecosystems where suitable spawning habitat is needed.
H53B-1231
Are Landscape Hierarchies of Groundwater-Surface Water Exchange Patterning Important in Spawning Site Selection by Atlantic Salmon (Salmo salar L.,)?
The Atlantic salmon's (Salmo salar L.,) native Scottish, headwater spawning grounds, can be viewed as dynamic hot spots of biological productivity set within a hierarchical landscape sculpted by complex physico-chemical processes. Traditionally controls on female spawning site selection, have mainly been attributed to the sedimentary and hydraulic characteristics of available spawning habitat. In the UK, the influence of physico-chemical landscape hierarchies on spawning site selection is poorly understood. This study aims to provide a preliminary insight into the importance stream hydrochemistry has at different hierarchical scales, on spawning site selection by Atlantic salmon in a Scottish braided river system. During the 2005 and 2006 spawning seasons, intensive surveys of dissolved oxygen, alkalinity, trace metals and continuous temperature monitoring were undertaken under high and low flow conditions, in the surface water network of the floodplain reaches. Using GPS data within a GIS framework, these data were related to the locations utilised by spawning fish surveyed on a daily basis in each year. Results indicated that patterns of groundwater – surface water exchange were spatially and temporally dynamic, occurring at a range of scales across the channel floodplain system. A hierarchy of channel types could be differentiated on the basis of contrasting surface water quality and source water characteristics. These included channels dominated by soures such as groundwater, hillslope drainage and main-stem river water. Although most channels contained good hydraulic and sedimentary conditions, spawning was concentrated in those locations which displayed strong chemical groundwater signatures. In 2005: 64 % and 2006: 44 % spawning occurred in groundwater channel types. This study suggests that GW-SW interaction hierarchies may play an important role in determining site selection by spawning Atlantic salmon and sea trout.
H53B-1232
Relationships Between Stream – Ground Water Exchange and Topography of the Channel, Valley, and Watershed
Stream flow gains and losses represent exchange with groundwater and are commonly associated with the topography of the stream channel and contributing area. The magnitude of stream gain, i.e. runoff generation, is thought to be related to the extent and geometry of the contributing surface area. At smaller scales, the magnitude of both stream gain and loss may be related to heterogeneity in the gradient of the stream channel or valley. To validate relationships such as these between streams and their topography, we compare measurements of stream reach gains and losses to the terrain analyses of corresponding channels, valleys, and contributing areas. Comparisons are made for 26, 100-m reaches that constitute a 2.6-km long headwater stream in the Tenderfoot Creek Experimental Forest (USFS), Montana. The stream drains a 5.5 km2 catchment with a riparian area of 0.073 km2, delineated by contributing area with elevation within 2 m of the stream channel. The study stream flows over 3 geological units with valley slopes around 6.7, 5.7, and 9.0%, from upstream to downstream. For each 100-m reach, upstream and downstream discharges were measured using conservative tracer (chloride) experiments and dilution gauging techniques. In addition, the upstream release was measured at the downstream end of each reach to determine tracer mass loss and to estimate gross hydrologic loss over the reach. To close the mass balance, gross gain was calculated from the net change in discharge and gross loss. The spatial distributions of gross gains and losses were determined at multiple times during the declining summer baseflows of the snowmelt driven hydrograph. At lower baseflow conditions, several net neutral or gaining reaches also showed a 5-15% tracer mass loss, indicating that gross gain and loss operate concurrently in these reaches. We use these water balance fluxes to indicate one scale of stream – ground water exchange, and we compare patterns in exchange with the surrounding topography. Elevation data for the stream and watershed were collected using traditional survey techniques and aerial laser swath mapping (ALSM, 1-m resolution). Topographic metrics, such as channel sinuosity, valley slope, riparian area, and lateral contributing area, are calculated through terrain analyses of elevation data. Comparing topography with stream water balance is a spatially explicit approach to linking watershed structure with stream – ground water interaction, which is important to understanding solute fate and transport among the stream and adjacent ecosystems. http://www.mines.edu/~mgooseff/web_research/hydroscapes.html
H53B-1233
Classification of Watersheds for Bioassessment Based on Hydrological Variables
A procedure for the classification of watersheds for bioassessment based on their streamflow regime and prediction of hydrologic class from watershed attributes is presented. We first identified a set of stream flow regime variables relevant to biota for the purposes of characterizing the invertebrate population in a stream, that can be abstracted from long term streamflow data measured at gauged sites. The selection of these variables was based on the past literature and discussions with stream ecologists. The following variables were selected: 1) base flow index (BFI) 2) daily coefficient of variation (DAYCV) 3) average daily flow (QMEAN), 4) Number of zero flow days (ZERODAY) 5) bank full flow (Q1.67) 6) Colwell's index 7) seven day minimum (7Qmin) 8) seven day maximum (7Qmax) 9) number of flow reversals (NOR) and 10) flood frequency. These variables were computed at 543 minimally impacted stream gage stations in the thirteen states of Western US. Principal Component Analysis (PCA) and K-means clustering analysis was then used to classify the watersheds into hydrologically different groups. Linear Discriminant Analysis (LDA), Classification and Regression Trees (CART) and Random Forests (RF) models were then developed to predict the class of an ungauged watershed from watershed attributes (climate, geomorphic, geology and soil attributes). We developed a series of classifications (with K equal to 4 to 8 in K-means clustering) that showed a strong geographical structure. The classification is sensitive to the quantity of water present in the stream and it also identified streams that appear similar at monthly time scale but are significantly different at the daily time scale. These differences are important to identify the variation in the biota. For the prediction of watershed class from watershed attributes we found that the RF model was slightly better than the other modeling approaches evaluated (LDA, CART). The class characterized by high BFI was difficult to predict in all models due to the lack of good watershed attribute, among those we considered, that is a reasonable surrogate for subsurface flow. The new hydrologic classification of watersheds provides opportunities to further examine the relationships between hydrology and stream ecology.
H53B-1234
The relative importance of headwater streams, tributary junctions, and floodplains to salmonid productivity and diversity
Recent literature has focused on the ecological importance of headwater streams, tributary junctions, and large river floodplains as dynamic and important hot spots of aquatic ecosystem productivity and diversity. Yet few studies attempt to address the relative importance of such network components at the watershed scale (>103 km2) to biological productivity and diversity. We examine the relative importance of each of these attributes to salmonid productivity and diversity. Specifically we compare and contrast the role of headwater streams, tributary junctions, and floodplains to salmonid life-history diversity and multiple species productivity. Preliminary results suggest that all three attributes are critical to specific life-histories and species. Aggregate life-history and species diversity increases with increasing drainage area. Lastly, evidence from several Pacific Northwest regions also suggest that life-history diversity and productivity is sustained by asynchronous cycles of high and low productivity among populations, and by migration across multiple, adjacent watersheds.