Interactions Between Physical and Biological Processes in Riverine Landscapes I: Posters
Presiding: J M Buffington, USDA Forest Service; C V Baxter, Colorado State University; A E Rosenberger, University of Idaho and USDA Forest Service
B33E-01 1330h
Physical and Biological Interactions of Intact Watersheds: Are they relevant?
Intact watersheds are rare or absent throughout most of the northern hemisphere. Southeast Alaska is an exception with many forested watersheds that either have not been or are lightly disturbed by human activity. The Kadashan River, a short (less than 20 km) river in an old growth watershed on Chichagof Island is one example of an intact watershed. Natural processes from the headwaters to the ocean interact to create a salmon rich environment. Steep headwater streams are sources for sediment recruitment, nutrients (organic material and invertebrates), and large wood. Large riparian trees entering the stream provide long term storage of sediment, create pools, and link the main stream to off-channel habitats. A natural discharge regime that floods a complex riparian zone with abundant beaver ponds is a key process that maintains this complex system. Examples of these processes and salmonid populations in the Kadashan River illustrate how they interact to support one of the more productive watersheds in southeast Alaska. Understanding and quantifying processes in intact watersheds provide an opportunity to test major river paradigms and provide reference points for watershed restoration projects elsewhere.
B33E-02 1330h
Development of a General Modeling Framework for Investigating Complex Interactions among Biological and Physical Ecosystem Dynamics
Historically, physical scientists have developed models with highly accurate governing equations, while biologists have excelled at abstraction (the strategic simplification of system complexity). These different modeling paradigms yield biological (e.g. food web) and physical (e.g. hydrologic) models that can be difficult to integrate. Complex biological dynamics may be impossible to represent with governing equations. Conversely, physical processes may be oversimplified in biological models. Using agent-based modeling, a technique applied widely in social sciences and economics, we are developing a general modeling system to integrate accurate representations of physical dynamics such as water and heat flux with abstracted biological processes such as nutrient transformations. The modeling system represents an ecosystem as a complex integrated network of intelligent physical and biological "agents" that store, transform, and trade ecosystem resources (e.g., water, heat, nutrients, carbon) using equations that describe either abstracted concepts and/or physical laws. The modular design of the system allows resource submodels to be developed independently and installed into the simulation architecture. The modeling system provides a useful heuristic tool to support integrated physical and biological research topics, such as the influence of hydrologic dynamics and spatio-temporal physical heterogeneity on trophic (food web) dynamics and/or nutrient cycling.
B33E-03 1330h
Hydrogeomorphic Classification of Benthic Macroinvertebrate Assemblages in the Pacific Northwest Mountains
Existing physical classifications for streams do not explicitly include considerations of how primary environmental drivers such as flow regime, geophysical setting, intermediate-scale geomorphic processes, and anthropogenic impacts vary in importance across spatial scales in structuring biotic communities. We present a multi-scale hydrogeomorphic classification of 307 sites in 6 mountainous ecoregions of the Pacific Northwest. Innovative descriptors of flow regime, geophysical setting, valley geomorphology, and substrate are used to construct both a priori and a posteriori hydrogeomorphic classifications. Clusters based on benthic macroinvertebrate data from R/EMAP sites provide a neutral-model classification representing the best achievable classification of benthic invertebrate assemblages and the basis for a posteriori classification. Given the emphasis on geographically-dependent classifications in bioassessments, we also group sites by ecoregion. The relative strengths of the four classification schemes are assessed by the extent to which average within-class assemblage similarities exceed the average similarity between classes based on the Sorensen-Dice (presence / absence) similarity measure. A better understanding of the key hydrogeomorphic processes that filter benthic macroinvertebrate communities provides a process-based framework for interpreting biomonitoring information from diverse watershed contexts and systematically selecting reference streams that are similar with regard to the most relevant physical drivers within a region.
B33E-04 1330h
Explanatory Power of Multi-scale Physical Descriptors in Modeling Benthic Indices Across Nested Ecoregions of the Pacific Northwest
Using over 300 R/EMAP sites in OR and WA, we examine the relative explanatory power of watershed, valley, and reach scale descriptors in modeling variation in benthic macroinvertebrate indices. Innovative metrics describing flow regime, geomorphic processes, and hydrologic-distance weighted watershed and valley characteristics are used in multiple regression and regression tree modeling to predict EPT richness, % EPT, EPT/C, and % Plecoptera. A nested design using seven ecoregions is employed to evaluate the influence of geographic scale and environmental heterogeneity on the explanatory power of individual and combined scales. Regression tree models are constructed to explain variability while identifying threshold responses and interactions. Cross-validated models demonstrate differences in the explanatory power associated with single-scale and multi-scale models as environmental heterogeneity is varied. Models explaining the greatest variability in biological indices result from multi-scale combinations of physical descriptors. Results also indicate that substantial variation in benthic macroinvertebrate response can be explained with process-based watershed and valley scale metrics derived exclusively from common geospatial data. This study outlines a general framework for identifying key processes driving macroinvertebrate assemblages across a range of scales and establishing the geographic extent at which various levels of physical description best explain biological variability. Such information can guide process-based stratification to avoid spurious comparison of dissimilar stream types in bioassessments and ensure that key environmental gradients are adequately represented in sampling designs.
B33E-05 1330h
Conditional Probability Approach for Assessing Fine Sediment Impacts on Aquatic Insects
We examine the response of benthic macroinvertebrates to deposited fine sediments using R/EMAP sites from the Southern Rockies Ecoregion of Colorado and data from a field experiment conducted in a gravel-bed mountain stream in Southern Wyoming in autumn 2004. A conditional probability approach based on selected benthic macroinvertebrate indices (EPT richness and others) is employed to express the likelihood of a reduced number of sensitive taxa given an exceedance of reach-scale percent sand and fines (less than or equal to 2mm), and deviation from regional mean levels of deposited fine sediments by stream type. We address natural variability in fine sediment resulting from differences in hydrogeomorphic context by stratifying R/EMAP sites according to relevant watershed-, valley-, and local-scale variables. This research demonstrates a transferable approach for 1) assessing the risk of impacts to stream biota associated with fine sediment deposition in various regional contexts, and 2) improving interpretation of qualitative sediment deposition standards.
B33E-06 1330h
Effects of Deposited Sediment on the Growth and Behavior of Physa Snails
Fine sediment loading is one of the leading causes of ecological impairment to rivers and streams in the United States, yet few studies have addressed how deposited silt affects benthic invertebrates. We determined how deposited silt affects the growth and behavior of Physa integra snails from Emmons Creek, WI by exposing them to 3 levels of silt (ambient, low, high) in streamside once-through flumes and recirculating chambers. While there were no significant differences in snail growth rates among silt treatments, the behavioral differences were striking. Snails that received added silt spent less time on algal-covered tiles than those not receiving silt. Elemental analysis of periphyton revealed higher nitrogen, carbon, and phosphorus levels for the silt addition treatments, which may be providing the snails with more nutritious food, although this was not reflected in their growth rates. Per capita egestion rates were significantly higher for the high-silt treatment. These snails may have ingested more of the algae and silt but the presence of silt may have decreased assimilation, nullifying any positive effect on growth. In conclusion, silt addition did not affect Physa integra growth rates but did influence their behavior and egestion rates.
B33E-07 1330h
The Relationship of Physicochemical Data, Macroinvertebrate Assemblages, and Land use in Rhode Island Wadeable Streams
The ability of streams to support aquatic organisms is inextricably connected to the physical structure and chemical conditions of the immediate environment. Our objective was to study the relationships among physical and chemical attributes, habitat, and the benthic macroinvertebrates from wadeable streams in Rhode Island subwatersheds along a gradient of residential land use. We used the U.S. EPA Rapid Bioassessment Protocols to measure biological, physicochemical, and habitat effects at stream sites. A Geographic Information System was used to delineate the subwatersheds and document land use surrounding the sampled stream reaches. By using principle components analysis of chemical metrics and the individual metrics for habitat assessment, we found patterns among the sites correlated with land use. Results showed that sites with low residential land use had high stream flow, high dissolved oxygen, and high habitat scores with low temperature and total dissolved solids. Sites with increased residential land use showed the opposite trends. The species composition of benthic macroinvertebrates changed from mostly insect taxa (sensitive species) to noninsect taxa (tolerant species) with increasing levels of residential land use. Thus, altered habitat associated with land use was highly correlated with degraded physicochemical conditions that affected the species composition of aquatic organisms.
B33E-08 1330h
Interspecific Relationships in Benthic Assemblages of a Large Lowland River : Co-existence or Competition as a Result of Habitat Characteristics
Macrobenthic assemblages of the lower, regulated and canalized Marne River (France) are dominated by several "exotic", recently introduced species. Arrival, installation and spread of such alien species were certainly promoted by (1) the close connection between French and other European river systems (Rhine, Danube), (2) the modification of natural river flow and benthic habitats. Both habitat characteristics (e.g. granulometric composition, organic content) and corresponding benthic assemblage structure were analysed to identify the substrate affinity of major taxa in the lower Marne River. The main aim of the study was to evaluate the possible biotic interactions (i.e. competition for food and/or space) among dominant taxa as function of habitat features and to predict the future development of newly established species, already known as potential invaders: Corbicula fluminea (Bivalvia), Hypania invalida (Polychaeta), Chelicorophium curvispinum (Amphipoda), etc. Most of the newly established species, coming from the same biogeographical area ("invasional meltdown" hypothesis?), are eurytopic, with high fecundity and large food spectrum. First results demonstrated the co-existence of such species in the Marne River. But the future ecological importance of these organisms in benthic assemblages of the river depends on their present population size, population dynamics, and ability to colonize bottom substrates.
B33E-09 1330h
Abiotic Factors Affecting Benthic Invertebrate Biomass and Community Structure in a Fourth-Order Rocky Mountain Watershed
The potential ecological impact of excess streambed sediment resulting from forest management activities is a persistent concern for land managers. This study examined the relationship between streambed sediment, along with other site- and reach-scale abiotic factors, and benthic macroinvertebrate community structure in a 272 km2 basin in the Colorado Front Range. Physical habitat parameters and invertebrates were sampled in late summer at 68 sites located in sixteen stream reaches. Invertebrate data were used to formulate twenty indices of community structure. Multiple regression identified site-level substrate particle size as the most important predictor of six indices, including total density (R2 = 0.22), biomass (R2 = 0.17), and taxa richness (R2 = 0.32). All of the remaining fourteen indices were most strongly predicted by reach-level variables, including discharge (percent shredders, R2 = 0.24; Plecoptera density, R2 = 0.29), and elevation (percent collector-filterers, R2 = 0.28; Trichoptera density, R2 = 0.37). Although the sites represented a wide range of substrate composition and embeddedness, no physical variable associated with fine sediment appeared as a strong predictor of any of the twenty indices. Thus, sediment is not among the most important factors associated with site-to-site variability of benthic community structure in this relatively pristine watershed.
B33E-10 1330h
How Far is Far Enough? Invertebrate Responses to Physical Constraints on Drift Distance
Many stream insects enter the drift and disperse downstream. Once entrained, however, the probability of settling in a patch of suitable habitat is a function of the physical properties and behavior of the drifting insect, as well as the hydrodynamic characteristics of the habitat through which the insect is drifting. The roles that taxa-specific morphology and behavior play in determining drift distance were examined for four mayflies with different habitat requirements; two rheophilous taxa (Baetis and Epeorus) and two pool-dwelling taxa (Ameletus and Paraleptophlebia). Larvae were released in an experimental channel in low and high water velocities. The total distances traveled by live mayfly larvae (+ behavior, + morphology) were compared to heat-killed larvae (- behavior, + morphology), and a series of low-density tracer particles (- behavior, - morphology). Live Baetis and Epeorus drifted similar distances, whereas the drift distances of the two pool taxa differed substantially (Ameletus < Epeorus, Baetis < Paraleptophlebia). The settlement distributions of dead larvae and passive tracer particles show that settlement behaviors allow drifting larvae to avoid becoming entrained in large-scale turbulent flow structures. These results suggest that stream insects have evolved strategies that facilitate dispersal between patches of suitable habitat.
B33E-11 1330h
Modeling Hierarchical Relationships Between Terrestrial Disturbance and Responses of Headwater Streams in the Southern Appalachians
Strong relationships have been demonstrated between catchment-wide logging practices and benthic community structure in headwater streams. Disturbance-induced changes in both abiotic and biotic factors often correlate with short term shifts in community dominance from collector-filterers to scrapers. We suggest a hierarchical model may apply where catchment-scale terrestrial disturbance modifies a suite of in-stream environmental factors (endogenous variables) that influence benthic macroinvertebrate community structure (response variable) at a variety of scales (e.g., reach, bedform, or substrate). Structural Equation Modeling was used to delineate possible hierarchical models of the dominant relationships among in-stream factors in forested headwater streams. Models developed from regional data (logged and un-logged headwater catchments) resolve which terrestrial-aquatic linkages (e.g., sediment input, allochthonous input) remain important predictors across the disturbance intensity gradient. Models generated with data from forested catchments were compared to those from logged catchments to predict how some linkages may become irrelevant, while others are accentuated following a disturbance. This study will provide testable hypotheses for forthcoming experimental clear cuts in headwater catchments in the southern Appalachians.
B33E-12 1330h
Physical and Biological Impacts of Changing Land-Uses and the Environment
A goal of the Changing Land Use and the Environment (CLUE) project is to characterize surface water quality impacted by land-use change in the Saluda and Reedy River watersheds of South Carolina. The CLUE project focuses on impacts common to urban development including 1. sedimentation from construction sites, 2. alteration of discharge and channel morphology due to increased impervious surfaces, 3. macroinvertebrate community response to sedimentation and habitat alteration, and 4. microbial contamination. We found that mean streambed particle size was reduced in developing areas. Stream cross-sectional areas enlarged in catchments with high percentages of impervious surfaces. Sedimentation and altered discharge resulted in the benthic macroinvertebrate community showing a general reduction in biotic integrity values and reductions in Plecoptera taxa richness. Fecal coliform levels were higher for both surface water and bottom sediments in and below urbanized areas during base flows. Levels of fecal coliform in samples collected during storm flows were significantly higher than in base flows, and were correlated with high sediment loads.
B33E-13 1330h
Woody Debris: Linking Stream Morphology and Aquatic Biota in Southeastern Coastal Plain Streams, USA
In-stream habitat including woody debris is widely recognized as an important aspect of aquatic health. Relationships between woody debris and the distribution and life history of aquatic species are increasingly documented in the literature. In southeastern coastal plain streams, woody debris may have a greater influence on stream form and function than in other regions due to highly mobile shifting sand substrates. In an effort to document baseline stream conditions, natural variability within biological communities, and to also link chemical, physical and biological parameters within the Gulf Coastal Plain physiographic region, aquatic surveys have been conducted on Eglin Air Force Base, Florida over the past four years. We developed quantitative indicators of habitat health and protocols for measuring habitat condition at various geographic scales. Physical variables included stream channel characteristics, habitat connectivity, riparian cover and in-stream habitat including woody debris. Biological data (fishes and aquatic insects) were compared with in-stream habitat data for 56 sites. Significant relationships between species and in-stream habitat presence/absence and abundance were detected. Associations between woody debris and bed morphology were also significant. These preliminary data suggest that woody debris is an important factor shaping biological communities and stream geomorphic features in sand-bottom coastal plain streams.
B33E-14 1330h
Riparian Vegetation Control of In-Stream Temperature
Stream temperature is an important physical characteristic that can control the structure of aquatic fauna and influence key ecological processes. The response of in-stream temperature to topographic and vegetative shade can be predicted with reasonable accuracy at the sub-catchment scale. A more difficult task is to determine the biotic response to thermal stress and therefore set guideline values. A series of LD50 experiments supported extensive worldwide literature on temperature tolerances of aquatic fauna. Temperature modelling was applied at different biogeographic zones of Australia. In Mediterranean climates where low flow corresponded with high summer air temperatures, in-stream thermal stress was highly seasonal. In the tropics the situation was reversed such that high air temperatures, during summer, corresponded with high stream flows that moderated the magnitude of seasonal variation. When combined with the estimated thermal tolerance of selected taxa, model analyses identified critical bottlenecks, in space and time, and enabled prioritisation of riparian restoration efforts.
B33E-15 1330h
Soil Microbial Community Contribution to Small Headwater Stream Metabolism.
The temporal dynamics of sediment respiration were examined in seven small headwater streams in forested catchments in 2004. A strong seasonal response was observed with higher respiration rates in depositional zones than in gravel runs. The data were also examined in the context of proportional habitat distributions that highlighted the importance of high flow events in shaping whole stream metabolic budgets. This study specifically examines the question of terrestrial soil respiration contribution to whole stream metabolism by the controlled inundation of terrestrial soils. The experiment included six experimentally inundated terrestrial zones, six terrestrial controls, and six in-stream depositional zones. Sediment bacterial respiration was measured using 14C leucine incorporation and cotton strip bioassays were also employed to provide an indicative measure of sediment microbial activity. Despite high variability and exhibiting significantly lower bacterial activity than in-stream sediments, modelling using flow data and habitat mapping illustrated the important contribution of terrestrial soil respiration to the whole stream metabolic budgets of small headwater streams. In addition, microbial community composition examined using phospholipid fatty acid analysis clearly differentiated between terrestrial and aquatic communities. Freshly inundated terrestrial communities remained similar to un-inundated controls after 28 days.
B33E-16 1330h
Developing a Method for Partitioning Transient Storage into Hyporheic and In-Channel Dead Zone Components
Transient storage is often represented as characterizing the hyporheic zone, but it may also represent storage in the channel. We attempted to separate the size and exchange rate of both storage types using a combination of rhodamine WT tracer injections and cross-sectional flow velocity measurements, coupled with inverse modeling using STAMMT-L (a one dimensional transient storage model that simulates a variety of different residence time distributions) and one- and two-compartment models of OTIS. Experiments comparing stream reach breakthrough curves were conducted while concurrently measuring lateral velocity profiles. Exchange rates between the main channel and in-channel dead zones were calculated with shear velocities and lateral eddy diffusion. The one-compartment OTIS model best described in-channel storage in reaches with pools and eddies while STAMMT-L better described storage due to hyporheic exchange. The two-compartment OTIS model was a good compromise between these models and adequately quantified in-channel and hyporheic storage when constrained by empirical data. The ability to separate hyporheic and in-channel storage can improve our understanding of nutrient cycling in streams that differ geomorphically because these storage zones differ in the rates and types of nutrient transformations.
B33E-17 1330h
Hyporheic Hydrology of the Umatilla River: Interactions among Physical and Biological Drivers
The hydrology of the hyporheic zone in the upper Umatilla River (Oregon, USA) is highly dynamic, due to the river's complex channel geometry and largely unregulated discharge regime. Coarse-scale variation in patterns of water exchange between the river channel and hyporheic zone are mediated by inundation and abandonment of flood channels during freshets. Our research, however, documents an important biological driver of hyporheic hydrology. Spring channels (perennial channels that derive their flow from hyporheic and/or phreatic water) on the flood plain provide excellent habitat for emergent aquatic vegetation during the summer. Plant growth increases flow resistance and alters the relationship between spring channel discharge and stage, influencing the water table elevation near the spring channel. Thus, the seasonal growth and senescence of emergent aquatic vegetation influence the rate and pattern of hyporheic discharge to the spring channel, affecting the hydrology of the hyporheic zone across the entire flood plain. Thus, our data provide a novel illustration of functional feedback between physical and biological processes, which ultimately contributes to habitat diversity within lotic ecosystems.
B33E-18 1330h
Application Of A Dynamic Model to Assess Geomorphic and Hydrologic Controls on Age-0 Colorado Pikeminnow Distribution in the Green River, Colorado And Utah
Analysis of field data and development and application of a dynamic model indicate that water releases from Flaming Gorge Dam have a large potential effect on larval drift and distribution of age-0 Colorado pikeminnow (Ptychocheilus lucius) in the middle Green River. The model predicts that high releases at the time of drift greatly increase the proportion of the population transported beyond the study area to unfavorable river environments. The model also predicts that channel simplification caused by flow regulation results in a more even longitudinal distribution of larval fish habitat. Colorado pikeminnow are a federally endangered species endemic to the Colorado River basin that utilize backwaters during their larval stage. The present agency-mandated field-sampling program for backwater habitats is probably inadequate, because it takes place at a time when the model predicts that most larval fish have drifted beyond the study area. Development of the model shows that the role of the geomorphic and hydraulic attributes that control larval drift and transport into backwaters, and that were parameterized in the model, are not well known.
B33E-19 1330h
The Influence of Dam Discharge Regime and Canyon Orientation on Ecosystem Metabolism in the Colorado River
Since the closure of Glen Canyon Dam and the beginning of flow regulation of the Colorado River in Grand Canyon in 1963, considerable efforts have been directed toward understanding the aquatic ecology of this altered ecosystem. Understanding what controls resource availability has been a central focus of these efforts because the Colorado River supports populations of sport fish and endangered humpback chub, both of which appear to be strongly resource limited. There is evidence that dam discharge regime and canyon orientation influence algal standing crop due to their effects on water velocity (scour) and solar insolation, respectively. We explored whether these physical factors influenced rates of primary production and ecosystem respiration, two different metrics of resource availability, in the clear tailwater section of the Colorado River by conducting whole system metabolism measurements across a range of discharge regimes and in reaches with different orientation (i.e. N-S vs. E-W). We found that while both discharge regime and canyon orientation influence rates of primary production, seasonal changes in light availability appear to have a far stronger influence on rates of primary production in the Colorado River. Water temperature appeared to be the main driver of ecosystem respiration.
B33E-20 1330h
Effect of Culverts on Predator-Prey Interactions in a Tropical Stream.
As part of a biocomplexity project in Puerto Rico, we use river and road networks as a platform to understand the interactions between stream biota, the physical environment, and human activity. Specifically, we ask if humans affect aquatic organisms through road building and recreational activities. Culverts have been documented to impede or slow migration of aquatic biota. This is especially important in these streams because all of the freshwater, stream species have diadramous life cycles. If culverts do act as bottlenecks to shrimp migrations, we expect altered predator-prey interactions downstream through density-dependent predation dynamics. In order to determine how roads may affect predation rates on upstream migrating shrimp, we parameterized functional response curves for mountain mullet (Agonostomus monticola) consuming shrimp (Xiphocaris sp.) using artificial mesocosm experiments. We then used data obtained from underwater videography to determine how culverts decrease the rate and number of shrimp moving upstream. These data were combined in a predator-prey model to quantify the effects of culverts on localized shrimp densities and fish predation.
http://biocomplexity.cnr.colostate.edu
B33E-21 1330h
Application of Whole Genome Expression Analysis to Assess Bacterial Responses to Environmental Conditions
Bacterial responses to environmental signals are multifactorial and are coupled to changes in gene expression. An understanding of bacterial responses to environmental conditions is possible using microarray expression analysis. In this study, the utility of microarrays for examining changes in gene expression in Escherichia coli under different environmental conditions was assessed. RNA was isolated, hybridized to Affymetrix E. coli Genome 2.0 chips and analyzed using Affymetrix GCOS and Genespring software. Major limiting factors were obtaining enough quality RNA (107-108 cells to get 10Îœg RNA)and accounting for differences in growth rates under different conditions. Stabilization of RNA prior to isolation and taking extreme precautions while handling RNA were crucial. In addition, use of this method in ecological studies is limited by availability and cost of commercial arrays; choice of primers for cDNA synthesis, reproducibility, complexity of results generated and need to validate findings. This method may be more widely applicable with the development of better approaches for RNA recovery from environmental samples and increased number of available strain-specific arrays. Diligent experimental design and verification of results with real-time PCR or northern blots is needed. Overall, there is a great potential for use of this technology to discover mechanisms underlying organisms' responses to environmental conditions.
B33E-22 1330h
Linking Forests and Fish: The Relationship Between Productivities of Salmonids and Forest Stands in Northern California
Productivities of resident salmonid populations, upland, and riparian areas in 25 small watersheds of coastal northern California were estimated and compared to determine if: 1) upland site productivity predicted riparian site productivity; 2) either upland or riparian site productivity predicted salmonid productivity; and 3) other parameters explained more of the variance in salmonid productivity than upland or riparian site productivity. Salmonid productivity was indexed by total salmonid biomass, length of age 1 fish, and percent habitat saturation. Upland and riparian site productivities were estimated using site indices for redwood (Sequoia sempervirens) and red alder (Alnus rubra), respectively. Upland and riparian site indices were correlated, but neither factor contributed to the best approximating models of salmonid biomass or fish length at age one. Salmonid biomass was best described by a positive relationship with drainage area, and length at age was best described by a positive relationship with percent of riparian hardwoods. Percent habitat saturation was not well described by any of the models constructed. Lack of a relationship between upland conifer and salmonid productivity suggests that management of land for timber productivity and component streams for salmonid production in these sites will require separate, albeit integrated, strategies.
B33E-23 1330h
Relative Influence of Catchment Landscape Indicators on Macroinvertebrate Communities in a Minimally Disturbed Watershed: Effects of Spatial Scale.
The management of watersheds and water resources requires the use and integration of many indicators including in-stream physical, chemical, and biotic criteria. Recent emphasis has also been placed on the use of landscape indicators to assess stream structure and function. In this study, the influence of landscape indicators on macroinvertebrate communities is assessed at multiple spatial scales to determine the relative effects of landscape influences at differing spatial scales. Preliminary results suggest the distance from stream origin to sample site and variables associated with urban landcover are the most significant predictors of macroinvertebrate abundance and community composition when landscape indicators are assessed at the catchment scale. These results will be compared to results acquired from quantification of landscape indicators within a 100m riparian buffer and a 100m buffer of a 1000m reach upstream of macroinvertebrate sample sites to determine the relative influence of landscape indicators at these scales.
B33E-24 1330h
The Infuence of Physical Variables on Whole-Stream Metabolism in an Arctic Tundra River
We examined the influence of light, temperature, nutrients, and discharge on whole-stream metabolism (WSM) in three experimental reaches of the Kuparuk River, Alaska, using the open-system, single-station method. Ambient PO4 levels in the reference reach were ~0.05ÎœM, while addition of phosphoric acid since 1983 in the fertilized reach and since 2004 in the ultra-fertilized reach increased PO4 levels to ~0.30ÎœM and ~0.90ÎœM, respectively. Among all reaches, gross primary production (GPP) was positively correlated with light, temperature, and PO4 and negatively correlated with discharge. Temperature explained most of the variance in GPP. Among all reaches, community respiration (CR) was weakly correlated with light, temperature, PO4, and discharge. However, CR showed a greater response to temperature in the fertilized reaches. Benthic respiration by mosses in the fertilized reaches responds to temperature while heterotrophic respiration in the hyporheic zone is similar in all reaches and does not respond to temperature due to thermal buffering. Light, temperature, and discharge were moderately intercorrelated. An Information-Theoretic approach using Akaike's Information Criterion (AIC) was used to examine the relative importance of each physical variable on WSM and to develop a photosynthesis model.
B33E-25 1330h
Advances in stream shade modelling. Accounting for canopy overhang and off-centre view
Riparian shade controls the stream thermal regime and light for photosynthesis of stream plants. The quantity difn (diffuse non-interceptance), defined as the proportion of incident lighting received under a sky of uniform brightness, is useful for general specification of stream light exposure, having the virtue that it can be measured directly with common light sensors of appropriate spatial and spectral character. A simple model (implemented in EXCEL-VBA) (Davies-Colley & Rutherford Ecol. Engrg in press) successfully reproduces the broad empirical trend of decreasing difn at the channel centre with increasing ratio of canopy height to stream width. We have now refined this model to account for (a) foliage overhanging the channel (for trees of different canopy form), and (b) off-centre view of the shade (rather than just the channel centre view). We use two extreme geometries bounding real (meandering) streams: the `canyon' model simulates an infinite straight canal, whereas the `cylinder' model simulates a stream meandering so tightly that its geometry collapses into an isolated pool in the forest. The model has been validated using a physical `rooftop' model of the cylinder case, with which it is possible to measure shade with different geometries.
B33E-26 1330h
Influence of Carbonate Crusts on Trophic Cascades in Big Sur Streams.
Top-down effects of fish predation in streams are influenced by habitat conditions and traits of intermediate consumers. Along the Big Sur coast, seasonal carbonate crusts form in many streams, and appear to alter insect communities. We conducted a small-scale field experiment in two adjacent tributaries (crusted and crust-free) to determine the effects of predation by juvenile steelhead (Oncorhynchus mykiss) on invertebrates and algae in these different stream types. We hypothesized that a trophic cascade would be more likely in the crusted stream, which was dominated by an herbivore vulnerable to steelhead predation (Baetis), than in the crust-free stream, which was dominated by a less vulnerable herbivore (Agapetus). However, excluding steelhead from paving bricks had no effect on algae or Baetis in the crusted stream. Algal abundance was low, suggesting possible bottom-up control. In the crust-free stream, steelhead predation did not affect Agapetus, but nevertheless produced a trophic cascade by causing Baetis to increase by 85%, which in turn depressed algae by 25%. These results suggest that trophic interactions may differ between crust forming and non-crust forming streams, which we plan to test with additional sampling and experiments replicated across stream types.
B33E-27 1330h
Influence of Ephemeral Sand Islands in a Large Prairie River on Diversity and Density of Benthic Macroinvertebrates.
The ecology of prairie rivers has rarely been investigated, with most aquatic studies focusing on reservoirs or intermittent streams. The Kansas River, a sand-bed river draining a watershed historically composed primarily of tallgrass and mixed grass prairie, contains in its channel numerous ephemeral sand-bar islands. Recent riverine research on relatively permanent, forested islands has shown a strong correlation between community attributes and hydrologic retention. However, our study examined potential relationships between the transitory slackwaters created by these ephemeral islands and the benthic macroinvertebrate community. We hypothesized that these slackwaters would contain greater species density and diversity than other parts of the river with higher current velocity. A total of 14 benthic habitats in 3 reaches were sampled with benthic cores in the summer of 2004. Density data confirm the importance of ephemeral slackwaters, with the highest total invertebrate densities occurring in low-velocity areas. Initial counts of chironomid midges, the dominant macroinvertebrate group, also support this conclusion, with greater diversity in areas of lower current velocity. Our results could help ensure the future health of grassland river ecosystems if the importance of ephemeral sand-bar islands and slackwaters are incorporated into future management plans.
B33E-28 1330h
Multi-scale Factors Affecting the Distribution of the Critically Imperiled Crayfish, Orconectes williamsi, in the Upper White River Drainage of Missouri, USA
Orconectes williamsi is known from only a few locations in the upper White River drainage of Missouri and Arkansas. We implemented a stratified (by stream order) random survey to sample for this crayfish in Missouri in 2003-04. Survey objectives were twofold. First, we estimated the overall proportion of stream order segments (segments of streams between points where stream order changes) in the drainage harboring O. williamsi, and then estimated the proportion of first, second and third (or larger) order stream segments harboring this crayfish. Second, we obtained means or ratio estimators (point estimates) and 95 % confidence limits (interval estimates) for 62 environmental variables and conducted overall contrasts of those estimates between segments that did and did not harbor O. williamsi. For significant overall contrasts, we then conducted contrasts partitioned by stream order. We detected O. williamsi at an overall proportion of 0.34 of 50 sampled segments, with no difference between first and second order stream segments. Variables exhibiting differences between segments that did and did not harbor O. williamsi included bankfull width:bankfull depth ratio, wetted width:wetted depth ratio, macrophyte abundance, elevation, and stream connectivity. Results will be used to drive future sampling and conservation efforts.
B33E-29 1330h
Scale-Dependent Thermal Patterns in a Groundwater-Influenced Stream
Thermal patterns vary on different time scales, with differing consequences for temperature-sensitive organisms. We monitored water temperature at 1/2hr intervals from mid-June through October 2003 along a 30km reach of the Straight River, a spring-fed stream in west-central Minnesota that begins with discharge from a warm-water lake. We analyzed temperature data at 20 sites to describe temperature patterns on scales ranging from hours to days to weeks. Patterns were similar among sites, despite differences in canopy cover or proximity to known springs. Daily thermal flux was typically 4-5° C, but maximum diel oscillations of 7-12° C were seen at most sites. Maximum daily temperatures resulted from net daily changes, and varied over periods of days in response to weather patterns. These diel and longer-term oscillations were superimposed on a broad seasonal pattern of summer warmth and autumnal cooling. Although the incipient lethal temperature for the resident population of Salmo trutta (24.7° C) was exceeded at all but one of 20 sites, only at the lake outlet did these thermal spikes exceed 5hr. Thus brown trout survival in this stream is facilitated both by a general lowering of temperature due to groundwater input, and by natural thermal oscillation on a diel time scale.