Search

Search Results

• 211. [Article] Transferability of models to predict selection of cover by coastal cutthroat trout in small streams in western Oregon, USA

  We assessed use and selection of cover by coastal cutthroat trout (Oncorhynchus clarkii clarkii) in six headwater streams in three watersheds in western Oregon, USA during the summer low flow period from ...

  Citation × Citation Citation Abstract Copy Title: Transferability of models to predict selection of cover by coastal cutthroat trout in small streams in western Oregon, USA Author: Andersen, Heidi V. We assessed use and selection of cover by coastal cutthroat trout (Oncorhynchus clarkii clarkii) in six headwater streams in three watersheds in western Oregon, USA during the summer low flow period from 1 August and September 30, 2007. We tagged 1037 coastal cutthroat trout (>100 mm) with passive integrated transponder (PIT) tags across all streams. Selection of cover was analyzed by comparing characteristics of locations used for concealment by relocated fish relative to characteristics of randomly available habitat that could be used for concealment. We measured habitat characteristics for 190 relocated individual fish using cover and 797 randomly points potentially available as cover. Of the latter points, only 235 of 797 were potential cover, based on characteristics of cover actually used by fish. In other words, 562 of the 797 randomly sampled points were unlikely to be used as cover by fish. Coastal cutthroat trout used substrate as cover (78%) more often than all other cover types combined (22%). Availability of different cover types was variable, but overall substrate made up 92% of available cover and the remaining 8% represented all other cover types combined. Habitat characteristics measured for both used and available cover included depth at fish location (cm), surface area of cover (m²), proximity to depth of 20 cm for fish located in < 20 cm in depth, b-axis (mm) for substrate >2 mm, and distance under substrate. Each of these habitat characteristics was different for used and available cover (Wilcoxon rank-sum p-values all < 0.0001). Analysis of selection using logistic regression models indicated that cover use was more likely with increasing depth and surface area of cover. A negative interaction effect between the influences of depth and surface area suggested fish were more likely to use cover with smaller surface areas in deeper water. We found good transferability (i.e. predictive capabilities) of the logistic regression models across streams using three different methods: "leave-one-stream-out" cross validation, Cohen's kappa statistic, and receiver operator characteristic curves. Our results suggested that characteristics of used cover were similar across six streams for coastal cutthroat trout in headwater streams. The strong and consistent influence of both depth and surface area of cover on selection of habitat by individual coastal cutthroat trout suggests these features of habitat may be critical to this species during summer low flows. Title: Transferability of models to predict selection of cover by coastal cutthroat trout in small streams in western Oregon, USA Author: Andersen, Heidi V. We assessed use and selection of cover by coastal cutthroat trout (Oncorhynchus clarkii clarkii) in six headwater streams in three watersheds in western Oregon, USA during the summer low flow period from 1 August and September 30, 2007. We tagged 1037 coastal cutthroat trout (>100 mm) with passive integrated transponder (PIT) tags across all streams. Selection of cover was analyzed by comparing characteristics of locations used for concealment by relocated fish relative to characteristics of randomly available habitat that could be used for concealment. We measured habitat characteristics for 190 relocated individual fish using cover and 797 randomly points potentially available as cover. Of the latter points, only 235 of 797 were potential cover, based on characteristics of cover actually used by fish. In other words, 562 of the 797 randomly sampled points were unlikely to be used as cover by fish. Coastal cutthroat trout used substrate as cover (78%) more often than all other cover types combined (22%). Availability of different cover types was variable, but overall substrate made up 92% of available cover and the remaining 8% represented all other cover types combined. Habitat characteristics measured for both used and available cover included depth at fish location (cm), surface area of cover (m²), proximity to depth of 20 cm for fish located in < 20 cm in depth, b-axis (mm) for substrate >2 mm, and distance under substrate. Each of these habitat characteristics was different for used and available cover (Wilcoxon rank-sum p-values all < 0.0001). Analysis of selection using logistic regression models indicated that cover use was more likely with increasing depth and surface area of cover. A negative interaction effect between the influences of depth and surface area suggested fish were more likely to use cover with smaller surface areas in deeper water. We found good transferability (i.e. predictive capabilities) of the logistic regression models across streams using three different methods: "leave-one-stream-out" cross validation, Cohen's kappa statistic, and receiver operator characteristic curves. Our results suggested that characteristics of used cover were similar across six streams for coastal cutthroat trout in headwater streams. The strong and consistent influence of both depth and surface area of cover on selection of habitat by individual coastal cutthroat trout suggests these features of habitat may be critical to this species during summer low flows. Close

• 212. [Article] Derivation of near-shore bathymetry from multispectral satellite imagery used in a coastal terrain model for the topographic analysis of human influence on coral reefs

  The analysis of material and energy exchange between the marine and terrestrial components of island ecosystems enables research into the impact of human population and land use on the health of coral ...

  Citation × Citation Citation Abstract Copy Title: Derivation of near-shore bathymetry from multispectral satellite imagery used in a coastal terrain model for the topographic analysis of human influence on coral reefs Author: Hogrefe, Kyle Richard The analysis of material and energy exchange between the marine and terrestrial components of island ecosystems enables research into the impact of human population and land use on the health of coral reef habitat. Satellite and acoustic remote sensing technologies enable the collection of data to produce high resolution bathymetry for integration with terrestrial digital elevation models (DEMs) into coastal terrain models. An integrated terrain surface that incorporates the land-sea interface, grounded by a geographic information system, is a powerful analytical tool for geomorphic studies of watersheds and coastal processes. The island of Tutuila, American Samoa is an ideal case study due to its high relief terrain, data availability and local interest in impacts to coral reef resources. The Tutuila model integrates a USGS DEM, multibeam bathymetry from 15 to 500 m and near shore bathymetric data from 0 to 15 m derived from IKONOS satellite imagery. The high spatial resolution of IKONOS imagery is suitable for detection of features with subtle relief and intricate structure. Shallow water bathymetry is derived by quantifying the relative attenuation of blue and green spectral band radiance as a function of depth. The procedure used to derive bathymetry, Lyzenga (1985), is identified as the most effective of several proposed in the recent literature. The product is error-checked using control points extracted from multibeam sonar data and collected during recent field surveys, as well as terrain profiles. The coastal terrain model provides morphological detail of fine resolution and high accuracy for terrain and land use analysis to enhance the study of ecosystem interconnectivity and the effects of anthropogenic inputs to coral reef habitats. Subsequent topographic analyses of the Tutuila model use drainage patterns to identify contiguous marine/terrestrial basins within which the marine environment is most directly impacted by land use through freshwater inputs from affiliated catchments. Human population density serves as an indicator of intensified land use and urbanization, which has been shown to increase pathogen and sediment loads in runoff, while percent coral cover, coral colony density and coral genera diversity are used as indicators of reef health. Spatiotemporal correlation analyses of population density against the three reef health indices within each of the marine/terrestrial basins reveal a decline in reef health associated with increased population density. This paper integrates and builds upon established methods of satellite imagery analysis and terrain modeling to create the Tutuila coastal terrain model and uses it to refine the scale of other studies linking human terrestrial activities to the physical condition of coral reefs. Title: Derivation of near-shore bathymetry from multispectral satellite imagery used in a coastal terrain model for the topographic analysis of human influence on coral reefs Author: Hogrefe, Kyle Richard The analysis of material and energy exchange between the marine and terrestrial components of island ecosystems enables research into the impact of human population and land use on the health of coral reef habitat. Satellite and acoustic remote sensing technologies enable the collection of data to produce high resolution bathymetry for integration with terrestrial digital elevation models (DEMs) into coastal terrain models. An integrated terrain surface that incorporates the land-sea interface, grounded by a geographic information system, is a powerful analytical tool for geomorphic studies of watersheds and coastal processes. The island of Tutuila, American Samoa is an ideal case study due to its high relief terrain, data availability and local interest in impacts to coral reef resources. The Tutuila model integrates a USGS DEM, multibeam bathymetry from 15 to 500 m and near shore bathymetric data from 0 to 15 m derived from IKONOS satellite imagery. The high spatial resolution of IKONOS imagery is suitable for detection of features with subtle relief and intricate structure. Shallow water bathymetry is derived by quantifying the relative attenuation of blue and green spectral band radiance as a function of depth. The procedure used to derive bathymetry, Lyzenga (1985), is identified as the most effective of several proposed in the recent literature. The product is error-checked using control points extracted from multibeam sonar data and collected during recent field surveys, as well as terrain profiles. The coastal terrain model provides morphological detail of fine resolution and high accuracy for terrain and land use analysis to enhance the study of ecosystem interconnectivity and the effects of anthropogenic inputs to coral reef habitats. Subsequent topographic analyses of the Tutuila model use drainage patterns to identify contiguous marine/terrestrial basins within which the marine environment is most directly impacted by land use through freshwater inputs from affiliated catchments. Human population density serves as an indicator of intensified land use and urbanization, which has been shown to increase pathogen and sediment loads in runoff, while percent coral cover, coral colony density and coral genera diversity are used as indicators of reef health. Spatiotemporal correlation analyses of population density against the three reef health indices within each of the marine/terrestrial basins reveal a decline in reef health associated with increased population density. This paper integrates and builds upon established methods of satellite imagery analysis and terrain modeling to create the Tutuila coastal terrain model and uses it to refine the scale of other studies linking human terrestrial activities to the physical condition of coral reefs. Close

• 213. [Article] Evolution and risk in conservation of Pacific salmon

  Identifying appropriate units for conservation requires knowledge of evolutionary patterns and risks of managing at different geographical and genetic scales. I examined genetic diversity at different ...

  Citation × Citation Citation Abstract Copy Title: Evolution and risk in conservation of Pacific salmon Author: Currens, Kenneth P. Identifying appropriate units for conservation requires knowledge of evolutionary patterns and risks of managing at different geographical and genetic scales. I examined genetic diversity at different geographical scales among 11,400 rainbow trout (Oncorhynchus mykiss) from 243 locations in 13 major river basins throughout much of their range and among coho salmon (O. kisutch) from 31 watersheds in Oregon, Washington, and northern California. I also developed a model of genetic vulnerability of managed populations that links sources of potential technological hazards, protective mechanisms and responses, and potential losses, using artificial propagation of Pacific salmon as an example. Across the range of rainbow trout, allozyme differences between inland and coastal populations were more localized than previously acknowledged. In contrast, evolutionary continuity was most related to stability and persistence of major river systems, such as upper Sacramento, Klamath, and Columbia rivers. Isolated, pluvial lake basins that contained divergent groups of redband trout (rainbow trout with plesiomorphic characteristics associated with cutthroat trout, O. clarki) were sources of evolutionary diversity within large river systems. Human effects on genetic organization occurred in local breeding populations to regional metapopulations. In coho salmon, regional differences in mitochondrial DNA existed among fish from Puget Sound, Columbia River, northern Oregon coastal streams, and southern coastal streams. Differences within regions lacked obvious geographical patterns but were most likely due to recent fish translocations and genetic drift. In the Umatilla River, Oregon, significant genetic differences were detected among rainbow trout, but temporal differences at sites were as great as differences among sites within tributaries. In 10 of 12 locations, rainbow trout became more similar to anadromous hatchery fish. Although small breeding sizes suggested a role for genetic drift, episodic gene flow from hatchery fish most likely explained temporal genetic changes. Program-specific genetic risk assessment of a large artificial propagation program in the Columbia River revealed that artificial supplementation would result in fewer hatchery-reared fish returning to the wild than were taken from the wild for brood stock, that proximate safeguards for reducing vulnerability were not available and appropriate, but that use of genetic reserves strengthened the program. Title: Evolution and risk in conservation of Pacific salmon Author: Currens, Kenneth P. Identifying appropriate units for conservation requires knowledge of evolutionary patterns and risks of managing at different geographical and genetic scales. I examined genetic diversity at different geographical scales among 11,400 rainbow trout (Oncorhynchus mykiss) from 243 locations in 13 major river basins throughout much of their range and among coho salmon (O. kisutch) from 31 watersheds in Oregon, Washington, and northern California. I also developed a model of genetic vulnerability of managed populations that links sources of potential technological hazards, protective mechanisms and responses, and potential losses, using artificial propagation of Pacific salmon as an example. Across the range of rainbow trout, allozyme differences between inland and coastal populations were more localized than previously acknowledged. In contrast, evolutionary continuity was most related to stability and persistence of major river systems, such as upper Sacramento, Klamath, and Columbia rivers. Isolated, pluvial lake basins that contained divergent groups of redband trout (rainbow trout with plesiomorphic characteristics associated with cutthroat trout, O. clarki) were sources of evolutionary diversity within large river systems. Human effects on genetic organization occurred in local breeding populations to regional metapopulations. In coho salmon, regional differences in mitochondrial DNA existed among fish from Puget Sound, Columbia River, northern Oregon coastal streams, and southern coastal streams. Differences within regions lacked obvious geographical patterns but were most likely due to recent fish translocations and genetic drift. In the Umatilla River, Oregon, significant genetic differences were detected among rainbow trout, but temporal differences at sites were as great as differences among sites within tributaries. In 10 of 12 locations, rainbow trout became more similar to anadromous hatchery fish. Although small breeding sizes suggested a role for genetic drift, episodic gene flow from hatchery fish most likely explained temporal genetic changes. Program-specific genetic risk assessment of a large artificial propagation program in the Columbia River revealed that artificial supplementation would result in fewer hatchery-reared fish returning to the wild than were taken from the wild for brood stock, that proximate safeguards for reducing vulnerability were not available and appropriate, but that use of genetic reserves strengthened the program. Close

• 214. [Article] Re‐establishment of the native oyster, Ostrea conchaphila, in Netarts Bay, Oregon, USA

  Olympia oysters, “Ostrea conchaphila,” were once common along the west coast of North America. A popular delicacy, native oyster populations began to decline in the late 1800’s due to over‐harvest, degraded ...

  Citation × Citation Citation Abstract Copy Title: Re‐establishment of the native oyster, Ostrea conchaphila, in Netarts Bay, Oregon, USA Author: Archer, Pamela Emily Olympia oysters, “Ostrea conchaphila,” were once common along the west coast of North America. A popular delicacy, native oyster populations began to decline in the late 1800’s due to over‐harvest, degraded water quality, and habitat loss. Interest in re‐establishing the native oyster in a small Oregon estuary, Netarts Bay, culminated in a partnership among The Nature Conservancy, the National Oceanic and Atmospheric Administration, the Oregon Watershed Enhancement Board, and Oregon State University. This study was designed to assess the reestablishment progress of the Olympia oyster restoration in Netarts Bay along with subsequent impacts of the restoration on eelgrass (“Zostera marina”), an important estuarine species. Two brood years (2005 & 2006) of cultch, consisting of O. conchaphila set on clean “Crassostrea gigas” shell substrate, were outplanted within an extensive, relatively uniform eelgrass bed. Cultch was placed in two experimental locations to determine the effect of cultch cover on native oyster survival, growth, and eelgrass abundance. The percent cover of cultch varied among treatments: “control” (no cultch), “low” (4% cultch cover), “medium” (11% cultch), and “high” (19% cultch). Research objectives were: (1) determination of O. conchaphila density, growth, and reproduction; and (2) quantification of the response of ”Z. marina” abundance and reproduction to cultch cover. Results from 2007 demonstrated that Olympia oysters were capable of growth, reproduction, and recruitment within their former habitat. Cultch cover within treatments did not change throughout the summer and there was minimal shell export out of the experimental location. Oyster size increased from March‐September, 2007: the mean size of the 2005 brood year increased by 10.5 mm, while the 2006 brood year increased by 16.2 mm. Sperm and larvae were found in individuals from both brood years, indicating that oysters were reproductively active. Declines in eelgrass mean percent leaf cover and shoot density were observed with increasing cultch cover. The mean eelgrass percent leaf cover was 15‐22% lower and shoot density was 27‐36% lower in high treatment (19% cultch) plots than in control plots. There were no discernable patterns in the eelgrass response variables of flowering shoot count, blade length, or blade width. The medium treatment (11% cultch), in which oyster densities were statistically similar to the high treatment (19% cultch), did not have statistically significant impacts on eelgrass percent cover or shoot density. We recommend continued testing of the medium treatment (11% cultch), as well as other cultch densities, such as a 50% cultch treatment. Additional monitoring will be needed to determine what, if any, long‐term impacts occur to the eelgrass bed. We also recommend long‐term monitoring of both oysters and eelgrass beds to detect any additional changes at the re‐establishment site. Title: Re‐establishment of the native oyster, Ostrea conchaphila, in Netarts Bay, Oregon, USA Author: Archer, Pamela Emily Olympia oysters, “Ostrea conchaphila,” were once common along the west coast of North America. A popular delicacy, native oyster populations began to decline in the late 1800’s due to over‐harvest, degraded water quality, and habitat loss. Interest in re‐establishing the native oyster in a small Oregon estuary, Netarts Bay, culminated in a partnership among The Nature Conservancy, the National Oceanic and Atmospheric Administration, the Oregon Watershed Enhancement Board, and Oregon State University. This study was designed to assess the reestablishment progress of the Olympia oyster restoration in Netarts Bay along with subsequent impacts of the restoration on eelgrass (“Zostera marina”), an important estuarine species. Two brood years (2005 & 2006) of cultch, consisting of O. conchaphila set on clean “Crassostrea gigas” shell substrate, were outplanted within an extensive, relatively uniform eelgrass bed. Cultch was placed in two experimental locations to determine the effect of cultch cover on native oyster survival, growth, and eelgrass abundance. The percent cover of cultch varied among treatments: “control” (no cultch), “low” (4% cultch cover), “medium” (11% cultch), and “high” (19% cultch). Research objectives were: (1) determination of O. conchaphila density, growth, and reproduction; and (2) quantification of the response of ”Z. marina” abundance and reproduction to cultch cover. Results from 2007 demonstrated that Olympia oysters were capable of growth, reproduction, and recruitment within their former habitat. Cultch cover within treatments did not change throughout the summer and there was minimal shell export out of the experimental location. Oyster size increased from March‐September, 2007: the mean size of the 2005 brood year increased by 10.5 mm, while the 2006 brood year increased by 16.2 mm. Sperm and larvae were found in individuals from both brood years, indicating that oysters were reproductively active. Declines in eelgrass mean percent leaf cover and shoot density were observed with increasing cultch cover. The mean eelgrass percent leaf cover was 15‐22% lower and shoot density was 27‐36% lower in high treatment (19% cultch) plots than in control plots. There were no discernable patterns in the eelgrass response variables of flowering shoot count, blade length, or blade width. The medium treatment (11% cultch), in which oyster densities were statistically similar to the high treatment (19% cultch), did not have statistically significant impacts on eelgrass percent cover or shoot density. We recommend continued testing of the medium treatment (11% cultch), as well as other cultch densities, such as a 50% cultch treatment. Additional monitoring will be needed to determine what, if any, long‐term impacts occur to the eelgrass bed. We also recommend long‐term monitoring of both oysters and eelgrass beds to detect any additional changes at the re‐establishment site. Close

• 215. [Article] Mercury distribution in sediments and uptake into an aquatic food web at Cottage Grove Reservoir, Oregon

  The Cottage Grove Reservoir has been impacted by a point source (Black Butte Mine), where mercury mining and processing has occurred within its watershed from the 1890's to the late 1960's. Total mercury ...

  Citation × Citation Citation Abstract Copy Title: Mercury distribution in sediments and uptake into an aquatic food web at Cottage Grove Reservoir, Oregon Author: Morgans, Donna L. The Cottage Grove Reservoir has been impacted by a point source (Black Butte Mine), where mercury mining and processing has occurred within its watershed from the 1890's to the late 1960's. Total mercury concentrations in surface sediments transect samples, core samples, tributary samples, and representative food web species were measured in 2002. Mercury stratigraphy in reservoir sediment cores and tributary surficial sediments from 2002 and 1995 (unpublished data) were compared and contrasted, core dating was performed to estimate sediment deposition rates and long-term trends for mercury accumulation, and mercury contamination in low trophic levels was assessed. Mercury concentrations in tributary stream sediments strongly correlate with its upstream or downstream proximity to the mine. Current results strongly suggest that the mine area is a continuing source of mercury to downstream tributaries with a strong concentration gradient from the area of the mine to the headwaters of the Coast Fork of the Willamette River. Mercury concentrations in surface sediment transects gradually increase from locations near the inlet to the dam, suggesting an increased mercury deposition rate as suspended solids approach the dam. Sediment cores from both time periods were collected from the deepest most stable region of the reservoir which is likely to have the highest sediment accumulation rates. Mercury concentrations from both cores increase from the surface to the bottom of the core. Core dating results indicate that approximately 40 years lapsed before mercury input to the reservoir stabilized and mercury input has remained relatively stable over the last 20 years. In an aquatic foodweb, methylmercury is the most important form of mercury because it is highly bioavailable for uptake into aquatic organisms. Bioaccumulation of mercury into aquatic organisms can occur through multiple pathways including uptake from sediment arid water, through the skin or cuticle, ventilation of gills, or consumption of contaminated sediment or prey. Biomagnification of methylmercury occurs at successively higher trophic levels of a foodweb. Total mercury concentrations were measured in organisms representative three trophic levels including benthic invertebrates, omnivorous/herbivorous amphibians and fish, and piscivorous fish. Mercury concentrations measured in benthic invertebrates (concentrations ranged from 0.035 mg/kg to 0.075 mg/kg wet-weight) and omnivorous/herbivorous amphibians and fish (concentrations ranged from less than 0.02 mg/kg to 0.043 mg/kg wet-weight) suggest little biomagnification has occurred at these trophic levels. Mercury concentrations in epaxial muscle of largemouth bass (concentrations ranged from 0.86 mg/kg to 1.6 mg/kg wet-weight) are at levels of concern for human health, and fish consumption advisories are posted at the Reservoir. Food web results indicate that significant biomagnification is occurring at higher trophic levels within the foodweb. Current studies suggest that elevated mercury concentrations in largemouth bass are strongly linked to the life history characteristics of the fish. Title: Mercury distribution in sediments and uptake into an aquatic food web at Cottage Grove Reservoir, Oregon Author: Morgans, Donna L. The Cottage Grove Reservoir has been impacted by a point source (Black Butte Mine), where mercury mining and processing has occurred within its watershed from the 1890's to the late 1960's. Total mercury concentrations in surface sediments transect samples, core samples, tributary samples, and representative food web species were measured in 2002. Mercury stratigraphy in reservoir sediment cores and tributary surficial sediments from 2002 and 1995 (unpublished data) were compared and contrasted, core dating was performed to estimate sediment deposition rates and long-term trends for mercury accumulation, and mercury contamination in low trophic levels was assessed. Mercury concentrations in tributary stream sediments strongly correlate with its upstream or downstream proximity to the mine. Current results strongly suggest that the mine area is a continuing source of mercury to downstream tributaries with a strong concentration gradient from the area of the mine to the headwaters of the Coast Fork of the Willamette River. Mercury concentrations in surface sediment transects gradually increase from locations near the inlet to the dam, suggesting an increased mercury deposition rate as suspended solids approach the dam. Sediment cores from both time periods were collected from the deepest most stable region of the reservoir which is likely to have the highest sediment accumulation rates. Mercury concentrations from both cores increase from the surface to the bottom of the core. Core dating results indicate that approximately 40 years lapsed before mercury input to the reservoir stabilized and mercury input has remained relatively stable over the last 20 years. In an aquatic foodweb, methylmercury is the most important form of mercury because it is highly bioavailable for uptake into aquatic organisms. Bioaccumulation of mercury into aquatic organisms can occur through multiple pathways including uptake from sediment arid water, through the skin or cuticle, ventilation of gills, or consumption of contaminated sediment or prey. Biomagnification of methylmercury occurs at successively higher trophic levels of a foodweb. Total mercury concentrations were measured in organisms representative three trophic levels including benthic invertebrates, omnivorous/herbivorous amphibians and fish, and piscivorous fish. Mercury concentrations measured in benthic invertebrates (concentrations ranged from 0.035 mg/kg to 0.075 mg/kg wet-weight) and omnivorous/herbivorous amphibians and fish (concentrations ranged from less than 0.02 mg/kg to 0.043 mg/kg wet-weight) suggest little biomagnification has occurred at these trophic levels. Mercury concentrations in epaxial muscle of largemouth bass (concentrations ranged from 0.86 mg/kg to 1.6 mg/kg wet-weight) are at levels of concern for human health, and fish consumption advisories are posted at the Reservoir. Food web results indicate that significant biomagnification is occurring at higher trophic levels within the foodweb. Current studies suggest that elevated mercury concentrations in largemouth bass are strongly linked to the life history characteristics of the fish. Close

• 216. [Article] Natural Variability of Nitrogen and Phosphorus in a Forested Headwater Stream System in the Oregon Coast Range

  Our understanding of nitrogen (N) and phosphorus (P) variability in forested stream systems with minimal disturbance is confounded by the complex interactions of hydrologic and biogeochemical controls. ...

  Citation × Citation Citation Abstract Copy Title: Natural Variability of Nitrogen and Phosphorus in a Forested Headwater Stream System in the Oregon Coast Range Author: Steadman, Casey L. Our understanding of nitrogen (N) and phosphorus (P) variability in forested stream systems with minimal disturbance is confounded by the complex interactions of hydrologic and biogeochemical controls. In this thesis, our studies evaluate the variability of stream water N and P in a forested, headwater system in the Oregon Coast Range. This type of system is underrepresented in the literature, though it serves as a reference against which more disturbed systems are compared. In the first study we evaluate instream N and P responses to precipitation across four catchments, Gus, Pothole, Rock, and Upper Main, using ~300 storm samples collected from a 1-year period. We characterized seasonal patterns and modeled relationships with cumulative precipitation and discharge for N and P species, nitrate-nitrogen (NO3--N), ammonium-nitrogen (NH4+-N), total dissolved nitrogen (TDN), soluble reactive phosphorus (SRP), and total dissolved phosphorus (TDP). There was significant spatial and temporal variability for all nutrient species concentrations with the exception of NH4+-N which showed little variability storm-to-storm. Maximum NO3--N concentration occurred at the onset of the wet season while SRP had the lowest concentration during the same period. Exports for all nutrient species across all catchments generally increased with cumulative precipitation. However, evidence was weak for relationships between nutrient concentrations and discharge. In the second study we evaluated instream N and P spatial variability in early summer with landscape characteristics using “snapshot” samples across 11 sub-catchments throughout the four catchments for eight years. Samples were taken during the end of the rainy season in June, a period in which the watershed is transitioning from wet to dry and during which nutrient relationships with landscape characteristics have not been specifically evaluated in the literature. We generated a series of simple linear regression models to assess the influence of landscape characteristics (geology, vegetation, sub-catchment slope, geology, elevation, and opportunities for biogeochemical activity) on each of the nutrient species evaluated in the first study, as well as dissolved organic nitrogen (DON), and dissolved organic phosphorus (DOP). To more fully illustrate the range of variability in stream water N and P, we modeled the mean, maximum, and coefficient of variation for each nutrient species. Results for means and maximums were comparable to each other. P concentrations varied with geology and elevation across catchments, with catchments composed of more intrusive geology showing less SRP and TDP. SRP and TDP also showed negative relationships with elevation. Mean distance from the hillslope to the stream and specific conductance, which was assumed to increase with increasing water residence time, served as metrics representing opportunities for biogeochemical activity. P species and, to a lesser degree, NH4+-N, varied with these metrics (positive relationships). NO3--N, DON, and TDN varied with Red alder (Alnus rubra) coverage on the hillslope (positive relationships) and was the only explanatory variable of significance for these nutrient species. Evidence was lacking for the importance of metrics summarizing vegetation in the riparian zone for any nutrient species. Similarly, evidence was weak to support the importance of sub-catchment slope for any nutrient species. Overall, there was little evidence to support the importance of the landscape characteristics on coefficient of variability. Our results suggest that characteristics influencing transportability of nutrients, such as sorptive properties, may play an important role in how landscape characteristics influence stream water nutrients during this season. This may be linked to the ease with which nutrient species are able to be flushed, which may affect the timing and degree in which drivers are related to stream water nutrients. Title: Natural Variability of Nitrogen and Phosphorus in a Forested Headwater Stream System in the Oregon Coast Range Author: Steadman, Casey L. Our understanding of nitrogen (N) and phosphorus (P) variability in forested stream systems with minimal disturbance is confounded by the complex interactions of hydrologic and biogeochemical controls. In this thesis, our studies evaluate the variability of stream water N and P in a forested, headwater system in the Oregon Coast Range. This type of system is underrepresented in the literature, though it serves as a reference against which more disturbed systems are compared. In the first study we evaluate instream N and P responses to precipitation across four catchments, Gus, Pothole, Rock, and Upper Main, using ~300 storm samples collected from a 1-year period. We characterized seasonal patterns and modeled relationships with cumulative precipitation and discharge for N and P species, nitrate-nitrogen (NO3--N), ammonium-nitrogen (NH4+-N), total dissolved nitrogen (TDN), soluble reactive phosphorus (SRP), and total dissolved phosphorus (TDP). There was significant spatial and temporal variability for all nutrient species concentrations with the exception of NH4+-N which showed little variability storm-to-storm. Maximum NO3--N concentration occurred at the onset of the wet season while SRP had the lowest concentration during the same period. Exports for all nutrient species across all catchments generally increased with cumulative precipitation. However, evidence was weak for relationships between nutrient concentrations and discharge. In the second study we evaluated instream N and P spatial variability in early summer with landscape characteristics using “snapshot” samples across 11 sub-catchments throughout the four catchments for eight years. Samples were taken during the end of the rainy season in June, a period in which the watershed is transitioning from wet to dry and during which nutrient relationships with landscape characteristics have not been specifically evaluated in the literature. We generated a series of simple linear regression models to assess the influence of landscape characteristics (geology, vegetation, sub-catchment slope, geology, elevation, and opportunities for biogeochemical activity) on each of the nutrient species evaluated in the first study, as well as dissolved organic nitrogen (DON), and dissolved organic phosphorus (DOP). To more fully illustrate the range of variability in stream water N and P, we modeled the mean, maximum, and coefficient of variation for each nutrient species. Results for means and maximums were comparable to each other. P concentrations varied with geology and elevation across catchments, with catchments composed of more intrusive geology showing less SRP and TDP. SRP and TDP also showed negative relationships with elevation. Mean distance from the hillslope to the stream and specific conductance, which was assumed to increase with increasing water residence time, served as metrics representing opportunities for biogeochemical activity. P species and, to a lesser degree, NH4+-N, varied with these metrics (positive relationships). NO3--N, DON, and TDN varied with Red alder (Alnus rubra) coverage on the hillslope (positive relationships) and was the only explanatory variable of significance for these nutrient species. Evidence was lacking for the importance of metrics summarizing vegetation in the riparian zone for any nutrient species. Similarly, evidence was weak to support the importance of sub-catchment slope for any nutrient species. Overall, there was little evidence to support the importance of the landscape characteristics on coefficient of variability. Our results suggest that characteristics influencing transportability of nutrients, such as sorptive properties, may play an important role in how landscape characteristics influence stream water nutrients during this season. This may be linked to the ease with which nutrient species are able to be flushed, which may affect the timing and degree in which drivers are related to stream water nutrients. Close

• 217. [Article] Clear Lake sediments: Anthropogenic changes in physical sedimentology and magnetic response

  Copyright by the Ecological Society of America.

  Citation × Citation Citation Abstract Copy Title: Clear Lake sediments: Anthropogenic changes in physical sedimentology and magnetic response Author: Zierenberg, Robert A., Suchanek, Thomas H., Osleger, David A., Adam, David P., Stoner, Joseph, Morgan, Sally Copyright by the Ecological Society of America. Title: Clear Lake sediments: Anthropogenic changes in physical sedimentology and magnetic response Author: Zierenberg, Robert A., Suchanek, Thomas H., Osleger, David A., Adam, David P., Stoner, Joseph, Morgan, Sally Copyright by the Ecological Society of America. Close

• 218. [Article] Regional Scale Modeling of Climate, Cryosphere, and Freshwater Discharge in Changing Coastal Mountain Environments

  The glaciated coastal mountain watersheds that drain into the Gulf of Alaska (GOA) provide a model laboratory to explore the challenges of hydrological modeling and study the impact of climate and glacier ...

  Citation × Citation Citation Abstract Copy Title: Regional Scale Modeling of Climate, Cryosphere, and Freshwater Discharge in Changing Coastal Mountain Environments Author: Beamer, Jordan P. The glaciated coastal mountain watersheds that drain into the Gulf of Alaska (GOA) provide a model laboratory to explore the challenges of hydrological modeling and study the impact of climate and glacier cover change on regional hydrology. The region is data-sparse and contains a complex assemblage of topography and land cover, including a system of mountain glaciers that are retreating at some of the highest rates on Earth. The high rates of runoff from precipitation and glacial melt delivered by coastal rivers influence ocean circulation patterns, rates of global sea level rise, and provide spawning habitat for the large salmon populations. Physically-based hydrological modeling of the major water budget components of the GOA, driven using historical reanalysis weather data and land cover, reveals that the modeled water budget components, particularly precipitation input, vary widely between commonly-used weather products. The majority of the large freshwater flux into the GOA is derived from distributed coastal streams rather than the large inland rivers. The modeled seasonal aggregated GOA hydrograph is dominated by the spring and early summer snowmelt, and supplemented by late summer glacial ice melt. Model results demonstrate good agreement with NASA Gravity Recovery and Climate Experiment (GRACE) satellite data in terms of annual amplitudes and long term losses (ice loss), and suggest that existing GRACE solutions, previously reported to represent glacier mass balance alone, are actually measuring the full water budget of land and ice surfaces. An ensemble of climate models and future emissions scenarios were paired with systematically altered land cover to test the sensitivity of the hydrologic system to changes in regional climate patterns and glacier coverage representative of late twenty first century conditions. Compared with the hindcast simulations, the model results forced with increased regional air temperatures and precipitation inputs and reduced glacier cover produce an increase in the annual GOA freshwater discharge volume. The seasonal GOA hydrograph is flattened due to increased winter runoff from more winter rainfall and less snow accumulation, and lower levels of snowmelt and glacier ice contribution. Large uncertainties exist in the direction of change in the glacier runoff component, primarily due to uncertainties that exist in predicting glacier response to climate change. Hydrological modeling with high resolution and inclusion of relevant physical processes can produce significantly improved products that are of high value to and in demand by numerous other scientific communities. However, the value and accuracy of the output from the hydrologic model is highly dependent on the weather forcing quality. Given the considerable importance of quality weather forcing for hydrologic modeling, it is imperative to assess the suitability of multiple products by evaluating local and regional performance and accounting for uncertainty. Additional efforts should be made to improve the spatial resolution of the reanalysis through downscaling and to strategically increase the number of weather stations at high elevations and incorporate that data into weather forcing datasets. Title: Regional Scale Modeling of Climate, Cryosphere, and Freshwater Discharge in Changing Coastal Mountain Environments Author: Beamer, Jordan P. The glaciated coastal mountain watersheds that drain into the Gulf of Alaska (GOA) provide a model laboratory to explore the challenges of hydrological modeling and study the impact of climate and glacier cover change on regional hydrology. The region is data-sparse and contains a complex assemblage of topography and land cover, including a system of mountain glaciers that are retreating at some of the highest rates on Earth. The high rates of runoff from precipitation and glacial melt delivered by coastal rivers influence ocean circulation patterns, rates of global sea level rise, and provide spawning habitat for the large salmon populations. Physically-based hydrological modeling of the major water budget components of the GOA, driven using historical reanalysis weather data and land cover, reveals that the modeled water budget components, particularly precipitation input, vary widely between commonly-used weather products. The majority of the large freshwater flux into the GOA is derived from distributed coastal streams rather than the large inland rivers. The modeled seasonal aggregated GOA hydrograph is dominated by the spring and early summer snowmelt, and supplemented by late summer glacial ice melt. Model results demonstrate good agreement with NASA Gravity Recovery and Climate Experiment (GRACE) satellite data in terms of annual amplitudes and long term losses (ice loss), and suggest that existing GRACE solutions, previously reported to represent glacier mass balance alone, are actually measuring the full water budget of land and ice surfaces. An ensemble of climate models and future emissions scenarios were paired with systematically altered land cover to test the sensitivity of the hydrologic system to changes in regional climate patterns and glacier coverage representative of late twenty first century conditions. Compared with the hindcast simulations, the model results forced with increased regional air temperatures and precipitation inputs and reduced glacier cover produce an increase in the annual GOA freshwater discharge volume. The seasonal GOA hydrograph is flattened due to increased winter runoff from more winter rainfall and less snow accumulation, and lower levels of snowmelt and glacier ice contribution. Large uncertainties exist in the direction of change in the glacier runoff component, primarily due to uncertainties that exist in predicting glacier response to climate change. Hydrological modeling with high resolution and inclusion of relevant physical processes can produce significantly improved products that are of high value to and in demand by numerous other scientific communities. However, the value and accuracy of the output from the hydrologic model is highly dependent on the weather forcing quality. Given the considerable importance of quality weather forcing for hydrologic modeling, it is imperative to assess the suitability of multiple products by evaluating local and regional performance and accounting for uncertainty. Additional efforts should be made to improve the spatial resolution of the reanalysis through downscaling and to strategically increase the number of weather stations at high elevations and incorporate that data into weather forcing datasets. Close

• 219. [Article] The conversion of allochthonous material by stream detritivores

  The role of detritivores in the conversion of allochthonous leaf material in two small streams in the Cascade Range of Oregon was studied. Their importance in relation to other consumers was estimated ...

  Citation × Citation Citation Abstract Copy Title: The conversion of allochthonous material by stream detritivores Author: Grafius, Edward John The role of detritivores in the conversion of allochthonous leaf material in two small streams in the Cascade Range of Oregon was studied. Their importance in relation to other consumers was estimated from emergence and benthic standing crop data. Ingestion rates and efficiencies of utilization of bigleaf maple (Acer macrophyllum) and red alder (Alnus rubra) leaves by several species of stream detritivores, one species of stonefly and several species of caddis lies, were measured under field or laboratory conditions. Mean consumption rates varied from 0. 04 mg/mg/day for Halesochila taylori larvae (Trichoptera:Limnephilidae) feeding on maple leaves to 0.60 mg /mg /day' for Lepidostoma sp. larvae (Trichoptera :Lepidostomatidae) feeding on alder and maple leaves. Ecclisomyia sp. larvae (Trichoptera:Limnephilidae) consumed 0.10 mg/mg/day of maple leaves in the field. All consumption rates were calculated assuming constant feeding rates by the insects and constant rates of leaf decomposition without insects, but a model is discussed that allows non-linear feeding and leaf decomposition rates. Efficiency of food utilization was measured either as assimilation efficiency, the ratio of assimilated food (ingestion minus egestion) to consumption; or as gross growth efficiency, the slope of the growth:consumption line calculated by linear regression analysis. Assimilation efficiencies ranged from 7.4% for Heteroplectron californicum larvae (Trichoptera:Calamoceratidae) fed on maple leaves to 12% for Lepidostoma sp. larvae fed on alder and maple leaves. In order to assess the role of microbial communities in leaf utilization by Pteronarcys princeps (Plecoptera:Pteronarcidae), nymphs were fed untreated maple leaves and maple leaves treated to reduce bacterial or fungal populations, for a period of 54 days. Gross growth efficiency when fed on untreated leaves was 4.94% compared with 1.03% for the insects fed antibacterial treated leaves. When fed fungicide treated leaves, the mean consumption rate was less than 10% of the consumption of untreated leaves and mortality was 85% compared with 15% for the insects fed untreated leaves. It was concluded that bacteria play a major role in the utilization of leaves by Pteronarcys princeps and that fungi are important as astimulus to feeding, although they may also aid in utilization of the leaves. To aid in classifying the emergence and benthos data into trophic categories, an extensive literature review of the foods of aquatic insects was compiled. The estimated yearly emergence of aquatic insects from Watershed 10 was 81 mg/m² /yr. of algivores, 253 mg/m²/yr. of detritivores, and 135 mg/m²/yr. of carnivores. The mean standing crop of aquatic insects in an old-growth location in Mack Creek was 0.85 g/m² of algivores, 3.10 g/m² of detritivores, and 2.31 g/m² of carnivores; and 0.83 g/m² of algivores, 1.57 g/m² of detritivores, and 4.04 g/m² of carnivores in a clear-cut location. A model is included describing the cycling of energy within the detrital component of the stream system, reflecting the importance of the fecal material and leaf fragments that return to the detrital pool. Title: The conversion of allochthonous material by stream detritivores Author: Grafius, Edward John The role of detritivores in the conversion of allochthonous leaf material in two small streams in the Cascade Range of Oregon was studied. Their importance in relation to other consumers was estimated from emergence and benthic standing crop data. Ingestion rates and efficiencies of utilization of bigleaf maple (Acer macrophyllum) and red alder (Alnus rubra) leaves by several species of stream detritivores, one species of stonefly and several species of caddis lies, were measured under field or laboratory conditions. Mean consumption rates varied from 0. 04 mg/mg/day for Halesochila taylori larvae (Trichoptera:Limnephilidae) feeding on maple leaves to 0.60 mg /mg /day' for Lepidostoma sp. larvae (Trichoptera :Lepidostomatidae) feeding on alder and maple leaves. Ecclisomyia sp. larvae (Trichoptera:Limnephilidae) consumed 0.10 mg/mg/day of maple leaves in the field. All consumption rates were calculated assuming constant feeding rates by the insects and constant rates of leaf decomposition without insects, but a model is discussed that allows non-linear feeding and leaf decomposition rates. Efficiency of food utilization was measured either as assimilation efficiency, the ratio of assimilated food (ingestion minus egestion) to consumption; or as gross growth efficiency, the slope of the growth:consumption line calculated by linear regression analysis. Assimilation efficiencies ranged from 7.4% for Heteroplectron californicum larvae (Trichoptera:Calamoceratidae) fed on maple leaves to 12% for Lepidostoma sp. larvae fed on alder and maple leaves. In order to assess the role of microbial communities in leaf utilization by Pteronarcys princeps (Plecoptera:Pteronarcidae), nymphs were fed untreated maple leaves and maple leaves treated to reduce bacterial or fungal populations, for a period of 54 days. Gross growth efficiency when fed on untreated leaves was 4.94% compared with 1.03% for the insects fed antibacterial treated leaves. When fed fungicide treated leaves, the mean consumption rate was less than 10% of the consumption of untreated leaves and mortality was 85% compared with 15% for the insects fed untreated leaves. It was concluded that bacteria play a major role in the utilization of leaves by Pteronarcys princeps and that fungi are important as astimulus to feeding, although they may also aid in utilization of the leaves. To aid in classifying the emergence and benthos data into trophic categories, an extensive literature review of the foods of aquatic insects was compiled. The estimated yearly emergence of aquatic insects from Watershed 10 was 81 mg/m² /yr. of algivores, 253 mg/m²/yr. of detritivores, and 135 mg/m²/yr. of carnivores. The mean standing crop of aquatic insects in an old-growth location in Mack Creek was 0.85 g/m² of algivores, 3.10 g/m² of detritivores, and 2.31 g/m² of carnivores; and 0.83 g/m² of algivores, 1.57 g/m² of detritivores, and 4.04 g/m² of carnivores in a clear-cut location. A model is included describing the cycling of energy within the detrital component of the stream system, reflecting the importance of the fecal material and leaf fragments that return to the detrital pool. Close

• 220. [Article] Soil and understory vegetation characteristics of a tractor-logged forest in eastern Oregon

  A 1.5 ha mixed conifer stand in the Blue Mountains of Oregon was intensively examined to assess the impact of logging disturbance on soils and herbaceous vegetation. Sampling was conducted six years after ...

  Citation × Citation Citation Abstract Copy Title: Soil and understory vegetation characteristics of a tractor-logged forest in eastern Oregon Author: Snider, Michael Duane A 1.5 ha mixed conifer stand in the Blue Mountains of Oregon was intensively examined to assess the impact of logging disturbance on soils and herbaceous vegetation. Sampling was conducted six years after much of the timber overstory was removed in a shelterwood cut and yarded by crawler tractor. A year after harvest slash was machine-piled and burned. Skid trails and other areas of disturbance were seeded with a mixture of perennial grasses. Logging impact was divided into five "disturbance classes'* as treatments: general, berm, slash fire ring, skid trail, and undisturbed. Analysis of variance was used to test treatment differences. Characteristics of skid trails were analyzed separately from those of non-skid trail areas. Soil compaction was evident on skid trails from substantially lower rates of infiltration and saturated hydraulic conductivity (SHC), and a higher average soil strength, relative to undisturbed areas. Skid trail compaction probably resulted from mixing of denser subsoils with the low-density volcanic ash soils at the surface, and from compression by tractor activity. Bulk density of skid trails was not statistically different from that of undisturbed areas. Soil densities of all classes of disturbance were generally lower than 0.9 g/cm³. Non-skid trail areas were not compacted. Soil water conductance rates markedly differed among disturbance classes, but were considerably higher than storm intensities projected for the region. High sample variability of both infiltration and SHC rates made interpreation of statistical comparisons among treatments difficult. Some watershed protection values were diminished on the disturbed areas. Litter cover of skid trails, general disturbance areas, and berms was 27, 45, and 68 percent lower, respectively, than that of undisturbed areas. Depth of litter decreased correspondingly among the disturbed treatments. Litter was nearly absent on the fire rings. Another agent in soil stabilization, belowground biomass, was two-thirds less on skid trails, in comparision with undisturbed areas. Amounts of soil organic matter in the surface 3 cm were significantly lower for the four types of ground disturbance, relative to undisturbed areas; however, organic matter was higher at some subsurface levels of berms, fire rings, and skid trails. Bulk density of non-skid trail soils was inversely correlated with organic matter content. There were no statistically significant differences among treatments for soil water retained at four potentials, for coarse mineral fragment content, or for fine earth particle size distribution. Aboveground vegetation production of the seeded skid trails was 66 percent greater than that of undisturbed areas. Understory plant cover of skid trails and undisturbed areas was equivalent. Correlation between aboveground herbaceous production and soil strength at the 5-10 cm depth of skid trails was not statistically significant. Title: Soil and understory vegetation characteristics of a tractor-logged forest in eastern Oregon Author: Snider, Michael Duane A 1.5 ha mixed conifer stand in the Blue Mountains of Oregon was intensively examined to assess the impact of logging disturbance on soils and herbaceous vegetation. Sampling was conducted six years after much of the timber overstory was removed in a shelterwood cut and yarded by crawler tractor. A year after harvest slash was machine-piled and burned. Skid trails and other areas of disturbance were seeded with a mixture of perennial grasses. Logging impact was divided into five "disturbance classes'* as treatments: general, berm, slash fire ring, skid trail, and undisturbed. Analysis of variance was used to test treatment differences. Characteristics of skid trails were analyzed separately from those of non-skid trail areas. Soil compaction was evident on skid trails from substantially lower rates of infiltration and saturated hydraulic conductivity (SHC), and a higher average soil strength, relative to undisturbed areas. Skid trail compaction probably resulted from mixing of denser subsoils with the low-density volcanic ash soils at the surface, and from compression by tractor activity. Bulk density of skid trails was not statistically different from that of undisturbed areas. Soil densities of all classes of disturbance were generally lower than 0.9 g/cm³. Non-skid trail areas were not compacted. Soil water conductance rates markedly differed among disturbance classes, but were considerably higher than storm intensities projected for the region. High sample variability of both infiltration and SHC rates made interpreation of statistical comparisons among treatments difficult. Some watershed protection values were diminished on the disturbed areas. Litter cover of skid trails, general disturbance areas, and berms was 27, 45, and 68 percent lower, respectively, than that of undisturbed areas. Depth of litter decreased correspondingly among the disturbed treatments. Litter was nearly absent on the fire rings. Another agent in soil stabilization, belowground biomass, was two-thirds less on skid trails, in comparision with undisturbed areas. Amounts of soil organic matter in the surface 3 cm were significantly lower for the four types of ground disturbance, relative to undisturbed areas; however, organic matter was higher at some subsurface levels of berms, fire rings, and skid trails. Bulk density of non-skid trail soils was inversely correlated with organic matter content. There were no statistically significant differences among treatments for soil water retained at four potentials, for coarse mineral fragment content, or for fine earth particle size distribution. Aboveground vegetation production of the seeded skid trails was 66 percent greater than that of undisturbed areas. Understory plant cover of skid trails and undisturbed areas was equivalent. Correlation between aboveground herbaceous production and soil strength at the 5-10 cm depth of skid trails was not statistically significant. Close