Concurrent Sessions A: Passage Effectiveness Monitoring in Small Streams I - Examination of Methods for Detecting Small, Non-Game Fish Passage at Road-Stream Crossings in the Southeastern U.S
As p~t of the Shenandoah National Park: Fish in Sensitive Habitats (SNP: FISH) project, movements... more As p~t of the Shenandoah National Park: Fish in Sensitive Habitats (SNP: FISH) project, movements of blacknose dace (Rhinichthys atratulus) and brook char (Salvelinusfontinalis) were examined during ex~9osure to ,artificial acidification in paired channels of a laboratory stream. The objective of this study was to determine the ability of the fish to avoid depressions in the ambient pH and recognize and use a neutral-pH microhabitat refuge dumlg acute reductions. Fish preference for a particular channel was stati~ically predictable based on the delivery of food. We les"led fish avoidance behavior by manipulating food and the delivery of a pulse of acidified water. Both blaekmose dace and brook d~ar avoided the acid pulse (ambient pH reduced from 7.2 to 5.1) by sheltering in the pHneulral refuge. Extensive field sampling in refuge microhabitats before and during episodic acidification is needed to determine changes in the distributional patterns of these species associated with acid precipitation events.
As part of a study of aquatic faunal community changes along riverine-lacustrine transition zones... more As part of a study of aquatic faunal community changes along riverine-lacustrine transition zones upstream of Lewis Smith Reservoir in northwest Alabama, USA, we collected crayfish from 60 sites in the Sipsey Fork, Brushy Creek, and selected tributaries (Black Warrior River system). After finding two unexpected and possibly-introduced crayfish species, we expanded our investigation of crayfish distributions to include crayfish obtained from stomachs of black bass (Micropterus spp.) caught at seven sites in the reservoir. To explore what crayfish species were in the drainage historically, we examined museum databases as well as stomach and intestinal contents of a variety of preserved fishes that were caught in the Sipsey Fork and Brushy Creek drainages upstream of the reservoir in the early 1990's. Of the seven crayfish species collected, one, Orconectes (Procericambarus) sp. nr ronaldi, was not previously reported from Alabama, and another, O. lancifer, was not reported from the Black Warrior River system prior to the study. Three are known or possibly introduced species. Upstream of the reservoir, the native species Cambarus obstipus, C. striatus, and O. validus were common. The same three species were found in fish collected in the 1990's. Orconectes perfectus was found only in the reservoir but may be native to the drainage. Orconectes lancifer was in the reservoir and in stream reaches influenced by the reservoir. Evidence points to O. lancifer being introduced in the drainage, but this is uncertain. Orconectes sp. nr ronaldi was found in a relatively small portion of Brushy Creek and its tributaries, in both flowing and impounded habitats, and may be introduced. Orconectes virilis is introduced in Alabama and was found only in stomachs of fish collected in the reservoir.
Citation for proceedings: Stringer, Christina E.; Krauss, Ken W.; Latimer, James S., eds. 2016. H... more Citation for proceedings: Stringer, Christina E.; Krauss, Ken W.; Latimer, James S., eds. 2016. Headwaters to estuaries: advances in watershed science and management—Proceedings of the Fifth Interagency Conference on Research in the Watersheds. March 2-5, 2015, North Charleston, South Carolina. e-Gen. Tech. Rep. SRS-211. Asheville, NC: U.S. Department of Agriculture Forest Service, Southern Research Station. 302 p. SPATIO-TEMPORAL VARIATION IN DISTRIBUTION OF AQUATIC SPECIES AND THEIR HABITATS IN A RESERVOIR TRANSITION ZONE
The US Fish and Wildlife Service is considering the addition of 374 riparian and aquatic species ... more The US Fish and Wildlife Service is considering the addition of 374 riparian and aquatic species in the southeastern United States to the federal Threated and Endangered Species List. This recommendation is a result of a 2011 petition, which recognized forest operations as having negative effects on 51 percent of the listed species, citing research conducted in the absence of Best Management Practices (BMPs) (Federal Register 76(187):59836-59862). We conducted a literature review to evaluate how BMPs might benefit these species, but found that information specific to these riparian species and forest operations was generally limited. Available literature pertaining to BMP effects and these riparian species generally contained broader conclusions, which were often conducted at higher taxonomic levels. We were able to develop some broad interpretations that support the benefits of BMP implementation to many of these species. Our review indicated that BMPs (i.e., streamside management ...
As part of a study of aquatic faunal community changes along riverine-lacustrine transition zones... more As part of a study of aquatic faunal community changes along riverine-lacustrine transition zones upstream of Lewis Smith Reservoir in northwest Alabama, USA, we collected crayfish from 60 sites in the Sipsey Fork, Brushy Creek, and selected tributaries (Black Warrior River system). After finding two unexpected and possibly-introduced crayfish species, we expanded our investigation of crayfish distributions to include crayfish obtained from stomachs of black bass (Micropterus spp.) caught at seven sites in the reservoir. To explore what crayfish species were in the drainage historically, we examined museum databases as well as stomach and intestinal contents of a variety of preserved fishes that were caught in the Sipsey Fork and Brushy Creek drainages upstream of the reservoir in the early 1990’s. Of the seven crayfish species collected, one, Orconectes (Procericambarus) sp. nr ronaldi, was not previously reported from Alabama, and another, O. lancifer, was not reported from the Bl...
Catadromous eels are found in more habitats than any other fish and are capable of inhabiting mar... more Catadromous eels are found in more habitats than any other fish and are capable of inhabiting marine, brackish and freshwater environments. In this study we used the American Eel (Anguilla rostrata) as a bioindicator organism to create a novel method of using spatial analysis to study species conservation over landscape scales. We built a model of the subwatersheds of the Chesapeake Bay using a Digital Elevation Model (DEM) and overlaid eel density data (> 1 million eels sampled), dam density data and land use in ArcGIS. Dam construction in the study area peaked between 1955 and 1975, possibly as a result of flood control measures. Effects of land use were localized and most pronounced in areas around Baltimore, Maryland, Washington, D.C. and Richmond, Virginia, USA. Results indicate the Potomac and Rappahannock rivers appear to be areas of lesser concern while the upper James and York rivers are ideal for follow-up studies, since these area rank poorly in both eel density and ba...
The U.S. Forest Service and other federal land managers are responsible for maintaining the produ... more The U.S. Forest Service and other federal land managers are responsible for maintaining the productivity of aquatic-riparian ecosystems, the associated native biota, and the ecosystem services they provide. These public lands are important sources of water, recreation opportunities, and habitat for a suite of animals and plants, including many that are protected under the Endangered Species Act. To meet these challenges and responsibilities, recent science suggests modifying practices to provide a broader array of habitat, biological conditions, and ecosystem functions than are associated with traditional management approaches. We suggest that by linking approaches based on natural disturbance and portfolio concepts, managers can achieve a robust strategy and desired outcomes more reliably and cost effectively. Locally complex habitat conditions created by natural disturbances provide the template for biological diversity to play out if provided enough time. Accordingly, natural disturbance regimes play an important role in creating and sustaining habitat and biological complexities on the landscape, suggesting that, to the extent possible, management actions should emulate natural disturbance processes at appropriate spatial and temporal scales. In concert with this approach, the portfolio effect (i.e., diversity that mitigates risk) provides justification for promoting connected heterogeneous habitats that reduce the risk of synchronous large-scale population and ecosystem collapse. In this article, we describe how disturbance and portfolio concepts fit into a broader strategy of conserving ecosystem integrity and dynamism and provide examples of how these concepts can be used to address a wide range of management concerns. Ultimately, the outcome for populations, habitats, and landscapes depends on how well environmental change is understood, the degree to which change is appropriately addressed by natural resource managers, and solutions that allow populations and ecosystems to persist in the presence of and be resilient to a growing scope of human influences.
Canadian Journal of Fisheries and Aquatic Sciences, 2015
Unregulated river systems are organized in a hierarchy in which large-scale factors (i.e., landsc... more Unregulated river systems are organized in a hierarchy in which large-scale factors (i.e., landscape and segment scales) influence local habitats (i.e., reach, meso-, and microhabitat scales), and both differentially exert selective pressures on biota. Dams, however, create discontinua in these processes and change the hierarchical structure. We examined the relative roles of hydrology and other instream factors, within a hierarchical landscape context, in organizing fish communities in regulated and unregulated tributaries to the Upper Tennessee River, USA. We used multivariate regression trees to identify factors that partition fish assemblages based on trait similarities, irrespective of spatial scale. We then used classical path analysis and structural equation modeling to evaluate the most plausible hierarchical causal structure of specific trait-based community components, given the data. Both statistical approaches suggested that river regulation affects stream fishes through...
Regional frameworks have been used extensively in recent years to aid in broad-scale management. ... more Regional frameworks have been used extensively in recent years to aid in broad-scale management. Widely used landscape-based regional frameworks, such as hydrologic landscape regions (HLRs) and physiographic provinces, may provide predictive tools of hydrologic variability. However, hydrologic-based regional frameworks, created using only streamflow data, are also available and have been created at various scales; thus, relating frameworks that share a common purpose can be informative. In addition, identifying how the relative importance of variables change in governing streamflow with respect to scale can also be informative. The purpose of this study was to determine whether landscape-based frameworks could explain variation in streamflow classifications and in the hydrologic variables used in their creation. We also evaluated how climate and watershed-based variables govern the divergence of different flow classifications at two different scales. HLRs and physiographic provinces poorly predicted flow class affiliation within our study and for the entire USA, although physiographic provinces explained more variability. We also found that HLRs explained very little variation in individual hydrologic parameters. Using variables summarized at the watershed scale, we found that climate will play a larger role in influencing hydrology across the entire US, whereas soils may govern variation in hydrology at smaller scales. Our results suggest that predictor variables, developed at the watershed scale, may be the most appropriate at explaining hydrology and that the variables used in creating regional landscape-based frameworks may be more useful than the frameworks themselves. In addition, managers should be careful when using landscape-based regional classifications for stream management because the scale of their construction may be too broad to capture differences in flow variability.
North American Journal of Fisheries Management, 2010
We assessed the efficacy of passive gravel addition at forming catostomid spawning habitat under ... more We assessed the efficacy of passive gravel addition at forming catostomid spawning habitat under various flow regimes in the Cheoah River, a high-gradient tailwater river in North Carolina. The purpose was to provide a case study that included recommendations for future applications. A total of 76.3 m 3 (162 tons) of washed gravel (10-50 mm) was passively dumped down the streambank and into the channel in four locations. Gravel sites differed in terms of average reach slope, bank slope, and the initial volume of gravel added, which could have influenced gravel entrainment. Maps of gravel movement under various flows suggested that large-magnitude discharges (113 m 3 /s) caused extensive migration; however, less obvious, smaller discharges (;28 m 3 /s) still caused substantial shifting, which may influence the stability of catostomid spawning substrates. Following gravel addition, the proportion of gravel in the streambed was significantly higher at all gravel sites. However, comparisons of sites to reference stream reaches suggested that sand, gravel, and cobble were still extremely deficient. Additionally, the volume of gravel was inadequate to create gravel depths that provided suitable habitat for catostomid spawning. Although periodic, passive gravel additions may take years to provide suitable spawning habitat for some fish species, we found that river chub Nocomis micropogon utilized the newly added gravel for spawning.
Journal of the North American Benthological Society, 2011
Consumer nutrient cycling supplies limiting elements to autotrophic and heterotrophic organisms i... more Consumer nutrient cycling supplies limiting elements to autotrophic and heterotrophic organisms in aquatic systems. However, the role of consumers in supplying nutrients may change depending on their diet and their own stoichiometry. We evaluated the stoichiometry, N and P excretion, and diets of the dominant macroinvertebrates and fish at 6 stream sites to determine if the nutritional composition of food alters nutrient excretion. We used Sterner's (1990) nutrient homeostasis model as a reference to gauge whether consumer nutrient excretion is influenced by diet. Body stoichiometry explained 61% of the variation in N:P excretion by macroinvertebrates but only 11% of the variation for fish. In both cases, the relationship was driven by 2 P-rich end-members, crayfish and mottled sculpin. Results of Akaike Information Criterion (AIC) analysis showed that family alone explained 71% of the variation in N:P excretion in macroinvertebrates and 31% of the variation in fish. Diet explained only 8% of the variation in both cases. Most consumers (9 of 11) had N:P excretion values that were well below predictions of Sterner's model. Two taxa, crayfish and sculpin, had N:P excretion that overlapped the model's predictions. Our results suggest that crayfish and sculpin may display strict homeostasis with respect to N and P and that their growth might be P-limited. Other consumers may be more flexible in their stoichiometry and not P-limited. We speculate that the extremely low excretion N:P measured for many consumers might have been the result of semiflexible homeostasis, inaccuracies in our assessment of dietary nutrients, growth-limiting nutrients other than N or P, or lack of egestion data. Our results suggest that crayfish and sculpin may alter N and P dynamics in streams by excreting low amounts of P relative to N compared to what is generally available in the water column.
For some time, ecologists have attempted to make generalizations concerning how disturbances infl... more For some time, ecologists have attempted to make generalizations concerning how disturbances influence natural ecosystems, especially river systems. The existing literature suggests that dams homogenize the hydrologic variability of rivers. However, this might insinuate that dams affect river systems similarly despite a large gradient in natural hydrologic character. In order to evaluate patterns in dam-regulated hydrology and associated ecological relationships, a broad framework is needed. Flow classes, or groups of streams that share similar hydrology, may provide a framework to evaluate the relative effects of dam regulation on natural flow dynamics. The purpose of this study was to use a regional flow classification as the foundation for evaluating patterns of hydrologic alteration due to dams and to determine if the response of rivers to regulation was specific to different flow classes. We used the US Geological Survey (USGS) database to access discharge information for 284 unregulated and 117 regulated gage records. For each record, we calculated 44 hydrologic statistics, including the Indicators of Hydrologic Alteration. We used a sub-regional flow classification for eight states as a way to stratify unregulated and regulated streams into comparable units. In general, our results showed that dam regulation generally had stronger effects on hydrologic indices than other disturbances when models were stratified by flow class; however, the effects of urbanization, withdrawals, and fragmentation, at times, were comparable or exceeded the effects of dam regulation. In agreement with the existing literature, maximum flows, flow variability, and rise rates were lower whereas minimum flows and reversals were higher in dam regulated streams. However, the response of monthly and seasonal flows, flow predictability, and baseflows were variable depending on flow class membership. Principal components analysis showed that regulated streams occupied a larger multivariate space than unregulated streams, which suggests that dams may not homogenize all river systems, but may move them outside the bounds of normal river function. Ultimately, our results suggest that flow classes provide a suitable framework to generalize patterns in hydrologic alterations due to dam regulation.
In order for habitat restoration in regulated rivers to be effective at large scales, broadly app... more In order for habitat restoration in regulated rivers to be effective at large scales, broadly applicable frameworks are needed that provide measurable objectives and contexts for management. The Ecological Limits of Hydrologic Alteration (ELOHA) framework was created as a template to assess hydrologic alterations, develop relationships between altered streamflow and ecology, and establish environmental flow standards. We tested the utility of ELOHA in informing flow restoration applications for fish and riparian communities in regulated rivers in the Upper Tennessee River Basin (UTRB). We followed the steps of ELOHA to generate univariate relationships between altered flows and ecology within the UTRB. By comparison, we constructed multivariate models to determine improvements in predictive capacity with the addition of non-flow variables. We then determined whether those relationships could predict fish and riparian responses to flow restoration in the Cheoah River, a regulated system within the UTRB. Although ELOHA provided a robust template to construct hydrologic information and predict hydrology for ungaged locations, our results do not suggest that univariate relationships between flow and ecology (step 4, ELOHA process) can produce results sufficient to guide flow restoration in regulated rivers. After constructing multivariate models, we successfully developed predictive relationships between flow alterations and fish/ riparian responses. In accordance with model predictions, riparian encroachment displayed consistent decreases with increases in flow magnitude in the Cheoah River; however, fish richness did not increase as predicted 4 years after restoration. Our results suggest that altered temperature and substrate and the current disturbance regime may have reduced opportunities for fish species colonization. Our case study highlights the need for interdisciplinary science in defining environmental flows for regulated rivers and the need for adaptive management approaches once flows are restored.
They always took time from their incredibly busy schedules to provide advice and commiseration. E... more They always took time from their incredibly busy schedules to provide advice and commiseration. Eric Hallerman had a fresh perspective and encouraged my efforts to enlarge my graduate school experience outside of research. Andy Dolloff generated ideas and insight for data collection when I initiated this project. He also provided advice and information that helped me launch my career. Tammy Newcomb's expertise on instream flow studies and habitat modeling was invaluable. I appreciate her open-minded interest and advice concerning the methods I chose to use to complete this dissertation. Eric Smith provided himself and his students for invaluable statistical advice and was extremely patient with my lack of expertise. The U.S. Fish and Wildlife Service and the Virginia Department of Game and Inland Fisheries provided funding for this project. Personnel from these agencies showed consistent interest and accompanied me on field excursions to learn more about the Roanoke logperch. I am particularly grateful to Mike Pinder, who arranged funding for this project and remained supportive and helpful throughout my tenure at Virginia Tech.
ABS1RACT : Even though beavers (Castor canadensis carolinensis) occur over most of the southeaste... more ABS1RACT : Even though beavers (Castor canadensis carolinensis) occur over most of the southeastern United States , the impacts of increasing beaver populations on riparian forests within the southern Appalachian mountains are. not been well documented. Long-term browsing and inundation by beaver may alter the composition and structure of riparian forests. A survey of 62 streams (7 4 mi) within the Chauga River drainage in the mountains of South Carolina was conducted during 1991-1992 to determine the level of beaver activity within the drainage and the amount of timber damaged by beaver activiti es. Thirty-six streams had evidence of significant beaver activity with a total of 5.3 mi (7 .2%) affected by beaver. Twenty-six streams (17.3 mi), primarily those with steep gradients and no flood plains, had no evidence of beaver activity. On beaver impacted areas , values of beaver damaged timber averaged $781.27 /ac for sawtimber and $36.01/ac for pulpwood. While high in terms of volume/acre within impacted riparian areas, timber damage was relatively minor for the entire drainage because of the small area (49 .2 ac) affected. Beavers occupied the entire state of South Carolina prior to European settlement, but populations were eliminated from South Carolina by the early 1800s (Salyer 1946, Golley 1966). In 1940 and 1950, beaver were reintroduced into South Carolina in Chesterfield County in the upper coastal plain on the North Carolina border and in Aiken County in the lower Piedmont on the Georgia border, respectively. During the 1950s, beaver probably migrated into South Carolina from Georgia. By 1966, 12 of the state's 46 counties were inhabited by beaver (Golley 1966); 26 counties had established beaver colonies by 197 4 (Hair and Woodward 1974). Presently, beaver occur in all counties of South Carolina (J. Witt, South Carolina Wildlife Marine Resources Department, pers. comm.) .
univ-st-etienne.fr RÉSUMÉ. Les documents multimédia composés de texte et d'images sont de plus en... more univ-st-etienne.fr RÉSUMÉ. Les documents multimédia composés de texte et d'images sont de plus en plus présents grâce à Internet et à l'augmentation des capacités de stockage. Cet article présente un modèle de représentation de documents multimédia qui combine l'information textuelle et l'information visuelle. En utilisant une approche par sac de mot, un document composé de texte et d'image peut être décrit par des vecteurs correspondant à chaque type d'information. Pour une requête multimédia donnée, une liste de documents pertinents est retournée en combinant linéairement les résultats obtenus séparément sur chaque modalité. Le but de cet article est d'étudier l'impact, sur les résultats, du poids attribué à l'information visuelle par rapport à l'information textuelle. Des expérimentations, réalisées sur la collection multimédia ImageCLEF extraite de l'encyclopédie Wikipedia, montrent que les résultats peuvent être améliorés après une première étape d'apprentissage de ce poids. ABSTRACT. Multimedia documents are increasingly used which involve to develop model to that kind of data. In this paper we present a multimedia model which combines textual and visual information. Using a bag of words approach, we can represent a textual and visual document with a vector for each modality. Given a multimedia query, our model lets us linearly combine scores obtained for each modality and return a list of relevant retrieved documents. This article aims at studying the influence of the weight given to the visual information according to the textual one. Experiments on the multimedia ImageCLEF collection extracted from Wikipedia show that results can be improved by learning this weight parameter.
Concurrent Sessions A: Passage Effectiveness Monitoring in Small Streams I - Examination of Methods for Detecting Small, Non-Game Fish Passage at Road-Stream Crossings in the Southeastern U.S
As p~t of the Shenandoah National Park: Fish in Sensitive Habitats (SNP: FISH) project, movements... more As p~t of the Shenandoah National Park: Fish in Sensitive Habitats (SNP: FISH) project, movements of blacknose dace (Rhinichthys atratulus) and brook char (Salvelinusfontinalis) were examined during ex~9osure to ,artificial acidification in paired channels of a laboratory stream. The objective of this study was to determine the ability of the fish to avoid depressions in the ambient pH and recognize and use a neutral-pH microhabitat refuge dumlg acute reductions. Fish preference for a particular channel was stati~ically predictable based on the delivery of food. We les"led fish avoidance behavior by manipulating food and the delivery of a pulse of acidified water. Both blaekmose dace and brook d~ar avoided the acid pulse (ambient pH reduced from 7.2 to 5.1) by sheltering in the pHneulral refuge. Extensive field sampling in refuge microhabitats before and during episodic acidification is needed to determine changes in the distributional patterns of these species associated with acid precipitation events.
As part of a study of aquatic faunal community changes along riverine-lacustrine transition zones... more As part of a study of aquatic faunal community changes along riverine-lacustrine transition zones upstream of Lewis Smith Reservoir in northwest Alabama, USA, we collected crayfish from 60 sites in the Sipsey Fork, Brushy Creek, and selected tributaries (Black Warrior River system). After finding two unexpected and possibly-introduced crayfish species, we expanded our investigation of crayfish distributions to include crayfish obtained from stomachs of black bass (Micropterus spp.) caught at seven sites in the reservoir. To explore what crayfish species were in the drainage historically, we examined museum databases as well as stomach and intestinal contents of a variety of preserved fishes that were caught in the Sipsey Fork and Brushy Creek drainages upstream of the reservoir in the early 1990's. Of the seven crayfish species collected, one, Orconectes (Procericambarus) sp. nr ronaldi, was not previously reported from Alabama, and another, O. lancifer, was not reported from the Black Warrior River system prior to the study. Three are known or possibly introduced species. Upstream of the reservoir, the native species Cambarus obstipus, C. striatus, and O. validus were common. The same three species were found in fish collected in the 1990's. Orconectes perfectus was found only in the reservoir but may be native to the drainage. Orconectes lancifer was in the reservoir and in stream reaches influenced by the reservoir. Evidence points to O. lancifer being introduced in the drainage, but this is uncertain. Orconectes sp. nr ronaldi was found in a relatively small portion of Brushy Creek and its tributaries, in both flowing and impounded habitats, and may be introduced. Orconectes virilis is introduced in Alabama and was found only in stomachs of fish collected in the reservoir.
Citation for proceedings: Stringer, Christina E.; Krauss, Ken W.; Latimer, James S., eds. 2016. H... more Citation for proceedings: Stringer, Christina E.; Krauss, Ken W.; Latimer, James S., eds. 2016. Headwaters to estuaries: advances in watershed science and management—Proceedings of the Fifth Interagency Conference on Research in the Watersheds. March 2-5, 2015, North Charleston, South Carolina. e-Gen. Tech. Rep. SRS-211. Asheville, NC: U.S. Department of Agriculture Forest Service, Southern Research Station. 302 p. SPATIO-TEMPORAL VARIATION IN DISTRIBUTION OF AQUATIC SPECIES AND THEIR HABITATS IN A RESERVOIR TRANSITION ZONE
The US Fish and Wildlife Service is considering the addition of 374 riparian and aquatic species ... more The US Fish and Wildlife Service is considering the addition of 374 riparian and aquatic species in the southeastern United States to the federal Threated and Endangered Species List. This recommendation is a result of a 2011 petition, which recognized forest operations as having negative effects on 51 percent of the listed species, citing research conducted in the absence of Best Management Practices (BMPs) (Federal Register 76(187):59836-59862). We conducted a literature review to evaluate how BMPs might benefit these species, but found that information specific to these riparian species and forest operations was generally limited. Available literature pertaining to BMP effects and these riparian species generally contained broader conclusions, which were often conducted at higher taxonomic levels. We were able to develop some broad interpretations that support the benefits of BMP implementation to many of these species. Our review indicated that BMPs (i.e., streamside management ...
As part of a study of aquatic faunal community changes along riverine-lacustrine transition zones... more As part of a study of aquatic faunal community changes along riverine-lacustrine transition zones upstream of Lewis Smith Reservoir in northwest Alabama, USA, we collected crayfish from 60 sites in the Sipsey Fork, Brushy Creek, and selected tributaries (Black Warrior River system). After finding two unexpected and possibly-introduced crayfish species, we expanded our investigation of crayfish distributions to include crayfish obtained from stomachs of black bass (Micropterus spp.) caught at seven sites in the reservoir. To explore what crayfish species were in the drainage historically, we examined museum databases as well as stomach and intestinal contents of a variety of preserved fishes that were caught in the Sipsey Fork and Brushy Creek drainages upstream of the reservoir in the early 1990’s. Of the seven crayfish species collected, one, Orconectes (Procericambarus) sp. nr ronaldi, was not previously reported from Alabama, and another, O. lancifer, was not reported from the Bl...
Catadromous eels are found in more habitats than any other fish and are capable of inhabiting mar... more Catadromous eels are found in more habitats than any other fish and are capable of inhabiting marine, brackish and freshwater environments. In this study we used the American Eel (Anguilla rostrata) as a bioindicator organism to create a novel method of using spatial analysis to study species conservation over landscape scales. We built a model of the subwatersheds of the Chesapeake Bay using a Digital Elevation Model (DEM) and overlaid eel density data (> 1 million eels sampled), dam density data and land use in ArcGIS. Dam construction in the study area peaked between 1955 and 1975, possibly as a result of flood control measures. Effects of land use were localized and most pronounced in areas around Baltimore, Maryland, Washington, D.C. and Richmond, Virginia, USA. Results indicate the Potomac and Rappahannock rivers appear to be areas of lesser concern while the upper James and York rivers are ideal for follow-up studies, since these area rank poorly in both eel density and ba...
The U.S. Forest Service and other federal land managers are responsible for maintaining the produ... more The U.S. Forest Service and other federal land managers are responsible for maintaining the productivity of aquatic-riparian ecosystems, the associated native biota, and the ecosystem services they provide. These public lands are important sources of water, recreation opportunities, and habitat for a suite of animals and plants, including many that are protected under the Endangered Species Act. To meet these challenges and responsibilities, recent science suggests modifying practices to provide a broader array of habitat, biological conditions, and ecosystem functions than are associated with traditional management approaches. We suggest that by linking approaches based on natural disturbance and portfolio concepts, managers can achieve a robust strategy and desired outcomes more reliably and cost effectively. Locally complex habitat conditions created by natural disturbances provide the template for biological diversity to play out if provided enough time. Accordingly, natural disturbance regimes play an important role in creating and sustaining habitat and biological complexities on the landscape, suggesting that, to the extent possible, management actions should emulate natural disturbance processes at appropriate spatial and temporal scales. In concert with this approach, the portfolio effect (i.e., diversity that mitigates risk) provides justification for promoting connected heterogeneous habitats that reduce the risk of synchronous large-scale population and ecosystem collapse. In this article, we describe how disturbance and portfolio concepts fit into a broader strategy of conserving ecosystem integrity and dynamism and provide examples of how these concepts can be used to address a wide range of management concerns. Ultimately, the outcome for populations, habitats, and landscapes depends on how well environmental change is understood, the degree to which change is appropriately addressed by natural resource managers, and solutions that allow populations and ecosystems to persist in the presence of and be resilient to a growing scope of human influences.
Canadian Journal of Fisheries and Aquatic Sciences, 2015
Unregulated river systems are organized in a hierarchy in which large-scale factors (i.e., landsc... more Unregulated river systems are organized in a hierarchy in which large-scale factors (i.e., landscape and segment scales) influence local habitats (i.e., reach, meso-, and microhabitat scales), and both differentially exert selective pressures on biota. Dams, however, create discontinua in these processes and change the hierarchical structure. We examined the relative roles of hydrology and other instream factors, within a hierarchical landscape context, in organizing fish communities in regulated and unregulated tributaries to the Upper Tennessee River, USA. We used multivariate regression trees to identify factors that partition fish assemblages based on trait similarities, irrespective of spatial scale. We then used classical path analysis and structural equation modeling to evaluate the most plausible hierarchical causal structure of specific trait-based community components, given the data. Both statistical approaches suggested that river regulation affects stream fishes through...
Regional frameworks have been used extensively in recent years to aid in broad-scale management. ... more Regional frameworks have been used extensively in recent years to aid in broad-scale management. Widely used landscape-based regional frameworks, such as hydrologic landscape regions (HLRs) and physiographic provinces, may provide predictive tools of hydrologic variability. However, hydrologic-based regional frameworks, created using only streamflow data, are also available and have been created at various scales; thus, relating frameworks that share a common purpose can be informative. In addition, identifying how the relative importance of variables change in governing streamflow with respect to scale can also be informative. The purpose of this study was to determine whether landscape-based frameworks could explain variation in streamflow classifications and in the hydrologic variables used in their creation. We also evaluated how climate and watershed-based variables govern the divergence of different flow classifications at two different scales. HLRs and physiographic provinces poorly predicted flow class affiliation within our study and for the entire USA, although physiographic provinces explained more variability. We also found that HLRs explained very little variation in individual hydrologic parameters. Using variables summarized at the watershed scale, we found that climate will play a larger role in influencing hydrology across the entire US, whereas soils may govern variation in hydrology at smaller scales. Our results suggest that predictor variables, developed at the watershed scale, may be the most appropriate at explaining hydrology and that the variables used in creating regional landscape-based frameworks may be more useful than the frameworks themselves. In addition, managers should be careful when using landscape-based regional classifications for stream management because the scale of their construction may be too broad to capture differences in flow variability.
North American Journal of Fisheries Management, 2010
We assessed the efficacy of passive gravel addition at forming catostomid spawning habitat under ... more We assessed the efficacy of passive gravel addition at forming catostomid spawning habitat under various flow regimes in the Cheoah River, a high-gradient tailwater river in North Carolina. The purpose was to provide a case study that included recommendations for future applications. A total of 76.3 m 3 (162 tons) of washed gravel (10-50 mm) was passively dumped down the streambank and into the channel in four locations. Gravel sites differed in terms of average reach slope, bank slope, and the initial volume of gravel added, which could have influenced gravel entrainment. Maps of gravel movement under various flows suggested that large-magnitude discharges (113 m 3 /s) caused extensive migration; however, less obvious, smaller discharges (;28 m 3 /s) still caused substantial shifting, which may influence the stability of catostomid spawning substrates. Following gravel addition, the proportion of gravel in the streambed was significantly higher at all gravel sites. However, comparisons of sites to reference stream reaches suggested that sand, gravel, and cobble were still extremely deficient. Additionally, the volume of gravel was inadequate to create gravel depths that provided suitable habitat for catostomid spawning. Although periodic, passive gravel additions may take years to provide suitable spawning habitat for some fish species, we found that river chub Nocomis micropogon utilized the newly added gravel for spawning.
Journal of the North American Benthological Society, 2011
Consumer nutrient cycling supplies limiting elements to autotrophic and heterotrophic organisms i... more Consumer nutrient cycling supplies limiting elements to autotrophic and heterotrophic organisms in aquatic systems. However, the role of consumers in supplying nutrients may change depending on their diet and their own stoichiometry. We evaluated the stoichiometry, N and P excretion, and diets of the dominant macroinvertebrates and fish at 6 stream sites to determine if the nutritional composition of food alters nutrient excretion. We used Sterner's (1990) nutrient homeostasis model as a reference to gauge whether consumer nutrient excretion is influenced by diet. Body stoichiometry explained 61% of the variation in N:P excretion by macroinvertebrates but only 11% of the variation for fish. In both cases, the relationship was driven by 2 P-rich end-members, crayfish and mottled sculpin. Results of Akaike Information Criterion (AIC) analysis showed that family alone explained 71% of the variation in N:P excretion in macroinvertebrates and 31% of the variation in fish. Diet explained only 8% of the variation in both cases. Most consumers (9 of 11) had N:P excretion values that were well below predictions of Sterner's model. Two taxa, crayfish and sculpin, had N:P excretion that overlapped the model's predictions. Our results suggest that crayfish and sculpin may display strict homeostasis with respect to N and P and that their growth might be P-limited. Other consumers may be more flexible in their stoichiometry and not P-limited. We speculate that the extremely low excretion N:P measured for many consumers might have been the result of semiflexible homeostasis, inaccuracies in our assessment of dietary nutrients, growth-limiting nutrients other than N or P, or lack of egestion data. Our results suggest that crayfish and sculpin may alter N and P dynamics in streams by excreting low amounts of P relative to N compared to what is generally available in the water column.
For some time, ecologists have attempted to make generalizations concerning how disturbances infl... more For some time, ecologists have attempted to make generalizations concerning how disturbances influence natural ecosystems, especially river systems. The existing literature suggests that dams homogenize the hydrologic variability of rivers. However, this might insinuate that dams affect river systems similarly despite a large gradient in natural hydrologic character. In order to evaluate patterns in dam-regulated hydrology and associated ecological relationships, a broad framework is needed. Flow classes, or groups of streams that share similar hydrology, may provide a framework to evaluate the relative effects of dam regulation on natural flow dynamics. The purpose of this study was to use a regional flow classification as the foundation for evaluating patterns of hydrologic alteration due to dams and to determine if the response of rivers to regulation was specific to different flow classes. We used the US Geological Survey (USGS) database to access discharge information for 284 unregulated and 117 regulated gage records. For each record, we calculated 44 hydrologic statistics, including the Indicators of Hydrologic Alteration. We used a sub-regional flow classification for eight states as a way to stratify unregulated and regulated streams into comparable units. In general, our results showed that dam regulation generally had stronger effects on hydrologic indices than other disturbances when models were stratified by flow class; however, the effects of urbanization, withdrawals, and fragmentation, at times, were comparable or exceeded the effects of dam regulation. In agreement with the existing literature, maximum flows, flow variability, and rise rates were lower whereas minimum flows and reversals were higher in dam regulated streams. However, the response of monthly and seasonal flows, flow predictability, and baseflows were variable depending on flow class membership. Principal components analysis showed that regulated streams occupied a larger multivariate space than unregulated streams, which suggests that dams may not homogenize all river systems, but may move them outside the bounds of normal river function. Ultimately, our results suggest that flow classes provide a suitable framework to generalize patterns in hydrologic alterations due to dam regulation.
In order for habitat restoration in regulated rivers to be effective at large scales, broadly app... more In order for habitat restoration in regulated rivers to be effective at large scales, broadly applicable frameworks are needed that provide measurable objectives and contexts for management. The Ecological Limits of Hydrologic Alteration (ELOHA) framework was created as a template to assess hydrologic alterations, develop relationships between altered streamflow and ecology, and establish environmental flow standards. We tested the utility of ELOHA in informing flow restoration applications for fish and riparian communities in regulated rivers in the Upper Tennessee River Basin (UTRB). We followed the steps of ELOHA to generate univariate relationships between altered flows and ecology within the UTRB. By comparison, we constructed multivariate models to determine improvements in predictive capacity with the addition of non-flow variables. We then determined whether those relationships could predict fish and riparian responses to flow restoration in the Cheoah River, a regulated system within the UTRB. Although ELOHA provided a robust template to construct hydrologic information and predict hydrology for ungaged locations, our results do not suggest that univariate relationships between flow and ecology (step 4, ELOHA process) can produce results sufficient to guide flow restoration in regulated rivers. After constructing multivariate models, we successfully developed predictive relationships between flow alterations and fish/ riparian responses. In accordance with model predictions, riparian encroachment displayed consistent decreases with increases in flow magnitude in the Cheoah River; however, fish richness did not increase as predicted 4 years after restoration. Our results suggest that altered temperature and substrate and the current disturbance regime may have reduced opportunities for fish species colonization. Our case study highlights the need for interdisciplinary science in defining environmental flows for regulated rivers and the need for adaptive management approaches once flows are restored.
They always took time from their incredibly busy schedules to provide advice and commiseration. E... more They always took time from their incredibly busy schedules to provide advice and commiseration. Eric Hallerman had a fresh perspective and encouraged my efforts to enlarge my graduate school experience outside of research. Andy Dolloff generated ideas and insight for data collection when I initiated this project. He also provided advice and information that helped me launch my career. Tammy Newcomb's expertise on instream flow studies and habitat modeling was invaluable. I appreciate her open-minded interest and advice concerning the methods I chose to use to complete this dissertation. Eric Smith provided himself and his students for invaluable statistical advice and was extremely patient with my lack of expertise. The U.S. Fish and Wildlife Service and the Virginia Department of Game and Inland Fisheries provided funding for this project. Personnel from these agencies showed consistent interest and accompanied me on field excursions to learn more about the Roanoke logperch. I am particularly grateful to Mike Pinder, who arranged funding for this project and remained supportive and helpful throughout my tenure at Virginia Tech.
ABS1RACT : Even though beavers (Castor canadensis carolinensis) occur over most of the southeaste... more ABS1RACT : Even though beavers (Castor canadensis carolinensis) occur over most of the southeastern United States , the impacts of increasing beaver populations on riparian forests within the southern Appalachian mountains are. not been well documented. Long-term browsing and inundation by beaver may alter the composition and structure of riparian forests. A survey of 62 streams (7 4 mi) within the Chauga River drainage in the mountains of South Carolina was conducted during 1991-1992 to determine the level of beaver activity within the drainage and the amount of timber damaged by beaver activiti es. Thirty-six streams had evidence of significant beaver activity with a total of 5.3 mi (7 .2%) affected by beaver. Twenty-six streams (17.3 mi), primarily those with steep gradients and no flood plains, had no evidence of beaver activity. On beaver impacted areas , values of beaver damaged timber averaged $781.27 /ac for sawtimber and $36.01/ac for pulpwood. While high in terms of volume/acre within impacted riparian areas, timber damage was relatively minor for the entire drainage because of the small area (49 .2 ac) affected. Beavers occupied the entire state of South Carolina prior to European settlement, but populations were eliminated from South Carolina by the early 1800s (Salyer 1946, Golley 1966). In 1940 and 1950, beaver were reintroduced into South Carolina in Chesterfield County in the upper coastal plain on the North Carolina border and in Aiken County in the lower Piedmont on the Georgia border, respectively. During the 1950s, beaver probably migrated into South Carolina from Georgia. By 1966, 12 of the state's 46 counties were inhabited by beaver (Golley 1966); 26 counties had established beaver colonies by 197 4 (Hair and Woodward 1974). Presently, beaver occur in all counties of South Carolina (J. Witt, South Carolina Wildlife Marine Resources Department, pers. comm.) .
univ-st-etienne.fr RÉSUMÉ. Les documents multimédia composés de texte et d'images sont de plus en... more univ-st-etienne.fr RÉSUMÉ. Les documents multimédia composés de texte et d'images sont de plus en plus présents grâce à Internet et à l'augmentation des capacités de stockage. Cet article présente un modèle de représentation de documents multimédia qui combine l'information textuelle et l'information visuelle. En utilisant une approche par sac de mot, un document composé de texte et d'image peut être décrit par des vecteurs correspondant à chaque type d'information. Pour une requête multimédia donnée, une liste de documents pertinents est retournée en combinant linéairement les résultats obtenus séparément sur chaque modalité. Le but de cet article est d'étudier l'impact, sur les résultats, du poids attribué à l'information visuelle par rapport à l'information textuelle. Des expérimentations, réalisées sur la collection multimédia ImageCLEF extraite de l'encyclopédie Wikipedia, montrent que les résultats peuvent être améliorés après une première étape d'apprentissage de ce poids. ABSTRACT. Multimedia documents are increasingly used which involve to develop model to that kind of data. In this paper we present a multimedia model which combines textual and visual information. Using a bag of words approach, we can represent a textual and visual document with a vector for each modality. Given a multimedia query, our model lets us linearly combine scores obtained for each modality and return a list of relevant retrieved documents. This article aims at studying the influence of the weight given to the visual information according to the textual one. Experiments on the multimedia ImageCLEF collection extracted from Wikipedia show that results can be improved by learning this weight parameter.
Uploads
Papers by Andy Dolloff