Vol. 4 No 2, 103-117 2004

Assessment of anthropogenic influences on littoral-zone aquatic communities of Lake Texoma, Oklahoma-Texas, USA

Michael A. Eggleton1, Keith B. Gido2, William J. Matthews3, Gary D. Schnell4 1Sam Noble Oklahoma Museum of Natural History, University of Oklahoma, 2401 Chautauqua, Norman 73072, Oklahoma USA. Current address: Aquaculture/Fisheries Center, University of Arkansas-Pine Bluff, Box 4912, Pine Bluff 71601, Arkansas USA. e-mail: meggleton@ uaex.edu. 2Division of Biology, Kansas State University, Ackert Hall, Manhattan, Kansas 66506, USA. e-mail: [email protected]. 3Sam Noble Oklahoma Museum of Natural History, University of Oklahoma, 2401 Chautauqua, Norman 73072, Oklahoma USA/University of Oklahoma Biological Station, HC 71, Box 205, Kingston 73439, Oklahoma USA/Department of Zoology, University of Oklahoma, Norman 73019, Oklahoma USA. e-mail: [email protected]. 4Sam Noble Oklahoma Museum of Natural History, University of Oklahoma, 2401 Chautauqua, Norman 73072, Oklahoma USA/Department of Zoology, University of Oklahoma, Norman 73019, Oklahoma USA. e-mail: [email protected].

Abstract From 1999-2001, we evaluated the effects of anthropogenic activities in and around Lake Texoma, Oklahoma-Texas, USA, on the structure of littoral-zone fish and benthic invertebrate assemblages at 20 potentially impacted sites relative to paired reference sites. Spatial structuring of both assemblages was strongly related to variables associated with water clarity, water-column chlorophyll-a levels, and degree of site exposure to wind and waves. Fish assemblages at reference sites and impact sites exhibited minor differences, but none that were considered indicative of severe anthropogenic stress. Conversely, benthic invertebrates exhibited greater differences between reference and impact sites. Procrustean analyses and Mantel tests indicated little concordance between reference site and impact site benthic invertebrate assemblages. Greater abundances of oligochaetes at impact sites and greater abundances of chironomids at reference sites contributed most to these differences, with the largest assemblage differences found at sites influenced by agriculture and sanitary dumps. Despite the fact that fishes and benthic invertebrates were structured along similar environmental gradients, little concordance was observed between assemblages. Wide annual fluctuations in the dominant taxa of each assemblage contributed most to the general discordance. Furthermore, discordant fish and invertebrate assemblages likely resulted because responses of each assemblage to anthropogenic impacts occurred at different scales of space and time. Key words: reservoirs, fishes, benthic macroinvertebrates, multitaxon assessment, assemblage structure, environmental impact, Procrustean analysis, Mantel test

104

M. A. Eggleton et al.

1. Introduction Concern over the condition or "ecological health" of reservoirs has increased over the past quarter century with a recognition that chemical standards have not protected aquatic resources (Adler et al. 1993). However, when studying environmental impact or biotic integrity (sensu Angermeier, Karr 1984), reservoirs provide a more difficult challenge than streams, rivers, or lakes because they are artificial systems that lack natural reference sites (Hickman, McDonough 1996). Furthermore, the temporary nature of reservoirs on an evolutionary time scale precludes expectations of natural communities (Jennings et al. 1995). These attributes have led some researchers to suggest the term "biotic integrity" may be inappropriate for reservoirs, recommending instead that effort be devoted to developing multimetric indices of reservoir health (e.g., "reservoir fish assemblage index"; Hickman, McDonough 1996). Despite the challenges presented in assessing the ecological health of reservoirs, they comprise a large proportion of freshwater resources in the United States (Miranda 1996). Although most contemporary methods focus on community-level assessments, several recent studies (e.g., Jackson, Harvey 1993; Allen et al. 1999a,b; O'Connor et al. 2000) have shown advantages in conducting community-level assessments on multiple assemblage types within the same environment. Measures of concordance between assemblage types implicit with this approach provide added insight into community-level responses to anthropogenic activities in aquatic systems (Jackson, Harvey 1993). Few studies have examined environmental impact in this manner, and none have been conducted in reservoirs. Our objective was to assess effects of anthropogenic activities on aquatic assemblages in the littoral-zone of a large reservoir. We assessed effects at the community level, considering only shifts in conventional assemblage measures (species richness, organism abundance, and assemblage structure) in response to anthropogenic influences. Most studies in reservoirs have focused on off-shore fish assemblages consisting of large sport fish that can move readily, whereas we focused on littoral-zone fish assemblages because they are an important functional component of aquatic systems (Northcote 1988). Littoral fishes tend to be small-bodied and have little capacity to move great distances and so were regarded as more likely to reflect local environmental stresses. Assemblages examined were those most likely to show demonstrable effects resulting from anthropogenic stressors that impinge on reservoir ecosystems: (1) juvenile/adult fishes (hereafter referred to as fishes), which include the littoral-zone "watercolumn" community (e.g., juvenile temperate

basses, atherinids, shads, and minnows) and the "benthic nesting" community (mainly centrarchids); and (2) benthic invertebrates, which include substratedwelling macroinvertebrates such as chironomids, oligochaetes, burrowing mayflies, and molluscs. If anthropogenic stresses were prevalent in Lake Texoma, our general expectation was that strong differences would exist with fish and benthic invertebrate assemblages between reference and impact sites. Further, we expected that assemblages might respond to anthropogenic influences differently, with invertebrates responding to more local impacts (e.g., substrate differences associated with anthropogenic activities) and fishes responding to broader-scale influences (e.g., turbidity differences related to land use practices). In addition, because human activities alter multiple facets of the environment (Karr 1991), these assemblage types represent very different taxonomic and functional groups and were expected to provide a more complete view of how anthropogenic activities impinge on the Lake Texoma ecosystem.

2. Materials and methods Study area Lake Texoma, Oklahoma-Texas, USA is a 35 200 ha impoundment in the southern Great Plains region of the United States, located at the confluence of the Red and Washita rivers on the Oklahoma-Texas boarder. The reservoir was constructed in 1944 by the U.S. Army Corps of Engineers (USACE) for flood control and hydroelectric power production. The watershed of the reservoir encompasses about 103 000 km2; land use is predominantly agriculture, ranching, and forest with relatively low human population densities. Highly saline inflows from the Red River occur due to natural salt sources in the headwaters and tributaries. As a result, conductance values in Lake Texoma are high for freshwater (700 - 1200 µS cm-1; Gelwick, Matthews 1990) and distinctly different between river arms (Red River arm, usually >1000 µS cm-1; Washita River arm, usually