1 Kansas Cooperative Fish and Wildlife Research Unit, Division of Biology, Kansas State University, Manhattan, Kansas
2 Division of Biology, Kansas State University, Manhattan, Kansas

Introduction

The Lower Colorado River Basin (LCRB; Fig. 1) is a highly altered system with regards to flow and temperature modifications, irrigation, land use, and non-native fish invasions (Mueller 2005) and the lower mainstem has “…the dubious distinction of being among the few major rivers of the world with an entirely introduced fish fauna” (Mueller and Marsh 2002). LCRB has one of the most unique but imperiled fish fauna in the nation (Mueller and Marsh 2002; Olden and Poff 2005). Of 31 native fish species, 75% are endemic, and 25 are extinct, extirpated, or listed as endangered or threatened under the Endangered Species Act (USFWS 1999). Although researchers have suggested full recovery of native fish communities in LCRB is not feasible for political, societal, and economic reasons (Minkley and Deacon 1991; Mueller 2005), the development of conservation areas will aid in future considerations to protect aquatic species. Previous Aquatic GAP efforts for inland streams have been in relatively mesic areas of the United States. The Lower Colorado River Basin Aquatic GAP (LCRGAP) will provide an opportunity to compare and contrast the role of environmental and biological variables in predicting fish distributions in the arid Southwest and more mesic regions in the Midwest and East.

Phase one of the LCRGAP was initiated in 2004 as a one-year feasibility study to gather existing datasets, and to evaluate stakeholder interest in participating in development and using Aquatic LCRGAP products. We are now in the second phase to develop species distributions and predictive models for the Verde River watershed within the LCRB (Fig. 2). Using methods refined in phase two, the final phase is to expand the process to the entire LCRB and define conservation areas for the basin based on native biodiversity and threats to the system with the intent to combine aquatic models with terrestrial models produced through the Southwest ReGAP. We report here on the methods that will be used for this project and our progress to date.

Conservation areas for the LCRB will be based on three primary factors: predicted distributions of native versus non-native fishes, ecosystem traits (e.g., stream hierarchy, land cover type, stream density, etc.), and threats. Each fish collection site is being attributed with habitat parameters in a geographical information system (GIS). Habitat parameters were selected based on relationships documented between specific variables and fish communities, such as elevation, distance to and presence of barriers, and land use practices (Mandrak 1995; Wang et al. 1997; Matthews 1998; Marchetti and Moyle 2001; Lamouroux et al. 2002; Zorn et al. 2002).

A hierarchical classification framework to conserve biodiversity will be created for LCRB following guidelines by Sowa et al. (2004) and Higgins et al. (2005). The upper levels will be derived from existing sources such as Maxwell et al. (1995) and Abell et al. (2000). The Ecological Drainage Units (EDU) and Aquatic Ecological Systems (AES) will be derived using biotic and abiotic data (e.g., geology, gradient). The number of EDU and AES will be determined through multivariate techniques such as clustering, principal components analysis, and nonmetric multidimensional scaling (Sowa et al. 2004). Expert reviewers will evaluate our hierarchical classification framework.

An anthropogenic threats classification will be created at the AES level and will include variables such as road density, dam locations, impaired streams, and/or other data deemed threats to aquatic resources. Once all metrics are identified, correlation analysis will be used to determine redundant metrics that can be eliminated from future analysis (Sowa et al. 2004). The remaining metrics will be combined to create one human stressor index for each AES.

After the hierarchal stream classification system and threats classification are developed and approved, conservation areas within each EDU will be determined. These conservation areas will be the EDU (and possibly AES) that have a high need (or most potential) of conservation. Priority areas will be selected to protect native biodiversity and underrepresented species/communities using factors such as native species richness, highest predicted target species richness (e.g., state or federally listed species), limited or no presence of non-native fishes, low human stresses, high proportion of public land, and overlap of existing conservation initiatives (Sowa et al. 2004). Resource professionals within the region will review our analyses and selected priority area. This methodology will incorporate a broad array of information (threats, land use, species and habitat data, expert opinion) in the decision-making process for selecting conservation areas (Wilson et al. 2005).

Progress to Date

Species Data

Fish location data have been gathered from several state and federal agencies, universities, online fish databases, and museums. Fish records with complete collection information (point location, species name, site description, and date collected) are being checked for accuracy and then entered into a database. Currently, we have over 1,500,000 individual records in the database encompassing 160 species. Distribution of records between native and non-native species is nearly even. Although data range from early 1900’s to present, about half the records were obtained from 1980–present.

Habitat and Supplementary Data

Numerous habitat data have been collected based on documented relationships between fish occurrence and habitat variables. We are in the process of deriving specific parameters (e.g., number of dams per watershed, percent land upstream that is used for agricultural purposes, etc.) for the basin. These are being linked to a stream layer and associated with species occurrence records.

The Verde Basin (Fig. 2) was selected for a pilot study because it is one of the few remaining perennial rivers within the LCRB (Averitt et al. 1994) and may serve as refugia for native fishes. The upper 60% of the Verde River is unimpounded and located within National Forest Service lands although scattered allotments of private lands occur throughout the reach. Approximately 65 km of the unimpounded reach have been designated as a Wild and Scenic River (US Forest Service 2004). In addition, the Verde River has been identified by Arizona Game and Fish Department as a focus area for fisheries planning in the future (Young et al. 2001; Larry Riley, Arizona Fisheries Branch Chief, personal communication).

Figure 2. Shading signifies the Verde Basin within the Lower Colorado River Basin

The Verde Basin currently contains 13 native fishes typically found outside the mainstem of the Lower Colorado. Of the native species > 50% are listed as of concern, endangered, or threatened by the State of Arizona or US Fish and Wildlife Service. Threats to the system include 41 dams, numerous stock ponds, >15 non-native fishes (outside the reservoirs), and surface and groundwater diversions for public supply, agriculture, and livestock uses.

Public Outreach

At the request of several stakeholders (e.g., Arizona Game and Fish Department, US Fish and Wildlife Service, Environmental Protection Agency), we developed an online literature database for products related to the LCRB that is searchable by author, title, year, and keywords. This database currently contains nearly 4,000 records of which approximately 1,300 are available to download. The database is accessible through the LCRGAP website (http://www.lcrgap.org/search.htm).

Web Page Development

A web page was developed to disseminate project updates and products (www.lcrgap.org). When the website went online, 135 cooperators and interested parties were notified of the web address. The webpage is our primary source of communication with our stakeholders (e.g., regional State and Federal government branches; university researchers, local interest groups).

Future Tasks and Challenges

The ultimate goal is to predict the distributions of fishes throughout the basin to develop conservation areas. We continue to build the species database through acquisition and correction of additional datasets and update those already acquired. We are coordinating with several organizations to obtain additional fish occurrence records. The entire stream network data are being corrected and attributed to use in predictive species modeling. A challenge will be to identify ephemeral drainages to minimize the overestimation of species distributions. Additional documents are being collected from online websites and through contacts with the cooperating agencies and organizations for inclusion in the literature database.

In the short term, our focus is on the Verde Basin. We posted an internet map server to display species records compiled to date so reviewers could examine these data and provide feedback. Attribution of environmental and anthropogenic data will be completed soon so species models can be developed by the end of summer 2006. These also will be posted online through our website.

Potential Outcomes

Products from this project may prove useful for a variety of conservation objectives. Examples from other Aquatic GAPs include using habitat models to locate an existing population of fish that was presumed extirpated, and utilizing distribution models and threat indices in State Comprehensive Wildlife Conservation Strategy (CWCS) plans. Although the CWCS plans have been completed for the states encompassing the Lower Colorado River, GAP products can be used to address specified information needs in those CWCS plans such as:

1. Develop habitat use models for species of conservation priority;
2. Investigate characteristics that drive whether a species will be invasive, an important keystone species, or more sensitive to stressors; and
3. Determine threats to vulnerable species.

Other organizations (Arizona Game and Fish Department, and two multi-agency species planning efforts) intend to utilize products from the LCRGAP to develop species-specific and regional conservation plans. Additional uses identified through discussions with regional management organizations have included identifying areas with appropriate habitat conditions for use as nursery areas, prioritizing research efforts, identifying potential re-introduction sites, and focusing sampling efforts to areas of high probability of occurrence of a target species or communities.

Literature Cited

Abell, R.A., P.T. Hurley, D.M. Olson, and E. Dinerstein, 2000. Freshwater ecoregions of North America: a conservation assessment. World Wildlife Fund – US, Island Press, Washington, D.C.

Averitt, E., F. Steiner, R.A. Yabes, and D. Patten. 1994. An assessment of the Verde River Corridor Project in Arizona. Landscape and Urban Planning 28: 161-178.

Higgins, J.V., M.T. Bryer, M.L. Khoury, and T.W. Fitzhugh. 2005. A freshwater classification approach for biodiversity conservation planning. Conservation Biology 19: 432-445.

Lamouroux, N., N.L. Poff, and P.L Angermeier. 2002. Intercontinental convergence of stream fish community traits along geomorphic and hydraulic gradients. Ecology53: 1792-1897.

Mandrak, N.E. 1995. Biogeographic patterns of fish species richness in Ontario lakes in relation to historical and environmental factors. Canadian Journal of Fisheries and Aquatic Sciences52: 1462-1474.

Marchetti, M.P., and P.B. Moyle. 2001. Effects of flow regime on fish assemblages in a regulated California stream. Ecological Applications11: 530-539.

Matthews, W. J. 1998. Patterns in Freshwater Fish Ecology. Chapman and Hall, New York, New York. 756 pages.

Maxwell, J.R., C.J. Edwards, M.E. Jensen, S.J. Paustian, H. Parrott, and D.M. Hill. 1995. A hierarchical framework of aquatic ecological units in North America (Nearctic Zone). US Forest Service, North Central Forest Experiment Station, St. Paul, Minnesota. General Technical Report NC-176.

McKenna, J.E., Jr. 2005. Application of neural networks to prediction of fish diversity and salmonid production in the Lake Ontario basin. Transactions of the American Fisheries Society 134: 28-43.

Minckley, W.L., and J.E. Deacon, editors. 1991. Battle against extinction, native fish management in the American West. The University of Arizona Press, Tucson.

Mueller, G.A. 2005. Predatory fish removal and native fish recovery in the Colorado River mainstem: what have we learned? Fisheries30: 10-17.

Mueller, G.A., and P.C. Marsh. 2002. Lost, a desert river and its native fishes: a historical perspective of the Lower Colorado River. U.S. Geological Survey USGS/BRD/ITR-2002-0010.

Oakes, R.M., K.B. Gido, J.A. Falke, J.D. Olden, and B.L. Brock. 2005. Modeling of stream fish assemblages in the Great Plains, U.S.A. Ecology of Freshwater Fish 14 : 361-374.

Olden, J.D., and D.A. Jackson. 2000. Torturing data for the sake of generality: How valid are our regression models? Ecoscience 7: 501–510.

Olden, J.D., and D.A. Jackson. 2002. A comparison of statistical approaches for modeling fish species distributions. Freshwater Biology 47: 1976-1995.

Olden, J.D., and N.L. Poff. 2005. Long-term trends of native and non-native fish faunas in the American Southwest. Animal Biodiversity and Conservation 28: 75-89.

Sowa, S.P., D.D. Diamond, R. Abbitt, G.M. Annis, T. Gordon, M.E. Morey, G.R. Sorensen, and D. True. 2004. The aquatic component of GAP analysis: a Missouri prototype. Final Report to United States Department of Defense Legacy Program, Project numbers 981713 and 991813.

Titus, K., J.A. Mosher, and B.K. Williams. 1984. Chance-corrected classification for use in discriminant analysis: Ecological applications. American Midland Naturalist 111: 1–7.

US Forest Service. 2004. Verde wild and scenic river: comprehensive river management plan. USDA, Southwestern Region.

US Fish and Wildlife Service (USFWS). 1999. Endangered and threatened wildlife and plants, 50 CFR 17.11 and 17.12. US Fish and Wildlife Service, Washington, D.C.

Wang, L., J. Lyons, P. Kanehl, and R. Gatti. 1997. Influences of watershed land use on habitat quality and biotic integrity in Wisconsin streams. Fisheries22: 6-12.

Wilson, K., R.L. Pressey, A. Newton, M. Burgman, H. Possingham, and C. Weston. 2005. Measuring and incorporating vulnerability into conservation planning. Environmental Management 35: 527-543.

Young, K.L., E.P. Lopez, and D.B. Dorum, editors. 2001. Integrated fisheries management plan for the Little Colorado River watershed. Nongame and Endangered Wildlife Program Technical Report 146. Arizona Game and Fish Department, Phoenix, Arizona.

Zorn, T.G., P.W. Seelbach, and M.J. Wiley. 2002. Distributions of stream fishes and their relationship to stream size and hydrology in Michigan’s lower peninsula. Transactions of the American Fisheries Society 131: 70-85.

Return to Table of Contents