Effects of Habitat Area and Complexity on Colorado River Cutthroat Trout Density in Uinta Mountain Streams

—Habitat degradation has reduced the complexity and connectivity of streams on the north slope of the Uinta Mountains in northeastern Utah. These changes have diminished the historical range of Colorado River cutthroat trout Oncorhynchus clarki pleuriticus, isolated the populations of this subspecies, and perhaps increased its risk of extinction. We assessed the effects of fragment area and habitat complexity on Colorado River cutthroat trout density. We studied 88 reaches in 4 isolated stream fragments. At the fragment scale, both the density of adults and habitat complexity increased significantly as fragment size increased. In the smaller fragments, the density of adults was lower while that of juveniles was higher. Habitat differed substantially among fragments. At the reach scale, the density of adults was positively related to elevation, the percentage of undercut banks, and mean substrate particle size and negatively related to residual pool depth and the extent of large woody debris. The density of juveniles was positively related to the extent of large woody debris and negatively related to residual pool depth and stream width. The habitat complexity index was weakly related to adult density at the reach scale. We were not able to distinguish the influence of habitat area or complexity on the density of adults, but a population living in an isolated stream fragment with low habitat complexity probably requires more area to persist than a population of the same size living in a highly complex habitat. Historically, Colorado River cutthroat trout Oncorhynchus clarki pleuriticus were widely distributed among the coldwater tributaries of the Green and upper Colorado rivers (Behnke and Benson 1980). Currently, these trout occupy less than 1% of their historical range (Behnke 1992). Federal and state agencies have responded to the species’ decline by granting it special status. Colorado River cutthroat trout were classified as a Category 2 species (i.e., a candidate species for listing under the U.S. Endangered Species Act) by the U.S. Fish and Wildlife Service until that classification code was abolished in 1996. The U.S. Forest Service (in Regions 2 and 4) and the states of Colorado, Utah, and Wyoming have granted Colorado River cutthroat trout special management status. A petition has also been filed to list the species under the U.S. Endangered Species Act. Habitat degradation has contributed to the decline of many salmonids throughout the western * Corresponding author: [email protected] Received March 26, 1999; accepted May 20, 2000 United States (Williams et al. 1989; Frissell 1993; Young 1995; Lee et al. 1997). Degradation can include habitat fragmentation and loss of habitat complexity. Habitat fragmentation results when a large area of habitat is subdivided into smaller, isolated patches (Wilcove et al. 1986) and is perhaps the most important problem threatening the survival of many species (Wilcox and Murphy 1985). Logging, mining, overgrazing, dams, and irrigation diversions have degraded aquatic habitats, eliminated migratory corridors, and isolated fish populations on the north slope of the Uinta Mountains in northeastern Utah (Kershner et al. 1997). Loss of migratory corridors and reduced habitat area may lead to local extinctions of cutthroat trout (Dunham et al. 1997). Habitat quality is also important to populations of aquatic animals. Complexity is one aspect of habitat quality that is thought to influence the size, structure, distribution, and stability of populations. Complexity has been described in terms of structural components (McMahon and Hartman 1989), hydraulic variation (Lamberti et al. 1989; Pearsons 1251 COLORADO RIVER CUTTHROAT TROUT DENSITY et al. 1992), and the diversity of depth, velocity, and substrate (Gorman and Karr 1978; Angermeier and Schlosser 1989). Cutthroat trout abundance has been positively associated with habitat complexity (Fausch and Northcote 1992). Complex habitats may be necessary to meet all life history needs of cutthroat trout as well as to provide refugia during extreme environmental events (Connell and Sousa 1983; Poff and Ward 1990; Sedell et al. 1990). Moreover, complex habitats may lessen predation risk by reducing predator efficiency (Crowder and Cooper 1982). Human activities have contributed to the loss of habitat complexity in Uinta Mountain streams. In the early 1900s, some drainages were logged, and succeeding floods often scoured stream channels to bedrock. Some of these streams still lack deep pools and late successional riparian vegetation, and streambeds are dominated by large, armored particles. The north slope of the Uinta Mountains has an extensive road network that is used for recreation and resource extraction. Poorly designed roads can affect aquatic biota by increasing sediment delivery to the stream, and improperly placed culverts can interfere with fish migration (Furniss et al. 1991). Many stream reaches are grazed by sheep and cattle. Stream reaches in the High Uintas Wilderness, however, have been less degraded than the downstream reaches, although this wilderness supports recreation and livestock grazing (Kershner et al. 1997). Management activities on the north slope of the Uinta Mountains have created a range of conditions within and among streams. Our aim was to study the role of habitat complexity and habitat area on Colorado River cutthroat trout (CRCT) populations in secondand third-order streams. We focused our analysis at two spatial scales, stream fragment and stream reach. Multiple-scale studies are important to better understand fish–habitat relations, as different spatial scales have unique physical and biological roles in the stream system (Frissell et al. 1986). Analyses that address both the stream and reach scales may also identify whether each scale affects a population independently or interactively (Dunham and Vinyard 1997). A stream fragment was defined as all contiguous streams and tributaries in a watershed that are physically isolated from other streams. Stream fragments vary geomorphically and vegetatively from the headwaters to downstream areas, thereby influencing the physical and chemical components of instream habitats. Habitat diversity, which typically varies over the entire length of a stream, can affect the abundance, structure, and distribution of a population. Studies at the fragment scale are often a useful means of assessing basinwide changes. Our first objective was to examine the relation of fragment area to the size and density of juvenile and adult CRCT. Many studies have looked at the effect of habitat area on animal density (Bender et al. 1998), but few have done so for fish (Lanka et al. 1987; Kozel and Hubert 1989). We also wished to test the merit of an index of habitat complexity in predicting adult density, describe habitat differences among study fragments, and evaluate fish–habitat relations. We defined stream reaches within fragments as lengths of stream having homogeneous morphological features that we evaluated qualitatively. Analyses done at the reach scale are useful for describing population characteristics (e.g., juvenile and adult density) as well as the physical variation within the reach, which can affect the temporal and spatial condition of channel form and fish habitat. At the reach scale, our objectives were to determine the most important variables for juvenile and adult CRCT and to test the index of habitat complexity against adult density.