Size-Dependent Territoriality of Mottled Sculpin in a Southern Appalachian Stream

—We quantified the space use behaviors of juvenile and adult mottled sculpin Cottus bairdii over a 3-year period in Shope Fork, western North Carolina. Our objectives were to (1) quantify home range size, (2) determine whether the fish exhibit territorial behaviors, (3) characterize the relative stability of territories, and (4) relate temporal variation in behaviors to environmental variability and population size structure. Adult behaviors were consistent with those of a strongly territorial organism. Adults exhibited nonrandom movements, restricted home ranges, and extremely low levels of spatial overlap with neighboring residents (<10% overlap). Territories were established in erosional microhabitats that were significantly more stable (as measured by seasonal shifts in dominant substrate composition) than randomly selected microhabitats in the study site. In contrast to adults, juveniles did not exhibit evidence of territoriality and instead occupied overlapping home ranges (16-36% overlap) in less-stable, depositional microhabitats along the stream margin. Mottled sculpin home range size, home range overlap, and territory abandonment rate were related to the density of large adults rather than flow variability or microhabitat stability. Adult territoriality and juvenile floating provide behavioral mechanisms capable of producing strong regulation of mottled sculpin populations in this system. Nonbreeding territoriality is common in marine fishes and appears to be less common in freshwater species (Barlow 1993). In marine fishes, territorial individuals defend areas of high-quality food resources, predator refuges, and nesting sites (Roberts and Ormond 1992; Pitcher 1993; Barlow 1993; Goncalves and Almada 1998; Letourneur 2000). This behavior is common to only one family of freshwater fishes, the Salmonidae (Grant and Noakes 1987; Grant 1990; Hughes 1992; Nakano 1995; Steingrimsson and Grant 1999). Competitively dominant salmonids defend highquality feeding territories and subdominant individuals are forced to use lower quality microhabitats. Territoriality in salmonids has been shown to influence population dynamics (Grant and Kramer 1990; Elliott 1994), interspecific interactions (Fausch and White 1981), and watershed-scale distribution patterns (Hughes and Reynolds 1994; Hughes 1998). Whereas territoriality has been extensively documented in salmonids, we have found no prior studies of territoriality in a North American benthic stream fish. The existence of foraging territories is determined in large part by the distribution of food resources and the stability of that distribution (Lott 1991; Maher and Lott 2000). Territoriality is favored by patchily distributed * Corresponding author: [email protected] 1 Present address: Division of Forestry and Natural Resources, West Virginia University, Morgantown, West Virginia 26506, USA. Received February 7, 2006; accepted October 2, 2006 Published online December 10, 2007 food resources and may function to ensure that a dominant individual maintains exclusive access to high-quality patches. Numerous researchers have found that territoriality will persist as long as stable, highquality food patches are present (Stamps and Krishnan 1995; Davies and Hartley 1996). By contrast, territoriality should not increase individual fitness if food resources are temporally variable (Lott 1991). In fact, Barlow (1993) suggested that territoriality should be rare in stream fishes because of highly variable resources in these habitats, although he also concluded that it would be logistically difficult to observe territorial behaviors in stream fishes. Many streams display high levels of variation in resource availability (Hildrew and Giller 1994; Allan 1995; Fausch et al. 2002). For example, benthic macroinvertebrate densities and physical habitat characteristics are patchily distributed in streams across a wide range of spatial scales (Frissell et al. 1986; Downes et al. 1993; Grossman et al. 1995a). In addition, streamflow and microhabitat stability may vary daily, seasonally, and over decades (Poff and Ward 1989; Grossman et al. 1995a, 2006). Our own studies in the Coweeta Creek drainage, western North Carolina, demonstrate the combined importance of spatial variability of food resources and temporal variability in flows on stream fish foraging behavior, population dynamics, and community structure (Grossman et al. 1995a, 1998, 2006; Thompson et al. 2001; Petty and Grossman 2004). Spatial and temporal variability of resources place