Abundance, size, and feeding success of larval shortnose suckers and Lost River suckers from different habitats of the littoral zone of Upper Klamath Lake Michael S. Cooperman & Douglas F. Markle Department of Fisheries and Wildlife, Oregon State University, 104 Nash Hall, Corvallis, OR 97331-3803, U.S.A. (e-mail: michael.cooperman@orst.edu) Received 2 July 2003 Accepted 2 March 2004 Key words: Catostomidae, Deltistes luxatus, Chasmistes brevirostris, early life, macrophyte Synopsis We examined near-shore habitat use by larval shortnose and Lost River suckers in the lower Wil- liamson River and Upper Klamath Lake of south-central Oregon. Emergent macrophytes (Scirpus, Sparganium and Polygonum) supported significantly more, larger, and better-fed larvae than submergent macrophytes, woody vegetation, or open water. Abundance, size, and gut fullness were similar for sucker larvae collected from different emergent macropytes. During the larval period, there was no evidence of density dependant effects or habitat shifts. Ranked catch per unit effort data indicated potential predators also were more likely to use emergent macrophytes, but ordination indicated larvae and potential predators were differentially distributed along a vegetation structure-water depth gradient with larvae in shallow vegetated areas. Between-habitat differences appeared to be due to larval sucker selection for, or better survival in, emergent macrophytes, rather than differential access or exclusion from other habitats. The importance of emergent macrophytes appears to be related to increased foraging success and reduced predation. Because larvae in emergent macrophytes have a size and gut fullness advantage, the amount of emergent habitat could affect early survival. However, interannual differences in recruitment to the adult population may or may not be dependent on larval dynamics. Our results suggest larval sucker access to emergent macrophytes may be necessary, but perhaps not su���cient, for promoting good year class formation. Introduction Habitat selection by young-of-year (YOY) fishes is a balancing act between maximizing growth rate and minimizing predation risk (Werner & Gilliam 1984, Werner 2002). In many ecosys- tems, aquatic vegetation provides YOY fish with abundant prey items and increased protection from predators (Heck & Crowder 1991, Dibble et al. 1996, Weaver et al. 1997), and many YOY fishes preferentially utilize vegetated habitats relative to other available habitats (Heck & Crowder 1991, Dibble et al. 1996). However, physical processes, physiological constraints, or biotic interactions may preclude YOY of some species from using vegetated areas even if they are a preferred habitat (Sogard 1994, Sinclair 1988, Jackson et al. 2001). Despite strong causal connections between YOY habitat use, growth, and survival, and between YOY survival and adult population dynamics, there are few de- tailed studies of YOY fish habitat utilization patterns and the potential fitness consequence at the scale of the immediate habitat surrounding the individual fish (the contingent habitat, sensu Werner 2002) (Dibble et al. 1996). Shortnose suckers and Lost River suckers are long-lived, highly fecund endemics (Perkins et al. Environmental Biology of Fishes 71: 365���377, 2004. �� 2004 Kluwer Academic Publishers. Printed in the Netherlands.

2000) of large shallow lakes in the upper Klamath River watershed (Figure 1). Both spe- cies were listed as endangered under the Federal Endangered Species Act in 1988, and the largest remaining populations of both species are in Upper Klamath Lake (UKL). UKL���s suckers spawn in the spring, and the majority run up the Williamson River. Preflexion protolarvae (10��� 12 mm standard length SL) make a rapid out- migration from the Williamson River and enter UKL (Cooperman & Markle 2003). The larval period ends approximately 45���60 days post- hatch, at about 22 mm SL, commensurate with absorption of the anal fin-fold and development of the full complement of fin rays (Cooperman, unpublished data). There is limited information on ontogenetic habitat change in larval suckers, mechanisms responsible for larval sucker habitat selection, or how fluctuations in lake surface elevation may affect habitat dynamics. In this paper, we describe larval shortnose and Lost River sucker habitat utilization in the littoral of the lower Williamson River and UKL. We ad- dress: (1) Are there differences in abundance, size, and/or gut fullness of larval suckers between dif- ferent habitat types (2) Do larval suckers shift their niche during ontogeny (3) What other fishes comprise the larval sucker fish assemblage, and (4) What factors affect larval sucker distributions. We consider potential mechanisms structuring the patterns and comment on endangered species management implications. Study location Our study occurred in the northeastern portion of UKL in the lower four km of the Williamson River Delta plus near-shore portions of UKL between Agency Straits and Goose (Figure 1). The delta ecosystem is the primary early nursery area for larval shortnose and Lost River suckers spawned in the Williamson River. Three emer- gent aquatic macrophytes are common in the delta: Scirpus acutus (hardstem bulrush), Polyg- onum coccineum (smart weed), and Sparganium eurycarpum (river burr reed) (Dunsmoor 2000). One submergent macrophyte, Potomogeton sp. (Dunsmoor 2000), was present but rare during our sampling. For simplicity, reference to these plants will be by genus name. An accounting of extensive habitat modifications and hydrologic manipulations of the delta ecosystem related to the Klamath Irrigation Project are provided in Cooperman & Markle (2003) and Markle & Cooperman (2003). Material and methods Sample gear, sample sites, and sample protocol Sample equipment and methodologies, sample sites, effort, and protocols were described in Cooperman & Markle (2003). Only daytime sam- ples collected between 08:00 and 16:30 hours were used in analyses herein. Salient points are reviewed below. Figure 1. Sample site locations. Site 1 is 0.5 km below Sprague River Dam. Sucker spawning occurs between Sites 1 and 2. Inset shows Klamath River watershed, south-central Oregon and Northern California. Map provided by Rip Shively, USGS, Klamath Falls, Oregon. 366