Armstrong et al.: Food habits of Leptocottus armatus 



457 



and in subtidal channels but within a few weeks can 

 be found only in areas that afford some refuge 4 

 (Dumbauld et al., 1993; Fernandez et al., 1993a). 

 First year survival, based on five years of trawl sur- 

 vey data in the estuary (Gunderson et al., 1990), has 

 been estimated at approximately 89c ( Wainwright et 

 al., 1992). Different assemblages of predators affect 

 survival of Dungeness crab during each phase of their 

 life history (Reilly, 1983; Stevens and Armstrong, 

 1985; Thomas, 1985). Estuarine and nearshore 

 fishes, wading birds, and older crab are known to 

 prey upon the young crab instars (Stevens et al., 1982; 

 Fernandez et al., 1993b), causing high mortality rates 

 in the summer months (Wainwright et al., 1992). 



Predation, including cannibalism, is considered to 

 be the prime cause of the rapid decline in crab abun- 

 dance through the summer. Stevens et al. (1982) 

 showed that during years of high 1+ crab abundance, 

 cannibalism can account for a significant portion of 

 0+ crab mortality. Even early settling 0+ crab are 

 capable of cannibalizing later settlers of the same 

 year class (Fernandez et al., 1993b). However, aside 

 from Reilly's study (1983) in California estuaries, 

 predation by estuarine fish and wading birds has 

 received little attention as a source of crab mortal- 

 ity. Although sea birds are known to take a toll on 

 megalopae 5 and on newly settled crab (Mace, 1983), 

 fish and crab predators are more likely to exert the 

 greatest predation pressure on the highly abundant, 

 small instars during settlement and early development. 



In this study, we discuss the estuarine summer 

 feeding patterns of staghorn sculpin with respect to 

 a broad spectrum of crustacean prey but focus on 

 their possible significance as predators of 0+ C. 

 magister and on the role of refuge habitat on the basis 

 of spatial patterns of predation in the estuary. A tem- 

 poral shift in consumption of certain prey taxa is also 

 discussed as indicative of opportunistic feeding by 

 staghorn sculpins with seasonal patterns of prey 

 abundance or with life history events which make 

 prey more susceptible to predators. 



Materials and methods 

 Sampling scheme 



Grays Harbor (46°55'N, 124°05'W) is a major Wash- 

 ington coastal estuary of about 8,545 hectares 



(Gunderson et al., 1988) marked by numerous 

 subtidal channels. These channels extend across ex- 

 tensive sandflats that become exposed and represent 

 679c of surface area during spring low tides. Refuge 

 habitat is provided by epibenthic shell exposed from 

 remnant Mya arenaria and Crassostrea gigas bivalve 

 populations and by eelgrass. Shell coverage has been 

 calculated to account for 199c of the total intertidal 

 area, 6 whereas eelgrass (Zostera marina and Z. noltii) 

 was reported to cover 42% of the tidal flat area. 7 



Staghorn sculpin were collected from Grays Har- 

 bor during six sampling trips from April through 

 August 1989. Sampling was timed to coincide with 

 the following periods: April-May prior to the main 

 pulse of annual crab settlement, June (two trips) 

 during recruitment of 0+ crab, and July— August dur- 

 ing the post-settlement summer growth period. Two 

 intertidal sandflat sites located about 10 km apart 

 were routinely trawled at high tide as a means of 

 contrasting diet in two common epibenthic habitats 

 of the estuary: an eelgrass (Zostera spp.) bed in North 

 Bay, and shell piles (Mya arenaria) in South Chan- 

 nel (Fig. 1). In order to ensure reasonable sample 

 sizes of fish for each trip and site (target of «>20), 

 two trawls each were conducted at both an intertidal 

 station (during slack flood tide) and at an adjacent 

 subtidal station (at low slack) within 6 h on the same 

 day, and fish were pooled for analyses of differences 

 in diet over time. Additional subtidal trawls were made 

 as time permitted for a total of 9— 11 trawls per trip. 



All trawl samples were collected with a 3-m beam 

 trawl (Gunderson and Ellis, 1986) deployed from a 

 7-m Boston Whaler. The net had an effective fishing 

 width of 2.3 m and a 4-mm codend liner to retain 

 juvenile crab and fish. Trawls in subtidal channels 

 were run for about four minutes at a speed of 3.7 to 

 5.6 km/h. Distance fished was determined by optical 

 range finder fixes between buoys deployed at the 

 beginning and end of each tow (see Gunderson et al., 

 1990, for details of trawl procedure); such distances 

 ranged from 200 to 350 m. Intertidal tows were made 

 between two staked points 160 m apart. Distance 

 fished and fishing width of net were used to estimate 

 area swept for calculation of catch per unit of effort 

 (CPUE; number per hectare). Trawl contents were 

 characterized as to type of vegetation (e.g. algae, 



4 Armstrong, D. A., L. Botsford, and G. S. Jamieson. 1990. Ecol- 

 ogy and population dynamics of juvenile Dungeness crab in 

 Grays Harbor estuary and adjacent nearshore waters of the 

 southern Washington coast. Rep. to U.S. Army Corps of Engi- 

 neers, Seattle District, Seattle, WA, 140 p. 



5 Armstrong, D. Personal observation at Whitcomb Flats sea gull 

 nesting site. Grays Harbor, WA, May 1988. 



6 Dumbauld, B. R., and D. A. Armstrong. 1987. Potential mitiga- 

 tion of juvenile Dungeness crab loss during dredging through 

 enhancement of intertidal shell habitat in Grays Harbor, Wash- 

 ington. Final Rep. FRI-UW-8714 to U.S. Army Corps of Engi- 

 neers, 64 p. 



7 Smith, J., L. D. R. Mudd, and L. W. Messmer. 1977. Impact of 

 dredging on the vegetation in Grays Harbor. Appendix F in 

 Maintenance of dredging and the environment of Grays Har- 

 bor, Washington. Final Rep. by U.S. Army Corps of Engineers, 

 Seattle District, 94 p. 



