466 



Fishery Bulletin 93(3). 1995 



'5 



a 

 a. 



i- 

 O 



J7 



J6 



J5 „ 



J4 



J3 



J2 



Jl 



60 



80 



100 



120 



140 



160 



180 



Sculpin total length (mm) 



Figure 8 



Relationship between body length (TL) of staghorn sculpin, Leptocottus armatus, 

 predators and carapace widths of Dungeness crab, Cancer magister, instars con- 

 sumed. Left axis shows mean sizes of instars J1-J7 (dotted lines) from Wainwright 

 and Armstrong ( 1993). 



megalopae settle and that Jl instars occur initially 

 on both open tideflats and in refuge materials (e.g. 

 shell, eelgrass) but are absent from the former within 

 several tidal cycles and are virtually never found on 

 open flats thereafter 9 (Dumbauld et al., 1993). De- 

 tailed excavation of shell patches at low tide reveal 

 post-settlement densities of Jl in excess of 100/m 2 , 

 but only a few per m 2 in eelgrass (Fernandez et al., 

 1993a), which suggests that trawl data collected at 

 high tide are likely a substantial underestimate of 

 0+ crab in shell compared with crab in eelgrass. 



These observations reflect a paradox indicated by 

 the data. Higher apparent crab consumption was 

 measured among sculpin collected from the shell 

 habitat (77% IRI) than from the eelgrass (5%) (Fig. 

 9), inconsistent with the notion that shell provides 

 critical refuge habitat for small crab instars. The 

 mean density of 0+ crab from eelgrass intertidal trawl 

 sites decreased by an order of magnitude from 2,455 

 crab/ha in early June to 280 crab/ha in mid-July (Fig. 

 10). During the same period, mean 0+ crab density 

 in shell habitat decreased from 9,452 crab/ha to 874 



crab/ha, reflecting migration into 

 the channels (Wainwright, 1994) 

 and the impact of predation in- 

 cluding cannibalism (1+ on 0+ 

 instars [Stevens et al., 1982]; 

 early 0+ on later 0+ [Fernandez 

 et al. 1993b] fish on 0+ crab 

 [Fernandez et al. 1993a]). There 

 appears to be a short time period 

 during peak crab settlement 

 when staghorn sculpin eat many 

 small instars, especially those 

 that settle on bare sand or mud- 

 flats. In this respect Dungeness 

 crab survival throughout much of 

 the bay is dependent upon the 

 availability of suitable refuge 

 habitat 210 (Fernandez et al., 

 1993a) as has been found for ju- 

 veniles of other decapod species 

 (Herrnkind and Butler, 1986; 

 Barshaw and Lavalli, 1988; 

 Howard, 1988; Warren, 1990). 



This pattern may be explained 

 by the short time scale of settle- 

 ment and rapid predation on 

 small instar crab (especially J1-J2). A possible ex- 

 planation for the difference in observed crab con- 

 sumption by sculpin between the two habitats is that 

 crab settlement in eelgrass or on open tideflats is 

 predated very rapidly and depleted from those areas 

 compared with crab settlement in the shell habitat. 

 Pulses of cohorts could be severely reduced in much 

 less time than the interval between sampling trips 

 and thereafter effectively be unavailable to sculpin 

 in certain areas of the estuary because of virtual re- 

 moval. Crab that recruit to areas of extensive shell 

 may provide a more stable and persistent prey basis 

 as the dynamics of small instars (agonistic interac- 

 tions and foraging) make them vulnerable on the 

 exterior of the shell matrix or as they move short 

 distances between shell piles. Predation by sculpin 

 on J1-J4 instars in shell habitats may occur over a 

 longer period, thereby increasing the likelihood that 

 sculpin in our samples contained crab later in the 

 summer, long after they were depleted from less pro- 

 tected areas of open mud and light cover of eelgrass. 

 This would be reflected both in greater numbers of 

 crab consumed and in greater frequency of occurrence 



Armstrong, D. A., K. A. McGraw, P. A. Dinnel, R. M. Thom, and 

 O. Iribarni. 1991. Construction dredging impacts on Dungeness 

 crab, Cancer magister, in Grays Harbor, Washington, and miti- 

 gation losses by development of intertidal shell habitat. Final 

 Rep. FRI-UW-9110 to U.S. Army Corps of Engineers, Seattle 

 District, Seattle, WA, 63 p. 



10 Doty, D., D. A. Armstrong, and B. R. Dumbauld. 1990. Com- 

 parison of carbaryl impacts on Dungeness crab (Cancer 

 magister) versus benefit of habitats derived from oyster cul- 

 ture in Willapa Bay, Washington. Univ. Washington, Fisheries 

 Res. Inst., Seattle, WA. Rep. FRI-UW-9020, 69 p. 



