Utilization of log-piling structures as artificial habitats for red king crab Paralithodes camtschaticus.

ABSTRACT Artificial habitats or reefs have been used to mitigate for alteration of marine habitats and increase populations of desirable marine species. In Kodiak, Alaska, breakwater construction covered 3.5 ha of sedimentary habitat potentially usable by commercially valuable red king crab Paralithodes camtschaticus. Juvenile king crab are common occupants of wooden dock pilings, suggesting that pilings could be used to mitigate for loss of natural habitat. To test this hypothesis, six log-piling structures were constructed from untreated spruce and placed in pairs at three different locations in ocean bays near Kodiak. Divers conducted a yearlong study of king crab recruitment by making quarterly counts of crab on the structures and adjacent seafloor areas. Abundance of juvenile (9-21 month-old) king crab increased steadily from June 1997 through March 1998 as crab recruited to the structures, then declined in June 1998. Crab abundance was significantly higher on piling structures than on the adjacent substratum. Site, season, and their interaction bad significant effects on abundance. Why juvenile king crab are attracted to pilings is unknown. Pilings are not structurally complex habitats, but provide hard surfaces for fouling organisms such as hydroids, which are preferred habitat for juvenile crab. Additionally, pilings do not persist in the environment, and may not be the best structure for habitat enhancement. For these reasons, and because there is no evidence that red king crab are habitat limited in our study area, we do not recommend the use of pilings as artificial habitats for red king crab.

KEY WORDS: king crab, habitat, ecology, settlement, artificial reefs. Paralithodes

INTRODUCTION

The use of artificial reefs for enhancement of marine populations has received considerable attention in recent years (for examples see Jensen 2002). Such structures may be assembled from a variety of components including natural (logs and rocks), seminatural (concrete blocks, modules, or concreted boulders), and non-natural (tires, coal ash, wrecked ships, junked cars, and derelict oil platforms). Reefs have been constructed for a variety of recreational and commercial uses, including fisheries enhancement, scuba diving, aquaculture, habitat restoration, environmental mitigation, resource conservation, and research (Seaman 2002). Over time, research on artificial reefs has developed from purely observational studies of colonization, to studies of their ecological function, design and placement, and performance evaluation (Seaman 2002). A major issue in research on enhancement structures has been the "attraction-production" debate: do artificial structures support increased production of target species, or simply attract them from somewhere else? Conflicting viewpoints often depend on assumptions about whether reef-associated species are limited by habitat availability (Grossman et al. 1997) or recruitment variability, making artificial reefs little more than elegant fishing tools (Bohnsack et al. 1997). Lindberg (1997) argued that the answer depends on the specific structure, its location and objectives, and the species considered. Generally, production can be improved by use of highly complex structures that provide high levels of structural heterogeneity at appropriate scales. Such structures probably function to reduce mortality during critical life stages by reducing predation, and providing improved foraging opportunities (Bohnsack et al. 1997).

Most studies of artificial reefs In date have focused on highly mobile fauna such as reef-associated or pelagic fishes. Those dealing with benthic resources have addressed fouling organisms (Foster et al. 1994), or epibenthic prey species (Jara & Cespedes 1994) of very low motility. Few have studied the use or impacts of artificial structures on commercially significant decapod crustaceans, which are highly motile within limited ranges. Notable exceptions include studies of Florida spiny lobsters Panulirus argus (Herrnkind & Butler 1986, Butler & Herrnkind 1997, Herrnkind et al. 1997a, Herrnkind et al. 1997b). These researchers have demonstrated that habitat availability limits the abundance of juvenile lobsters, and enhancement with appropriate artificial substrates can greatly increase abundance, whereas artificially increasing the abundance of juveniles by seeding is not effective. Such increases occur primarily through settlement rather than by immigration (Herrnkind et al. 1997a).

In 1997, a new breakwater was constructed by the United States Army Corps of Engineers (USACE) at the south end of St. Herman Harbor, Near Island, Kodiak, Alaska. During construction of the breakwater >3.5 ha of sedimentary habitat were covered with rock, at depths from 5-20 m, raising concerns that such habitat alterations might displace economically important marine species, particularly red king crab Paralithodes camtschaticus (Tilesius, 1815) (hereafter referred to as RKC).

At Kodiak, RKC up to 1.5 y of age are commonly found on wooden pilings covered with a variety of fauna as biologic structure, suggesting that pilings associated with piers are good habitat for juvenile crab. Based on the results of previous studies, (Dew et al. 1992) the National Marine Fisheries Service (NMFS) convinced the United States Coast Guard not to remove a condemned pier (Marginal Pier) in Womens Bay, Kodiak Island, Alaska. which contains hundreds of pilings and is a common site for juvenile king crab of 10-25 mm carapace length (CL). The USACE also installed 50 pilings at a site in Womens Bay as additional crab habitat: although these pilings were not routinely surveyed, we occasionally observed RKC there, though most of the pilings were destroyed by ice within a few years.

This study was undertaken to determine whether juvenile RKC would recruit to submerged log piling structures, and if they could be used to mitigate for habitat alterations caused by construction of breakwater and marina facilities.

MATERIALS AND METHODS

Six log piling structures (Fig. 1) were built and deployed in pairs. Each structure consisted of four 30-cm diameter corner posts (3 m spruce logs with intact bark) sunk into 0.6 x 0.6 x 0.5 m concrete blocks. Adjacent and opposing pairs of corner posts were connected by horizontal 5 x 20 cm ([less than or equal to] 2 x [less than or equal to] 8) beams of rough milled spruce, and similar beams connected the base of each post to the upper cross beam (see Fig. 1). Each structure had a surface area of approximately 35 [m.sup.2]. Two structures, labeled North (N) and South (S), were placed in approximately 10 m of water at each of 3 sites (Fig. 2): inside the breakwater (IB), outside the breakwater (OB), and in Womens Bay (WB), approximately 8 km away. The paired structures at the IB and OB sites were placed 27.4 m and 57.3 m apart, respectively. Although both sites were within 100 m of the rock breakwater, we did not survey it because the profusion of kelp prevented adequate sampling of the rock surface and crevices. In Womens Bay, the North and South structures were separated by about 600 m but were placed 34.1 m and 39.3 m, respectively, from Marginal Pier. Marginal Pier is a +60-year-old dock consisting of hundreds of creosote treated wooden pilings, many of which have decayed, and most are covered with fauna including sponges, anemones, hydroids, bryozoans, barnacles, and sea stars. A fourth pair of survey sites included four Marginal Pier pilings nearest to each of the WBN and WBS habitats; these were labeled MPN and MPS, respectively. All log-piling structures were placed on the seafloor between May 19 and 22, 1997 prior to settlement of the 1997 year-class of RKC.

[FIGURES 1-2 OMITTED]

To compare counts of organisms on the structures to the surrounding environment, benthic transect lines were staked out on opposite sides of each structure. These also served as guidelines between the structures at the IB and OB sites. Because the IB structures were closest together, the standard transect length was defined as half the distance between them, or 13.7 m. Transects of similar length were established on the opposite sides of each structure as well. Surveys were conducted by scuba every 3 mo from June 1997 to June 1998 for a total of five quarterly samples. During each survey, divers descended a marker line to the bottom and counted all RKC on the outside and inside of the structure. Divers then counted RKC within a 1 m swath along both sides of the benthic transect lines. Each transect count was treated as an individual replicate. The total bottom area surveyed around each structure was 54.8 [m.sup.2] (2 transects x 2 m x 13.7 m). Divers then swam to the second structure at the site and repeated the structure and transect counts. Both structures at a site were surveyed on a single day, and each site on separate days within a week. At Marginal Pier, divers also counted RKC on the bottom 3 m of the tour pilings nearest to each structure. The total data matrix consisted of 100 samples (5 seasons, 3 sites, 2 structures, 3 locations (structure. transect 1, transect 2), plus two MP samples in each season). To prevent disturbance or displacement, sizes of RKC observed by one diver (JEM) were estimated in 5-mm increments of carapace length (CL). During each dive, a series of replicate photographs were taken at specific locations, including piling tops, sides, crossbeams, concrete bases, and on the transects. Qualitative observations of substrate characteristics and sessile epibenthic species on the piling structures and transects were made from these photographs.

Preliminary data analysis showed that variances among samples were not homogeneous, indicating severe departures from normality that were not remedied by square root or angular transformation. For this reason, and because the data are counts of (relatively) rare events, we chose to use a generalized linear model (GEM) type of ANOVA, based on a Poisson distribution rather than the normal distribution. The GLM allowed us to model RKC counts as a function of several factors. Only the data from the structures was analyzed in this manner; transect counts were excluded because of their low numbers, and counts from Marginal Pier were excluded because they were a different type of piling (older, creosote treated). The latter were only compared with the adjacent structures in Womens Bay. Factors included in the analysis were quarter, site, structure (north or south within pairs), and 2-way interactions between those 3 variables. Following this, a nonparametric Mann-Whitney U-test was used to make comparisons between pairs of sample sites. Sites compared were Marginal Pier versus site WB, site IB versus site OB, site IB versus WB, and OB versus WB. Statistical analysis was conducted using S-Plus version 4.5 or SPSS version 10.

RESULTS

A total of 136 RKC were observed and counted (Table 1). The largest number of RKC (90 crab, or 73% of the total excluding Marginal Pier) were observed at the OB site, followed by WH (25 or 20%), Marginal Pier (12), and IB (9). Excluding Marginal Pier samples, virtually all RKC (99%) were observed on the structures, and only two RKC were observed on the transects: one on kelp and one on a boulder. At each site, over twice as many RKC occurred on the southern structure (85) as on the northern (37). Only one RKC was observed (on the OB benthic transec0 during the first 2 quarters (June and September 1997). By December 1997, young-of-the-year RKC <15 mm CL (9 mo post-hatch) were present on the structures, having probably settled there as post larvae (Table 2). Numbers increased through March 1998, but most were still <15 mm CL. A few crab >25 mm CL, representing the previous yearclass (age +1, 21 mo post-hatch) were present on the structures, and probably arrived by immigration. Numbers of RKC observed on structures declined in June 1998; by this time, most crab were in the 20-24 mm CL range.

The Poisson-based GLM showed that the effects of quarter. site, and structure were all significant (Table 3). The interaction of quarter x site was significant because the highest counts occurred in March 1998 at the OB and WB sites, but counts were highest in June 1998 at the IB site. The interactions of structure x quarter and structure x site were not significant. There were not enough degrees of freedom remaining to include a 3-way interaction, and it probably would not have been significant because no 2-way inter actions involving structure were significant. The Mann-Whitney U-test showed no significant difference in counts of RKC between the WB and MP sites (U = 45.0, P = 0.678), WB and IB (U = 39.5, P = 0.369), or WB and OB (U = 36.0, P = 0.267). Differences between IB and OB were much greater, but still marginally non-significant (U = 27.0, P = 0.057).

Sediments inside the breakwater were silty and supported a dense community of tube-building polychaetes (possibly Spiochaetopterus costarum), whereas sediments outside the breakwater were sandier with scattered rocks and were devoid of epibenthos, except for occasional sea stars. Sediments near the WB structures were also silty, but scattered with mussel shells from Marginal Pier, and occasional anthropogenic debris from 50 y of military use. Calcareous tube-building polychaetes (probably Serpula vermicularis) and plumose anemones (Metridium senile) were abundant on Marginal Pier and adjacent debris.

By July 1997, approximately 3 wk after placement of the structures, barnacles of 1-2 mm diameter had colonized most of the structures. Mottled sea stars (Evasterias troschellii), sunflower stars (Pvcnopodia helianthoides), green sea urchins (Strongylocentrotus droebachiensis) and sculpins (family Cottidae) arrived on the structures via immigration. By September 1997, small clumps of green algae and hydroids were growing on most of the structures, and filamentous red algae were common on the WB structures. Densities of epibenthos were lower inside the breakwater than outside. By December, the structures had been colonized by encrusting bryozoans, decorator crab (Oregonia gracilis), RKC, calcareous tubeworms, and nudibranchs (Flabellina fusca). In March, red algae (Palmaria sp.), Laminaria sp., tunicates (probably Molgula sp.), and hydroids were abundant. Filamentous red algae obscured the sheltered sediments inside the breakwater, but were absent outside the breakwater. By June, piling tops were once again covered with green algae, and helmet crab (Telmessus cheiragonus), hermit crab (Pagurus sp.), and RKC were common on the structures.

DISCUSSION

Our data demonstrate that log piling structures may serve as potential habitats for juvenile RKC. Numbers of crab on the structures were two orders of magnitude greater than on the adjacent seafloor. As a result of this work, we recommended to the USACE that new dock structures in the Kodiak area be built with pilings rather than sheet-metal bulkheads with backfill, as proposed by some developers. Use of pilings preserves the underlying seafloor substratum that would be covered by fill, and adds hard vertical structure useful to crab and other fauna. We cannot recommend that piling structures be deployed to mitigate for habitat loss at this time for several reasons. First, the reason why juvenile RKC are attracted to pilings is still not understood, though settlement among fouling organisms seems most likely. Second, pilings have limited surface area and low fractal dimension; that is, they are devoid of highly complex interstitial spaces, which is apparently the structural feature that makes various natural and artificial habitats attractive to juvenile RKC (Stevens & Kittaka 1998 and Stevens 2003). The relationship between the scale of refuge spaces and the range of body sizes is an important factor in survival of reef-associated decapods (Eggleston et al. 1992). If habitat enhancement is deemed worthwhile for future research, then other types of structures might be more effective by providing a greater variety of interstices in a more compact structure. Furthermore, wooden pilings do not last long in a marine environment unless treated with toxic chemicals, and steel or concrete pilings may not attract the same fouling community. Moreover, in this study we evaluated only one kind of structure; better structures might include crushed rock or gravel, or specifically designed man-made substrata. It is possible that the new breakwater provides more habitat of better quality than the original sedimentary substratum that it replaced.

In our study, RKC were more abundant at a site (OB) that was exposed to prevailing weather and current, and less abundant at a nearby site (IB) protected behind a large breakwater, or at the head of Womens Bay, several kilometers from open water. This distribution could result from recruitment processes involving the transport of larvae by ocean currents until they reach suitable exposed habitat sites. As more larvae settle on exposed sites, fewer would remain in the water to settle at protected sites. The breakwater and its many crevices may also have "filtered" out many larvae prior to their arrival at the IB site. This hypothesis could also account for the higher numbers of RKC on the southernmost habitats within each pair. At each site, exposure to the open ocean decreased, and distance from it increased, along an axis from south to north. The abundance of infaunal polychaete tubes at the IB site is indicative of sheltered habitat, whereas lack of their presence at the OB site was probably the result of wave action and current scour. RKC probably arrived on the structures by settlement as postlarvae during the late summer and fall of 1997, but were too small to be seen by divers until December. Observed numbers of RKC increased through March 1998, then declined the following summer. By that time, the earliest arrivals were 15 mo old and 15-20 mm CL. At that size, they are less vulnerable to predation, so may have emigrated from the structures. Newer recruits either did not replace them at the same rate, or were not yet large enough to be seen by divers. One of the most important species of the epibenthic community are hydroids because they are important habitat for newly settled RKC glaucothoe (postlarvae), which choose them over alternative habitats due to their complex 3-dimensional structure (Stevens 2003). Hydroids were abundant among artificial collectors made of onion sacs stuffed with monofilament line (Donaldson et al. 1992 and Blau & Byersdorfer 1994), and may be one reason that glaucothoe settle there en masse. Hydroid colonies do not develop until late summer, and their presence on the pilings and buoy lines may attract settling RKC glaucothoe to the pilings. However, hydroids were not highly abundant when the first recruits began to appear, and most crab were observed on bare wood.

Biologic structures are important settlement habitats for juvenile RKC and are practically the only location where they are found in many parts of their geographical range, such as the Bering Sea, where physical structure is scarce. This distribution is the result of habitat selection by settling postlarvae (glaucothoe) rather than of predation. During their first year of life, juvenile RKC are associated with sponge and bryozoan colonies (Sundberg & Clausen 1977), mussel and hydroid colonies (see photo in Stevens 2003), stalked ascidians and polychaete tubes (Stevens & MacIntosh 1991), and shell debris and cobble (McMurray et al. 1986 and Loher & Armstrong 2000). Glaucothoe will settle in large numbers on various types of artificial collectors (Donaldson et al. 1991 and Blau & Byersdoffer 1994). In the laboratory RKC glaucothoe prefer structurally complex habitats, whether artificial (aquarium filter material) or biologic (hydroids and complex red algae), to those with less structure (gravel) and will not settle on structure-less open sand (Stevens & Kittaka 1998 and Stevens 2003). Selection for such habitats is probably an adaptive response to high predation levels. At sizes of 10-15 mm CL, RKC are often observed "hitch-hiking" on sea stars (Dew 1990). At sizes >25 mm CL, RKC start to exhibit aggregative (podding) behavior (Powell & Nickerson 1965 and Dew 1990).

Crab and lobsters are attracted to, and use, various artificial habitats. In Chile, Cancer edwardsi and Homalaspis plana use the hollow spaces in artificial reefs made of concrete blocks (Jara & Cespedes 1994). Numbers of both species increased significantly on the reef and in quadrat samples taken from the surrounding substratum, in contrast to our data, which showed virtually no RKC present on surrounding substrata. Adult stone crab Menippe sp. were more abundant among concrete block reefs spaced at 60 m intervals than among those spaced at 2 m intervals, and abundance was strongly seasonal (Frazer & Lindberg 1994). Stone crab apparently arrived by immigration as adults and foraged among the surrounding substratum, in contrast to RKC that settle on pilings as postlarvae, and were rarely observed "off structure". Reefs made of cemented coal-ash were used by spider crab Maja squinado, velvet swimming crab Liocarcinus puber, and European lobster Homarus gammarus, the latter of which were predominantly adults, indicating that recruitment occurred through immigration rather than by settlement (Jensen et al. 1994). Caribbean spiny lobster Panulirus argus are attracted to "casitas" consisting of concrete blocks (Lozano-Alvarez et al. 1994), or concrete over a PVC frame (Eggleston et al. 1992). Despite the presence of crab and octopus, artificial shelters may reduce predation on small juvenile lobsters, and thus increase production of that sizegroup (Eggleston et al. 1992). In Delaware Bay, the biomass associated with concrete artificial reefs was 2 to 3 orders of magnitude greater than that of the surrounding substratum, although no large decapods were observed there (Foster et al. 1994 and Steimle et al. 2002).

Population enhancement is probably feasible for commercially valuable decapods like spiny and American lobster because their abundance is limited by the availability of their preferred habitats. Postlarval spiny lobsters settle preferentially in clumps of red algae (Laurencia sp.) (Herrnkind & Butler 1986). Enhancement of natural habitat (algal clumps) or addition of artificial shelters (concrete blocks) is an effective way to increase the abundance of juvenile lobster, whereas seeding with postlarvae or juveniles is not effective (Butler & Herrnkind 1997 and Herrnkind et al. 1997a). Enhancement of natural substrata with gravel plots caused densities of postlarval American lobsters (Homarus umericanus) to increase by a factor of 10, whereas supplemental postlarvae declined by a factor of 5 (Wahle 1991). At present, them is no convincing evidence that RKC populations are limited by lack of available settlement habitat, so population enhancement by the addition of artificial habitats may not be effective for this species. Further research on settlement behavior of postlarval RKC, and shelter use by juveniles, both in the laboratory and in their natural environment, is warranted before enhancement of habitat or populations is undertaken with this species.

TABLE 1.
Numbers of red king crab counted on log-piling structures,
transacts and Marginal Pier pilings, in four seasons, at
three sites: IB (inside breakwater), OB (outside breakwater),
and WB (Womens Bay).

Sample Period   Sub-stratum     IB   OB   WB   Total

June 1997       Structures      0     0    0      0
                Transects       0     1    0      1
                Marginal Pier              0      0
                Subtotal        0     1    0      1

Sept 1997       Structures      0     0    0      0
                Transects       0     0    0      0
                Marginal Pier              0      0
                Subtotal        0     0    0      0

Dec 1997        Structures      1    15    2     18
                Transects       0     0    0      0
                Marginal Pier              1      1
                Subtotal        1    15    3     19

March 1998      Structures      0    47   17     64
                Transects       0     0    0      0
                Marginal Pier              9      9
                Subtotal        0    47   26     73

June 1998       Structures      8    26    6     40
                Transects       0     1    0      1
                Marginal Pier              2      2
                Subtotal        8    27    8     43

Grand Total                     9    90   37    136

TABLE 2.
Numbers of red king crabs in 5-mm categories of estimated
carapace length (CL), observed on piling structures. Only
one diver estimated sizes, so this table dues not include
all crabs observed.

CL Range   June   Sept   Dec    March   June
  (mm)     1997   1997   1997   1998    1998

 10-14      1             3      20       1
 15-19                    1       4       1
 20-24                    1              28
 25-29                            1
 30-34
 35-39                            1       1
 40-44                                    1

TABLE 3.
GLM analysis for effects of factors on counts of red king crabs
on piling structures (N or S) in five quarters, at three sites.
Data from transect counts and marginal pier are excluded. All
data were transformed to log (x + 1) prior to analysis.

Source            df   Res. Dev.     P

Null              29     293.24    <0.001
Quarter (QTR)     25     141.21    <0.001
Site (SIT)        23      56.83    <0.001
Structure (STR)   22      33.94    <0.001
QTR x SIT         14      15.31     0.017
QTR x STR         10      12.15     0.532
SIT x STR          8       7.93     0.121

ACKNOWLEDGMENTS

This project was partially funded by the United States Army Corps of Engineers under MIPR Nos. E86954046-0001 and F12961362-0001. The piling structures were built and installed by Majdic and Sons, Inc The authors thank W. E. Donaldson for assistance with diving surveys, I. Vining for help with statistical analysis and interpretation, and C. Armistead for production of maps. The manuscript benefited greatly from reviews by R. S. Otto, A. Stoner, and R. Stone.

LITERATURE CITED

Blau, S. F. & S. C. Byersdorfer. 1994. Sausage-shaped artificial collector developed in Alaska to study young-of-year red king crabs. In: Fifth International Conference on Aquatic Habitat Enhancement. Long Beach. CA (USA). November 3-7, 1991. Bull, Mar. Sci. 55:878-886.

Bohnsack, J. A., A. M. Ecklund & A. M. Szmant. 1997. Artificial reef research: is there more than the attraction-production issue? Fisheries. 22:14-16.

Butler, M. J. & W. F. Herrnkind. 1997. A test of recruitment limitation and the potential for artificial enhancement of spiny lobster (Panulirus argus) populations in Florida. Can. J. Fish. Aquat. Sci. 54:452-463.

Dew, C. B. 1990. Behavioral ecology of podding red king crab, Paralith odes camtschatica. Can. J. Fish. Aquat. Sci. 47:1944-1958.

Dew, C. B., P. A. Cummiskey & J. E. Munk. 1992. The behavioral ecology and spatial distribution of red king crab and other target species: implications for sampling design and data treatment. Kodiak, AK: National Marine Fisheries Service. 51 pp.

Donaldson, W. E., S. Byersdorfer & F. Blau. 1991. Use of artificial collectors to study growth of small red king crab. J. Shellfish Res. 10:305.

Donaldson, W. E., S. Byersdorter, D. Pengilly & S. F. Blau. 1992. Growth of red king crab, Paralithodes camtschaticus, (Tilesius, 1815), in artificial habitat collectors at Kodiak, Alaska. J. Shellfish Res. 11:85-89.

Eggleston, D. B., R. N. Lipcius & D. L. Miller. 1992. Artificial shelters and survival of juvenile Caribbean spiny lobster Panulirus argus: spatial, habitat, and lobster size effects. Fish. Bull. 90:691-702.

Foster. K. L., F. W. Steimle, W. C. Muir, R. K. Kropp & B. E. Conlin. 1994. Mitigation potential of habitat replacement--concrete artificial reef in Delaware Bay--preliminary results. Bull. Mar. Sci. 55:783-795.

Frazer, T. K. & W. J. Lindberg. 1994. Refuge spacing similarly affects reef-associated species from 3 phyla. Bull. Mar. Sci. 55:388-400.

Grossman, G. D., G. P. Jones & W. J. Seaman. 1997. Do artificial reefs increase regional fish production? A review of existing data. Fisheries. 22:17-23.

Hermkind, W. F. & M. J. Butler. 1986. Factors regulating postlarval settlement and juvenile microhabitat use by spiny lobsters Panulirus argus. Mar. Ecol. Prog. Set. 34:23-30.

Herrnkind, W. F., M. J. Butler & J. H. Hunt. 1997a. Can artificial habitats that mimic natural structures enhance recruitment of Caribbean spiny lobster? Fisheries. 22:24-27.

Herrnkind, W. F., M. J. Butler, J. H. Hunt & M. Childress. 1997b. Role of physical refugia: implications from a mass sponge die-off in a lobster nursery in Florida. Mar. Freshw. Res. 48:759-769.

Jara, F. & R. Cespedes. 1994. An experimental evaluation of habitat enhancement on homogeneous marine bottoms in southern Chile. Bull. Mar. Sci. 55:295-307.

Jensen, A. C. 2002. The seventh international conference an artificial reefs and related aquatic habitats: Introduction. ICES J. Mar. Sci. 59(Suppl.):S1-S2.

Jensen, A. C., K. J. Collins. A. P. M. Lockwood, J. J. Mallinson & W. H. Turnpenny. 1994. Colonization and fishery potential of a coal-ash artificial reef, Poole-Bay, United-Kingdom. Bull. Mar. Sci. 55:1263-1276.

Lindberg, W. J. 1997. Can science resolve the attraction production issue? Fisheries. 22: 10-13.

Loher, T. & D. A. Armstrong. 2000. Effects of habitat complexity and relative larval supply on the establishment of early benthic phase red king crab (Paralithodes camtshaticus Tilesius, 1815) populations in Auke Bay, Alaska. J. Exp. Mar. Biol. Ecol. 245:83-109.

Lozano-Alvarez, E., P. Briones-Fourzan & F. Negrete-Soto. 1994. An evaluation of concrete block structures as shelter for juvenile Caribbean spiny lobsters. Panulirus argus. Bull. Mar. Sci. 55:351-362.

McMurray, G., A. H. Vogel, P. A. Fishman, D. A. Armstrong & S. C. Jewett. 1986. Distribution of larval and juvenile red king crabs (Paralithodes camtschatica) in Bristol Bay. pp. 267-477.

Powell, G. C. & R. B. Nickerson. 1965. Aggregations among juvenile king crabs (Paralithodes camtschatica, Tilesius) Kodiak. Alaska. Anita. Behav. 13:374-380.

Seaman. W., Jr. 2002. Unifying trends and opportunities in global artificial reef research, including evaluation. ICES J. Mar. Sci. 59(Suppl.):S14-S16.

Steimle, F., K. Foster, R. Kropp & B. Conlin. 2002. Benthic macrofauna productivity enhancement by an artificial reef in Delaware Bay, USA. ICES J. Mat; Sci. 59:(Suppl.)S100-S105.

Stevens. B. G. 2003. Settlement, substrate preference, and survival of red king crab Paralithodes camtschaticus (Tilesius, 1815) glaucothoe on natural substrata in the laboratory. J. Exp. Mar. Biol. Ecol. 283:63-78.

Stevens, B. G. & J. Kittaka 1998. Postlarval settling behavior, substrate preference, and time to metamorphosis for red king crab Paralithodes camtschaticus. Mar. Ecol. Prog. Ser. 167:197-206.

Stevens, B. G. & R. A. MacIntosh. 1991. Cruise results supplement, Cruise 91-1 Ocean Hope 3: 1991 eastern Bering Sea juvenile red king crab survey, May 24 to June 3, 1991. Kodiak, AK: National Marine Fisheries Service, Alaska Fisheries Science Center. Kodiak Fisheries Research Center.

Sundberg, K. A. & D. Clausen. 1977. Post-larval king crab (Paralithodes camtschatica) in Kachemak Bay, Lower Cook Inlet, Alaska 1976. In: L. L. Trasky, L. B. Flagg & D. C. Burbank, editors, Environmental studies of Kachemak Bay and Lower Cook Inlet. Vol. 5. Anchorage: Alaska Department of Fish and Game pp. 1-36.

Wahle, R. A. 1991. Implications for lobster fishery enhancement f natural benthic recruitment, hatchery-reared "blues" and experimental cobbles. J. Shellfish Res. 10:287.

BRADLEY G. STEVENS, * J. ERIC MUNK, AND PETER A. CUMMISKEY National Marine Fisheries Service, Alaska Fisheries Science Center, Kodiak Fishery Research Center, 301 Research Ct., Kodiak, Arkansas

* Corresponding author. E-mail: [e-mail]