North American Journal of Fisheries Management 26:702–713, 2006 Ó Copyright by the American Fisheries Society 2006 DOI: 10.1577/M05-029.1
If You Build It, Will They Come? Fish and Angler Use at a Freshwater Artificial Reef SARA M. CREQUE,* MATTHEW J. RAFFENBERG,1 WAYNE A. BROFKA,
JOHN M. DETTMERS
Illinois Natural History Survey, Lake Michigan Biological Station, 400 17th Street, Zion, Illinois 60099, USA Abstract.—In November 1999, an artificial reef composed of granite rubble was built in southwestern Lake Michigan to attract smallmouth bass Micropterus dolomieu. Adult fish communities were sampled at the artificial reef site and a nearby reference site before (1999) and after reef construction (2000–2003) to determine whether the artificial reef attracted sport fishes. The total number of fish observed along scuba transects was higher at the artificial reef than at the reference site during 2000–2003. Smallmouth bass, rock bass Ambloplites rupestris, round goby Neogobius melanostomus, and yellow perch Perca flavescens were most commonly observed by divers at the artificial reef site, whereas the round goby was the most prevalent species observed at the reference site. Mean annual total gill-net catch per unit effort (CPUE) did not differ at the two sites after reef construction. Freshwater drum Aplodinotus grunniens, gizzard shad Dorosoma cepedianum, yellow perch, and salmonines were commonly caught at both locations. The presence of several of these taxa was related to water temperature but not location. Smallmouth bass presence was related to location; CPUE was greater at the artificial reef than at the reference site during 2000–2002. Rock bass CPUE also was greater at the artificial reef than at the reference site during 2002. Smallmouth bass association with the reef was seasonal and correlated with temperature. Although anglers were aware of the artificial reef, fishing effort and success were low, in part because few anglers targeted black bass Micropterus spp. Because water temperature strongly influences the use of structure by centrarchids in deep, cold lakes like Lake Michigan, care must be taken to site artificial reefs in zones of the most suitable water temperature for these species.
Artificial habitat construction is included in fisheries management plans of many states (Stone 1985) to attract desirable fish species, increase angler effort and success, redirect fishing pressure away from areas with critical natural habitat, or provide spawning habitat for fish species (Sheey 1985; Bohnsack 1989; Gannon 1990). Use of artificial reefs in larger freshwater bodies, such as the Great Lakes, was limited until the 1980s and is still considered experimental (Kevern et al. 1985; Gannon 1990; Kelch et al. 1999), in part because research on the ecology and success of freshwater artificial reefs is sparse (Prince et al. 1985; McGurrin et al. 1989; Bohnsack et al. 1991). Some of the first studies on artificial habitats in the Great Lakes examined riprap associated with water intake structures at a nuclear power plant and along breakwalls and jetties along the eastern shoreline of Lake Michigan. Although these rock structures were not specifically placed for habitat considerations, fish (including alewives Alosa pseudoharengus and yellow perch Perca flavescens) used these structures season* Corresponding author: [email protected]
1 Present address: Florida Power and Light, PGD Environmental, 700 Universe Boulevard GPA/JB, Juno Beach, Florida 33408, USA. Received February 22, 2005; accepted January 24, 2006 Published online August 14, 2006
ally (Liston et al. 1985; Rutecki et al. 1985). In fall 1980, the first artificial reef built specifically to attract fish for anglers in the Great Lakes was completed in Lake Michigan near the mouth of the Muskegon River (Kevern et al. 1985). One year after construction, yellow perch, alewives, and johnny darter Etheostoma nigrum were the most commonly observed fish at the reef. Of the few published studies on artificial habitat structures in the Great Lakes, almost all began assessment of fish populations only after construction of the artificial structures and lack long-term assessment data, presenting only 1–3 years of data (Binkowski 1985; Gannon et al. 1985; Kevern et al. 1985; Liston et al. 1985; Kelch et al. 1999). The exception, Rutecki et al. (1985), monitored limestone riprap at a nuclear power plant from 1973 to 1982. However, those authors made no mention of reference site results or preconstruction data; thus, the role of the structure in attracting fish was inconclusive. Most Great Lakes studies also used only one survey (scuba) or collection (nets) method to assess fish use of artificial structures (Gannon et al. 1985; Liston et al. 1985; Kelch et al. 1999). Both methods have their own biases that can influence findings, and neither method presents a complete a picture of the fish community when used alone. Most of these studies also generated too little data to correlate fish use and water temperature. This is
FRESHWATER ARTIFICAL REEF USE
an important factor to consider because water temperature can influence spatial and temporal distributions of fish in the Great Lakes and other systems (Dahlberg 1981; Olson et al. 1988; Haynes 1995; Ho¨o¨k et al. 2004). In addition, studies on artificial reefs in Lake Erie during the 1990s were the first to present data following the introduction of two exotic species, zebra mussels Dreissena polymorpha and round goby Neogobius melanostomus (Kelch et al. 1999). Most recently, artificial reef construction in the Great Lakes has occurred in Lake Erie. During 1986–1989, several artificial reefs of concrete, rocks, and other rubble were placed within 1.2 km of shore in the central basin, which has a relatively flat and featureless bottom habitat (Kelch et al. 1999). The impetus for the Lake Erie reefs was to attract sport fishes and create angling opportunities closer to shore and harbors, as well as to study the feasibility and effects of reef construction in the Great Lakes (Hushak et al. 1999; Kelch et al. 1999). Unlike early efforts to provide artificial habitat in Lake Michigan (Liston et al. 1985; Rutecki et al. 1985), the Lake Erie artificial reefs attracted large numbers of fish, especially smallmouth bass Micropterus dolomieu, rock bass Ambloplites rupestris, and walleyes Sander vitreus (Kelch et al. 1999). Due to the large numbers of fish and anglers attracted to these Lake Erie reefs, as well as the resulting economic value to surrounding communities, additional artificial reefs were constructed off of Cleveland in 1997 (Kelch et al. 1999). The objective of this study was to gain a better understanding of the factors (e.g., water temperature and distance from shore) that can influence the success of freshwater artificial reefs in attracting fish and their anglers. We evaluated an artificial reef of granite rubble constructed in November 1999 in southwestern Lake Michigan near Chicago, Illinois, to determine whether it was effective at attracting smallmouth bass and providing additional angling opportunities. Our approach was to (1) compare fish community structure and relative abundance of key species at the artificial reef and a nearby reference site both before and after construction using a combination of collection and survey methods and (2) examine angler use of the artificial reef through a creel survey. We evaluated whether use of the artificial reef by smallmouth bass and other species was significantly impacted by water temperature. We also compared our results with those of studies on other Great Lakes artificial reefs. Methods Study area.—Our study area was located in southwestern Lake Michigan off the coast of Chicago (Figure 1). Bottom bathymetry in this portion of the
FIGURE 1.—Location of artificial reef and reference sites in the nearshore waters of southwestern Lake Michigan. Both locations were 2.8 km offshore in 7.5 m of water. Creel survey launch ramp locations are also indicated.
lake is very gradual; depths greater than 10 m typically do not occur until more than 5 km offshore. Bottom substrate consists of patches of clay, sandy silt, sand, and mixed sand and gravel (Chrzastowski et al. 1998). Because of the shallow depth and extensive mixing, a persistent thermocline does not develop during summer. Coldwater upwelling does occur but is infrequent and not as severe as that found in areas with steeper bathymetry. The artificial reef was constructed 2.8 km offshore in 7.5 m of water, using approximately 4,500 tons of granite rock of various sizes (0.25 to . 1.5 m in diameter) in November 1999. The reef lacked smaller-sized pebbles and gravel, but the variety of sizes of larger boulders created large interstitial spaces and vertical relief. The rock was placed in a manner to ensure the rock piles were continuous and connected at the base. Side-scan sonar indicated that the completed reef was 256 m long and had a mean height of 2.1 m and mean width of 15.5 m; total bottom area covered by the artificial reef was 5,440 m2 (S. Anderson, Applied Marine Acoustics, Inc., personal communication.). A reference site that
CREQUE ET AL.
was located approximately 3.3 km south of the reef site and that had comparable biological characteristics, depth, and distance from shore also was sampled to allow us to make comparisons between sites before and after reef construction. Substrate complexity at the reference site is minimal; the substrate is predominately sand (.80%) and small gravel (S. M. Creque, unpublished data) with very intermittent rocks and boulders, overlaying a thick layer of clay. Field surveys.—Both the artificial reef and reference sites were sampled in 1999, before reef construction. Gill-net sampling and scuba surveys occurred every 2 weeks from June through late October 1999–2003 when weather permitted. On each sampling date, water temperature profiles (1-m intervals from surface to bottom) were recorded at each site. Data loggers placed at the bottom and at midwater depths recorded hourly water temperatures at the artificial reef site in 2001– 2003. Scuba divers conducted a visual survey using the transect method (Bortone and Kimmel 1991; Bortone et al. 2000) at the artificial reef and reference sites when weather conditions and water visibility permitted. During 1999, two divers swam 100-m transects at both sites to estimate fish composition and abundance before reef construction. During 2000–2003, divers swam the entire length of the artificial reef (256 m) and for a timed duration at the reference site; average (6SD) transect time was 27 6 9 min at the artificial reef and 14 6 4 min at the reference site (Table 1). Variability in transect time resulted from bottom current, equipment problems, diver illness, poor visibility, or a combination thereof. Divers swam side by side, identifying and counting fish within a 2-m radius of each diver. Divers moved along transects at the same speed to maintain equal encounter rates. Fish counts were recorded and then permanently transcribed on data sheets at the surface as the pooled total from both divers. Monofilament gill nets were set at both sites every 2 weeks to gather additional information on relative TABLE 1.—Scuba sampling effort (min) at artificial reef and reference sites in southwestern Lake Michigan during 1999– 2003. During 1999, a 100-m distance was observed at each site; NA ¼ not applicable. Artificial reef site
Number of transects
Number of transects
1999 2000 2001 2002 2003
2 6 3 7 7
NA 190 101 179 155
2 4 2 7 6
NA 61 46 85 80
abundance and composition of the fish assemblage. Gill nets used in 1999 totaled 46 m in length and were composed of 3.8- and 4.4-cm stretch mesh; these two mesh sizes are hereafter referred to as ‘‘small.’’ We used different nets after artificial reef construction to capture adult sport fish more efficiently than in 1999, when the small mesh was used. Two monofilament gill nets measuring 61 3 1.52 m (two 30.5-m panels, 10.2and 11.5-cm stretch mesh) were set at each site during 2000–2001. During the 2002 and 2003 sampling seasons, two 30.5-m panels (5.1- and 7.6-cm stretch mesh) were added to the nets (i.e., total gill-net length ¼ 122 m). The 10.2- and 11.5-cm mesh sizes are referred to as ‘‘large,’’ and the 5.1- and 7.6-cm mesh sizes used in 2002 and 2003 are referred to as ‘‘medium.’’ On each sampling date, paired nets were fished on the bottom at each site from about 1 h before sunset to 1 h after sunrise. All fish were identified, measured, and released. A sportfishing survey gauged angler awareness and use of the artificial reef. From April 1 to September 30 of each year, we collected data on directed sportfishing effort and harvest of launched-boat anglers at two locations near the artificial reef. Anglers at the Calumet Park and Burnham Park launch ramps (Figure 1) were asked two specific questions about the artificial reef: (1) are you aware of the artificial reef and (2) did you fish at the artificial reef? Data analysis.—For analysis of scuba data, the number of fish observed on each dive was standardized by the amount of time swum for that transect. Using 15 min of swimming observation as our unit of effort, we calculated observations per unit effort (OPUE), the number of fish observed per 15 min. We chose 15 min as the unit of effort because it was similar to the mean time taken to swim the reference site. Observations per unit effort served as our standard metric to compare the relative abundance of fish seen among all dives and locations. Round goby relative abundance was represented as estimated percent cover rather than as individuals because of their large numbers, behavior, and movements, which precluded accurate individual counts. Because of the qualitative nature of the round goby estimates, they were not included in analyses of dive data. Alewife schools were also not included in analyses because numbers of fish in each school were a rough estimate. Individual alewives were noted separately and included in analyses of total and mean number of fish. Student’s t-tests, suited for small sample sizes, were used to test for within-year differences between sites in OPUE of all fish combined and of the most common species. Simple linear regression analysis was also run to determine whether
FRESHWATER ARTIFICAL REEF USE
surface water temperature was related to smallmouth bass OPUE. Catch per unit effort (CPUE) for gill nets was calculated as mean number of fish caught per net-night. Daily CPUEs were averaged to develop the annual mean for each sampling year. Because of the various mesh sizes used in different years, mean total annual CPUE was analyzed with general linear models in which location and year were used as main effects and year 3 location was used as the interaction term. Differences in the CPUE of individual species between locations were tested separately by year using Student’s t-tests because catch rates of individual species varied among the different mesh sizes. For example, rock bass were caught only in medium-mesh nets; thus, we detected them by this method only in 2002 and 2003. Gill-net and dive data were combined to determine the presence or absence of fish species at each location on all sampling dates in 2000–2003. Logistic regression models were run for the six most common taxa— freshwater drum Aplodinotus grunniens, gizzard shad Dorosoma cepedianum, rock bass, smallmouth bass, and salmonines (brown trout Salmo trutta, Chinook salmon Oncorhynchus tshawytscha, and lake trout Salvelinus namaycush). Presence or absence was used as the binary response variable (present ¼ 1, absent ¼ 0) to predict the probability that each taxon was present in our samples. Surface water temperature on the corresponding sampling date, location, and the temperature 3 location interaction were the explanatory variables of interest. Logistic regression was run using forward selection and full-rank parameterization with reference coding; the Hosmer and Lemeshow goodness-of-fit test was run to determine the adequacy of the fitted model. We considered P-values less than or equal to 0.10 to be significant for all analyses because of small sample sizes and seasonal variability of CPUE and OPUE. Data were loge(x þ 1) transformed for data analysis to meet assumptions of normality, and test statistics reported are those from transformed data. However, for ease of presentation, actual counts and means are presented in the text, tables, and figures. Means are reported with SDs in the text and tables and with 2 SEs in the figures. Two-sample t-tests assuming unequal variances were used for those years and species lacking homogeneity of variances after transformations. Results Scuba Observations The round goby was the only fish species observed by divers at each site before artificial reef construction (Table 2). Five additional species (alewives, common carp, rock bass, smallmouth bass, and yellow perch)
were observed at the reference site during 2000–2003. Those same six species, along with largemouth bass and freshwater drum, were observed at the artificial reef site after construction. Round goby was the only species present on all dives at both locations both before and after reef construction. Annual mean OPUE of all fishes pooled at the artificial reef ranged from 6.4 6 7.1 fish/15 min in 2003 to 21.5 6 17.0 fish/15 min in 2002 (Table 2). In contrast, mean OPUE at the reference site never exceeded 3.0. Mean annual OPUE at the artificial reef was significantly higher than at the reference site in 3 of the 4 years after construction (2000: t ¼ 2.76, df ¼ 8, P , 0.03; 2002: t ¼ 3.24, df ¼ 6.03, unequal variances, P , 0.02; 2003: t ¼ 3.35, df ¼ 11, P , 0.01). Due to very low sample sizes, numbers observed in 2001 did not differ (t ¼ 2.04, df ¼ 2, unequal variances, P . 0.18); weather conditions during 2001 permitted only three dives at the artificial reef and two at the reference site, which took place between June and early August. Smallmouth bass were observed at the artificial reef site during all years after construction but only at the reference site in 2000 (Table 2). Smallmouth bass were present at the reef beginning in mid-July, and numbers observed generally were highest during late July through mid-September (Figure 2). The lowest average smallmouth bass OPUE (0.8 6 1.4 fish/15 min) occurred in 2001; diving stopped on August 1, 2001, and observations during the usual peak smallmouth bass abundance did not occur. After reef construction, the observed number of smallmouth bass was significantly higher at the artificial reef in all years except 2001 (2000: t ¼ 2.27, df ¼ 8, P , 0.05; 2002: t ¼ 6.39, df ¼ 6, unequal variances, P , 0.001; 2003: t ¼ 3.26, df ¼ 6, unequal variances, P , 0.02). Rock bass also were observed at the artificial reef every year after reef construction but at the reference site only during 2000 and 2002 (Table 2). Rock bass were first observed at the artificial reef in mid-June to early July; highest numbers occurred around August 1 in 2001 and 2002 (Figure 2). Rock bass OPUE at the artificial reef was significantly higher than at the reference site in 2002 and 2003 (2002: t ¼ 2.48, df ¼ 12, P , 0.03; 2003: t ¼ 2.73, df ¼ 6, unequal variances, P , 0.03). Although rock bass mean OPUE was highest at the artificial reef in 2001, differences were nonsignificant due to low sample size and high variability in numbers seen at the reef. Sixteen alewife schools of varying numbers (generally 100–500 fish) were observed on dives at the artificial reef during 2000–2003; seven of these occurred on one sampling date in June 2001. Four alewife schools were observed at the reference site during 2000–2003.
CREQUE ET AL.
TABLE 2.—Total number of fish observed annually and annual mean (SD) number of fish observed per 15 min (observations per unit effort [OPUE]) at Lake Michigan artificial reef and reference sites along scuba transects in 1999–2003. Round goby numbers are presented as percent coverage rather than OPUE. Artificial reef site
Round goby (%) Smallmouth bass Rock bass Yellow perch Common carp Round goby (%) All Smallmouth bass Rock bass Yellow perch Round goby (%) All Smallmouth bass Rock bass Yellow perch Common carp Cyprinus carpio Largemouth bass Mictropterus salmoides Alewife (individuals) Round goby (%) All Smallmouth bass Rock bass Yellow perch Alewife (individuals) Freshwater drum Round goby (%) All
Total number of fish
OPUE or %
75 154 72 5
15.0 8.9 0.8 0.1 0.1 6.7 11.3 0.8 8.0 2.7 13.3 11.5 12.8 5.5 0.4
8.0 1.1 0.4 0.3 5.1 7.4 1.4 12.0 1.6 5.7 13.8 9.9 8.9 0.6
5.1 21.5 2.3 0.7 3.3 0.1 0.1 4.4 6.4
4.9 17.0 2.2 0.7 7.7 0.2 0.2 2.7 7.1
96 12 3 2 113 5 51 19
267 19 7 50 1 1 78
Gill-Net Catches General linear model analysis showed differences in mean annual total CPUE (F9, 131 ¼ 14.89, P , 0.001). Location was not a significant factor before or after reef construction (Figure 3). The effect of year, which allowed us to assess the different mesh sizes used, was a significant factor. Catch per unit effort in 2002 and 2003 was higher than in previous years at both locations due to the addition of the medium-mesh panels (Bonferroni t . 2.86, df ¼ 122, P , 0.05). Examination of CPUEs for individual taxa provided a more detailed picture of which fishes used each site. In 1999, yellow perch were collected at both sites, while round goby were caught at the artificial reef and alewives were captured at the reference site (Figure 4). Common carp, freshwater drum, gizzard shad, rock bass, round goby, smallmouth bass, salmonines, and yellow perch were collected in gill nets at both locations after reef construction (Figures 4, 5). Fish caught infrequently and included in the ‘‘other’’ category were alewives, channel catfish Ictalurus punctatus, burbot Lota lota, rainbow smelt Osmerus mordax, and white suckers Catostomus commersonii. Rock bass, alewives, and round goby were not captured
Total number of fish
OPUE or %
15.0 1.5 0.8
1 1 1
0.1 0.1 0.2
0.3 0.3 0.6
0.2 3.2 0.7
0.5 2.0 0.8
in large-mesh gill nets due to gear selectivity; there were no gill-net data for these species in 2000–2001. Of the nine taxa common to both locations, the CPUEs of four differed between the two locations in at least 1 year. Mean annual CPUE of smallmouth bass was higher at the artificial reef in 2000 (t ¼ 1.87, df ¼ 30, P , 0.07), 2001 (t ¼ 1.87, df ¼ 13, unequal variances, P , 0.09) and 2002 (t ¼ 2.63, df ¼ 30, P , 0.02). Rock bass CPUE was higher at the artificial reef in 2002 (t ¼ 2.94, df ¼ 20.8, unequal variances, P , 0.01) but not in 2003. Catch rates of round goby differed during 1999 (t ¼ 2.04, df ¼ 5, unequal variances, P , 0.10) but not during 2002 or 2003 (Figures 4, 5). Freshwater drum CPUE differed only in 2003 (t ¼ 2.27, df ¼ 15.5, unequal variances, P , 0.04), and the freshwater drum was the only species that was caught more frequently at the reference site than at the artificial reef (Figure 5). Presence/Absence and Temperature Relationships Logistic regression models were not significant for rock bass, yellow perch, or gizzard shad, indicating that surface water temperatures and location were not good predictors of their presence. Surface temperatures predicted salmonine presence (Table 3); the likelihood
FRESHWATER ARTIFICAL REEF USE
FIGURE 3.—Annual mean (62 SEs) number of fish caught per net-night in gill nets at Lake Michigan artificial reef and reference sites during 1999–2003. The numbers of samples collected are indicated above the bars. The dashed line indicates the date of artificial reef construction. Stretch mesh sizes are as follows: small (3.8 and 4.4 cm), medium (5.1 and 7.6 cm), and large (10.2 and 11.5 cm).
FIGURE 2.—Seasonal trends in the numbers of smallmouth bass and rock bass observed per 15 min (observations per unit effort [OPUE]) along scuba transects at a Lake Michigan artificial reef site in 2000–2003, and trends in surface water temperature (8C) in 2002 and 2003. Scuba sampling during 2001 ended in early August due to poor weather conditions.
of salmonines being present at either location increased when temperatures decreased. Salmonines were captured or observed only when water temperatures were below 188C. The best-fit logistic model for freshwater drum (Table 3) indicated that their presence was positively associated with surface water temperatures. With one exception, freshwater drum were present only when surface water temperatures were above 178C. The smallmouth bass was the only species whose presence was related to both surface water temperature and location (Table 3). Adult smallmouth bass were not observed at the artificial reef until water temperatures reached 228C; the exception was an adult observed on July 14, 2003, when surface temperature was 18.88C (Figures 2, 6). During fall, smallmouth bass were observed at the reef on six dates when water temperature was below 228C. Smallmouth bass were present at the artificial reef on all 13 sampling dates during 2000–2003 when surface water temperature was at or above 228C. Smallmouth bass were 5.2 times more likely to be present at the artificial reef than at the
reference site, a result consistent with the patterns we observed with dive and gill-net data. Surface water temperatures also helped to explain how many smallmouth bass were present at the artificial reef (Figure 6). Linear regression revealed that smallmouth bass OPUE was positively related to surface water temperature (adjusted R2 ¼ 0.36, df ¼ 23, P , 0.002). For example, water temperatures during 2002 warmed quickly; 54 d between July 1 and September 30 were at or above 228C, whereas only 35 d in 2003 were above the threshold (Figure 2). Larger numbers of smallmouth bass were observed during 2002 than during 2003. A reduction in smallmouth bass numbers during late August 2002 probably was related to a surface water temperature decline to 16.88C caused by an upwelling event. Creel Survey The percentage of interviewed anglers aware of the Lake Michigan artificial reef ranged from 53% to 79% during 2000–2003 (Table 4). No interviewed anglers used the artificial reef in 2002, and less than 6% used the reef in 2000, 2001, and 2003. Of the three parties that fished the reef in 2000, two targeted salmon and one did not target specific taxa. In 2001, the party that fished the reef targeted and caught yellow perch. The party of two anglers that fished at the artificial reef in 2003 targeted smallmouth bass; they caught and released 4 smallmouth bass and 20 rock bass on July 29. Anglers who were aware of the reef but did not fish
CREQUE ET AL.
FIGURE 4.—Annual mean (62 SEs) number of fish caught per net-night at Lake Michigan artificial reef (black bars) and reference sites (white bars) in (A) small-mesh (3.8- and 4.4cm) gill nets during 1999, (B) large-mesh (10.2- and 11.5-cm) gill nets during 2000, and (C) large-mesh gill nets during 2001. Species codes are as follows: CAP ¼ common carp, FRD ¼ freshwater drum, GBY ¼ round goby, GZS ¼ gizzard shad, RKB ¼ rock bass, SAL ¼ salmonines, SMB ¼ smallmouth bass, YEP ¼ yellow perch, and OTH ¼ other. Asterisks indicate significant differences between locations for a given species.
it primarily targeted salmonines, followed by yellow perch and then black bass Micropterus spp. (Table 4). Discussion Our results demonstrated that fish did begin using the artificial reef quickly—within 6 months of construction. Although smallmouth bass and rock bass did preferentially use the reef, they did so only on a seasonal basis. In fact, CPUE for all fish did not differ between the artificial reef and our nearby reference site. Water temperature played a more-important role in determining presence than did structure for most of the fishes we encountered. Structurally, the reef is holding up well, and there is no evidence of it sinking into the bottom or silting in.
FIGURE 5.—Annual mean (62 SEs) number of fish caught in medium-mesh (5.1- and 7.6-cm) and large-mesh (10.2- and 11.5-cm) gill nets at Lake Michigan artificial reef and reference sites during (A) 2002 and (B) 2003. Species codes are defined in Figure 4. Asterisks indicate significant differences between locations for a given species.
Colonization of artificial reefs by fish occurs rapidly after deployment in marine systems (Bohnsack and Sutherland 1985; Bohnsack et al. 1994). We observed adult fish using the artificial reef in Lake Michigan within 6 months after construction, similar to other studies in the Great Lakes (Gannon et al. 1985; Kevern et al. 1985; Kelch et al. 1999). However, species richness was similar at both locations after construction of the artificial reef. Although fish species diversity was high at rock jetties in eastern Lake Michigan, diversity was also high at natural areas similar to our reference site (Liston et al. 1985). Artificial reefs are often noted for increasing catch rates and attracting more fish (Brown 1986; Bohnsack et al. 1991; Stone et al. 1991; Grossman et al. 1997), a primary motive for reef construction. We observed more fish at the artificial reef during scuba transects, but gill-net CPUE did not differ between locations in our study. Artificial reefs also did not significantly increase gill-net CPUE relative to that of reference sites in Lakes Erie, Ontario (Bohnsack et al. 1991), and Michigan (Liston et al. 1985). In terms of individual species, the artificial reef did
FRESHWATER ARTIFICAL REEF USE
TABLE 3.—Results of significant logistic regression models for the probability that each taxon would be present in gill-net and dive samples at Lake Michigan artificial reef and reference sites using surface water temperature, location and temperature 3 location as explanatory variables (CI ¼ confidence interval; temperature ¼ surface water temperature [8C]); models for yellow perch, gizzard shad, and rock bass yielded no explanatory variables. Odds ratio Response variable Salmonines Freshwater drum Smallmouth bass
Parameter Intercept Temperature Intercept Temperature Intercept Location Temperature
Estimate (SE) 3.37 0.25 6.94 0.35 4.65 1.65 0.17
(1.65) (0.09) (1.88) (0.09) (1.77) (0.59) (0.08)
4.19 7.42 13.58 13.67 6.94 7.76 4.23
attract larger numbers of smallmouth bass and rock bass than did the reference site. Although these differences were not always statistically significant due to low sample sizes and seasonal variability, the empirical differences were clear. These two species strongly prefer structure, and numerous studies recognize that centrarchids are the freshwater fishes most attracted to artificial habitat (Prince et al. 1985; Bohnsack et al. 1991; Kelch et al. 1999). Rock bass were observed in close association with this reef, similar to observations at other artificial reefs (Gannon et al. 1985). In Lake Ontario, rock bass did not remain at one location for extended periods (Storr et al. 1983). This was also the case at the artificial reef, where numbers of rock bass fluctuated throughout the sampling season. Smallmouth bass and rock bass accounted for the majority of all fish observed during scuba transects at the artificial reef in Lake Michigan and at two artificial reefs in Lake Erie (Kelch et al. 1999). However,
P 0.041 0.007 0.001 0.001 0.009 0.005 0.040
Hosmer–Lemeshow test P
numbers of smallmouth bass associated with artificial structure were much higher in Lake Erie, despite having similar dimensions, size of material, and depth of placement as this artificial reef in Lake Michigan. The Lakewood Reef in Lake Erie was constructed with scrap concrete ranging from a diameter of 20–30 cm to the size of cars and was placed in 8.5 m of water; its length was 243 m, mean width was approximately 18.3 m, and height was approximately 2.4 m (D. Kelch, Ohio Sea Grant, personal communication). Because transect times were approximately equal in both studies, we compared mean annual number of fish per dive. Annual mean number of smallmouth bass observed ranged from 1.7 6 2.9 to 22.0 6 17.0 fish/ dive at the Lake Michigan artificial reef over 4 years. At the Lakewood Reef in Lake Erie, mean numbers of smallmouth bass observed in 1994 and 1995 were 369.3 6 656.2 and 69.5 6 42.5 fish/dive, respectively (12.8 and 3.5 fish/dive, respectively, at the reference site; Kelch et al. 1999). Mean annual number of rock bass observed at the Lake Michigan artificial reef (1.0– 17.0 fish/dive) fell within the range observed at the two Lake Erie reefs (0.8–70.5 fish/dive; Kelch et al. 1999). Smallmouth bass were observed at artificial reefs in Lake Erie during May, a full 2.5 months earlier than their appearance at the Lake Michigan artificial reef TABLE 4.—Results of angler interviews at Calumet and Burnham Park launch ramps to gauge awareness and use of a Lake Michigan artificial reef during 2000–2003. Target percentages were calculated based on anglers who were aware of the reef but did not fish it.
FIGURE 6.—Relationship between the loge(x þ 1) transformed number of adult smallmouth bass observed per 15 min along scuba transects and surface water temperatures at an artificial reef site in Lake Michigan during 2000–2003.
Total interviewed Anglers aware of the reef (%) Anglers who used the reef (%) Anglers targeting black bass (%) Anglers targeting yellow perch (%) Anglers targeting salmonines (%) Anglers targeting walleyes (%)
189 53 6 21 34 44 0
181 79 2 27 52 21 0
319 63 0 16 33 49 0
188 72 1.5 30 37 35 0
CREQUE ET AL.
(Kelch et al. 1999). Lake Erie warms more quickly during spring and reaches higher temperatures in summer than does Lake Michigan. Thus, smallmouth bass were most numerous during spring and fall at the Lake Erie reefs (Kelch et al. 1999) and during summer at the Lake Michigan artificial reef. Smallmouth bass spawn nearshore at temperatures of 15–18.38C (Armour 1993); no smallmouth bass spawning activity was observed at the Lake Michigan artificial reef. Instead, adult smallmouth bass were attracted to the artificial reef from shallow nearshore areas on a seasonal basis after nest guarding was complete and after surface water temperatures warmed above 228C. Sharp midsummer temperature declines due to upwellings caused smallmouth bass to temporarily leave the artificial reef. This relationship between temperature and artificial reef use is supported by other studies in Lakes Michigan and Ontario; very few or no smallmouth bass and rock bass occurred at artificial structures in areas with frequent cold water upwellings that reduced summer temperatures as low as 4–158C (Gannon et al. 1985; Kevern et al. 1985; Liston et al. 1985). Based on our observations, smallmouth bass left the reef in October when temperatures declined to 14– 178C. This pattern was consistent with riverine and lacustrine data indicating that smallmouth bass initiated winter migrations and hibernate when temperatures fell below 10–168C (Armour 1993). Both our results and those of Kelch et al. (1999) demonstrated that smallmouth bass will be attracted to artificial reefs in the Great Lakes during periods of preferred water temperatures, although the numbers observed differed greatly from system to system. Structural habitat is not lacking in the Chicago area because the entire shoreline area is armored with riprap, concrete, or sheet piling. Many structural features, such as breakwalls, marinas, and water intake cribs, are also present. Fish attracted to structure in this region of the lake may be using these features more than the reef. In addition, Lake Erie is more productive than Lake Michigan (Fahnenstiel et al. 1998; Nicholls et al. 2001); as a result, populations of smallmouth bass available to utilize artificial reefs are larger in Lake Erie. Lake Michigan is more oligotrophic (Fahnenstiel et al. 1998; Makarewicz et al. 1998; Carrick et al. 2001) and is dominated by coldwater species (including salmonines) that are less dependent on structural habitats. Most freshwater fishes are mobile and respond to changes in temperature, currents, food supply, or a combination thereof, rather than to available structure. Freshwater drum and salmonines exhibited clear responses to temperature rather than location. Freshwater drum CPUE was higher at the reference site
in 2003, indicating structure was not driving their presence in the nearshore area. The presence of salmonines at both sites was low but was correlated with water temperatures below 188C. Salmonines are highly mobile piscivores and may not depend on rocky substrates except for spawning. In other Great Lakes studies, use of artificial structures by salmonines was seasonally limited and did not differ from use of reference areas (Gannon et al. 1985; Kevern et al. 1985; Liston et al. 1985). The numbers of alewives and gizzard shad were not significantly higher at our artificial reef or other artificial structures in Lake Michigan than the numbers observed in natural habitats (Kevern et al. 1985; Liston et al. 1985), indicating no additional incentive exists at artificial structures for these pelagic species. For example, alewives were only loosely associated with artificial reefs near two power plants in eastern Lake Michigan because they fed on zooplankton in the currents from the intake structures (Rutecki et al. 1985). Yellow perch CPUE declined in early summer at both sites when surface water temperatures rose above 188C, and it exhibited similar temperature-related patterns at other artificial structures in Lake Michigan (Kevern et al. 1985; Liston et al. 1985; Rutecki et al. 1985). The Lake Michigan artificial reef also created prime habitat for the round goby, an aggressive invasive benthic fish. Round goby began expanding from Lake St. Clair in 1993 and were well established in the central basin of Lake Erie in August 1994 and in the Chicago region of Lake Michigan by 1995 (Marsden and Jude 1995). However, studies at the Lake Erie Lakewood Reef in 1994 and 1995 observed no round goby (Kelch et al. 1999); the species was first seen in 1997 at this reef (D. Kelch, personal communication). During our scuba observations, round goby were most numerous in May and June and congregated on the tops of rocks. By July, they were primarily hidden in crevices, possibly due to predator avoidance after the arrival of smallmouth bass. Although each of our sampling methods had its own bias, these were offset by the methods’ complementary benefits. The presence of divers can influence the behavior of some taxa and can therefore affect detection (Bortone et al. 2000). However, we set up stationary cameras on several dates in 2002 to record fish in the absence of divers. No fish species were recorded that were not observed by divers during transects, just as Kelch et al. (1999) found in Lake Erie. Smallmouth bass and rock bass are well suited for dive observations (Mueller 2003) because they show little reaction to divers and their movement is greatest at midday; nocturnal activity is rare outside of the spawning season (Gerber and Haynes 1988; Hinch
FRESHWATER ARTIFICAL REEF USE
and Collins 1991). Setting gill nets overnight allowed us to capture nocturnally active species and those not easily observed by divers. Gill-net catches were limited by the range of mesh sizes and were not as efficient as dive observations at capturing less-nocturnally-active centrarchids. Our study is one of the few in the Great Lakes that have evaluated artificial reefs using more than one collection or survey technique. Using the two gear types in combination provided a more complete picture of fish attraction to the artificial reef versus the reference site than could be provided by either gear alone. Both methods were not as efficient at evaluating juvenile abundance as they were at evaluating adult fish. However, because the artificial reef was constructed to attract sport fish and anglers, we concentrated on fish in the catchable population. Management Implications The artificial reef in southern Lake Michigan was installed to attract smallmouth bass and increase angling opportunities. However, angler use of the Lake Michigan reef has been minimal. The majority of Lake Michigan anglers target salmonines and yellow perch (Brofka and Dettmers 2004), which were not attracted to the reef in large numbers or for long periods of time. Our creel results indicated that only 16–30% of anglers who were aware of the artificial reef were targeting black bass; of the few anglers (11) who actually fished the reef, only two targeted bass. The relatively few smallmouth bass and rock bass attracted to the reef were present only during mid-July–October and may not be a sufficiently strong incentive to draw anglers. Smallmouth bass anglers in the Chicago area already have success fishing close to shore near harbors and breakwalls. As Talhelm (1985) noted, ‘‘an increase in angling quality over a reef would have little or no benefits if the same kind of angling were already available at a closer location.’’ In the central basin of Lake Erie, however, most sportfishing was previously done far offshore (Hushak et al. 1999; Kelch et al. 1999). Therefore, the abundant smallmouth bass present from May to October at these nearshore artificial reefs did attract anglers in large numbers. During 1992, 4 years after construction of the Lorain artificial reef in Lake Erie, 87% of people interviewed were aware of the reef; of those, 64% had used the artificial reef for fishing or diving and 40% of anglers fishing the reef targeted bass (Hushak et al. 1999). To maximize angler use in large aquatic systems such as the Great Lakes, artificial reefs designed to attract centrarchids should be constructed in easily accessible nearshore areas with nearby launch ramps (Radonski et al. 1985; Kelch et al. 1999). The Lake Michigan artificial reef is 2.8 km offshore, 1.6 km
farther offshore than the Lorain Reef. The closest launch ramp to the artificial reef at Jackson Park was unusable by many boats during the past several years because of low water levels and various construction projects, making angler access to the artificial reef difficult. In addition, the Lake Michigan artificial reef has been poorly marked. Lake Erie anglers enjoyed fishing at the Lorain Reef because it was close to shore and launch ramps; was well marked; and had high catch rates (Hushak et al. 1999). The Lake Erie reefs were specifically placed in locations that were easily accessible to those with small boats (,6.7 m; D. Kelch, personal communication). Thus, future structures for attracting smallmouth bass in Lake Michigan should be positioned closer to shore and launch facilities and should be well marked to attract more anglers. Our study demonstrated that the success of artificial reefs in the Great Lakes is not guaranteed. Managers considering artificial reef construction in large oligotrophic lakes must recognize that most species will be transients rather than year-round residents of artificial reefs. Managers must also realize that artificial reefs could create additional habitat for exotic species, such as the round goby and dreissenid mussels. Variables such as water depth, temperature, food supply, building materials, and configuration need to be taken into account so that future reefs can be constructed in optimal locations to meet particular management goals. Based on our results, we offer the following insights for managers considering the use of artificial structures to enhance black bass in the Great Lakes and other large water bodies. Reef planners should ensure heterogeneity in the width and height of artificial reefs to attract the largest number of fish species. We observed adult smallmouth bass most frequently at the highest peaks and in the middle of the widest sections of the artificial reef, whereas rock bass and juvenile smallmouth bass and largemouth bass were more commonly observed in the lower valleys and along the edges. Our results demonstrate that reefs located at a depth of 7.5 m are suitable to attract both of these species as well as to ensure structural stability. To date, we have observed no evidence of silting in, ice-scour, or wave-action damage at the artificial reef. Because smallmouth bass use of artificial reefs was so closely correlated with warm temperatures, reefs for centrarchids should be located in areas that are less susceptible to upwelling and thermoclines to maintain warmwater temperatures throughout the summer. In addition to these physical and biological considerations, artificial reefs should also be located in areas that are easily accessible to shore and boat anglers to maximize use. Our experience has shown that a distance of more than 7 km was likely a deterrent to boaters fishing the Lake
CREQUE ET AL.
Michigan artificial reef. Constructing reefs nearshore and within 4 km of launch ramps may result in higher angler use. Artificial reefs properly located and constructed will attract fish. However, the degree of attraction varies substantially within and across lakes. Thus, managers need to set realistic expectations for reef performance based on system characteristics and the primary goal(s) for each structure. Acknowledgments We thank B. Pientka, D. Glover, A. Jaeger, S. Miehls, and numerous staff at the Lake Michigan Biological Station for their help with data collection. We also thank D. Kelch of Ohio Sea Grant for providing us with additional information on artificial reefs in Lake Erie. D. Clapp, S. Robillard, and three anonymous reviewers provided useful comments on earlier drafts of this manuscript. Funding was provided by the Federal Aid in Sport Fish Restoration Act under projects F-52-R and F-138-R administered through the Illinois Department of Natural Resources, and by the Illinois Natural History Survey. References Armour, C. L. 1993. Evaluating temperature regimes for protection of smallmouth bass. U.S. Fish and Wildlife Service, Resource Publication 191, Fort Collins, Colorado. Binkowski, F. P. 1985. Utilization of artificial reefs in the inshore areas of Lake Michigan. Pages 349–362 in F. M. D’Itri, editor. Artificial reefs: marine and freshwater applications. Lewis Publishers, Chelsea, Michigan. Bohnsack, J. A. 1989. Are high densities of fishes at artificial reefs the result of habitat limitation or behavioral preference? Bulletin of Marine Science 44:631–645. Bohnsack, J. A., and D. L. Sutherland. 1985. Artificial reef research: a review with recommendations for future priorities. Bulletin of Marine Science 37:11–39. Bohnsack, J. A., D. E. Harper, D. B. McClellan, and M. Hulsbeck. 1994. Effects of reef size on colonization and assemblage structure of fishes at artificial reefs off southeastern Florida, U.S.A. Bulletin of Marine Science 55:796–823. Bohnsack, J. A., D. L. Johnson, and R. F. Ambrose. 1991. Ecology of artificial reef habitats and fishes. Pages 61– 108 in W. Seaman, Jr. and L. M. Sprague, editors. Artificial habitats for marine and freshwater fisheries. Academic Press, San Diego, California. Bortone, S. A., and J. J. Kimmel. 1991. Environmental assessment and monitoring of artificial habitats. Pages 177–236 in W. Seaman, Jr. and L. M. Sprague, editors. Artificial habitats for marine and freshwater fisheries. Academic Press, San Diego, California. Bortone, S. A., M. A. Samoilys, and P. Francour. 2000. Fish and macroinvertebrate evaluation methods. Pages 127– 164 in W. Seaman, Jr., editor. Artificial reef evaluation: with application to natural marine habitats. CRC Press, New York.
Brofka, W. A., and J. M. Dettmers. 2004. A survey of sport fishing in the Illinois portion of Lake Michigan. Illinois Natural History Survey, Aquatic Ecology Technical Report 04/02, Champaign. Brown, A. M. 1986. Modifying reservoir fish habitat with artificial structures. Pages 98–102 in G. E. Hall and M. J. Van Den Avyle, editors. Reservoir fisheries management: strategies for the 80 0 s. Southern Division American Fisheries Society, Reservoir Committee, Bethesda, Maryland. Carrick, H., R. P. Barbiero, and M. Tuchman. 2001. Variation in Lake Michigan plankton: temporal, spatial, and historical trends. Journal of Great Lakes Research 27:467–485. Chrzastowski, M. J., D. B. Ketterling, and C. J. Stohr. 1998. 1998 bathymetric survey in the vicinity of the IDNRproposed artificial reef for recreational fishing off the Chicago lakeshore. Illinois State Geological Survey, Open File Series 1998–4, Champaign. Dahlberg, M. D. 1981. Nearshore spatial distribution of fishes in gill net samples, Cayuga Lake, New York. Journal of Great Lakes Research 7:7–14. Fahnenstiel, G. L., A. E. Krause, M. J. McCormick, H. J. Carrick, and C. L. Schelske. 1998. The structure of the planktonic food-web in the St. Lawrence Great Lakes. Journal of Great Lakes Research 24:531–554. Gannon, J. E., R. J. Danehy, J. W. Anderson, G. Merritt, and A. P. Bader. 1985. The ecology of natural shoals in Lake Ontario and their importance to artificial reef development. Pages 113–140 in F. M. D’Itri, editor. Artificial reefs: marine and freshwater applications. Lewis Publishers, Chelsea, Michigan. Gannon, J. E., editor. 1990. International position statement and evaluation guidelines for artificial reefs in the Great Lakes. Great Lakes Fishery Commission, Special Publication 90–2, Ann Arbor, Michigan. Gerber, G. P., and J. M. Haynes. 1988. Movements and behavior of smallmouth bass, Micropterus dolomieu, and rock bass, Ambloplites rupestris, in south-central Lake Ontario and two tributaries. Journal of Freshwater Ecology 4:425–440. Grossman, G. D., G. P. Jones, and W. J. Seaman, Jr. 1997. Do artificial reefs increase fish production? A review of existing data. Fisheries 22(4):17–23. Haynes, J. M. 1995. Thermal ecology of salmonids in Lake Ontario. Great Lakes Research Review 2:17–22. Hinch, S. G., and N. C. Collins. 1991. Importance of diurnal and nocturnal nest defense in the energy budget of male smallmouth bass: insights from direct video observations. Transactions of the American Fisheries Society 120:657– 663. Ho¨o¨k, T. O., E. S. Rutherford, S. J. Brines, D. J. Schwab, and M. J. McCormick. 2004. Relationship between surface water temperature and steelhead distributions in Lake Michigan. North American Journal of Fisheries Management 24:211–221. Hushak, L. J., D. O. Kelch, and S. J. Glenn. 1999. The economic value of the Lorain County, Ohio artificial reef. Pages 348–362 in L. R. Benaka, editor. Fish habitat: essential fish habitat and rehabilitation. American Fisheries Society, Symposium 22, Bethesda, Maryland. Kelch, D. O., F. L. Snyder, and J. M. Ruetter. 1999. Artificial
FRESHWATER ARTIFICAL REEF USE
reefs in Lake Erie: biological impacts of habitat alteration. Pages 335–347 in L. R. Benaka, editor. Fish habitat: essential fish habitat and rehabilitation. American Fisheries Society, Symposium 22, Bethesda, Maryland. Kevern, N. R., W. E. Biener, S. R. VanDerLann, and S.D. Cornelius. 1985. Preliminary evaluation of an artificial reef as a fishery management strategy in Lake Michigan. Pages 443–458 in F. M. D’Itri, editor. Artificial reefs: marine and freshwater applications. Lewis Publishers, Chelsea, Michigan. Liston, C. R., D. C. Brazo, J. R. Bohr, and J. A. Gulvas. 1985. Abundance and composition of Lake Michigan fishes near rock jetties and a breakwater, with comparisons to fishes in nearby natural habitats. Pages 491–516 in F. M. D’Itri, editor. Artificial reefs: marine and freshwater applications. Lewis Publishers, Chelsea, Michigan. Makarewicz, J. C., P. Bertram, and T. W. Lewis. 1998. Changes in phytoplankton size-class abundance and species composition coinciding with changes in water chemistry and zooplankton community structure of Lake Michigan, 1983 to 1992. Journal of Great Lakes Research 24:637–657. Marsden, J. E., and D. J. Jude. 1995. Round gobies invade North America. Illinois-Indiana Sea Grant Program, ILIN-SG-95-10, West Lafayette, Indiana. McGurrin, J. M., R. B. Stone, and R. J. Sousa. 1989. Profiling United States artificial reef development. Bulletin of Marine Science 44:1004–1013. Mueller, K. W. 2003. A comparison of electrofishing and scuba diving to sample black bass in Western Washington lakes. North American Journal of Fisheries Management 23:632–639. Nicholls, K. H., G. J. Hopkins, S. J. Standke, and L. Nakamoto. 2001. Trends in total phosphorus in Canadian near-shore waters of the Laurentian Great Lakes: 1976– 1999. Journal of Great Lakes Research 27:402–422. Olson, R. A., J. D. Winter, D. C. Nettles, and J. M. Haynes. 1988. Resource portioning in summer by salmonids in south-central Lake Ontario. Transactions of the American Fisheries Society 117:552–559.
Prince, E. D., O. E. Maughan, and P. Brouha. 1985. Summary and update of the Smith Mountain Lake artificial reef project. Pages 401–430 in F. M. D’Itri, editor. Artificial reefs marine and freshwater applications. Lewis Publishers, Chelsea, Michigan. Radonski, G. C., R. G. Martin, and W. P. DuBose IV. 1985. Artificial reefs: the sport fishing perspective. Pages 529– 536 in F. M. D’Itri, editor. Artificial reefs marine and freshwater applications. Lewis Publishers, Chelsea, Michigan. Rutecki, T. L., J. A. Dorr, III, and D. J. Jude. 1985. Preliminary analysis of colonization and succession of selected algae, invertebrates, and fish on two artificial reefs in inshore southeastern Lake Michigan. Pages 459– 490 in F. M. D’Itri, editor. Artificial reefs: marine and freshwater applications. Lewis Publishers, Chelsea, Michigan. Sheey, D. J. 1985. New approaches in artificial reef design and applications. Pages 253–264 in F. M. D’Itri, editor. Artificial reefs: marine and freshwater applications. Lewis Publishers, Chelsea, Michigan. Stone, R. B. 1985. History of artificial reef use in the United States. Pages 3–12 in F. M. D’Itri, editor. Artificial reefs: marine and freshwater applications. Lewis Publishers, Chelsea, Michigan. Stone, R. B., J. M. McGurrin, L. M. Sprague, and W. Seaman, Jr. 1991. Artificial habitats of the world: synopsis and major trends. Pages 31–60 in W. Seaman, Jr. and L. M. Sprague, editors. Artificial habitats for marine and freshwater fisheries. Academic Press, San Diego, California. Storr, J. F., P. J. Hadden-Carter, and J. M. Myers. 1983. Dispersion of rock bass along the south shore of Lake Ontario. Transactions of the American Fisheries Society 112:618–628. Talhelm, D. R. 1985. The economic impact of artificial reefs on Great Lakes sport fisheries. Pages 537–544 in F. M. D’Itri, editor. Artificial reefs: marine and freshwater applications. Lewis Publishers, Chelsea, Michigan.