Ecological Archives A017-037-A2

Julie B. Kellner, Irene Tetreault, Steven D. Gaines, and Roger M. Nisbet. 2007. Fishing the line near marine reserves in single and multispecies fisheries. Ecological Applications 17:1039–1054.

Appendix B. Empirical study of fish densities across a no-take marine reserve boundary.

To explore the effect of harvesting near a marine reserve boundary, I. Tetreault sampled the spatial pattern of fish densities as a function of distance from the boundary, both inside and outside of a no-take marine reserve, the Catalina Marine Science Center Marine Life Refuge (CMLR). The goals of the study were to determine (1) whether harvesting impacted fish densities outside the CMLR boundary and (2) whether there was a significant relationship between fish densities outside the CMLR and distance from the boundary. If the reserve protects fish from harvest within its boundaries, there would be significantly more fish inside the reserve compared to outside. An alternative hypothesis is that the habitat within the reserve supports higher densities of fish. Because there is a peak fishing season, any effects from harvest would be more apparent during or shortly after the fishing season, but less so during the off season. A finding of significantly more fish inside the reserve during both the fishing and off seasons would suggest that any differences in habitat are driving fish density patterns across the boundary. Thus the comparison of fish densities during two different seasons presents a natural experiment of a fishing effect and a control.

The CMLR is located off the northwest coast of Santa Catalina Island in southern California (33º26' N, 118º28' W). The 0.13 km2 reserve was established in 1988, and California Fish and Game Code 10932 established its official boundaries. Substrate habitat, kelp coverage and relief of the CMLR and nearby reefs consist of rocky reefs, mud/sand substrate and kelp beds (Lowe et al. 2003). Carr (1994) has shown that fish density is correlated to the habitat characteristics of substrate relief and density of giant kelp (Macrocystis pyrifera). To minimize variation in fish densities related to habitat type, the study site (both inside and outside the reserve) was limited to shallow rocky reefs with giant kelp, excluding the sandy-bottom habitat common in Big Fisherman Cove within the CMLR. Thus the study site spanned the contiguous reefs across the eastern reserve boundary at Blue Cavern, extending 700 m west of the boundary (inside) and 500 m southeast (outside) (Fig. B1). Density estimates were focused on two common nearshore harvested fish species in this region: Paralabrax clathratus (kelp bass) and Semicossyphus pulcher (California sheephead). Harvesting consists mainly of recreational angling, and for California sheephead there is also a nominal commercial fishery. The minimum size limit for both species is 30.5 cm (12 inches) total length.

Scuba divers sampled densities of kelp bass and California sheephead in September and October 2001 (“Fall” or “fishing season”) and in January 2002 (“Winter” or “off season”). Sampling in the Fall took place from September 9 through October 9, 2001; Winter from January 22 to 24, 2002. Winter sampling was limited to January due to weather conditions. A total of 222 distinct transects were sampled at distances 0 to 700 m from the reserve boundary. Divers conducted standardized visual underwater surveys, recording the number and size of all kelp bass and California sheephead on benthic transects within a 180 m3 volume of water 2 m wide, 3 m high, and 30 m long, while swimming for approximately 2 min. Transect depths ranged from 3 to 13 m deep. Three to five transects were sampled at pre-determined depths within a defined distance from the reserve boundary (e.g., 4.3, 7.9, and 11.6 m deep within the area 60–90 m outside the reserve). Visibility for all dives was at least 3 m. To avoid disturbing the area prior to sampling (which might bias density estimates), transect tapes were allowed to reel out behind divers. Divers recorded only those fish in front of them to avoid double counting individuals. All divers were trained to estimate fish total length to within 10% of actual length. Divers began training with plywood fish-shaped cutouts; each diver estimated lengths of “fish” of various sizes and shapes, from different distances. Divers then estimated and measured actual fish by placing a ruled underwater slate next to California sheephead and other docile species. Fish size calibration continued on each dive using the ruled slates.

Impact of Fishing Outside the Reserve

To explore whether density differences inside versus outside the reserve were due to fishing impacts on the system or to habitat differences, ANOVAs were used to test for spatial and temporal differences in fish densities of legal-sized individuals (≥ 30 cm TL). A reciprocal transformation [-1/(Y+1)2] equalized within-group variances of the extremely skewed raw data. The fixed effects of reserve and season and their interaction were tested using a full two-factor model with transects as replicates. The interaction term of reserve by season was significant for both species combined (df = 1, 435, F = 4.236, P = 0.040, see Fig. B2), thus the effect of reserve within each season was analyzed separately, using post hoc tests. There were significantly more fish inside versus outside the reserve in Fall (df = 1, 435, F = 45.160, P<0.001). Mean densities (per 180 m3) for legal-sized kelp bass in the Fall were 1.67 ± 0.18 inside the reserve and 0.51 ± 0.12 outside; for California sheephead, densities were 1.04 ± 0.14 inside the reserve and 0.22 ± 0.07 outside. In contrast, no significant difference was found in mean densities of fish during Winter (df = 1, 435, F = 1.536, P = 0.216). The mean densities for kelp bass in Winter were 3.07 ± 0.61 inside the reserve and 1.61 ± 0.32 outside; for California sheephead, 1.87 ± 0.32 inside the reserve and 1.39 ± 0.38 outside.

These findings suggest that, in Fall 2001 (fishing season), the reduced fish densities for both species outside the reserve (compared to inside) were due to fishing. Next the data were examined for any spatial differences in the densities of California sheephead outside the reserve.

Density in Relation to Distance from the Reserve Boundary

The model described here predicts that if harvesters are fishing-the-line in a multispecies fishery, then densities of fish with more limited mobility will be lower immediately adjacent to the reserve boundary than further away. In contrast, the more mobile species will not exhibit this pattern. For these analyses, transect data were categorized into 100-m sections with respect to distance from the reserve boundary, and by season. Each category includes at least six transects per section per season sampled at a minimum of three depths. Spatial differences were tested by comparing whether fish densities differed among the five 100-m sections outside the reserve in Fall. The ANOVA model included the single, fixed factor of “section,” using transects as replicates. Mean densities among sections were very similar for kelp bass outside the reserve in the Fall (Fig. 5C). However, for sheephead, there was a significant difference among mean densities of the five 100-m sections (df = 4, 58, F = 2.558, P = 0.048). A post hoc test showed that mean fish density within the first 100-m section from the reserve boundary was significantly lower than the combined means of the outer four 100-m sections (df = 1, 58, F = 5.666, P = 0.021). In the area outside the reserve, from the boundary to 100 m south, no legal-sized sheephead were observed (n = 21 transects); for the section 100–200 m from the boundary, mean density was 0.33 ± 0.26; within 200–300 m, 0.25 ± 0.18, within 300–400 m, 0.50 ± 0.19, for 400–500 m, 0.167 ± 0.167. Note that the 100-m outside section closest to the reserve showed the greatest rebound in sheephead density between seasons (from 0/180 m3 in Fall to mean 2.67 ± 0.84 fish/180 m3 in Winter, n = 6). These findings suggest that fishing-the-line behavior occurred within the 100-m section close to the reserve boundary during Fall 2001 (the fishing season).

In summary, results from analyses of the density data suggest that fishing is impacting fish densities outside the reserve for both species, and that for California sheephead the area immediately adjacent to the reserve has lower density than areas farther from the boundary.

   FIG. B1. Map of study area at Catalina Marine Life Refuge. Dashed lines mark the reserve boundary, solid lines transecting the coastline mark 100-m sections in relation to the southeast boundary at Blue Cavern Point (700 m inside and 500 m outside).


   FIG. B2. Mean fish densities for kelp bass and California sheephead combined, depicting the significant interaction between the factors “reserve” and “season.” Fish densities were significantly greater inside the reserve in Fall 2001 (fishing season), but not in Winter 2002 (off- fishing season). Data are transformed mean densities and standard errors of the means from the ANOVA model.




Carr, M. H. 1994. Effects of macroalgal dynamics on recruitment of a temperate reef fish. Ecology 75:1320–1333.

Lowe, C. G., D. T. Topping, D. P. Cartamil, and Y. P. Papastamatiou. 2003. Movement patterns, home range, and habitat utilization of adult kelp bass Paralabrax clathratus in a temperate no-take marine reserve. Marine Ecology Progress Series 256:205–216.

[Back to A017-037]