Contributed Paper
Biomass-based targets and the management of multispecies coral reef fisheries T. R. McClanahan,∗ ¶ N. A. J. Graham,† M. A. MacNeil,‡ and J. E. Cinner† ∗
Wildlife Conservation Society, Bronx, NY 10460, U.S.A. †Australian Research Council Centre of Excellence for Coral Reef Studies James Cook University, Townsville, QLD 4811, Australia ‡Australian Institute of Marine Science, Townsville, QLD 4810, Australia
Abstract: The failure of fisheries management among multispecies coral reef fisheries is well documented and has dire implications for the 100 million people engaged in these small-scale operations. Weak or missing management institutions, a lack of research capacity, and the complex nature of these ecosystems have heralded a call for ecosystem-based management approaches. However, ecosystem-based management of coral reef fisheries has proved challenging due to the multispecies nature of catches and the diversity of fish functional roles. We used data on fish communities collected from 233 individual sites in 9 western Indian Ocean countries to evaluate changes in the site’s functional composition and associated lifehistory characteristics along a large range of fish biomass. As biomass increased along this range, fish were larger and grew and matured more slowly while the abundance of scraping and predatory species increased. The greatest changes in functional composition occurred below relatively low standing stock biomass (15 years old and larger than 5 km2 or lightly fished locations (e.g., the Maldives), we estimated the reference point biomass in the region (B0 ) at 1200 kg/ha (±110, 95% CI; n = 47) (McClanahan et al. 2011). To examine how our results fit with targets for fisheries management and conservation, we overlaid the multispecies maximum sustainable yield (BMMSY ) window estimates for the region with our estimated biomass. The BMMSY is an extension of the well-known single species maximum sustainable yield (MSY) estimate, where fisheries targets are based on historical stock biomass and lightly fished areas in the seascape (Hilborn 2002; Worm et al. 2009). While single-species MSY estimates vary by species and environment, BMMSY has a wider confidence interval or window; values typically range from 0.25 to 0.50 B0 (Thorson et al. 2012). While this larger window lacks precision, it is a conservative estimate for reef fisheries, which are typically data poor. Furthermore, the lower end of this window (0.25 B0 ) corresponds to many of the detrimental ecological impacts of fishing on reefs (McClanahan et al. 2011). Consequently, the convergence of maximum yields for many species and the indirect ecological impacts of fishing create a useful heuristic for evaluating the state of the fish stocks relative to fisheries production goals (Worm et al. 2009). Based on the pristine biomass (B0 ) estimate for the region, we
McClanahan et al.
5
Figure 3. Marginal estimated relationship between modeled log biomass and 9 feeding categories based on a hierarchical normal model (black line, line of best fit; open circles, sites on atolls; filled circles, sites not on atolls). used 300 kg/ha (±28, 95% CI) to 600 kg/ha (±54, 95% CI) as the heuristic window of approximately 0.25–0.50 B0 . Above this BMMSY window, fish biomass values are often associated with ecological stability created by high levels of herbivory and predation. For example, McClanahan et al. (2011) found that when fish biomass exceeds 1130 kg/ha in the western Indian Ocean, reef ecosystems display low between-site variability in algae and high and uniform levels of herbivory and predation. Additionally, the 95% CIs for the lower estimate of pristine biomass would be 1090 kg/ha. Therefore, combining these 2 observations, we chose 1150 kg/ha as an approximate conservation target for recovered reef fish biomass. This target is expected to represent reefs with high predation or herbivory and ecological stability, which are often the focus of conservation. Because biomass values are
frequently strongly right skewed, which could bias target estimates, we modeled log biomass values, but backtransformed biomass values and presented the averages of the quantile median values for presentation in summary figures. Results and Discussion Site-specific fish biomass varied by more than 3 orders of magnitude across 331 site-time replicates throughout the Indian Ocean. Reef fish functional composition and life-history characteristics revealed log-linear trends across this range of biomass-based estimates of fishing impact. As a result, life histories and ecological functions showed the greatest absolute differences among sites at relatively low to moderate levels of biomass (
Biomass-based targets and the management of multispecies coral reef fisheries T. R. McClanahan,∗ ¶ N. A. J. Graham,† M. A. MacNeil,‡ and J. E. Cinner† ∗
Wildlife Conservation Society, Bronx, NY 10460, U.S.A. †Australian Research Council Centre of Excellence for Coral Reef Studies James Cook University, Townsville, QLD 4811, Australia ‡Australian Institute of Marine Science, Townsville, QLD 4810, Australia
Abstract: The failure of fisheries management among multispecies coral reef fisheries is well documented and has dire implications for the 100 million people engaged in these small-scale operations. Weak or missing management institutions, a lack of research capacity, and the complex nature of these ecosystems have heralded a call for ecosystem-based management approaches. However, ecosystem-based management of coral reef fisheries has proved challenging due to the multispecies nature of catches and the diversity of fish functional roles. We used data on fish communities collected from 233 individual sites in 9 western Indian Ocean countries to evaluate changes in the site’s functional composition and associated lifehistory characteristics along a large range of fish biomass. As biomass increased along this range, fish were larger and grew and matured more slowly while the abundance of scraping and predatory species increased. The greatest changes in functional composition occurred below relatively low standing stock biomass (15 years old and larger than 5 km2 or lightly fished locations (e.g., the Maldives), we estimated the reference point biomass in the region (B0 ) at 1200 kg/ha (±110, 95% CI; n = 47) (McClanahan et al. 2011). To examine how our results fit with targets for fisheries management and conservation, we overlaid the multispecies maximum sustainable yield (BMMSY ) window estimates for the region with our estimated biomass. The BMMSY is an extension of the well-known single species maximum sustainable yield (MSY) estimate, where fisheries targets are based on historical stock biomass and lightly fished areas in the seascape (Hilborn 2002; Worm et al. 2009). While single-species MSY estimates vary by species and environment, BMMSY has a wider confidence interval or window; values typically range from 0.25 to 0.50 B0 (Thorson et al. 2012). While this larger window lacks precision, it is a conservative estimate for reef fisheries, which are typically data poor. Furthermore, the lower end of this window (0.25 B0 ) corresponds to many of the detrimental ecological impacts of fishing on reefs (McClanahan et al. 2011). Consequently, the convergence of maximum yields for many species and the indirect ecological impacts of fishing create a useful heuristic for evaluating the state of the fish stocks relative to fisheries production goals (Worm et al. 2009). Based on the pristine biomass (B0 ) estimate for the region, we
McClanahan et al.
5
Figure 3. Marginal estimated relationship between modeled log biomass and 9 feeding categories based on a hierarchical normal model (black line, line of best fit; open circles, sites on atolls; filled circles, sites not on atolls). used 300 kg/ha (±28, 95% CI) to 600 kg/ha (±54, 95% CI) as the heuristic window of approximately 0.25–0.50 B0 . Above this BMMSY window, fish biomass values are often associated with ecological stability created by high levels of herbivory and predation. For example, McClanahan et al. (2011) found that when fish biomass exceeds 1130 kg/ha in the western Indian Ocean, reef ecosystems display low between-site variability in algae and high and uniform levels of herbivory and predation. Additionally, the 95% CIs for the lower estimate of pristine biomass would be 1090 kg/ha. Therefore, combining these 2 observations, we chose 1150 kg/ha as an approximate conservation target for recovered reef fish biomass. This target is expected to represent reefs with high predation or herbivory and ecological stability, which are often the focus of conservation. Because biomass values are
frequently strongly right skewed, which could bias target estimates, we modeled log biomass values, but backtransformed biomass values and presented the averages of the quantile median values for presentation in summary figures. Results and Discussion Site-specific fish biomass varied by more than 3 orders of magnitude across 331 site-time replicates throughout the Indian Ocean. Reef fish functional composition and life-history characteristics revealed log-linear trends across this range of biomass-based estimates of fishing impact. As a result, life histories and ecological functions showed the greatest absolute differences among sites at relatively low to moderate levels of biomass (
