Marine reserves: long-term protection is required for full recovery of predatory fish populations.

No-take marine reserves are advocated widely as a potential solution to the loss of marine biodiversity and ecosystem structure, and to over-fishing. We assess the duration of protection required for unfished populations of large predatory reef fish to attain natural states. We have monitored two marine reserves at Sumilon and Apo Islands, Philippines, regularly for 17 years (1983-2000). The biomass of large predatory fish was still increasing exponentially after 9 and 18 years of protection at Sumilon and Apo reserves, respectively. There was little evidence that the rate of accumulation of biomass inside the reserves was slowing down even after so many years of protection. This suggests that the length of time to full recovery will be considerable. We made two assumptions in order to estimate this period. Firstly, that biomass growth will follow the logistic model. Secondly, the conservative assumption that biomass had already attained 90% of the local carrying capacity of the environments in the reserves. We conclude that the time required for full recovery will be 15 and 40 years at Sumilon and Apo reserves, respectively. Such durations of recovery appear consistent with known life history characteristics of these fish, and with empirical data on recovery rates of heavily exploited fish stocks. By the time the full fisheries or ecosystem benefits from such reserves are apparent, human populations and impacts will have doubled in much of the developing world. Thus, networks of such reserves need to be implemented immediately. Furthermore, the management mechanisms for the reserves need to be successful over timescales of human generations.