The use of marine reserves throughout the world's oceans is gaining prominence as a fisheries management tool, but the full potential for their use in management is unrealized. Here, I developed a collaborative fisheries research program and designed a series of management models to evaluate the use of marine reserves in the management of small-scale, data-limited fisheries.

Biological sampling of grass rockfish (Sebastes rastrelliger) populations was conducted inside and outside of marine reserves in the Santa Barbara Channel, CA, USA. Model fits to the data demonstrated geographic differences in growth, reproductive maturity and length frequency distributions in two environmentally distinct regions of the channel. A simulation model was developed to test the use of bounded catch curve models to estimate natural and fishing mortality rates using biological data from inside and outside of marine reserves. The models were shown to be robust across life history strategies, variable fishing mortality rates, and stochastic recruitment. These models were applied to collaborative research data on grass rockfish to estimate fishery relevant metrics and the status of the resource in the Santa Barbara Channel. Results indicated that populations in the high and low productivity regions of the channel are undergoing overfishing. The incorporation of marine reserves into these management models reduced the probability of overfishing to minimal levels in the high productivity region, but did not reduce the probability of overfishing to acceptable levels in the low productivity region.

A management strategy for using marine reserves to set sustainable catch levels was developed for small-scale, data-poor fisheries. The marine reserve-based decision tree uses length frequency and catch per unit effort data from scientific sampling inside and outside of marine reserves to modify allowable catches on an annual basis to achieve target reference points of fisheries management. Management strategy evaluation (MSE) was performed to demonstrate the utility of this model. Over a 30 year time period, simulation results demonstrated that the use of the reserve-based decision tree consistently achieved higher yields than the current precautionary approach to management while meeting conservation targets.