Marine Protected Areas
Our interest in marine protected areas developed in the mid-90s when marine resource scientists were beginning to ask general, strategic questions about how population persistence and fishery yield were going to behave when fish populations were distributed over a heterogeneous seascape with fishing mortality in some locations, but not others. Conventional fishery management had paid little attention to spatial variability. We first used a simple age-structured model to show that conventional fishery management and management with MPAs were equivalent in the sense that the yield from MPAs would be about the same as that from proper conventional management.
Next we used another simple model to show that a marine species with dispersing larvae, inhabiting a network of periodic MPAs could persist in one of two ways: 1. Self-persistence, by which a single subpopulation would persist in an isolated MPA if the dispersal distance was less than the dimension of the MPA, and 2. Network persistence, by which species without big enough MPAs to persist alone, could persist if a specific fraction of the coastline was in MPAs. This result for spatial population dynamics turned out to be linked to results from non-spatial population dynamics: for heavy fishing between the MPAs, the fraction of the coastline required for network persistence was the same as the fraction of unfished lifetime egg production required for persistence of populations with conventional fishing, i.e., a value of FLEP (fraction of lifetime egg production) or in equivalent fisheries terminology SPR (spawning potential ratio) of about 35-40 percent. With less fishing the required fraction would be less.
In a similar parallel effort to understand what determined persistence of spatially distributed populations linked by dispersal, we showed that patches without adequate reproduction for replacement by themselves could persist by depending on replacement loops that went through other patches. Those replacement loops could involve several generations before returning home with an adequate value of replacement.
Another step in understanding persistence and yield in systems of MPAs was to add the effects of juvenile and adult swimming movement within a home range, to the effects of larval dispersal described in Botsford, et al. (2001). This effort was led by a student, Liz Moffit, who showed how greater protection was required as the size of the home range increased.
The state of California passed the Marine Life Protection Act in 1999, which mandated MPAs along the California coast, and our lab became involved in the process of evaluating the pros and cons of proposed MPAs during the decision-making process. This led to development of spatially explicit, tactical (as opposed to strategic) models for that evaluation. These tactical models addressing specific locations and species confirmed the results of more general, strategic models above, regarding MPAs and conventional management and the dependence of persistence on dispersal distances.
A number of empirical larval researchers began to take advantage of the developing technologies for determining where marine larvae metamorphosing at a location had originated. This was done to aid in the design of MPAs, and to understand persistence of marine populations in general. A pervasive problem with these studies was that they often reported their results in terms of “self-recruitment,” which was defined to be the fraction of larvae settling at a location that actually originated there. This quantity is actually not useful in determining replacement, hence it is not useful in determining persistence for the design of MPAs. What was needed was measures of the fraction leaving a location that ends up at each other location. We collaborated with a number of empirical colleagues to point this out.
In spite of this publication, his shortcoming was still not heeded by most empiricists attempting to measure larval connectivity, so we undertook a second effort that included a review of empirical work, and a more extensive explanation of the information needed to express the dynamics of connectivity. This effort was led by Scott Burgess, at the time a postdoc at UC Davis.
After California’s MPAs were implemented they were required to be managed adaptively, through a process of monitoring population responses to the MPAs to determine whether they met predicted performance. Since the modeling we had used in the decision-making process had been used to examine long-term results, we needed to examine the transient responses of populations to understand and project their short-term responses to MPAs. We first obtained a general description of how the timing and magnitude of responses depended on life histories and the amount of fishing. This effort was led by Will White, at the time an assistant professor at the University of North Carolina, Wilmington, and formerly a post doc at UC Davis.
Next we analyzed the expected responses to MPA implementation when sampling at different times after implementation and distances from the MPA, both inside and outside the MPA. This effort was led by Liz Moffit, at the time a graduate student at UC Davis.
