Spatio-temporal Ecology and Management of Temperate Reef Fish Population

The demography of marine reef fishes with a dispersive larval stage can be highly variable, with several processes that govern reproduction, dispersal, and recruitment operating across multiple spatial and temporal scales. Characterizing the patterns of this variation and the underlying processes responsible for them can help guide efforts to conserve and manage human impacts on these species. In this dissertation, I explore how associations between temperate reef habitat and fish reproductive potential may vary across spatial scales, how spatial and temporal variation in larval production can influence spatial management strategies, and how stochastic variation in larval dispersal and settlement may affect the ability to detect the impacts of management actions.;Habitat attributes may have both separate and combined effects on the reproductive potential of reef fishes across spatial scales, but the patterns and ecological processes governing these relationships do not necessarily "scale up" from small-scale, in-situ observations to whole reefs and seascapes. In Chapter 1, I combine in-situ dive surveys with high-resolution habitat maps to investigate how associations between reproductive potential of the kelp bass (Paralabrax clathratus) and kelp forest habitat attributes translate between within- and among-reef spatial scales. Macroalgae and benthic rugosity explain the most variation in reproductive potential within reefs, but the configuration of available habitat explains more variation among reefs. I propose that a mismatch between processes operating at different spatial scales is responsible for these results and suggest that future efforts to clarify among-reef relationships are necessary to estimate reproductive potential of fishes on rocky reefs.;Marine reserves are an integral management tool to protect biomass and rebuild depleted fisheries, particularly when important locations for larval production are protected. However, given the uncertainty and temporally dynamic nature of spatially-structured populations, rotating closures might provide a more adaptive solution to both conserve and utilize built-up biomass. In Chapter 2, I construct an age-structured population model with spatially-structured larval productivity to compare the relative efficacy of marine reserves and rotating closures at achieving conservation and fishery management goals. Given a population with a static larval production location, rotating closures outperform permanently fishing the larval production location but underperform permanently protecting the larval production location in the source in terms of total biomass and yield. When the location of the larval production varies in time, permanent reserves generally sustain higher biomass but rotating closures allow for higher fishery yield, and these results are robust to uncertainty in the location of the larval production in any given year. Rotating closures may therefore better buffer against spatial uncertainty in production locations while permanent reserves buffer against temporal uncertainty.;Many marine populations are subject to considerable interannual variation in critical demographic rates such as larval dispersal and survival, and such variation may obscure short-term population responses to protection by marine reserves. In Chapter 3, I consider an age-structured population with stochastic variation in larval survival and population openness to explore the conditions under which managers might expect to observe population responses contrary to deterministic projections through the first 10 years of reserve protection. While the shape of stochasticity in larval survival does not have a strong effect on the ability to detect population changes due to reserve protection, temporal autocorrelation in survival does increase the risk of failing to detect a reserve effect. The degree of population openness may indirectly affect recovery trajectories by changing intrinsic population growth rates, but resolving overall larval survival matters more to detecting reserve effects than does resolving larval origin.