We develop and test models for the population dynamics of species that undergo regular alternations of generations between independent, free-living, haploid and diploid phases. The models are patterned after the dioecious, haploid-diploid lifecycle of many marine algae. If the two phases have equal demographic rates, all models (with or without density dependence) predict a stable distribution with a ratio of root2:1 (similar to60% haploids and 40% diploids). We find that observable deviations from this distribution can occur when the demographic rates (mortality, fecundity) between phases vary widely. Field surveys of three macroalgal species with independent, free-living haploids and diploids, Mazzaella flaccida, M. laminarioides, and M. splendens, reveal a consistent pattern of haploid dominance that significantly exceeds this null model expectation (72%, 76%, and 66%, respectively). These species also exhibit significant variation in haploid frequency among sites. We estimated the per capita fecundity and mortality rates for each phase of both M. flaccida and M. laminarioides and used this information to address two basic questions: (1) Which parameter better explains the overall high relative haploid abundance for each species? and (2) Which parameter better explains the variation in relative haploid abundance among sites? Differences in haploid and diploid mortality rates did not explain the high relative abundance of haploids for either M. flaccidet or M. laminarioides. However, differences in fecundity rates between phases were consistent with the high relative abundance of haploids for M. flaccida; diploids usually had significantly higher per capita fecundity rates than haploids. For M. laminarioides, there were no significant differences in fecundity rates between phases. For M. flaccida, among-site variation in relative abundance of haploids can be explained by differences in mortality rates. The relative abundance of haploids was positively correlated with the ratio of diploid:haploid mortality rates; the sites with the highest ratio of diploid:haploid mortality had the highest percentage of haploids. Differences in mortality rates for M. laminarioides, and fecundity rates for M. flaccida and M. laminarioides, could not explain the observed variation in haploid abundance among sites.

Many algal species have life cycles that involve an obligate alternation of generations between nearly identical, free-living haploid and diploid individuals. The percentages of haploids to diploids, and the spatial and temporal scales of variation of these percentages, must be examined before hypotheses about the relative importance of haploid and diploid demographic processes on population dynamics can be evaluated. In this paper, we investigated the spatial and temporal dynamics of the percentage of haploid and diploid stages in 4 congeners of the marine alga Mazzaella (M capensis, M flaccida, M laminarioides, and M splendens) across large portions of their geographical ranges. For the 3 species that were sampled over multiple years (M. flaccida, M. laminarioides, M splendens), the percentage of haploids varied significantly among sites. However, 2 of them (M flaccida, M. splendens) did not exhibit significant temporal (interannual) variation in the percentage of haploids. In addition, more fine-scale temporal studies on M flaccida populations revealed that there were also no significant intra-annual fluctuations in the percentage of haploids, although there was significant population turnover on this timescale. This spatial variability but temporal stability of haploid-diploid ratios (for M flaccida and M splendens) can be used to generate, and subsequently test, hypotheses on the regulation of haploid-diploid ratios and the relative importance of haploid and diploid demographic processes on population dynamics.

Marine reserves affect areas outside reserve boundaries via the displacement of fishing effort and the export of production. Here we focus on how these key factors interact to influence the results seen once reserves are created. For a settlement-limited fishery, export of increased production from within reserves can offset the effects of displaced fishing effort. We develop simple mathematical models that indicate net fisheries benefits can accrue at closures up to and perhaps beyond 50% of total stock area through the export of production, given documented average increases in biomass within reserves. However, reserve monitoring programs face problems identifying independent control sites because the spatial extent of export is unknown. Efforts to monitor reserve impacts on recruitment are further complicated by the fact that large reserve closures are likely necessary before significant changes in recruitment can be detected above normal interannual fluctuations. Resolving these limitations requires comprehensive monitoring data before reserves are implemented. Fortunately, studies of reserves that used Before-After, Control-Impact (BACI) experimental designs show that control and reserve sites were equivalent prior to protection, and that control sites improved after reserves were in place. Consequently, any bias in our current perception of reserve impacts likely underestimates their effect.

Benthic marine organisms are characterized by a bipartite life history in which populations of sedentary adults are connected by oceanic transport of planktonic propagules. In contrast with the terrestrial case, where 'long distance dispersal' (LDD) has traditionally been viewed as a process involving rare events, this creates the possibility for large numbers of offspring to travel far relative to the spatial scale of adult populations. As a result, the concept of LDD must be examined carefully when applied in a marine context. Any measure of LDD requires reference to an explicit 'local' scale, often defined in terms of adult population demography, habitat patchiness, or the average dispersal distance. Terms such as 'open' and 'closed' are relative, and should be used with caution, especially when compared across different taxa and systems. We use recently synthesized data on marine propagule dispersal potential and the spread of marine invasive species to draw inferences about average and maximum effective dispersal distances for marine taxa. Foremost, our results indicate that dispersal occurs at a wide range of scales in marine communities. The nonrandom distribution of these scales among community members has implications for marine community dynamics, and for the implementation of marine conservation efforts. Second, in agreement with theoretical results, our data illustrate that average and extreme dispersal scales do not necessarily covary. This further confounds simple classifications of 'short' and 'long' dispersers, because different ecological processes (e.g. range expansion vs. population replenishment) depend on different aspects of the dispersal pattern (e.g. extremes vs. average). Our findings argue for a more rigorous quantitative view of scale in the study of marine dispersal processes, where relative terms such as 'short' and 'long', 'open' and 'closed', 'retained' and 'exported' are defined only in conjunction with explicit definitions of the scale and process of interest. This shift in perspective represents an important step towards unifying theoretical and empirical studies of dispersal processes in marine and terrestrial systems.

Many nearshore fish and invertebrate populations are overexploited even when apparently coherent management structures are in place. One potential cause of mismanagement may be a poor understanding and accounting of stochasticity, particularly for stock recruitment. Many of the fishes and invertebrates that comprise nearshore fisheries are relatively sedentary as adults but have an obligate larval pelagic stage that is dispersed by ocean currents. Here, we demonstrate that larval connectivity is inherently an intermittent and heterogeneous process on annual time scales. This stochasticity arises from the advection of pelagic larvae by chaotic coastal circulations. This result departs from typical assumptions where larvae simply diffuse from one site to another or where complex connectivity patterns are created by transport within spatially complicated environments. We derive a statistical model for the expected variability in larval settlement patterns and demonstrate how larval connectivity varies as a function of different biological and physical processes. The stochastic nature of larval connectivity creates an unavoidable uncertainty in the assessment of fish recruitment and the resulting forecasts of sustainable yields.