One of the least understood and most fundamental processes in marine ecology is dispersal, or transport of larvae. To what extent are local adult population dynamics influenced by the dispersal of larvae from neighboring source populations? Do self-recruiting populations exist? How far do larvae disperse? The sources and destinations of larvae have been unknown for nearly all commercially important species, yet any model of resource management or biodiversity preservation depends on this information. We wish to understand the interdependency of commercially important California coastal populations by identifying the sources of larval recruits. To do this, we will use environmental signatures, consisting of trace metal inclusions in the hard structures of larvae (e.g., fish otoliths) that are laid down over time. Elements from the surrounding seawater become permanently incorporated into the structures, which then act as daily recorders of the environment experienced during the larval phase. Along coastal margins, concentrations of many trace elements are significantly higher than in the open ocean, and differences in trace element concentrations among coastal water masses occur as well. Recently developed high-precision mass spectrometers and laser microprobes allow a sequential analysis of past environments, down to a resolution of a few days.

We propose the following interrelated approaches to the problem of reconstructing larval paths:

• Validation: How well and how consistently is water chemistry reflected in the trace element composition of hard parts produced by organisms? Can the techniques be extended to other commercially important species?

• Documenting the temporal and spatial nature of environmental signatures: constructing an “atlas” of potential source populations.

• Tracing larval sources through microchemical analyses of new recruits by matching core signatures of hard parts to the “atlas”, and through a pilot mark-recapture study.

• Integrating the results into ongoing students of population genetics and ocean circulation models.

An estimate of the degree of connectivity among local populations through larval dispersal can help to define the appropriate geographic scale of resource management (e.g., spatial extent of fishery closures, placement of marine reserves).

Humans have altered the ocean on a global scale. From large-scale extractive activities such as fishing, to the unintentional introduction and spread of invasive species via shipping vessels, human activities shape marine communities worldwide, having myriad effects on ecosystem functions and resilience. Even recreational activities that are assumed to have lesser impacts, such as SCUBA diving, are intensifying and becoming increasingly concentrated inside marine protected areas. Research into a wide range of direct and obvious human impacts has informed the effective management of marine systems for both conservation goals and human uses. However, understanding the more nuanced pathways for the effects of human disturbance is critical, especially in systems facing multiple stressors, as they may serve to tip the balance between healthy ecosystems and degraded ones. In particular, there is recent and growing evidence that human activities are altering predation risk in marine systems. When humans change the abundance, distribution, or behavior of marine predators, whether through fishing or other forms of disturbance, those changes can ripple through marine ecosystems due to the comparatively large effects that predators have on other trophic levels. Predators exert both consumptive and non-consumptive effects on prey populations, and can thus influence other species interactions and trigger cascades that can result in the fundamental restructuring of marine systems. However, although risk effects have been well studied in terrestrial systems, these effects have been largely neglected in marine systems. In this thesis, I explore two different pathways for the effects of human alteration of predation risk. My first two chapters examine the non-consumptive, or fear effects of recreational spearfishing, diving, and snorkeling to marine fish. My third chapter examines the consumptive effects of an introduced predator on both a native and an invasive prey population, and the dynamics between them.

Chapters One and Two: Diver fear effects

The ecology of fear is based on evidence demonstrating that predation risk is a strong motivating factor in individual prey behavior, and has consequences at the population, community and ecosystem level. While predator escape theory was developed with natural predators in mind, humans can act as predators (e.g. while fishing) or mimic predators (e.g. while diving). The recent testing of predator escape theory by the measurement of flight initiation distance (FID) - the distance at which a fish flees from approaching diver - represents a promising new avenue for future research on diver fear effects to marine fish and in marine ecosystems. My synthesized review of the available empirical evidence suggested that fish react to divers with increases in vigilance and flight where they are exposed to spearfishers as compared to areas where they are protected from fishing. In contrast, I found that the effects of recreational diving on the flight and vigilance behavior of fish have not been well studied or effectively measured to date. In order to explore the effects of long-term recreational diving on the behavior of a common reef fish, I conducted a series of experiments on reefs in the Cayman Islands where recreational diving activity has been consistent and spearfishing has been banned for decades. By measuring the FID and vigilance behavior of over 250 individual Stoplight Parrotfish (Sparisoma viride), I found evidence for habituation to divers by this species in areas where recreational diving is most concentrated. The near lack of a flight response in fish exposed to chronic diving activity represents the lowest FID recorded for the species in either protected or fished areas elsewhere in the world. The significant decrease in flight and vigilance in individuals in the highly dived area in comparison to those in less frequently dived areas also suggests that the frequency of diving activity drives this response. However, significant declines in the bite rates and frequency of cleaning interactions of individuals in the immediate presence of a diver also suggested that there may be fitness costs associated with recreational diver encounters, even for habituated fish. I therefore used these decreased bite rates to estimate a cumulative cost of daily diver encounters to an individual fish in terms of lost feeding. Where habituation to divers was extreme at my study site, estimated feeding loss for an individual fish was low. For scenarios in which I simulated lower levels of fish habituation and different diver behavior, estimated feeding loss due to diver encounters were much higher. This suggests that the cumulative costs of recreational diver encounters may be significant if, for example, larger flight distances must be maintained due to the occasional or nearby presence of spearfishers. Based on my findings, I make recommendations for future research into the behavioral effects of spearfishing, recreational diving, and the potential interaction of these two activities on the behavior of marine fish. I also recommend the incorporation of these behavioral effects into marine spatial planning and management for multiple human uses, particularly where spearfishing and recreational diving co-occur.

Chapter Three: Predation effects of an introduced predator

As multiple invasions accumulate in marine systems, identifying the species traits and interactions that affect invasion success is critical for predicting the outcomes of multiple species introductions. Facilitation is a major mechanism for invasion success that can increase the magnitude of impact of an existing invasion. Marine predators can facilitate invasions and may be more likely to tip the balance of dynamics between species in an invaded system via top-down effects. I examined the effect of predation by an introduced crab, Carcinus maenas, on the native snail Cerithidea californica and the invasive snail Ilyanassa obsoleta. These two common snail species co-occur in the San Francisco Bay estuary, where there is evidence for the competitive displacement of the native snail by the invasive. Laboratory experiments revealed that Carcinus maenas predation was strongly and significantly greater on the native species as compared to the invasive. While the frequency of crab attacks was not significantly different between snail species, the predator was 5.45 times more successful at killing the native C. californica than the invasive I. obsoleta given an attack. This differential predation success drives the higher overall mortality in native snails. In light of the competitive displacement of the native snail by the invasive in this system,these results demonstrate that the introduced predator C. maenas can have both direct negative impacts on native species and indirect positive impacts on other invaders via facilitation. My results provide unique information about a predator’s potential to indirectly influence the success of an invasion, and contribute to our understanding of the role of introduced predators in influencing the dynamics of multiple invasions in marine systems.

Body size has long been recognized as a key driver of species interactions and of an individual’s role in the ecosystem. Body size determines the amount, species, and sizes of prey resources an individual can consume, as well as its own susceptibility to predators. Human harvest of predators can result in severe truncations in predator body size that can have cascading consequences on food webs. Where small and large individuals of the same species differ greatly in their diets, as is common in aquatic systems, the absence of large predators may functionally eliminate a key predator-prey linkage.

Recently management agencies have begun to include size-based metrics as targets. As various harvest strategies differentially affect predator size and biomass, the research presented in this dissertation aims to understand the conditions under which truncations in predator size structure will result in additional loss of predator function than would be predicted from predator biomass alone, and where it will therefore be important to maintain predator size distributions. I specifically examine how the type of ontogenetic shift in diet (e.g. prey species or size class), and the shape of the diet switching function (e.g. gradual or abrupt) will affect the consequences of the loss of the largest predators, and the relative utility of various management strategies in maintaining predator function.

In Chapter 1, I examined the tradeoffs between fishery yield and predator function in the ecosystem when preferentially fishing the largest predators. I found that fisheries that delay harvest until large predator sizes maximize fishery yield but that this virtually eliminates predation on focal prey eaten late in life history when diet shifts are abrupt and occur at or after the size at maturity. In this case, there is a clear tradeoff between fisheries and ecosystem objectives. Instead, where shifts in diet toward late prey are more gradual, targeting the largest predators can achieve a win-win by maximizing yield and achieving predation rates similar to that with other strategies that harvest predators earlier. As such, the optimal fishing strategy to achieve both single-species and ecosystem benefits depends strongly on the interaction between the fishery selectivity pattern and the changes in predator diet with size.

In Chapter 2, I quantified the size-dependence of the predator-prey interaction between herbivorous sea urchins and one of their important predators in southern California kelp forests, California Sheephead. I further examined the consequences of changes in sheephead size and abundance in marine reserves at Catalina Island on size-specific urchin mortality in field predation trials. In my observations of predation of sheephead on urchins, sheephead smaller than 20cm TL do not eat urchins of any size. Thereafter, small sheephead only consumed small urchins, with larger sheephead sizes needed to successfully consume larger urchins, and the largest sheephead preferentially targeted the largest urchins. Inside marine reserves at Catalina, the greater abundance of large sheephead in combination with the observed size-specific capacities for urchin predation led to higher urchin mortality with marine reserve protection, particularly for the largest urchins. Ultimately, by restoring predator size structure, reserves may serve to enhance the resilience of southern California kelp forests.

In Chapter 3, I examined how variation in predator body size distributions and biomass affects the likelihood of size escapes in situations where predators begin eating prey at some threshold size and thereafter consume increasingly larger prey. We focus on California sheephead because of the size dependence of its interaction with herbivorous sea urchins (Chapter 2), and the natural variation in demography where sheephead achieve smaller maximum sizes but higher biomass in the south of its range. We evaluate the consequences of smaller predator body size on top-down control of urchin populations in two scenarios: 1) when overall predator abundance is the same as the population with larger body size, and 2) when predator biomass is the same. With the same numbers of predators, top-down control was significantly weakened by the lack of large sheephead. However, when sheephead biomass was maintained, the absence of large sheephead did not lead to greater urchin abundance, despite lower predation rates overall and much lower predation on large urchins. Higher predation rates on the smallest urchin size classes served as a bottleneck that kept total urchin population at similar levels and prevented a size escape for the largest urchins. This suggests that where predators switch prey size classes in the same species, the loss of the largest individuals does not inherently result in weaker top-down control, if biomass is maintained, but effective control is sensitive to prey growth rates.

The results of this research suggest that the ignoring shifts in predator size structure can under-estimate the effects of fishing on predator function, especially when large predators eat different species than their smaller counterparts. High predator biomass can compensate when diet shifts are to different prey size classes of the same species. Concordance between diet shifts and fishery selectivity can help identify where it will be important to consider changes in predator size in addition to biomass.