Harvesting at the Community Level

Most fisheries are not directed at individual species alone. Rather, in many instances, species within a community are caught together and are also part of competitive networks and food webs. Species that are caught together are subject to technical interactions. Species that compete or are connected through predator–prey interactions (or other types of interactions) are subject to biological interactions. Ignoring either of these forms of interaction in management can lead to unintended consequences. Technical solutions can help to avoid some species while targeting others, but a comprehensive solution requires creating the right economic incentives and some incidental catch is still inevitable. Accounting for trophic interactions means that biological reference points depend on the abundance of other taxa. Single-species approaches are invalid in a multispecies or community context where biological interactions are important. Technical interactions can make it impossible to achieve target exploitation rates even if biological interactions are relatively unimportant.

Production at the Cohort and Population Levels

The dominant focus on production processes in fisheries science sets it apart from other areas of population ecology in which population numbers are the principal currency for analysis. This chapter extends consideration of individual growth and mortality rates provided in earlier chapters to broaden the context for understanding cohort and population processes. A cohort is a group of organisms born within a given time period (e.g. year). How a fish population will respond to harvesting requires not only accurate accounting of its effective reproductive output but an understanding of the relative importance of compensatory mechanisms operating at different points in the life cycle. Recruitment (the number in a cohort surviving to a specified life stage or age) emerges as a dominant component of production at the population level. A dominant theme in this chapter concerns population regulation as embodied in the recruitment process and the high variability in this process.

Spatial Processes

Aquatic populations are patchily distributed. The full implications of this statement for the dynamics of these populations depend very strongly on movement and dispersal patterns. The characteristically heterogeneous distribution of exploited aquatic species is of course essential to harvesting strategies employed by fishers. It can also present important challenges to management when species distributions contract to core habitat areas and these concentrations can be readily located and exploited. The types of models described in this chapter, including metapopulation models, provide an initial framework for considering the dynamics of spatially structured populations. Dispersal can provide a stabilizing force by providing a subsidy or rescue effect for depleted populations. Realistic representation of spatial processes in models of aquatic populations is an evolving art. Quantifying movement and connectivity of aquatic species entails special challenges. Spatially explicit models should account for exchange among subpopulations in relation to their size, distance, and degree of separation.

Interspecific Interactions II: Competition and Mutualism

Competition and mutualism are important forms of biotic interaction in aquatic communities. Quantification of the population and community-level effects of these interactions has historically been less common in fisheries analyses than predation. In part, this reflects the difficulties in conducting controlled experiments for larger-bodied organisms in aquatic environments. Documenting competition entails not only identifying patterns of shared resource use but evidence that these resources are limiting. Inferences concerning competitive interactions in non-experimental settings may be possible if histories of population change for putative competitors are available and quantifiable interventions involving the addition of a species (through deliberate or inadvertent introductions) or a differential reduction in abundance of the species through harvesting is undertaken. Care must be taken to account for other changes in the environment in these uncontrolled quasi-experiments. Mutualistic interactions are widely recognized in aquatic ecosystems but far less commonly quantified than competition.

Density-Dependent Population Growth

The observation that no population can grow indefinitely and that most populations persist on ecological timescales implies that mechanisms of population regulation exist. Feedback mechanisms include competition for limited resources, cannibalism, and predation rates that vary with density. Density dependence occurs when per capita birth or death rates depend on population density. Density dependence is compensatory when the population growth rate decreases with population density and depensatory when it increases. The logistic model incorporates density dependence as a simple linear function. A population exhibiting logistic growth will reach a stable population size. Non-linear density-dependent terms can give rise to multiple equilibria. With discrete time models or time delays in density-dependent regulation, the approach to equilibrium may not be smooth—complex dynamical behavior is possible. Density-dependent feedback processes can compensate, up to a point, for natural and anthropogenic disturbances; beyond this point a population will collapse.

Harvesting at the Ecosystem Level

Energy entering at the base of the food web places inherent constraints on total harvest from an ecosystem. Empirical relationships have been established between fishery yield and factors such as chlorophyll concentration, primary production, and other lower trophic level variables to guide management actions. Extension of network models to include harvesting has a long lineage and these models are now being employed worldwide to help guide management decisions. These static mass-balance models have been augmented with a fully dynamic modeling component to explore management options. Biomass spectrum models for exploited ecosystems have also been developed. In addition to direct effects on target species, fisheries can affect the structure and function of ecosystems through habitat damage and incidental catch of non-target organisms, including threatened and endangered species. Increasingly, the effects of climate change are being addressed in ecosystem models through their potential effect on production at all levels of the ecosystem.

Empirical Dynamic Modeling

Empirical Dynamic Modeling offers a flexible complement to standard mechanistic modeling approaches. Because it makes no assumptions concerning the structural form of ecological processes, it can provide an effective approach to dealing with model uncertainty. The method uses non-linear, non-parametric models. It can accommodate a wide spectrum of dynamical behaviors and makes no equilibrium assumptions. The approach is predicated on the idea that within a time series of observations of an ecological variable (e.g. population or species abundance) is encoded information on the factors that have affected that variable over time (e.g. the effects of predators or prey, competitors, environmental change, etc.). The method employs state-space reconstruction to decode this embedded information, and applies nearest-neighbor and kernel regression methods of forecasting. Forecast skill is used directly as a criterion for model selection and validation. It has been proven effective in application to fisheries forecasting problems, often outperforming standard modeling approaches.

Harvesting at the Cohort and Population Levels

This chapter explores dynamical behaviors that go beyond globally stable outcomes to include alternate stable states, and non-equilibrium behaviors. The possibility of multiple equilibria emerges quite readily in models with non-linear harvesting functions. In practice, most fisheries management protocols at least implicitly assume that harvested populations have well-behaved stable equilibrium properties. If this is not the case, then sudden changes (including collapse) can occur and be totally unanticipated. This chapter describes the spectrum of single-species harvesting models from biomass dynamics models that do not include age or size structure, to delay–difference models with a simple demographic structure, to full age-structured models. Dynamic-pool models combine yield per recruit and egg-per-recruit with a stock-recruitment model to obtain an equilibrium yield curve. These single-species models are used to estimate biological reference points with which to assess stock status.

Production at the Ecosystem Level

The development of ecosystem models can be size-based, species-based, or trophocentric. In all cases, equilibrium mass-balance descriptions of ecosystems can be translated to dynamic models. Linear network models trace the flow of energy through food webs. Starting from the base of the food web, they can be solved from the bottom up to calculate how many predators can be supported for a given level of primary production. Conversely, the food web can be solved from the top down to calculate how much primary production is required to support fisheries yield, given the dietary requirements of top predators. These models typically employ species-level and/or trophic-level designations for the nodes in the model. Biomass-spectrum models in contrast are based on body size dimensions (typically weight) rather than any taxonomic designation. Biogeochemical models provide another approach to developing ecosystem production models by making the connection between the availability of key nutrients and ecosystem production.

Community Dynamics

Ecological theory indicates that increasing the number of species, the number of interactions, and the strength of these interactions all tend to make communities less stable. Conversely, stability is enhanced by strong intraspecific density dependence, low connectivity, or weak trophic links. These theoretical predictions are borne out in many fish communities. Species diversity is an important metric for ecological communities. Organizing species into groups according to size, function, or diet composition can reduce the dimensionality of fish community models. Analyses of fish communities from around the world lend support to the prediction of strong compensation within functional groups, with weaker predator–prey links among groups. Size spectra describe the distribution of individuals across size classes irrespective of their species. Qualitative models can be used to assess the indirect effects of species on each other and the overall stability of the community.