Kenneth A. Rose

WHY ARE QUANTITATIVE RELATIONSHIPS BETWEEN ENVIRONMENTAL QUALITY AND FISH POPULATIONS SO ELUSIVE

Despite the ecological and economic importance of fish, fisheries manage- ment has generally failed to achieve its principal goal of sustainability. Management is hindered because most exploited fish are long-lived species that utilize a variety of habitats and exhibit high interannual fluctuations in abundance. Effective management requires that we understand how natural and anthropogenic sources of variability in abiotic variables (termed environmental quality (EQ)) affect fish population dynamics. Quantifying the ef- fects of anthropogenic changes in EQ on fish populations has remained elusive and con- troversial. I illustrate, with examples, six issues related to quantifying EQ effects on fish populations. These examples also serve as demonstrations of how modeling can be used to address these issues. The six issues are: (1) detectability-high interannual variation and interaction effects among climatic variables that affect population dynamics make isolating effects of individual stressors difficult; (2) complex habitat and nonintuitive responses- spatial heterogeneity in habitat can result in population responses that are disproportionate to the changes in EQ; (3) regional predictions-biological realism is often sacrificed un- necessarily when broad spatial scale predictions are needed; (4) community interactions- too little attention is paid to how community-level interactions can affect population-based analyses; (5) sublethal effects-sublethal effects are often ignored but can have large effects on population dynamics; and (6) cumulative effects-the combined effect of multiple stress- ors can be much different than expected from the sum of their individual effects. Examples include a variety of freshwater and marine species. Quantifying EQ effects on fish popu- lations can be improved by considering these issues in analyses, and by taking a true multidisciplinary approach that combines individual-based modeling and life history theory.

End-To-End Models for the Analysis of Marine Ecosystems: Challenges, Issues, and Next Steps

Abstract There is growing interest in models of marine ecosystems that deal with the effects of climate change through the higher trophic levels. Such end-to-end models combine physicochemical oceanographic descriptors and organisms ranging from microbes to higher-trophic-level (HTL) organisms, including humans, in a single modeling framework. The demand for such approaches arises from the need for quantitative tools for ecosystem-based management, particularly models that can deal with bottom-up and top-down controls that operate simultaneously and vary in time and space and that are capable of handling the multiple impacts expected under climate change. End-to-end models are now feasible because of improvements in the component submodels and the availability of sufficient computing power. We discuss nine issues related to the development of end-to-end models. These issues relate to formulation of the zooplankton submodel, melding of multiple temporal and spatial scales, acclimation and adaptation, behavioral movement, software and technology, model coupling, skill assessment, and interdisciplinary challenges. We urge restraint in using end-to-end models in a true forecasting mode until we know more about their performance. End-to-end models will challenge the available data and our ability to analyze and interpret complicated models that generate complex behavior. End-to-end modeling is in its early developmental stages and thus presents an opportunity to establish an open-access, community-based approach supported by a suite of true interdisciplinary efforts.

Bioenergetics Modeling of Stream Trout Growth: Temperature and Food Consumption Effects

Abstract We investigated bioenergetics modeling of growth as an approach for assessing the effects of temperature changes on stream dwelling rainbow trout Oncorhynchus mykiss. Study objectives were (1) to determine the relative effect of temperature versus food consumption on model-predicted growth and (2) to identify relationships between model-predicted food consumption and commonly measured environmental variables. A bioenergetics model for rainbow trout was calibrated to apparent age-1 growth in summer and fall–spring periods for 10 years at eight Sierra Nevada, California, study sites. Model analyses showed that the observed year-to-year variation in summer growth was related to food consumption but not to temperature and that temperature was more important, but still of secondary importance, to observed variation in fall–spring growth. Growth at all sampling sites appeared lower and more variable in summer than in other seasons, and variation among sites and years in the food consumption parameter P...

Why Do Most Fish Produce so Many Tiny Offspring

A simulation model containing size-based rules for foraging, growth, and probability of survival was created to track the fates of pelagic larval fish. The relative success of cohorts comprised of equivalent initial biomass but containing different numbers and sizes of first-feeding larvae was compared in environments having different levels of patchiness and densities of food resources. In environments containing randomly distributed prey, the rate of growth and probability of survival were always greater, and duration of the larval stage was shorter for larger larvae due to the size-based rules. Broods comprised of fewer but larger first-feeding larvae resulted in the greatest number of survivors in low-prey-density environments. In prey-rich environments, broods containing the same initial biomass divided into greater numbers of small first-feeding larvae resulted in more survivors. This result occurred despite the fact that, on a per-individual basis, survival rate was much lower initially in larger broods. We used a negative binomial algorithm to generate encounter probabilities with patchily distributed prey on small spatial scales in which the previous days experiences had no relationship to current probabilities for encounter. In prey-poor environments, the strategy of producing fewer large larvae was superior at all levels of small-scale prey patchiness. In prey-rich environments, broods containing larger numbers of smaller larvae resulted in more survivors in simulations for randomly distributed and moderately clumped prey. With greater clumping of prey, the greatest number of survivors resulted from the strategy of producing fewer but larger larvae. To examine the effect of large-scale prey patchiness, we solved for the percentage of a larval cohort that would have to settle and remain within a prey-rich patch, in order for a strategy of producing many small larvae to yield more survivors than a strategy of producing fewer larger larvae under the same conditions. When prey patches contained 200 prey/L (compared with 50 prey/L outside) and as few as 1% of the brood settled into the prey-rich patches, large broods comprised of 3-mm larvae yielded more survivors than small broods comprised of 10-mm larvae. Our simulations indicate that the superior evolutionary strategy is the investment in larger numbers of smaller eggs when resources are patchy on a relatively large spatial scale.

The Pattern and Influence of Low Dissolved Oxygen in the Patuxent River, a Seasonally Hypoxic Estuary

Increased nutrient loadings have resulted in low dissolved oxygen (DO) concentrations in bottom waters of the Patuxent River, a tributary of Chesapeake Bay. We synthesize existing and newly collected data to examine spatial and temporal variation in bottom DO, the prevalence of hypoxia-induced mortality of fishes, the tolerance of Patuxent River biota to low DO, and the influence of bottom DO on the vertical distributions and spatial overlap of larval fish and fish eggs with their gelatinous predators and zooplankton prey. We use this information, as well as output from watershed-quality and water-quality models, to configure a spatially-explicit individual-based model to predict how changing land use within the Patuxent watershed may affect survival of early life stages of summer breeding fishes through its effect on DO. Bottom waters in much of the mesohaline Patuxent River are below 50% DO saturation during summer. The system is characterized by high spatial and temporal variation in DO concentrations, and the current severity and extent of hypoxia are sufficient to alter distributions of organisms and trophic interactions in the river. Gelatinous zooplankton are among the most tolerant species of hypoxia, while several of the ecologically and economically important finfish are among the most sensitive. This variation in DO tolerances may make the Patuxent River, and similar estuaries, particularly susceptible to hypoxia-induced alterations in food web dynamics. Model simulations consistently predict high mortality of planktonic bay anchovy eggs (Anchoa mitchilli) under current DO, and increasing survival of fish eggs with increasing DO. Changes in land use that reduce nutrient loadings may either increase or decrease predation mortality of larval fish depending on the baseline DO conditions at any point in space and time. A precautionary approach towards fisheries and ecosystem management would recommend reducing nutrients to levels at which low oxygen effects on estuarine habitat are reduced and, where possible, eliminated.

Nutrient enrichment and fisheries exploitation: interactive effects on estuarine living resources and their management

Both fisheries exploitation and increased nutrient loadings strongly affect fish and shellfish abundance and production in estuaries. These stressors do not act independently; instead, they jointly influence food webs, and each affects the sensitivity of species and ecosystems to the other. Nutrient enrichment and the habitat degradation it sometimes causes can affect sustainable yields of fisheries, and fisheries exploitation can affect the ability of estuarine systems to process nutrients. The total biomass of fisheries landings in estuaries and semi-enclosed seas tends to increase with nitrogen loadings in spite of hypoxia, but hypoxia and other negative effects of nutrient over-enrichment cause declines in individual species and in parts of systems most severely affected. More thoroughly integrated management of nutrients and fisheries will permit more effective management responses to systems affected by both stressors, including the application of fisheries regulations to rebuild stocks negatively affected by eutrophication. Reducing fishing mortality may lead to the recovery of depressed populations even when eutrophication contributes to population declines if actions are taken while the population retains sufficient reproductive potential. New advances in modeling, statistics, and technology promise to provide the information needed to improve the understanding and management of systems subject to both nutrient enrichment and fisheries exploitation.

PREDICTING CLIMATE CHANGE EFFECTS ON APPALACHIAN TROUT: COMBINING GIS AND INDIVIDUAL‐BASED MODELING

We coupled an individual-based model of brook trout (Salvelinusfontinalis) and rainbow trout (Oncorhynchus mykiss) with a geographic information system (GIS) database to predict climate change effects on southern Appalachian stream populations. The model tracked individuals of both species through the daily processes of spawning, growth, feeding, mortality, and movement for 30 years in a stream reach consisting of connected pools, runs, and riffles. The southern Appalachian Plateau was divided into 101 watershed elevation band zones. Model simulations were performed for a representative stream reach of each stream order in each zone. Trout abundance was estimated by mul- tiplying predicted trout densities (measured in number of trout per meter) by the total length of streams of each order in each watershed elevation zone. Three climate change scenarios were analyzed: temperature only (1.5-2.5?C warmer stream temperatures); temperature and flow (warmer stream temperatures and lower baseline flows with threefold higher peak flows); and temperature, flow, and mortality episodes (warmer stream temperatures, changed flows, and flow-related scouring of redds). Increased temperature alone resulted in increased abundances of brook and rainbow trout. The temperature-and-flow scenario resulted in a complex mosaic of positive and negative changes in abundances in zones, but little change in total abundance. Addition of episodic mortality in the form of floods that scour redds and kill eggs and fry caused a net loss of rainbow trout. Predicted changes in habitat (based on simulation results and temperature alone) were, at best, weakly correlated with predicted changes in abundance. The coupling of individual-based models to GIS databases, in order to scale up environmental effects on individuals to regional population responses, offers a promising approach for regional assessments.

Analysis of an Estuarine Striped Bass Population: Effects of Environmental Conditions During Early Life

Estuarine fish populations are exposed to a variety of environmental conditions that cause both short-term variability and long-term trends in abundance. We analyzed an extensive data set for striped bass (Morone saxatilis) in the San Francisco Estuary to refine our understanding of how environmental variability influences recruitment. We examined the effects of environmental variability during early life stages on subsequent recruitment (age 3 yr), and the degree to which conditions in early life may have contributed to a long-term decline in abundance of adult striped bass in the San Francisco Estuary. Survival from egg to young-of-the-year varied strongly with freshwater flow; this effect apparently occurred within the first week or two of life, a time period that encompasses transport of eggs and larvae from the rivers to rearing areas and the onset of feeding. The rate of freshwater flow to pumping facilities that export freshwater from the system had small or sporadic effects on survival during the first month or two of life. Although many young striped bass between ages 2 and 8 mo were entrained in export pumping facilities, the resulting high mortality was unrelated to total mortality rates determined from field data on young striped bass. This lack of effect was apparently due to strong density-dependent mortality occurring between ages 1 mo and 3 yr (Kimmerer et al. 2000). The available data do not support previously suggested relationships between recruitment and freshwater flow during early life, or between gross estimates of pesticide input and survival of early life stages. We used a simple life-cycle model to show that various combined factors could have led to a decline in adult abundance, particularly a large and increasing adult mortality, but that events early in life probably did not contribute substantially to the decline. These results demonstrate that several decades of monitoring data from numerous life stages are needed to distinguish among alternative hypotheses about environmental influences on populations of estuarine fish.

Survival skills required for predator evasion by fish larvae and their relation to laboratory measures of performance

Our objective was to determine which survival skills were most informative of larval fishs escape potential, and thereby better understand larval fish strategies for predator evasion and the nature of selective pressures during early life. We examined evasive performance of red drum larvae, Sciaenops ocellatus, in predation trials and related their escape potential to 12 putative survival skills measured in nonlethal assays. The putative survival skills measured were spontaneous swimming speed and the responsiveness, timing and magnitude of escape responses elicited by artificial stimuli. Responsiveness to the predator was the primary determinant of success in predation trials, accounting for 86% of the variation in escape potential. Of 12 putative survival skills and composite variables that summarized all putative skills, only visual responsiveness (proportion of stimulus presentations responded to) was significantly correlated with escape potential. Regression tree analysis showed that three putative survival skills could be applied in sequence to coarsely filter larvae according to escape potential. Previous work showed that batches of larvae (spawns) could be grouped into clusters whose members showed similar levels of performance in most putative skills. Predation trials showed that escape potential varied as predicted by these clusters. Our results imply that some putative survival skills measured in nonlethal assays reflect responsiveness to an actual predator, but they do not relate directly to their analogues in actual predator–prey interactions. Nevertheless, the ability to distinguish unlikely survivors from more fit individuals, even coarsely, can be valuable for designing experiments, and careful interpretation of putative survival skills can improve ecological models designed to project the fate of larval cohorts.

BROWN SHRIMP ON THE EDGE: LINKING HABITAT TO SURVIVAL USING AN INDIVIDUAL-BASED SIMULATION MODEL

In many coastal areas, natural habitats are being fragmented and lost to encroaching human development. These landscape changes can affect the production of recreationally and commercially important fisheries because many exploited species of fish and shellfish are estuarine dependent and utilize coastal marshes as nursery grounds. Brown shrimp are an example of a commercially exploited species that may be highly affected by changes in the spatial distribution of habitat types. We used a spatially explicit, individual-based simulation model to explore the role of marsh vegetation and edge habitat in brown shrimp survival. The model simulated shrimp movement, mortality, and growth of individual shrimp from arrival as postlarvae to 70-mm body length, when they emigrate offshore. Simulations were performed on 100 × 100 m spatial grid of 1-m2 cells, with each cell labeled as “water” or “vegetation”. Predation mortality was influenced by shrimp size, movement, and habitat. Simulated shrimp growth depended on temperature, habitat, and local shrimp density. We examined the relationships between shrimp survival and marsh attributes (amount of vegetation and edge habitat) by simulating a series of four habitat maps that we created from aerial photographs. Biological parameters were derived from published estimates and from field data. We corroborated the model by comparing the simulated shrimp abundance with summary statistics from long-term monitoring data, by comparing the simulated density with fine-scale patterns observed in field studies, and by comparing simulated and measured stable-isotope values. Surviving shrimp grew faster, moved less, spent more time in vegetation, and experienced slightly higher local densities than shrimp that died during the simulation. Habitat maps with more edge habitat invariably produced higher simulated shrimp survival rates. High-edge habitats increased survival by providing shrimp more direct access to vegetation without additional movement-related mortality and density-dependent growth costs associated with low-edge habitats. Model predictions were robust to higher numbers of initial postlarvae and to alterations to the movement rules. The results of this study suggest that the management of brown shrimp should be extended from protecting the spawning stock through catch regulations to also protecting the estuarine life stages through habitat conservation and restoration.