Modelling and interpreting fish bioenergetics: a role for behaviour, life-history traits and survival trade-offs.

Bioenergetics is used as the mechanistic foundation of many models of fishes. As the context of a model gradually extends beyond pure bioenergetics to include behaviour, life-history traits and function and performance of the entire organism, so does the need for complementing bioenergetic measurements with trade-offs, particularly those dealing with survival. Such a broadening of focus revitalized and expanded the domain of behavioural ecology in the 1980s. This review makes the case that a similar change of perspective is required for physiology to contribute to the types of predictions society currently demands, e.g. regarding climate change and other anthropogenic stressors.

Eco-evolutionary dynamics induced by massive mortality events.

An eco-genetic model tuned on a population of marble trout Salmo marmoratus subject to periodic flood events was used to explore how the evolution of growth rates interacting with density-dependent processes can modify size at age and population structure and in turn influence the resilience of populations. Fish with greater growth potential were assumed to have higher mortality rates. The results of simulations were compared between two scenarios, one in which populations may evolve growth rates and the other one in which the distribution of growth rates within a population is kept fixed. Evolving populations had a greater proportion of age 1 year individuals in the population, greater median length at age 3 years (the typical age at sexual maturity for S. marmoratus) and lower population sizes. The slightly smaller population sizes did not affect realized extinction risk. Resilience, defined as the number of years necessary to rebound from flood-induced population collapse, was on average from 2 to 3 years in both scenarios, with no significant difference between them. Moderate heritability of growth, relaxation of density-dependent processes at low densities and rapid recovery to a safe population size combine to limit the capacity to evolve faster recovery after flood-induced population collapses via changing growth rates.

Using scale characteristics and water temperature to reconstruct growth rates of juvenile steelhead Oncorhynchus mykiss.

Juvenile steelhead Oncorhynchus mykiss from a northern California Central Valley population were reared in a controlled laboratory experiment. Significantly different rates of growth were observed among fish reared under two ration treatments and three temperature treatments (8, 14 and 20°C). Wider circulus spacing and faster deposition was associated with faster growth. For the same growth rate, however, circulus spacing was two-fold wider and deposited 36% less frequently in the cold compared to the hot temperature treatment. In a multiple linear regression, median circulus spacing and water temperature accounted for 68% of the variation in observed O. mykiss growth. These results corroborate previous research on scale characteristics and growth, while providing novel evidence that highlights the importance of water temperature in these relationships. Thus, this study establishes the utility of using scale analysis as a relatively non-invasive method for inferring growth in salmonids.

Top-down and bottom-up control of life-history strategies in coho salmon (Oncorhynchus kisutch).

Sexual maturation profoundly affects population dynamics, but the degrees to which genetic, top-down, and bottom-up controls affect age at maturity are unclear. Salmonid fishes have plastic age at maturity, and we consider genetic and environmental effects on this trait by developing fitness functions for coho salmon (Oncorhynchus kisutch). The functions are based on size-specific survival and reproductive success, where reproductive success is the product of fecundity and ability to defend nests (females) or the product of sperm volume and ability to mate (males). We model genetic and bottom-up controls (e.g., food availability) with an environmentally explicit growth function and top-down control (predation mortality) with survival functions that consider both size-dependent and size-independent mortality. For females, we predict that early maturation rarely maximizes fitness, but males can maximize fitness by maturing early if they grow well in freshwater. We predict that early maturation is most affected by the bottom-up effects of resource distribution at sea, followed by bottom-up and genotypic effects in freshwater. Top-down processes are predicted to have strong effects on the likelihood of delayed maturation.

Relationship of the photosensitivity of bilayer lipid membranes and the aqueous acceptor. Studies using complex ions of amino acids.

The photocurrent in photosensitive bilayer lipid membranes has been studied as a function of the aqueous acceptor. Correlations are observed between the relative photocurrent and the position of the complex ion visible absorption band and the dipole moment of the ligand. The effect of the ligands is nondirectional: they may be added to either side of the membrane with a corresponding effect on the photocurrent. The effects of the ligands are interpreted using an energy barrier model.

Egg maturation, egg resorption and the costliness of transient egg limitation in insects.

Although there is widespread agreement that the cost of oviposition underlies selective oviposition in insects, there is no consensus regarding which factors mediate the cost of oviposition. Models have suggested that egg costs are often paramount in those insects that do not continue to mature eggs during the adult stage (pro-ovigenic insects). Here we address the hypothesis that egg costs are generally less significant in synovigenic insects, which can replenish oocyte supplies through continuous egg maturation. A dynamic optimization model based on the biology of a highly synovigenic parasitoid, Aphytis aonidiae, suggests that the maximum rate of egg maturation is insufficient to balance the depletion of eggs when opportunities to oviposit are abundant. Transient egg limitation therefore occurs, which imposes opportunity costs on reproducing females. Thus, whereas the most fundamental constraint acting on the lifetime reproductive success of pro-ovigenic species is the fixed total number of eggs that they carry at eclosion, the most fundamental constraint acting on a synovigenic species is the maximum rate of oocyte maturation. Furthermore, the ability of synovigenic species to reverse the flow of nutrients from the soma to oocytes (i.e. egg resorption) has a dramatic influence on the cost of oviposition. Whereas females in hostrich environments may experience oviposition-mediated egg limitation, females in host-poor environments may experience oosorption-mediated egg limitation. Both forms of egg limitation are costly. Contrary to initial expectations, the flexibility of resource allocation that typifies synovigenic reproduction actually appears to broaden the range of conditions under which costly egg limitation occurs. Egg costs appear to be fundamental in mediating the trade-off between current and future reproduction, and therefore are an important factor favouring selective insect oviposition.

Age and longevity in fish, with consideration of the ferox trout.

In the first part of this paper, we review the evolutionary aspects of age and longevity in fish and then summarize the theory of maturity due to Ray Beverton. This theory allows one to predict age at maturity (and thus a putative point for the onset of senescence) from information on growth rate and mortality rate. We illustrate the application of this theory with data on tilapia species and then discuss the limitations of the theory. In the second part of the paper, we develop an individual based model for the ferox trout. This is a morph of brown trout Salmo salar that is an exception to the common notion that caloric restriction extends lifespan, in the sense that ferox trout achieve long life by eating more, not less. The model allows one to identify the role that ecological and biochemical adaptations play in the longevity of the ferox trout.

Predation on adult Aphytis parasitoids in the field.

We report on predation on adult females of the parasitoids Aphytis aonidiae and A. vandenboschi (Hymenoptera: Aphelinidae) foraging in the field. During 89.6 h of observation, we witnessed 18 encounters with predators, 6 of which resulted in parasitoid capture. Three classes of generalist predators attacked Aphytis: spiders (unidentified Salticidae and Thomisidae), workers of the Argentine ant, Linepithema humile, and nymphs of the assassin bug Zelus renardii (Hemiptera: Reduviidae). Although observations were conducted during most months of the year, encounters with predators occurred only during September, October and November. During these months, encounters with predators occurred on average every 3.2 h of observation, with one in three encounters resulting in parasitoid capture. Peaks in predation coincided with population peaks of A. aonidiae, but were unrelated to population dynamics of any of the predators. We compare these results with previously published laboratory studies on longevity of Aphytis parasitoids, and conclude that predation pressure has the potential to severely limit parasitoid fitness in the field.

A dynamic habitat selection game.

A patch selection game is formulated and analyzed. Organisms can forage in one of H patches. Each patch is characterized by the cost of foraging, the density and value of food, the predation risk, and the density of conspecifics. The presence of conspecifics affects the finding and sharing of food, and the predation risk. Optimal foraging theory can be viewed as a "1-person" game against nature in which the optimal patch choice of a specific organism is analyzed assuming that the number of conspecifics in other patches is fixed. In the general game theoretic approach, the behavior of conspecifics is included in the determination of the distinguished organism's strategy. An iterative algorithm is used to compute the solution of the "n-person" game or dynamic ESS, which differs from the optimal foraging theory solution. Experiments to test the proposed theory using rodents and seed trays are briefly discussed.

Complex adaptive systems, aging and longevity.

Mortality and reproduction are intimately entwined in the study of aging and longevity. I apply the modern theory of complex adaptive systems (nonlinear, stochastic, dynamic methods) to questions of aging and longevity. I begin by highlighting major questions that must be answered in order to obtain a deeper understanding of aging. These are: (i) What should (in an evolutionary sense) mortality trajectories look like? (ii) Why does caloric restriction slow aging? (iii) Why does reproduction cause delayed mortality? (iv) Why does compensatory growth cause delayed mortality? I show how dynamic state variable models based on stochastic dynamic programming (Clark & Mangel, 2000) can be used to embed genetic theories of senescence (either mutation accumulation or antagonistic pleiotropy) in the somatic environment, as George Williams called for in 1957, and how they make the disposable soma theory of aging operational. Such models will allow unification of genetic and phenotypic theories of aging.

Evolutionary optimization and neural network models of behavior.

One of the main challenges to the adaptionist program in general and the use of optimization models in behavioral and evolutionary ecology, in particular, is that organisms are so constrained by ontogeny and phylogeny that they may not be able to attain optimal solutions, however those are defined. This paper responds to the challenge through the comparison of optimality and neural network models for the behavior of an individual polychaete worm. The evolutionary optimization model is used to compute behaviors (movement in and out of a tube) that maximize a measure of Darwinian fitness based on individual survival and reproduction. The neural network involves motor, sensory, energetic reserve and clock neuronal groups. Ontogeny of the neural network is the change of connections of a single individual in response to its experiences in the environment. Evolution of the neural network is the natural selection of initial values of connections between groups and learning rules for changing connections. Taken together, these can be viewed as "design parameters". The best neural networks have fitnesses between 85% and 99% of the fitness of the evolutionary optimization model. More complicated models for polychaete worms are discussed. Formulation of a neural network model for host acceptance decisions by tephritid fruit flies leads to predictions about the neurobiology of the flies. The general conclusion is that neural networks appear to be sufficiently rich and plastic that even weak evolution of design parameters may be sufficient for organisms to achieve behaviors that give fitnesses close to the evolutionary optimal fitness, particularly if the behaviors are relatively simple.

Dynamic information in uncertain and changing worlds.

A general theory for information processing by organisms living in uncertain and changing worlds is developed. The three fundamental properties of the theory are: (i) the use of a memory parameter that allows the organism to forget the more distant past, (ii) a succinct representation of encounters and information and (iii) flexibility in the estimates of parameters by including the uncertainty in these estimates in a consistent manner. The theory is developed using Bayesian methods (but can also be applied to maximum likelihood estimation) and is applied to the encounter models standardly used in ecology (Poisson, binomial, and negative binomial). Two applications are discussed: (i) patch selection and the matching rule and (ii) superparasitism by a parasitoid.

Solution of functional difference equations from behavioral theory.

Behavioral models based on Markovian decision processes lead to functional difference equations for quantities such as the mean lifetime of the forager and the probability of reproductive success of the forager. In this paper, asymptotic and iterative methods are developed for the solution of such equations. The asymptotic methods are compared with numerical simulations. The iterative methods can be proved by a simple application of contraction mapping theorems.

A simple direct method for finding persistence times of populations and application to conservation problems.

The computation of persistence times of populations has become a central focus in conservation biology. We describe a simple, direct method for finding the statistics of persistence times by assuming that there is a maximum population size. Thus, even though the population dynamics may be very complex for population sizes below the maximum, it is possible to write a finite set of equations from which the mean and second moment of the persistence time can be found by using simple, algebraic methods. We apply the method to compute the mean and coefficient of variation of persistence times of populations that suffer large decrements (catastrophes). Our results show that in the presence of catastrophes, the increase in mean persistence time with large populations is not nearly as rapid as other theories suggest and that catastrophes occurring at even modest rates can considerably increase the risk of extinction.

When can a clonal organism escape senescence?

Some clonal organisms may live for thousands of years and show no signs of senescence, while others consistently die after finite life spans. Using two models, we examined how stage-specific life-history rates of a clone's modules determine whether a genetic individual escapes senescence by replacing old modules with new ones. When the rates of clonal or sexual reproduction and survival of individual modules decline with age, clones are more likely to experience senescence. In addition, the models predict that there is a greater tendency to find senescence in terms of a decline in the rate of sexual reproduction with clone age than in terms of an increase in the probability of clone mortality, unless rates of sexual reproduction increase dramatically with module stage. Using a matrix model modified to represent the clonal lifestyle, we show how a trade-off between sexual and clonal reproduction could result in selection for or against clonal senescence. We also show that, in contrast to unitary organisms, the strength of selection on life-history traits can increase with the age of a clone even in a growing population, countering the evolution of senescence.

Parent-offspring conflict and life-history consequences in herbivorous insects.

We develop an evolutionarily stable strategy theory of parent-offspring conflict in insect herbivores for the case in which offspring can choose to leave host plants on which they have been deposited by their mother. We find that a fundamental parent-offspring conflict in larval leaving rates occurs because individual larvae are more related to themselves than to their siblings whereas mothers are equally related to each of their offspring. Several patterns emerge: (1) The optimal probability of movement from the mother's perspective, P*(mom), is always greater than or equal to the optimal probability of movement from the offspring's perspective, P*(off), (2) a consequence of this difference in optimal probabilities of movement is that the mother's fitness for a given clutch is always greater for P*(mom) than P*(off), (3) as the payoff for leaving a plant decreases, (i) the optimal movement rates decrease and (ii) clutches become smaller, (4) as relatedness increases, optimal movement probabilities increase and this causes an increase in optimal clutches, and (5) the clutch size that maximizes the mother's lifetime fitness will frequently diverge from that which the mother would produce were the offspring to move at her optimal rate (i.e., P*(mom)).

The shape of things to come: using models with physiological structure to predict mortality trajectories.

If mortality rate is viewed as the outcome of processes of behavior, growth and reproduction, then it should be possible to predict mortality rate as a result of those processes. We provide two examples of how this may be done. In the first, we use the method of linear chains to treat mortality that is the result of multiple physiological processes, some of which may have delays. In the second, we assume that mortality is the result of damage associated with growth and metabolism. Both approaches lead to a rich diversity of predicted mortality trajectories. Although many of these look Gompertzian at young ages, the behavior at older ages depends upon the details of the physiological models.