Summary1. Accumulating evidence suggests that the average body size of many organisms is declining in response to climate warming. This phenomenon has been suggested to represent a universal response to warming that may impose significant adverse effects on ecosystem functioning and services. 2. However, we do not have a thorough understanding of why body sizes are commonly declining, and why some organisms show the opposite response. Because ectotherms constitute the vast majority of organism biomass and about 99% of species worldwide, it is particularly important to understand how ectotherms respond to a warming climate. 3. This review discusses the underlying physiological mechanisms of changes in ectotherm body size and addresses observed responses within a broad ecological context at different levels of organization, from individuals to communities, particularly in aquatic systems. 4. Warming-induced responses in average body size are not only determined by changes in rates of individual growth and development, but also mediated through size-dependent feedbacks at the population level, as well as competitive and predatory interactions within the community. Emergent properties at higher organizational levels have already been observed in both experimental and natural systems. 5. Various approaches will be required for enhancing our knowledge about the importance of such processes in natural systems. These include controlled semi-natural experiments and phylogenetic comparisons as well as statistical models of time-series data and theoretical models linking climate effects at the individual, population and community levels. 6. Understanding causes of observed changes in organism body sizes and how these depend on the ecological context is essential for improving our predictions and the management of ecosystems in the face of a warming climate.

Current understanding of animal population responses to rising temperatures is based on the assumption that biological rates such as metabolism, which governs fundamental ecological processes, scale independently with body size and temperature, despite empirical evidence for interactive effects. Here, we investigate the consequences of interactive temperature-and size scaling of vital rates for the dynamics of populations experiencing warming using a stage-structured consumer-resource model. We show that interactive scaling alters population and stage-specific responses to rising temperatures, such that warming can induce shifts in population regulation and stage-structure, influence community structure and govern population responses to mortality. Analysing experimental data for 20 fish species, we found size-temperature interactions in intraspecific scaling of metabolic rate to be common. Given the evidence for size-temperature interactions and the ubiquity of size structure in animal populations, we argue that accounting for size-specific temperature effects is pivotal for understanding how warming affects animal populations and communities.

The demographic structure of populations is affected by life history strategies and how these interact with natural and anthropogenic factors such as exploitation, climate change, and biotic interactions. Previous work suggests that the mean size and age of some North American populations of Chinook salmon (Oncorhynchus tshawytscha, Salmonidae) are declining. These trends are of concern because Chinook salmon are highly valued commercially for their exceptional size and because the loss of the largest and oldest individuals may lead to reduced population productivity. Using long‐term data from wild and hatchery populations, we quantified changes in the demographic structure of Chinook salmon populations over the past four decades across the Northeast Pacific Ocean, from California through western Alaska. Our results show that wild and hatchery fish are becoming smaller and younger throughout most of the Pacific coast. Proportions of older age classes have decreased over time in most regions. Simultaneously, the length‐at‐age of older fish has declined while the length‐at‐age of younger fish has typically increased. However, negative size trends of older ages were weak or non‐existent at the southern end of the range. While it remains to be explored whether these trends are caused by changes in climate, fishing practices or species interactions such as predation, our qualitative review of the potential causes of demographic change suggests that selective removal of large fish has likely contributed to the apparent widespread declines in average body sizes.

Metabolism constitutes a fundamental property of all organisms. Metabolic rate is commonly described to scale as a power function of body size and exponentially with temperature, thereby treating the effects of body size and temperature independently. Mounting evidence shows that the scaling of metabolic rate with body mass itself depends on temperature. Across‐species analyses in fishes suggest that the mass‐scaling exponent decreases with increasing temperature. However, whether this relationship holds at the within‐species level has rarely been tested. Here, we re‐analyse data on the metabolic rates of four freshwater fish species, two coregonids and two cyprinids, that cover wide ranges of body masses and their naturally experienced temperatures. We show that the standard metabolic rate of the coregonids is best fit when accounting for a linear temperature dependence of the scaling of metabolic rate with body mass, whereas a constant mass‐scaling exponent is supported in case of the cyprinids. Our study shows that phenotypic responses to temperature can result in temperature‐dependent scaling relationships at the species level and that these responses differ between taxa. Together with previous findings, these results indicate that evolutionarily adaptive and phenotypically plastic responses to temperature affect the scaling of metabolic rate with body mass in fishes.

Global warming impacts virtually all biota and ecosystems. Many of these impacts are mediated through direct effects of temperature on individual vital rates. Yet how this translates from the individual to the population level is still poorly understood, hampering the assessment of global warming impacts on population structure and dynamics. Here, we study the effects of temperature on intraspecific competition and cannibalism and the population dynamical consequences in a size-structured fish population. We use a physiologically structured consumer-resource model in which we explicitly model the temperature dependencies of the consumer vital rates and the resource population growth rate. Our model predicts that increased temperature decreases resource density despite higher resource growth rates, reflecting stronger intraspecific competition among consumers. At a critical temperature, the consumer population dynamics destabilize and shift from a stable equilibrium to competition-driven generation cycles that are dominated by recruits. As a consequence, maximum age decreases and the proportion of younger and smaller-sized fish increases. These model predictions support the hypothesis of decreasing mean body sizes due to increased temperatures. We conclude that in size-structured fish populations, global warming may increase competition, favor smaller size classes, and induce regime shifts that destabilize population and community dynamics.

Swimming performance is considered a main character determining survival in many aquatic animals. Body morphology highly influences the energetic costs and efficiency of swimming and sets general limits on a species capacity to use habitats and foods. For two cyprinid fishes with different morphological characteristics, carp (Cyprinus carpio L.) and roach (Rutilus rutilus (L.)), optimum swimming speeds (U(mc)) as well as total and net costs of transport (COT, NCOT) were determined to evaluate differences in their swimming efficiency. Costs of transport and optimum speeds proved to be allometric functions of fish mass. NCOT was higher but U(mc) was lower in carp, indicating a lower swimming efficiency compared to roach. The differences in swimming costs are attributed to the different ecological demands of the species and could partly be explained by their morphological characteristics. Body fineness ratios were used to quantify the influence of body shape on activity costs. This factor proved to be significantly different between the species, indicating a better streamlining in roach with values closer to the optimum body form for efficient swimming. Net swimming costs were directly related to fish morphology.

Fluctuations of fish populations abundances are shaped by the interplay between population dynamics and the stochastic forcing of the environment. Age-structured populations behave as a filter of the environment. This filter is characterised by the species-specific life cycle and life-history traits. An increased mortality of mature individuals alters these characteristics and may therefore induce changes in the variability of populations. The response of a generic age-structured model was analysed to investigate the expected changes in the fluctuations of fish populations in response to decreased adult survival. These expectations were then tested on an extensive dataset. In accordance with theory, the analyses revealed that decreased adult survival and mean age of spawners were linked to an increase in the relative importance of short-term fluctuations. It suggests that intensive exploitation can lead to a change in the variability of fish populations, an issue of central interest from both conservation and management perspectives.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. Online enhancements: appendixes, R code file.abstract: Predicted universal responses of ectotherms to climate warming include increased maximum population growth rate and changes in body size through the temperature-size rule. However, the mechanisms that would underlie these predicted responses are not clear. Many studies have focused on proximate mechanisms of physiological processes affecting individual growth. One can also consider ultimate mechanisms involving adaptive explanations by evaluating temperature effects on different vital rates across the life history and using the information in a population dynamical model. Here, we combine long-term data for a top predator in freshwater ecosystems (pike; Esox lucius) with a stochastic integral projection model to analyze concurrent effects of temperature on vital rates, body size, and population dynamics. As predicted, the net effect of warming on population growth rate (fitness) is positive, but the thermal sensitivity of this rate is highly size-and vital rate-dependent. These results are not sensitive to increasing variability in temperature. Somatic growth follows the temperature-size rule, and our results support an adaptive explanation for this response. The stable length structure of the population shifts with warming toward an increased proportion of medium-sized but a reduced proportion of small and large individuals. This study highlights how demographic approaches can help reveal complex underlying mechanisms for population responses to warming.