A functional approach to understanding how temperature and habitat dimensionality drive universality and variation in ecological systems

Understanding the processes that form, drive and maintain ecological systems is fundamental to understanding basic and applied biology. In particular, identifying mechanisms that regulate the structure and functioning of ecological systems over a variety of levels, from individuals and populations, through communities and ecosystems, is a central goal of ecology. Understanding how the physical environment influences these mechanisms, and therefore why communities differ between habitats, is also a key goal. Significant advances on these research fronts have been made, but scientists are still a long way from a mechanistic understanding of ecological systems. The aim of this thesis is to improve understanding how local-scale interactions of individuals unite to drive and stabilise patterns at higher levels of ecological organisation, and to determine what role the physical environment plays. Using an approach that combines meta-analysis of published data, ecoinformatics, and mechanistic theory, this thesis explores how the physical environment drives universality and variation in ecological systems. I begin in Chapter 1 with a brief overview of ecological systems and approaches to identifying, studying and understanding mechanisms that determine their organisation. Ecological communities are the framework within which life operates, while individuals and their interactions with each other and the physical environment are the fundamental building blocks of communities. Consumer–resource interactions are especially important because they determine energy flow between individuals, and fluxes and stability in communities. The physical environment can have profound effects on ecological organisation, at the local scale (e.g., species interactions) and on entire systems (e.g., body size and abundance distribution, food web architecture). Together, these factors demand a functional approach, which explicitly links mechanisms acting across multiple levels of organisation, to understanding ecological systems and their dependence on the physical environment. Environmental temperature has strong and systematic effects on biological processes at all levels of organisation, ranging from cells to ecosystems, so it is surprising that little is known about the general mechanisms by which temperature affects biological systems. In Chapter 2 I present a dataset on how diverse biological rates and times respond to temperature, which I then analyse in two subsequent chapters to aid in the search for general mechanisms of thermal dependence. For nearly a century, intraspecific studies (within single species' populations) of thermal responses have been conducted on a wide range of organismal traits. Comparative studies of these data are essential for elucidating mechanisms underlying thermal response curves. However, such comparative intraspecific studies have been limited because of a lack of a comprehensive database that organizes these data with consistent units and trait definitions. I present a database of 2,352 thermal responses for 220 traits for microbes, plants, and animals compiled from 270 published sources. This represents the most diverse and comprehensive thermal response dataset ever compiled. The traits in this database span levels of biological organisation from internal physiology to species interactions, and were measured in marine, freshwater, and terrestrial habitats for 411 species. Although I include some physiological rates, most data are for ecological traits, which I define to mean any organismal trait that directly determines interactions between individuals within or between species. Analyses of this dataset should provide new insights into generalities and deviations in the thermal dependence of biological traits, and thus how biological systems, from cells to ecosystems, respond to temperature change. Such insights are essential for understanding how natural biological systems function, and for how life is responding to Earth's complex and rapidly changing thermal landscape. In Chapter 3 I analyse this database to understand patterns in the temperature dependence of physiological and ecological traits. After removal of data to only retain high quality thermal responses, I analyse responses ranging over 66°C and representing 112 traits, 309 species, and spanning 15 orders of magnitude in body size. I analyse three components of the thermal response: the initial increase in the trait value with temperature (rise), its ultimate decrease at higher temperatures (fall), and the transition between the rise and fall components (unimodal). The diversity and number of traits, species, and habitats allows me to identify and quantify novel features of the temperature response of biological traits. Analysis of the rising component of within-species (intraspecific) responses reveals that 87% fit the Boltzmann-Arrhenius model. The Boltzmann-Arrhenius model is a model of the thermal dependence of traits based on chemical reaction kinetics, and can be used to estimate the activation energy required for biochemical reactions. The rate of rise (or fall) of a Boltmann- Arrhenius model is described by an activation energy, which is essentially the amount of energy required for biochemical reactions to occur with a higher activation energy meaning that the temperature dependence is stronger (see Chapter 3 for more detail). I found that the mean activation energy for rises is 0.66±0.05 eV, similar to the often-reported across-species (interspecific) value of 0.65 eV. However, systematic variation in the distribution of activation energies is evident, including previously unrecognised right-skewness. This right-skewness exists across levels of organisation, taxa, trophic groups, and habitats, and can be partly explained by a thermal version of the life-dinner principle—stronger selection to run for your life than to run for your dinner. For unimodal responses, habitat (marine, freshwater, or terrestrial) largely explains the mean temperature at which trait values peak but not the variation around this mean. Results of this chapter highlight generalities and deviations in the temperature response of biological traits and help provide a basis to better predict how ecological systems, from individuals to communities, respond to temperature. In Chapter 4 I combine the empirical thermal scaling's of ecological traits (Chapter 3) with consumer-resource allometric theory to present a mechanistic model for the thermal response of consumer-resource interactions. I focus on how temperature affects species interactions via key traits—body velocity, detection distance, search rate, and handling time—that underlie per-capita consumption rate. The model is general because it applies to all foraging strategies: active-capture (both consumer and resource velocity are important), sit-and-wait (resource velocity dominates), and grazing (consumer velocity dominates). The model predicts that temperature influences consumer-resource interactions primarily through its effects on body velocity (either of the consumer, resource, or both), which determines how often consumers and resources encounter each other, and that asymmetries in the thermal responses of interacting species can introduce qualitative, not just quantitative, changes in consumer-resource dynamics. Using the model, I also derive predictions for the effect of temperature on equilibrium population densities. Using the database described in Chapter 2, I analyse temperature responses for 309 species that span 15 orders of magnitude in body size and live in terrestrial, marine, and freshwater habitats, and find extensive evidence for asymmetries in consumer-resource thermal responses. In particular, I identify three general types of asymmetries: i) different magnitude of response, ii) different rates of response (e.g., activation energies), and iii) different peak temperatures. Such asymmetries should occur more frequently as the climate changes and species' geographical distribution and phenology are altered, such that previously non-interacting species come into contact. By using characteristics of trophic interactions that are often well known, such as body size, foraging strategy, thermy, and environmental temperature, this framework should allow more accurate predictions about the thermal dependence of food-web and ecosystem dynamics, including how natural systems will respond to current and future temperature change. Spatial dimensionality is a fundamental physical property of all environments, and is probably one of the most important differences between aquatic and terrestrial habitats. Therefore, in Chapter 5 I use additional data obtained from the literature, combined with the database described in Chapter 2, to explore in detail how habitat dimensionality influences ecological organisation. I show how substantial variation in consumption rate data, and thus trophic interaction strengths, arises because (for a given resource density) consumers generally encounter resources more frequently in three dimensions (3D) than two dimensions (2D). By extending the allometric theory developed in Chapter 4, and literature data for 376 species, I show that consumption rates scale sub-linearly with consumer body mass (exponent ~0.85) for 2D interactions, but super-linearly (~1.06) in 3D, contrasting with the current usage of a single exponent (~0.75) in food web research. A further analysis of 2,930 consumer-resource interactions shows that interaction dimensionality is potentially a major driver of species coexistences, population stability, and abundances in local communities. In the final chapter (Chapter 6) I synthesize the results of previous chapters, and suggest directions for future research. Each chapter provides insights into how the physical environment constrains the organisation of ecological systems. Detailed analysis of how the environment constrains species interactions, specifically temperature (Chapter 2 – Chapter 4) and habitat dimensionality (Chapter 5), show how physical drivers can alter the nature of species interactions, which are central to community organisation. These results clearly show that constraints on species interactions, such as from environmental drivers, can drive the organisation of populations and communities. These processes are central for understanding how natural ecosystems will respond to natural and anthropogenic-caused variation in the physical environment