Growth and development rates have different thermal responses

Growth and development rates are fundamental to all living organisms. In a warming world, it is important to determine how these rates will respond to increasing temperatures. It is often assumed that the thermal responses of physiological rates are coupled to metabolic rate and thus have the same temperature dependence. However, the existence of the temperature-size rule suggests that intraspecific growth and development are decoupled. Decoupling of these rates would have important consequences for individual species and ecosystems, yet this has not been tested systematically across a range of species. We conducted an analysis on growth and development rate data compiled from the literature for a well-studied group, marine pelagic copepods, and use an information-theoretic approach to test which equations best describe these rates. Growth and development rates were best characterized by models with significantly different parameters: development has stronger temperature dependence than does growth across all life stages. As such, it is incorrect to assume that these rates have the same temperature dependence. We used the best-fit models for these rates to predict changes in organism mass in response to temperature. These predictions follow a concave relationship, which complicates attempts to model the impacts of increasing global temperatures on species body size.     

Building integral projection models with nonindependent vital rates

Population dynamics are functions of several demographic processes including survival, reproduction, somatic growth, and maturation. The rates or probabilities for these processes can vary by time, by location, and by individual. These processes can co‐vary and interact to varying degrees, e.g., an animal can only reproduce when it is in a particular maturation state. Population dynamics models that treat the processes as independent may yield somewhat biased or imprecise parameter estimates, as well as predictions of population abundances or densities. However, commonly used integral projection models (IPMs) typically assume independence across these demographic processes. We examine several approaches for modelling between process dependence in IPMs and include cases where the processes co‐vary as a function of time (temporal variation), co‐vary within each individual (individual heterogeneity), and combinations of these (temporal variation and individual heterogeneity). We compare our methods to conventional IPMs, which treat vital rates independent, using simulations and a case study of Soay sheep (Ovis aries). In particular, our results indicate that correlation between vital rates can moderately affect variability of some population‐level statistics. Therefore, including such dependent structures is generally advisable when fitting IPMs to ascertain whether or not such between vital rate dependencies exist, which in turn can have subsequent impact on population management or life‐history evolution.     

Incorporating spatial autocorrelation into species distribution models alters forecasts of climate‐mediated range shifts

Species distribution models (SDMs) are widely used to forecast changes in the spatial distributions of species and communities in response to climate change. However, spatial autocorrelation (SA) is rarely accounted for in these models, despite its ubiquity in broad‐scale ecological data. While spatial autocorrelation in model residuals is known to result in biased parameter estimates and the inflation of type I errors, the influence of unmodeled SA on species' range forecasts is poorly understood. Here we quantify how accounting for SA in SDMs influences the magnitude of range shift forecasts produced by SDMs for multiple climate change scenarios. SDMs were fitted to simulated data with a known autocorrelation structure, and to field observations of three mangrove communities from northern Australia displaying strong spatial autocorrelation. Three modeling approaches were implemented: environment‐only models (most frequently applied in species' range forecasts), and two approaches that incorporate SA; autologistic models and residuals autocovariate (RAC) models. Differences in forecasts among modeling approaches and climate scenarios were quantified. While all model predictions at the current time closely matched that of the actual current distribution of the mangrove communities, under the climate change scenarios environment‐only models forecast substantially greater range shifts than models incorporating SA. Furthermore, the magnitude of these differences intensified with increasing increments of climate change across the scenarios. When models do not account for SA, forecasts of species' range shifts indicate more extreme impacts of climate change, compared to models that explicitly account for SA. Therefore, where biological or population processes induce substantial autocorrelation in the distribution of organisms, and this is not modeled, model predictions will be inaccurate. These results have global importance for conservation efforts as inaccurate forecasts lead to ineffective prioritization of conservation activities and potentially to avoidable species extinctions.     

Stream isotherm shifts from climate change and implications for distributions of ectothermic organisms

Stream ecosystems are especially vulnerable to climate warming because most aquatic organisms are ectothermic and live in dendritic networks that are easily fragmented. Many bioclimatic models predict significant range contractions in stream biotas, but subsequent biological assessments have rarely been done to determine the accuracy of these predictions. Assessments are difficult because model predictions are either untestable or so imprecise that definitive answers may not be obtained within timespans relevant for effective conservation. Here, we develop the equations for calculating isotherm shift rates (ISRs) in streams that can be used to represent historic or future warming scenarios and be calibrated to individual streams using local measurements of stream temperature and slope. A set of reference equations and formulas are provided for application to most streams. Example calculations for streams with lapse rates of 0.8 °C/100 m and long‐term warming rates of 0.1–0.2 °C decade^?1 indicate that isotherms shift upstream at 0.13–1.3 km decade^?1 in steep streams (2–10% slope) and 1.3–25 km decade^?1 in flat streams (0.1–1% slope). Used more generally with global scenarios, the equations predict isotherms shifted 1.5–43 km in many streams during the 20th Century as air temperatures increased by 0.6 °C and would shift another 5–143 km in the first half of the 21st Century if midrange projections of a 2 °C air temperature increase occur. Variability analysis suggests that short‐term variation associated with interannual stream temperature changes will mask long‐term isotherm shifts for several decades in most locations, so extended biological monitoring efforts are required to document anticipated distribution shifts. Resampling of historical sites could yield estimates of biological responses in the short term and should be prioritized to validate bioclimatic models and develop a better understanding about the effects of temperature increases on stream biotas.     

Soil respiration does not acclimatize to warmer temperatures when modeled over seasonal timescales

Soil warming studies have generally demonstrated an ephemeral response of soil respiration to warming suggesting acclimatization to increased temperatures. Many of these studies depict acclimatization as an empirical temperature-respiration model with data collected from late spring through early autumn. We examined the apparent temperature sensitivity of soil respiration in chronically warmed soils over three different timescales: annually, during the growing season, and seasonally during winter, spring, summer, and fall. Temperature sensitivity was evaluated by fitting exponential and flexible temperature functions as mixed effects models. From model coefficients, we estimated annual, growing season, and season-specific Q ₁₀ values, and assessed the ability of model coefficients to predict daily soil respiration rates over a two-year period. We found that respiration in warmed soils can exhibit characteristics of acclimatized temperature sensitivity depending on the timeframe and the function (exponential or flexible) used. Models using growing season data suggested acclimatization while models using data collected in winter or spring indicated enhanced temperature sensitivity with 5 °C of warming. Differences in temperature sensitivity affected predicted daily soil respiration rates, particularly in winter and spring. Models constructed over longer timescales overestimated daily respiration rates by as much 10–40 % whereas season-specific predictions were generally within 2 % of actual values. Failure to use season-specific models to depict changes in temperature dependence may over- or under-estimate carbon losses due to climate warming, especially during the colder months of the year.     

Matching the forecast horizon with the relevant spatial and temporal processes and data sources

Most phenomenological, statistical models used to generate ecological forecasts take either a time-series approach, based on long-term data from one location, or a space-for-time approach, based on data describing spatial patterns across environmental gradients. However, the magnitude and even the sign of environment–response relationships detected using these two approaches often differs, leading to contrasting predictions about responses to future environmental change. Here we consider how the forecast horizon determines whether more accurate predictions come from the time-series approach, the space-for-time approach or a combination of the two. As proof of concept, we use simulated case studies to show that forecasts for short and long forecast horizons need to focus on different ecological processes, which are reflected in different kinds of data. First, we simulated population or community dynamics under stationary temperature using two simple, mechanistic models. Second, we fit statistical models to the simulated data using a time-series approach, a space-for-time approach or a weighted average. We then forecast the response to a temperature increase using the statistical models, and compared these forecasts to temperature effects simulated by the mechanistic models. We found that the time-series approach made accurate short-term predictions because it captured initial conditions and effects of fast processes such as birth and death. The space-for-time approach made more accurate long-term predictions because it better captured the influence of slower processes such as evolutionary and ecological selection. The weighted average made accurate predictions at all time scales, including intermediate time-scales where the other two approaches performed poorly. A weighted average of time-series and space-for-time approaches shows promise, but making this weighted model operational will require new research to predict the rate at which slow processes begin to influence dynamics.     

Rates of biotic interactions scale predictably with temperature despite variation

Most biological processes are temperature dependent. To quantify the temperature dependence of biotic interactions and evaluate predictions of metabolic theory, we: 1) compiled a database of 81 studies that provided 112 measures of rates of herbivory, predation, parasitism, parasitoidy, or competition between two species at two or more temperatures; and 2) analyzed the temperature dependence of these rates in the framework of metabolic ecology to test our prediction that the “activation energy,”E, centers around 0.65 eV. We focused on studies that assessed rates or associated times of entire biotic interactions, such as time to consumption of all prey, rather than rates of components of these interactions, such as prey encounter rate. Results were: 1) the frequency distribution of E for each interaction type was typically peaked and right skewed; 2) the overall mean is E= 0.96 eV and median E= 0.78 eV; 3) there was significant variation in E within but not across interaction types; but 4) average values of E were not significantly different from 0.65 eV by interaction type and 5) studies with measurements at more temperatures were more consistent with E= 0.65 eV. These synthetic findings suggest that, despite the many complicating factors, the temperature‐dependence of rates of biotic interactions broadly reflect of rates of metabolism, a relationship with important implications for a warming world.     

Opening the black box of plant nutrient uptake under warming predicts global patterns in community biomass and biological carbon storage

The effect of climate change on the amount of carbon stored in the different biological compartments of complex natural communities is relevant for a range of ecosystem functions and services. Temperature‐dependency of many physiological and ecological processes drives this storage capacity. As opposed to other physiological rates, the temperature‐dependence of nutrient uptake by plants has, to date, not been thoroughly investigated and therefore was not explicitly included in food web models. In a meta‐study, we extracted experimental data to establish the temperature‐dependence of the parameters determining plant nutrient uptake. Overall, we found an increase in the maximum uptake rate, as well as the half‐saturation density. As the respiration rates of plants (biomass loss) increase more strongly than the nutrient uptake rates (driving biomass gain under nutrient limitation), our results suggest that warming should decrease plant biomass. We applied these temperature‐dependent nutrient uptake rates by plants to a model of a three‐level food‐chain composed of two nutrients, a plant pool, and an herbivore pool. Having established plant nutrient uptake rates based on real data to replace the previously used assumption of logistic growth, we were able to use realistic natural nutrient deposition rates as the input variables in this model. This mechanistic model approach allowed us to show the quantitative responses of natural communities to realistic fertilization rates for the first time. We ran the model under realistic nutrient supply scenarios based on deposition data from the literature, adding a scenario of anthropogenic fertilization. We found decreases in overall community biomass with increasing temperature, but the intensity of this decrease varied strongly depending on the nutrient supply scenario. Our findings highlight the importance of including other global change drivers besides warming, as they can mediate the temperature impact on changes in global carbon storage and thus biomass‐related ecosystem services.     

The sensitivity of soil respiration to soil temperature,moisture, and carbon supply at the global scale

Soil respiration (Rs) is a major pathway by which fixed carbon in the biosphere is returned to the atmosphere, yet there are limits to our ability to predict respiration rates using environmental drivers at the global scale. While temperature, moisture, carbon supply, and other site characteristics are known to regulate soil respiration rates at plot scales within certain biomes, quantitative frameworks for evaluating the relative importance of these factors across different biomes and at the global scale require tests of the relationships between field estimates and global climatic data. This study evaluates the factors driving Rs at the global scale by linking global datasets of soil moisture, soil temperature, primary productivity, and soil carbon estimates with observations of annual Rs from the Global Soil Respiration Database (SRDB). We find that calibrating models with parabolic soil moisture functions can improve predictive power over similar models with asymptotic functions of mean annual precipitation. Soil temperature is comparable with previously reported air temperature observations used in predicting Rs and is the dominant driver of Rs in global models; however, within certain biomes soil moisture and soil carbon emerge as dominant predictors of Rs. We identify regions where typical temperature‐driven responses are further mediated by soil moisture, precipitation, and carbon supply and regions in which environmental controls on high Rs values are difficult to ascertain due to limited field data. Because soil moisture integrates temperature and precipitation dynamics, it can more directly constrain the heterotrophic component of Rs, but global‐scale models tend to smooth its spatial heterogeneity by aggregating factors that increase moisture variability within and across biomes. We compare statistical and mechanistic models that provide independent estimates of global Rs ranging from 83 to 108 Pg yr^?1, but also highlight regions of uncertainty where more observations are required or environmental controls are hard to constrain.     

Viscosity of supercooled aqueous glycerol solutions, validity of the Stokes-Einstein relationship, and implications for cryopreservation

The viscosity of supercooled glycerol aqueous solutions, with glycerol mass fractions between 0.70 and 0.90, have been determined to confirm that the Avramov-Milchev equation describes very well the temperature dependence of the viscosity of the binary mixtures including the supercooled regime. On the contrary, it is shown that the free volume model of viscosity, with the parameters proposed in a recent work (He, Fowler, Toner, J. Appl. Phys. 100 (2006) 074702), overestimates the viscosity of the glycerol-rich mixtures at low temperatures by several orders of magnitude. Moreover, the free volume model for the water diffusion leads to predictions of the Stokes-Einstein product, which are incompatible with the experimental findings. We conclude that the use of these free volume models, with parameters obtained by fitting experimental data far from the supercooled and glassy regions, lead to incorrect predictions of the deterioration rates of biomolecules, overestimating their life times in these cryopreservation media.     

The combined effects of reactant kinetics and enzyme stability explain the temperature dependence of metabolic rates

A mechanistic understanding of the response of metabolic rate to temperature is essential for understanding thermal ecology and metabolic adaptation. Although the Arrhenius equation has been used to describe the effects of temperature on reaction rates and metabolic traits, it does not adequately describe two aspects of the thermal performance curve (TPC) for metabolic rate—that metabolic rate is a unimodal function of temperature often with maximal values in the biologically relevant temperature range and that activation energies are temperature dependent. We show that the temperature dependence of metabolic rate in ectotherms is well described by an enzyme‐assisted Arrhenius (EAAR) model that accounts for the temperature‐dependent contribution of enzymes to decreasing the activation energy required for reactions to occur. The model is mechanistically derived using the thermodynamic rules that govern protein stability. We contrast our model with other unimodal functions that also can be used to describe the temperature dependence of metabolic rate to show how the EAAR model provides an important advance over previous work. We fit the EAAR model to metabolic rate data for a variety of taxa to demonstrate the model's utility in describing metabolic rate TPCs while revealing significant differences in thermodynamic properties across species and acclimation temperatures. Our model advances our ability to understand the metabolic and ecological consequences of increases in the mean and variance of temperature associated with global climate change. In addition, the model suggests avenues by which organisms can acclimate and adapt to changing thermal environments. Furthermore, the parameters in the EAAR model generate links between organismal level performance and underlying molecular processes that can be tested for in future work.     

Investigating biological control over soil carbon temperature sensitivity

Understanding the temperature sensitivity of soil respiration is critical for predicting the response of ecosystems to climate change, yet the microbial communities responsible are rarely considered explicitly in studies or models. In this study, we assessed total microbial community composition, quantified bacterial respiration temperature response, and investigated the temperature dependence of bacterial carbon substrate utilization in tropical, temperate, and taiga soils (from Puerto Rico, California, and Alaska). Microbial community composition was characterized using phospholipid fatty acid analysis. Bacterial community respiration on a standardized set of substrates was ascertained using the BiOLOG^™ substrate utilization assay incubated at four temperatures: 4, 12, 28, and 40 °C. First, we found that microbial communities from the three latitudes were compositionally distinct and that the bacterial component of the three communities had markedly different respiration temperature–response curves corresponding with their experienced temperature regimes. We use these data to highlight limitations of widely used temperature–response equations and investigate temperature-dependent patterns of substrate utilization. We found that temperature response, in terms of both respiration rates and substrate use, varied for these bacterial communities independent of substrate quality or quantity interactions such as labile depletion. In contrast to the common assumption of heterotrophic microbial ubiquity, we found that bacterial community differences from these diverse systems appeared to determine both rates of respiration and patterns of carbon substrate usage. We suggest that microbial community composition-specific responses to changing climate may be important in predicting the long-term role of ecosystems in atmospheric CO₂ dynamics.     

Strong Discrepancies between Local Temperature Mapping and Interpolated Climatic Grids in Tropical Mountainous Agricultural Landscapes

Bridging the gap between the predictions of coarse-scale climate models and the fine-scale climatic reality of species is a key issue of climate change biology research. While it is now well known that most organisms do not experience the climatic conditions recorded at weather stations, there is little information on the discrepancies between microclimates and global interpolated temperatures used in species distribution models, and their consequences for organisms’ performance. To address this issue, we examined the fine-scale spatiotemporal heterogeneity in air, crop canopy and soil temperatures of agricultural landscapes in the Ecuadorian Andes and compared them to predictions of global interpolated climatic grids. Temperature time-series were measured in air, canopy and soil for 108 localities at three altitudes and analysed using Fourier transform. Discrepancies between local temperatures vs. global interpolated grids and their implications for pest performance were then mapped and analysed using GIS statistical toolbox. Our results showed that global interpolated predictions over-estimate by 77.5±10% and under-estimate by 82.1±12% local minimum and maximum air temperatures recorded in the studied grid. Additional modifications of local air temperatures were due to the thermal buffering of plant canopies (from −2.7°K during daytime to 1.3°K during night-time) and soils (from −4.9°K during daytime to 6.7°K during night-time) with a significant effect of crop phenology on the buffer effect. This discrepancies between interpolated and local temperatures strongly affected predictions of the performance of an ectothermic crop pest as interpolated temperatures predicted pest growth rates 2.3–4.3 times lower than those predicted by local temperatures. This study provides quantitative information on the limitation of coarse-scale climate data to capture the reality of the climatic environment experienced by living organisms. In highly heterogeneous region such as tropical mountains, caution should therefore be taken when using global models to infer local-scale biological processes.     

Thermodynamic theory explains the temperature optima of soil microbial processes and high Q10 values at low temperatures

Our current understanding of the temperature response of biological processes in soil is based on the Arrhenius equation. This predicts an exponential increase in rate as temperature rises, whereas in the laboratory and in the field, there is always a clearly identifiable temperature optimum for all microbial processes. In the laboratory, this has been explained by denaturation of enzymes at higher temperatures, and in the field, the availability of substrates and water is often cited as critical factors. Recently, we have shown that temperature optima for enzymes and microbial growth occur in the absence of denaturation and that this is a consequence of the unusual heat capacity changes associated with enzymes. We have called this macromolecular rate theory – MMRT (Hobbs et al., 2013 , ACS Chem. Biol. 8:2388). Here, we apply MMRT to a wide range of literature data on the response of soil microbial processes to temperature with a focus on respiration but also including different soil enzyme activities, nitrogen and methane cycling. Our theory agrees closely with a wide range of experimental data and predicts temperature optima for these microbial processes. MMRT also predicted high relative temperature sensitivity (as assessed by Q₁₀ calculations) at low temperatures and that Q₁₀ declined as temperature increases in agreement with data synthesis from the literature. Declining Q₁₀ and temperature optima in soils are coherently explained by MMRT which is based on thermodynamics and heat capacity changes for enzyme‐catalysed rates. MMRT also provides a new perspective, and makes new predictions, regarding the absolute temperature sensitivity of ecosystems – a fundamental component of models for climate change.     

Interactions between climate and habitat loss effects on biodiversity: a systematic review and meta‐analysis

Climate change and habitat loss are both key threatening processes driving the global loss in biodiversity. Yet little is known about their synergistic effects on biological populations due to the complexity underlying both processes. If the combined effects of habitat loss and climate change are greater than the effects of each threat individually, current conservation management strategies may be inefficient and at worst ineffective. Therefore, there is a pressing need to identify whether interacting effects between climate change and habitat loss exist and, if so, quantify the magnitude of their impact. In this article, we present a meta‐analysis of studies that quantify the effect of habitat loss on biological populations and examine whether the magnitude of these effects depends on current climatic conditions and historical rates of climate change. We examined 1319 papers on habitat loss and fragmentation, identified from the past 20 years, representing a range of taxa, landscapes, land‐uses, geographic locations and climatic conditions. We find that current climate and climate change are important factors determining the negative effects of habitat loss on species density and/or diversity. The most important determinant of habitat loss and fragmentation effects, averaged across species and geographic regions, was current maximum temperature, with mean precipitation change over the last 100 years of secondary importance. Habitat loss and fragmentation effects were greatest in areas with high maximum temperatures. Conversely, they were lowest in areas where average rainfall has increased over time. To our knowledge, this is the first study to conduct a global terrestrial analysis of existing data to quantify and test for interacting effects between current climate, climatic change and habitat loss on biological populations. Understanding the synergistic effects between climate change and other threatening processes has critical implications for our ability to support and incorporate climate change adaptation measures into policy development and management response.     

Oil palm expansion increases the vectorial capacity of dengue vectors in Malaysian Borneo

Changes in land-use and the associated shifts in environmental conditions can have large effects on the transmission and emergence of mosquito-borne disease. Mosquito-borne disease are particularly sensitive to these changes because mosquito growth, reproduction, survival and susceptibility to infection are all thermally sensitive traits, and land use change dramatically alters local microclimate. Predicting disease transmission under environmental change is increasingly critical for targeting mosquito-borne disease control and for identifying hotspots of disease emergence. Mechanistic models offer a powerful tool for improving these predictions. However, these approaches are limited by the quality and scale of temperature data and the thermal response curves that underlie predictions. Here, we used fine-scale temperature monitoring and a combination of empirical, laboratory and temperature-dependent estimates to estimate the vectorial capacity of Aedes albopictus mosquitoes across a tropical forest–oil palm plantation conversion gradient in Malaysian Borneo. We found that fine-scale differences in temperature between logged forest and oil palm plantation sites were not sufficient to produce differences in temperature-dependent demographic trait estimates using published thermal performance curves. However, when measured under field conditions a key parameter, adult abundance, differed significantly between land-use types, resulting in estimates of vectorial capacity that were 1.5 times higher in plantations than in forests. The prediction that oil palm plantations would support mosquito populations with higher vectorial capacity was robust to uncertainties in our adult survival estimates. These results provide a mechanistic basis for understanding the effects of forest conversion to agriculture on mosquito-borne disease risk, and a framework for interpreting emergent relationships between land-use and disease transmission. As the burden of Ae. albopictus-vectored diseases, such as dengue virus, increases globally and rising demand for palm oil products drives continued expansion of plantations, these findings have important implications for conservation, land management and public health policy at the global scale.     

Theoretical predictions for how temperature affects the dynamics of interacting herbivores and plants

Concern about climate change has spurred experimental tests of how warming affects species' abundance and performance. As this body of research grows, interpretation and extrapolation to other species and systems have been limited by a lack of theory. To address the need for theory for how warming affects species interactions, we used consumer-prey models and the metabolic theory of ecology to develop quantitative predictions for how systematic differences between the temperature dependence of heterotrophic and autotrophic population growth lead to temperature-dependent herbivory. We found that herbivore and plant abundances change with temperature in proportion to the ratio of autotrophic to heterotrophic metabolic temperature dependences. This result is consistent across five different formulations of consumer-prey models and over varying resource supply rates. Two models predict that temperature-dependent herbivory causes primary producer abundance to be independent of temperature. This finding contradicts simpler extensions of metabolic theory to abundance that ignore trophic interactions, and is consistent with patterns in terrestrial ecosystems. When applied to experimental data, the model explained 77% and 66% of the variation in phytoplankton and zooplankton abundances, respectively. We suggest that metabolic theory provides a foundation for understanding the effects of temperature change on multitrophic ecological communities.     

The migration of sessile organisms: a simulation model with measurable parameters

Abstract. Palaeoecologists have shown that trees migrated at rates of 100–1000 m/yr in response to post-glacial warming. In order to predict the impact of forecast anthropogenic climate changes upon forest ecosystems we need to simulate how trees may migrate in response to the changes predicted for the next 1–2 centuries. These predictions must take account of the impacts upon migration of human land-use and habitat fragmentation. We have developed a spatially-explicit mechanistic model (MIGRATE) able to simulate the migration of a single species across a realistically heterogeneous landscape. MIGRATE uses biological parameters that readily may be estimated from data in the literature or from field studies, and represents the landscape as a grid of cells, each with an associated carrying capacity. A one-dimensional version of MIGRATE has been compared both with Skellam's (1951) diffusion model and with the more recent analytical models of van den Bosch et al. (1990, 1992); despite its fundamentally different approach, MIGRATE provides comparable estimates of migration rates, given equivalent input parameters. An example is described that demonstrates the ability of the two-dimensional version of MIGRATE to simulate the likely pattern of spread of a species across a heterogeneous landscape. It is argued that MIGRATE, or models like it, will play a central role in a spatially-hierarchic modelling strategy that must be developed if we are to achieve the goal of simulating the likely response of trees, and other organisms, to both global climate change and the increasing pressures of human land-use.     

Empirical modelling of lake water-temperature relationships: a comparison of approaches

1.  As a result of the role that temperature plays in many aquatic processes, good predictive models of annual maximum near-surface lake water temperature across large spatial scales are needed, particularly given concerns regarding climate change. Comparisons of suitable modelling approaches are required to determine their relative merit and suitability for providing good predictions of current conditions. We developed models predicting annual maximum near-surface lake water temperatures for lakes across Canada using four statistical approaches: multiple regression, regression tree, artificial neural networks and Bayesian multiple regression.
2.  Annual maximum near-surface (from 0 to 2 m) lake water-temperature data were obtained for more than 13 000 lakes and were matched to geographic, climatic, lake morphology, physical habitat and water chemistry data. We modelled 2348 lakes and three subsets thereof encompassing different spatial scales and predictor variables to identify the relative importance of these variables at predicting lake temperature.
3.  Although artificial neural networks were marginally better for three of the four data sets, multiple regression was considered to provide the best solution based on the combination of model performance and computational complexity. Climatic variables and date of sampling were the most important variables for predicting water temperature in our models.
4.  Lake morphology did not play a substantial role in predicting lake temperature across any of the spatial scales. Maximum near-surface temperatures for Canadian lakes appeared to be dominated by large-scale climatic and geographic patterns, rather than lake-specific variables, such as lake morphology and water chemistry.