A global analysis of patterns of aquatic ectotherm growth rate variation

Emma R. Moffett, Ecology and Evolutionary Biology, University of California Irvine, Irvine, CA, Stephan B. Munch, Southwest Fisheries Science Center, National Oceanic and Atmospheric Administration, Santa Cruz, CA, David Fryxell, Ecology and Evolutionary Biology, University of California Santa Cruz, Santa Cruz, CA, K B Oke, University of Alaska Fairbanks, AK and Celia C. Symons, Ecology and Evolutionary Biology, University of California - Irvine, Irvine, CA

Background/Question/Methods
Physiological responses to temperature have often been used to infer how species will respond to climate change. In ectotherms, growth within a species typically increases with temperature to a certain optimum, after which growth decelerates. Measurements made in lab growth assays have been extrapolated to climate change scenarios to make predictions of future growth; however, lab assays ignore the possibility of local adaptation to environmental temperatures. Such adaptive responses may counteract the effect of slow growth environments, such as colder areas at high latitude and altitude. However, these adaptive responses may be difficult to detect in nature where genetic change dampens the effect of the environment (countergradient variation), resulting in negligible phenotypic differences across temperature gradients. To understand the prevalence of countergradient growth, we collated growth rate data from laboratory and field studies on aquatic organisms. Here, we focus on patterns within species to 1) determine the relationship between growth rate and temperature in wild-caught organisms globally, and 2) compare temperature sensitivity of wild populations to lab assays to infer the extent of countergradient variation in growth rate. This will allow us to determine if local adaptation disrupts the expectation of increasing growth rate with rising temperatures.
Results/Conclusions
We compiled over 16,000 observations of growth rate data across 250 families, and have lab and wild growth data from 51 species. Across all individuals, when plotted as an Arrhenius function, growth rate increased with temperature, but the slope of this relationship was close to zero (0.007), and there was substantial variation among species (slopes range from -6.5 to 5.8). In contrast, when growth rates were measured in a common environment, they demonstrated stronger and more consistent patterns with temperature. For example, juvenile growth rates increased strongly with temperature (Arrhenius slope; ~-0.402). Similarly, development time decreased strongly (~0.341) with temperature, and there was less variation in slopes among species. Our data show that wild-caught organisms had a similar growth rate, no matter the temperature they experienced. In contrast, the expected positive relationship between growth and temperature was recapitulated when only considering lab studies that exposed one population to different temperatures, consistent with countergradient variation. Thus, growth rates commonly counteract the phenotypic effect of temperature in the wild, moderating how changes in growth rates are expressed. Ultimately, our results suggest that failing to account for such countergradient changes may hinder accurate predictions of life-history responses to warming.