FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2897714406> ?p ?o ?g. }

• W2897714406 endingPage "257" @default.
• W2897714406 startingPage "247" @default.
• W2897714406 abstract "The response of individuals to climate warming and, ultimately, the combined effects on populations and species, is mediated through interactions between physiology and behaviour (Pörtner & Farrell, 2008; Somero, 2012; Sunday et al., 2014). Organisms eat, move and reproduce within a thermal window bounded by their thermal tolerance, or performance limits, generally known as critical thermal maximum (CTmax) and minimum (CTmin) (Cowles & Bogert, 1944; Huey & Stevenson, 1979). When environmental temperatures are at or around critical limits, biological activity is constrained and, if temperature extremes are prolonged, they might compromise individual and population survival (Clarke, 2014; Somero, 2011). The policy consequences of these interactions, in the context of both impacts and adaptation, have been stressed by the Intergovernmental Panel on Climate Change in its Fifth Assessment Report: “…surpassing temperature thresholds means going into time-limited tolerance, and more extreme temperature changes lead to exceedance of thresholds that cause metabolic disturbances and ultimately onset of cell damage” (Field et al., 2014). Thermal metrics, including upper thermal limits, have been documented for many species and compiled into global databases (Bennett et al., 2018). Collectively, these data constitute a valuable resource for research aiming at determining how clines in thermal tolerance might affect biotic responses to local, regional and global environmental change (Angilletta, 2009; Deutsch et al., 2008; Sunday, Bates, & Dulvy, 2012; we review examples focusing on climate change in Supporting Information Table S1). For instance, conservatism of CTmax implies that many species might have exhausted the evolutionary potential for developing further tolerance to climate warming (Araújo et al., 2013; Grigg & Buckley, 2013; Hoffmann, Chown, & Clusella-Trullas, 2013). And extinction risk due to increased temperatures is expected to peak in the tropics because tropical species seem to be closer to their upper thermal limits than temperate species (Deutsch et al., 2008; Gunderson & Stillman, 2015; Sunday, Bates, & Dulvy, 2010). Such research is based on sizeable datasets, typically collated from multiple primary-literature sources, whereby the thermal tolerance of individual species is most often characterized based on one population (Figure 1). The approach is based on the implicit working assumption that interspecific variation is larger than intraspecific variation or, the least, that variation among populations is so small as to have negligible impact on overall patterns. Invariably, a potentially large source of physiological variation at the species level has been to date rarely considered to forecast biological responses to environmental change (Valladares et al., 2014; see below). This aspect is important because spatial mosaics of thermal tolerance might reduce the vulnerability of species to climate warming, consequently altering estimates of impact and plans for conservation and management (Cochrane, Yates, Hoyle, & Nicotra, 2015; Sears, Raskin, & Angilletta, 2011). The thermal performance of species along environmental gradients is expected to vary considerably across both altitude and latitude and might include ontogenic and evolutionary components (Clusella-Trullas & Chown, 2014; Sinclair et al., 2016). Theoretical developments have demonstrated that incorporating intraspecific variability in life-history traits into modelling could substantially improve forecasts of how species ranges might respond to environmental shifts such as climate change (Ikeda et al., 2017; Valladares et al., 2014). This has been corroborated in a few applied studies, particularly with plants (Benito Garzón, Alía, Robson, & Zavala, 2011; Cochrane, Hoyle, Yates, Wood, & Nicotra, 2015; Smith, Alsdurf, Knapp, Baer, & Johnson, 2017), and including thermal traits in ectotherms such as insects (Lancaster, 2016), reptiles (Artacho, Saravia, Perret, Bartheld, & Le Galliard, 2017) and amphibians (Kolbe, Kearney, & Shine, 2010). However, little effort has been made to quantify the magnitude and the likely predictive importance of population differences in CTmax for multiple species in a common biogeographical context. To address this issue, we quantify intraspecific variability in CTmax across 59 populations of 15 species of Iberian lizards, and determine the extent to which among-population variability is likely to modify forecasts and predictions of annual restricted activity when environmental temperatures exceed CTmax. Through spring and summer 2013 and 2014, we sampled 304 adult males from 59 populations and 15 Iberian species of lacertid lizards (Table 1). Our study populations comprised wide- and restricted-range taxa (nine endemic to the Iberian Peninsula) covering all of the major climatic regions through Spain and Portugal (Supporting Information Figures S1 and S2). Given the substantial sampling effort involved, sampling was tailored to cover a grid of localities maximizing species richness, minimizing travelled distance (see sampling map in Supporting Information Figure S2), and capturing in most cases more than 50% of the length of the climate gradient occupied by each species in the Iberian Peninsula (climate gradients reconstructed in Supporting Information Figure S3). Individuals were caught by noosing. We sampled two to five populations per species (median = 4 populations per species with 90% interquartile ranges of [3, 5]), and 3–10 males per population (Table 1). The former intrapopulation sample size is within the range of that used in many studies of lizard thermal tolerance (e.g., Beal, Lattanzio, & Miles, 2014; Muñoz et al., 2016). However, we assessed optimal sample size from six datasets, including population CTmax from 8 to 25 adult lizards from the Americas, Asia and Europe (Andrango, Sette, & Torres-Carvajal, 2016; Belasen et al., 2016; Buckley, Ehrenberger, & Angilletta, 2015; Huang & Tu, 2008; Muñoz et al., 2014; and unpublished data), and concluded that a sample size of three males should accurately capture the mean population CTmax estimated from larger population sizes (Supporting Information Table S2; Figures S4–S9). To minimize local-ambient effects prior to CTmax estimation, all lizards were housed at 25°C (8:00–18:00 hr, lights-on) and 15°C (18:00–8:00 hr, lights-off) for 14 days. Throughout, lizards were kept individually in terraria layered with coconut fibre and a ceramic brick for shelter. We fed lizards daily with house crickets (Acheta domesticus) and provided them daily with spring water in a Petri dish. Following the former acclimation period, we determined CTmax through the “dynamic method” after Lutterschmidt and Hutchison (1997) as follows. We inserted a thermocouple some 1–2 cm (depending on body size) into the cloaca of each lizard—to avoid internal damage, the thermocouple was covered with pure vaseline and the apical sensor was rounded with a minute drop of dry nail polish. The thermocouple was wired to a HH-25TC Omega Thermometer via a 1-m-long cable made up of flexible polytetrafluoroethylene. For each lizard, we fastened a 2 × 5 cm2 strip of parafilm that immobilized the distal part of the cable (equipped with the thermocouple) ventrally against the cloacal and genital area. Thereafter, lizards were kept some 30 cm from a red-light lamp, and body temperatures were ramped up at 1°C × min−1. Throughout each trial, we monitored body temperature and behaviour, and ultimately estimated CTmax as the cloacal temperature at which a lizard lost its righting response (Lutterschmidt & Hutchison, 1997). Ramping rates affect CTmax estimation (Terblanche, Deere, Clusella-Trullas, Janion, & Chown, 2007), so we controlled for this factor by using a single rate across all individuals examined. Immediately after thermal shocks, we immersed lizards down to their necks in a bath of cool water to prevent physiological/physical damage, thereafter providing water and food ad libitum to all individuals for up to five consecutive days following the day of CTmax estimation. We released all individuals at their point of capture. We ranked statistical support for two models through the Akaike's information criterion adjusted to finite sample size (Sugiura, 1978), using model probabilities (wAICc, which are scaled to a 0–1 interval), and evidence ratios of wAICc of the first-ranked model to the wAICc of the other model (Burnham & Anderson, 2002). For each species separately, the model set contrasted the hypothesis that CTmax varies according to the geographic location of populations (locality effect: CTmax ~ loc, i.e., a unique intercept or mean CTmax for each population, with loc being a categorical variable coding for population identity) against the alternative (null) hypothesis that CTmax is homogeneous across all populations studied (region effect: CTmax ~ 1, i.e., a unique intercept or mean CTmax across all populations within a species). Throughout, we assumed Gaussian errors (corroborated in the saturated model [CTmax ~ loc] by means of Q-Q and fit-versus-residual plots) and discarded heteroscedasticity effects using Bartlett's tests (Bartlett, 1937). The “lm” function in the R package “stats” was used for the linear models, and all data plots hereafter were generated in “graphics” (R Development Core Team 2016). Since we sampled 3–10 males per population (Table 1, see above), statistical support for our two models might be influenced by different sample sizes among populations. We controlled for this factor by sampling randomly three individuals per population (the minimum sample size collected from any one population) and contrasting our model set for each study species (locality vs. region effects on CTmax as above), and repeating this procedure 100 times each including a different triplet of individuals per population. To assess the extent to which single-population estimates generate varying predictions of lizard responses to climatic variation, we used activity restriction times as the biological response. Sinervo et al. (2010) defined “restriction time” for lizards as the amount of time per day in which “operative temperatures” (Te, see below) are above preferred body temperatures. We use a more conservative metric whereby “restriction time” equates with the amount of hours over which a lizard's Te exceeds CTmax; this metric is biologically realistic because diurnal lizards are bound to be active when Te surpasses their preferred temperature (Gunderson & Leal, 2015). Effectively, our metric equates with diurnal restriction times because there is no solar radiation, nor (in our species set) lizard activity, at night. Overall, our purpose was to generate a metric of restriction time that is biologically meaningful and comparable among populations, rather than to construct a fully parameterized model of the thermal niche (e.g., Kearney & Porter, 2004), or a model characterizing the magnitude of the erosion of the thermal niche (e.g., Sinervo et al., 2010). Prior to calculating restriction times, we estimated Te by solving the biophysical model by Porter, Mitchell, Beckman, and DeWitt (1973), namely QE = (QS + QI) − (QT + QC), which includes climate (fluxes of ground radiation, air temperature and wind speed) and biological (e.g., skin emissivity and absorbance) information. Succinctly, this model formulates the heat flow between an organism and its environment (QE) as a function of the solar shortwave (QS) and infrared (QI) radiations absorbed by an individual, minus the heat the animal exchanges by thermal radiation (QT) and convection (QC). Te is one of the factors in the four-order equation for QT given the surface area (S) and emissivity (ε) of the animal, and the Stefan–Boltzmann constant (σ), that is, QT = S × ε × σ × Te4. Equations and variables are fully described in Supporting Information Table S4. For each population, we estimated Te every hour daily for an entire calendar year, resulting in 8,784 maps (24 hr per day × 366 days per year, where 366 days account for leap years in our model of regional climate, see below) using the R package “RnetCDF” (Michna & Woods, 2013). To do so, we obtained hourly estimates of the climate variables (averaged over the period 1955–2010) from a “Regional Climate Model” specifically elaborated for the Iberian Peninsula at a 5 × 5 km2 grid-cell resolution (Dasari, Salgado, Perdigao, & Challa, 2014); the former represents an improvement for the study area relative to other datasets with coarser resolution and global scope (e.g., Kearney, Isaac, & Porter, 2014). For every 5 × 5 km2 grid cell, we summed the number of hours when Te was larger than CTmax for each population separately (= annual restriction time) using the “Climate Data Operators” package (Kaspar, Schulzweida, & Mueller, 2010). The resulting maps were clipped to the geographical distribution of each species and converted to geo-tiff format through the package “GDAL/OGR” (GDAL/OGR Contributors 2016). Species distributions (presence–absence data in 10 × 10 km2 grid cells) were obtained from Araújo, Guilhaumon, Neto, Pozo, and Calmaestra (2011) and from additional sources for recently described species (see Supporting Information). We therefore downscaled species distribution data to 5 × 5 km2 grid cells to match the resolution of the climate data, and standardized annual restriction times by the total number 5 × 5 km2 grid cells occupied by each species in the R packages “raster” (Hijmans & Etten, 2012) and “rgdal” (Keitt, Bivand, Pebesma, & Rowlingson, 2010). The median CTmax across all 304 individuals (59 populations, 15 species) was 42.8°C with 90% interquartile limits of [41.0, 46.0]. The CTmax across all individual lizards varied by 8°C, from 40.2°C (Catalonian wall lizard Podarcis liolepis from Cuellar, Segovia-Castilla and León) to 48.3°C (spiny-footed lizard Acanthodactylus erythrurus from Matalascañas, Huelva-Andalucía) (Figure 2a). The median of the population CTmax medians varied from 41.2 [41.0, 41.9] in the western green lizard Lacerta bilineata to 45.6 [44.5, 47.3] °C in A. erythrurus (Table 1). CTmax medians tended to decrease from arid- to mesic-adapted species, with climate generalists in between (Table 1). We found strong statistical support for the hypothesis that CTmax differed among populations within each species separately (locality effect: different mean CTmax), against the alternative hypothesis that CTmax did not differ (region effect: same mean CTmax among populations)—we report AIC metrics per species in Supporting Information Table S3. Across species, the locality- and region-effect hypotheses had a median probability of 0.83 [0.52, 0.96] and 0.17 [0.04, 0.48], respectively (Figure 2b), and we obtained similar support for both models when we controlled for intrapopulation sample size (see Materials and Methods, Supporting Information Figure S10). The locality-effect model explained up to 32% of the variation in CTmax and scored between 2 and 32 times higher model probabilities (wAICc, see Materials and Methods) than the region-effect model for 12 of the 15 study species (Supporting Information Table S3). Both models had similar support for the Vaucher's wall lizard Podarcis vaucheri, the Geniez's wall lizard Podarcis virescens and the Spanish large psammodromus Psammodromus algirus (Supporting Information Table S3). Median and mean differences in CTmax between populations remained mostly between 1 and 3°C (Table 1). The median annual restriction time across the 59 sampled populations was 281 [63, 670] hr per 5 × 5 km2 grid cell (Figure 3) in the known Iberian distribution of the 15 study species. Assuming an average daily activity of 7 hr, those restriction times mean lack of diurnal activity due to overheating over 40 [9, 96] days in a calendar year. Annual restriction times across the 59 populations varied from 31 (common wall lizard P. muralis from Navacerrada, Madrid) to 596 (Schreiber's green lizard Lacerta schreiberi from Gata, Caceres-Extremadura) hours per grid cell. Within species, the median population-level range of restriction times was 83 [43,201], varying from 43 (P. algirus) to 232 (P. virescens) hours per grid cell (Table 1). In particular, for 6 of the 15 study species (A. erythrurus, L. schreiberi, P. virescens, L. bilineata, P. liolepis, P. vaucheri) predicted annual restriction times differed by > 140 hr per grid cell among populations (Table 1, Figure 3); for the former species, therefore, depending on which population-level CTmax was included in the biophysical model, intraspecific discrepancies of > 20 days in the number of seven-hour days free of lizard activity could result. Additionally, for the four study species with the lowest predicted annual restriction times (median among populations < 90 hr per grid cell; P. muralis, Carbonell's wall lizard Podarcis carbonelli, Iberian rock lizard Iberolacerta monticola, Spanish psammodromus Psammodromus hispanicus), the range of restriction times among populations was larger or nearly equal to the smallest restriction time estimated for a single population (Table 1, Figure 3). We show that intraspecific heat tolerance varies considerably across the Iberian Peninsula, with CTmax differences of up to 3°C (median) among populations of single lizard species. Those differences comprise ~75% of the CTmax range observed across our 15 study species (Table 1). This pattern holds true across Iberian lineages of lacertids—all represented in our species set (see Pyron, Burbrink, & Wiens, 2013). In contrast, whenever multiple population data are available, in those studies that have related heat tolerance to climate change across multiple ectothermic species (Figure 1, Supporting Information Table S1), authors often average or take the maximum CTmax of a limited number of populations (e.g., Araújo et al., 2013; Clusella-Trullas, Blackburn, & Chown, 2011; Sinervo et al., 2010), or otherwise use several CTmax values for some species and, mostly, only one for others (e.g., Gunderson & Stillman, 2015; Hoffmann et al., 2013; Huey et al., 2009). Our results demonstrate, however, that the population, or population estimate, taken to represent a full species can result in starkly different cross-taxon patterns of heat tolerance, be they for CTmax or, potentially, any of its composites such as thermal performance curves (Sinclair et al., 2016), “safety margins” (e.g., Gunderson & Stillman, 2015; Sunday et al., 2014) or “warming tolerance” (e.g., Deutsch et al., 2008; Hoffmann et al., 2013). The rationale behind a “one species, one thermal trait” approach (Sinclair et al., 2016) is partly the fact that thermal limits have been estimated only once for most species whose thermal physiology has been investigated and partly because popular statistical methods to analyse those datasets while accounting for phylogenetic relatedness among species (e.g., phylogenetic least-squares regression) were originally designed to deal with one data point per species. However, from a purely analytical viewpoint (see Ives, Midford, & Garland, 2007; Stone, Nee, & Felsenstein, 2011), the effect of heterogeneous sampling can be as important as phylogenetic bias and should be considered in comparative studies (Garamszegi & Møller, 2010). For six of the 15 study species, our prediction of annual restriction times, using different population-level CTmax estimates, translated into species-level variation of 20 days, and longer, during which environmental (operative) temperatures would be above CTmax. This variation can dramatically modify inferences about behavioural patterns and fitness (Adolph & Porter, 1993) and, therefore, estimations of climate impacts accounting for trait (Cochrane, Yates, et al., 2015) or niche (Pearman, D'Amen, Graham, Thuiller, & Zimmermann, 2010) variation within species. For instance, phenotypic and species distributions can track different aspects of climate change (e.g., temperature vs. precipitation) and both complement impact assessments (Smith et al., 2017), climate effects can be masked by thermoregulation interacting with topography (Sears et al., 2011) and spatial scale (Barton, Clusella-Trullas, & Terblanche, 2018), latitudinal enhancement of heat tolerance might signal recent poleward range expansions (Lancaster, 2016), while species distribution models can fail to detect constrains in population expansion (Kolbe et al., 2010) or area of occupancy (Benito Garzón et al., 2011) owing to trait plasticity. To illustrate this point, we predicted annual restriction times of 596, 668, 697 and 715 hr per 5 × 5 km2 grid cell using the CTmax of each of four populations of the Schreiber's green lizard L. schreiberi (CTmax range = 40.5–42.9°C across individuals, and 41.0–42.1°C among population medians). This species is endemic to the Iberian Peninsula, where it is mostly restricted to riparian shrubs in mountain ranges and low temperate forests close to mountain slopes (Monasterio, Shoo, Salvador, Iraeta, & Díaz, 2013). Our population-based modelling variation above would imply differences in predicted annual restriction times for the species from 85 to 102 seven-hour days per grid cell during which individuals would be forced to shelter from overheating. This is biologically important because body growth and reproductive rates are compromised by high temperatures in this species (Monasterio et al., 2013), while extended periods of shelter might come at the expense of feeding and reproductive opportunities for ectothermic fauna (Sinervo et al., 2010). Our study supports the idea that single populations are bound to represent a snapshot of the continuum of phenotypic variability in species’ heat tolerances, and reinforces the proposition that intraspecific variability in thermal limits is a key factor to project reliably how species might respond to climate warming (Valladares et al., 2014). Physiological flexibility represents an important mechanism by which species can persist in variable environments (Chevin & Hoffmann, 2017) and adapt to environmental change (Chown et al., 2010). This is a major area of research with regard to the ways in which species responses to climate change are mediated by the interplay between microevolutionary processes and plasticity (Gienapp, Teplitsky, Alho, Mills, & Merila, 2008). Considering the heterogeneity of climatic conditions across the Iberian Peninsula and within the range of most Iberian lacertids, and the projected loss of climate space (higher temperatures, lower precipitation) for reptiles in the south-west of Europe (Araújo, Thuiller, & Pearson, 2006), it is unsurprising that current and future variability in CTmax might be the result of natural selection and adaptation at the population level. The genetic basis of this variation in lizards still needs to be disentangled. Heritability of some thermal traits may be low (Logan, Cox, & Calsbeek, 2014; Martins, Kruuk, Llewelyn, Moritz, & Phillips, 2019), and determining how common such low heritability might be requires wider taxonomical coverage and larger sample sizes. Recent studies do illustrate that ectotherm thermal limits can evolve in a few generations in response to climate events (Campbell-Staton et al., 2017; Geerts et al., 2015), while rapid climate change can also elicit complex evolutionary patterns combining directional (Gilbert & Miles, 2017; Logan et al., 2018) and correlational (Artacho, Saravia, Ferrandière, Perret, & Le Galliard, 2015; Gilbert & Miles, 2017) selection across populations (Logan et al., 2014; Martins et al., 2019). These mechanisms can shed light on the role of plasticity and evolutionary forcing in high vs. low latitudes (Lancaster, 2016), and at the margins of the distribution of species where extinction and dispersal processes in response to environmental change might be most frequent and intense (Valladares et al., 2014). It is also unclear whether intraspecific variability in heat tolerance could buffer proposed rates of extinction of ectothermic species under climate change (see Sinervo et al., 2010), and interact with other life-history traits to constrain activity times and elicit concomitant effects on population fitness (Kearney, 2012, 2013). For instance, in insects intraspecific variation in thermal breadth (CTmax − CTmin) seems latitudinally invariant only for stable and declining species, so those taxa currently not experiencing range expansions might be unfit to tolerate further warming at high altitudes, especially in the case of insular, endemic or, overall, narrowly distributed species (Lancaster, 2016). An assessment of the biological implications of this phenomenon seems highly relevant for the ecology and conservation of species, particularly those with long generation times and slow genetic responses to rapid environmental change (Hoffmann & Sgrò, 2011), and could be integrated into modelling approaches available for the study of climate-change impacts on biodiversity (Pacifici et al., 2015). The generation of novel physiological data for more species from the tree of life is indeed a pressing endeavour (Pörtner & Farrell, 2008). Thus, the recent collation of the GlobTherm database (Bennett et al., 2018: species-level metrics of thermal tolerance for 2133 taxa globally) highlights voids of knowledge in the most diverse eukaryotic groups (algae, plants, invertebrates) and in vast stretches of northernmost Africa, America and Eurasia. Bennett et al. (2018) foresaw the inclusion of intraspecific variation in future versions of the dataset, and those updates will quantify the existing bias in thermal-trait information for species vs. populations. All the same, acquiring estimates of thermal traits over broader expanses of the geographical ranges of single species seems essential (Chown & Gaston, 2016) before we might be able to address cross-taxa patterns and drivers of heat tolerance in a biologically comprehensive way. This endeavour could be logistically prohibitive for very wide-ranging species (which are an uncommon phenomenon), but could be realized for species with restricted distributions, and represents an important area of future development. We thank the Biological Station of “El Ventorrillo” for hosting the field team and for providing the thermal experimental facilities essential for this research. SHP and MBA partly funded through CGL2011-26852 project of the Spanish Ministry of Economy and Competitiveness. SHP also funded through British Ecological Society “Research Grant” 4496-5470. Field and experimental work funded by IC&DT 1/SAESCTN/ALENT-07-0224-FEDER-001755 project led by MBA. SLC funded by Australian Research Council DP170101046. We also thank Clive McMahon for useful comments on an earlier version of the manuscript; Tim Leerschool, Filipe Serrano and Matthijs Hollanders for their support in the field; and Mike Angilletta, Shu-Ping Huang, Martha Muñoz and Omar Torres-Carvajal for sharing their CTmax datasets. Samples, experiments and use of experimental animals supported for Portuguese populations by permits 360 to 362/2014/CAPT and 550 to 552/2014/CAPT (Instituto da Conservação da Natureza e das Florestas), and for Spanish populations by permits (autonomous communities in parenthesis hereafter) SGYB/EF/FJRH Re-9H/13 & SGYB/AF/DBP Re-79y131/14 (Andalucía), 2014-500201724/2014/02343 & INAGA/5000201/24/2013/04434 (Aragón), DGMEN/SEN/avp_13_025_aut & DGMEN/SEN/avp_14_020_aut (Castilla La Mancha), P/CYL/101/2013 & EP/CYL/106/2014 (Castilla and León), CN0023/14/ACA1587 (14) (Extremadura), 2566/RX131316//clave031/2013 & 2241/RX123724//clave018/2014 (Galicia) and 10/033298.9/13 & 10/013907.9/14 (Madrid). S.H.P., D.R.V. and M.B.A. conceived the idea. M.B.A. and C.M. designed the project and the full set of experiments. C.M., V.G. and W.B. did field and experimental work. S.H.P. run hypothesis contrasts and assessed optimal sample size, F.F.Y. modelled restriction times, and W.B. generated polygon data of species distributions. S.H.P. led the writing of the manuscript, and all authors contributed critically to the drafts and gave final approval for publication. All physiological data associated with this study (CTmax for 304 male lizard individuals) have been deposited in the Dryad Digital Repository: https://doi.org/10.5061/dryad.1553pc3 (Monasterio et al., 2018). As further data analyses are in progress, release of data has been embargoed for 1 year from the date of publication. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article." @default.
• W2897714406 created "2018-10-26" @default.
• W2897714406 creator A5009196818 @default.
• W2897714406 creator A5013652427 @default.
• W2897714406 creator A5041520132 @default.
• W2897714406 creator A5057675287 @default.
• W2897714406 creator A5060972449 @default.
• W2897714406 creator A5069682620 @default.
• W2897714406 creator A5077491329 @default.
• W2897714406 creator A5088295071 @default.
• W2897714406 creator A5088770554 @default.
• W2897714406 date "2018-11-02" @default.
• W2897714406 modified "2023-10-12" @default.
• W2897714406 title "Intraspecific variation in lizard heat tolerance alters estimates of climate impact" @default.
• W2897714406 cites W1503900815 @default.
• W2897714406 cites W1557624815 @default.
• W2897714406 cites W1598813349 @default.
• W2897714406 cites W1632979976 @default.
• W2897714406 cites W1963839947 @default.
• W2897714406 cites W1973140802 @default.
• W2897714406 cites W1976726371 @default.
• W2897714406 cites W1985712136 @default.
• W2897714406 cites W1992244678 @default.
• W2897714406 cites W1995983584 @default.
• W2897714406 cites W1996301170 @default.
• W2897714406 cites W2008951609 @default.
• W2897714406 cites W2017677393 @default.
• W2897714406 cites W2019097212 @default.
• W2897714406 cites W2019845843 @default.
• W2897714406 cites W2022776523 @default.
• W2897714406 cites W2028562213 @default.
• W2897714406 cites W2038950179 @default.
• W2897714406 cites W2046659947 @default.
• W2897714406 cites W2066743889 @default.
• W2897714406 cites W2069028965 @default.
• W2897714406 cites W2074352485 @default.
• W2897714406 cites W2082455532 @default.
• W2897714406 cites W2083660279 @default.
• W2897714406 cites W2089345118 @default.
• W2897714406 cites W2090834035 @default.
• W2897714406 cites W2099699510 @default.
• W2897714406 cites W2103072403 @default.
• W2897714406 cites W2105635509 @default.
• W2897714406 cites W2106590499 @default.
• W2897714406 cites W2112776483 @default.
• W2897714406 cites W2114837299 @default.
• W2897714406 cites W2115556561 @default.
• W2897714406 cites W2116417654 @default.
• W2897714406 cites W2120529009 @default.
• W2897714406 cites W2124163774 @default.
• W2897714406 cites W2125270554 @default.
• W2897714406 cites W2126675709 @default.
• W2897714406 cites W2133951380 @default.
• W2897714406 cites W2135019697 @default.
• W2897714406 cites W2145911870 @default.
• W2897714406 cites W2153330886 @default.
• W2897714406 cites W2158763052 @default.
• W2897714406 cites W2160891704 @default.
• W2897714406 cites W2162629614 @default.
• W2897714406 cites W2162832384 @default.
• W2897714406 cites W2164414466 @default.
• W2897714406 cites W2165626063 @default.
• W2897714406 cites W2168689121 @default.
• W2897714406 cites W2168788964 @default.
• W2897714406 cites W2170597372 @default.
• W2897714406 cites W2195161482 @default.
• W2897714406 cites W2285040739 @default.
• W2897714406 cites W2304200276 @default.
• W2897714406 cites W2515425212 @default.
• W2897714406 cites W2515542550 @default.
• W2897714406 cites W2516198602 @default.
• W2897714406 cites W2524360775 @default.
• W2897714406 cites W2525809904 @default.
• W2897714406 cites W2547805142 @default.
• W2897714406 cites W2588943316 @default.
• W2897714406 cites W2614083928 @default.
• W2897714406 cites W2744970791 @default.
• W2897714406 cites W2745493855 @default.
• W2897714406 cites W2792542239 @default.
• W2897714406 cites W2801502667 @default.
• W2897714406 cites W2803780610 @default.
• W2897714406 cites W2912029870 @default.
• W2897714406 cites W4237754067 @default.
• W2897714406 cites W4243398269 @default.
• W2897714406 cites W4253182300 @default.
• W2897714406 cites W4255655703 @default.
• W2897714406 doi "https://doi.org/10.1111/1365-2656.12914" @default.
• W2897714406 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/30303530" @default.
• W2897714406 hasPublicationYear "2018" @default.
• W2897714406 type Work @default.
• W2897714406 sameAs 2897714406 @default.
• W2897714406 citedByCount "49" @default.
• W2897714406 countsByYear W28977144062019 @default.
• W2897714406 countsByYear W28977144062020 @default.
• W2897714406 countsByYear W28977144062021 @default.
• W2897714406 countsByYear W28977144062022 @default.
• W2897714406 countsByYear W28977144062023 @default.
• W2897714406 crossrefType "journal-article" @default.

• next