SEARCH

SEARCH BY CITATION

FilenameFormatSizeDescription
ece3298-sup-0001-FigureS1.docxWord document4470KFigure S1. Standardized BIC-weighted effect sizes for four life-history traits (Body size, Longevity, Fertility) as predictors of variation in census-derived strength of compensatory density feedback (response), over all time series supported for Gompertz growth: (i) all taxa (a,b: aquatic invertebrates, amphibians, birds, fish, insects, mammals, plants, reptiles), and the subsets of (ii) mammal (c,d) and (iii) bird (e,f) species. Left panels (a,c,e) come from a model set controlling for number of generations monitored (G), and right panels (b,d,f) from a model set controlling for G and Body size. Bold lines represent wBIC medians as obtained from 100 bootstrapped samples. Fits were obtained using GLMM which accounted for phylogenetic nonindependence at the Linnaean taxonomical level of Class, and GLM for the subset analyses of mammals and birds. We show model sets in Table 1, sample sizes in Table S1, and BIC metrics in Table S4.
ece3298-sup-0002-FigureS2.docxWord document4470KFigure S2 Standardized BIC-weighted effect sizes for four life-history traits (Body size, Longevity, Age at first reproduction, Fertility) as predictors of variation in census-derived strength of compensatory density feedback (response), over high-quality time series supported for Gompertz growth: (i) all taxa (a,b: aquatic invertebrates, amphibians, birds, fish, insects, mammals, plants, reptiles), and the subsets of (ii) mammal (c,d) and (iii) bird (e,f) species. Left panels (a,c,e) come from a model set controlling for time-series length (q), and right panels (b,d,f) from a model set controlling for q and Body size. Bold lines represent wBIC medians as obtained from 100 bootstrapped samples. Fits were obtained using GLMM which accounted for phylogenetic nonindependence at the Linnean taxonomical level of Class, and GLM for the subset analyses of mammals and birds. We show model sets in Table 1, sample sizes in Table S1, and BIC metrics in Table S5.
ece3298-sup-0003-FigureS3.docxWord document4470KFigure S3. Standardized BIC-weighted effect sizes for four life-history traits (Body size, Longevity, Fertility) as predictors of variation in census-derived strength of compensatory density feedback (response), high-quality time series supported for Gompertz growth: (i) all taxa (a,b: aquatic invertebrates, amphibians, birds, fish, insects, mammals, plants, reptiles), and the subsets of (ii) mammal (c,d) and (iii) bird (e,f) species. Left panels (a,c,e) come from a model set controlling for number of generations monitored (G), and right panels (b,d,f) from a model set controlling for G and Body size. Bold lines represent wBIC medians as obtained from 100 bootstrapped samples. Fits were obtained using GLMM which accounted for phylogenetic nonindependence at the Linnean taxonomical level of Class, and GLM for the subset analyses of mammals and birds. We show model sets in Table 1, sample sizes in Table S1, and BIC metrics in Table S6.
ece3298-sup-0004-TableS1.docxWord document4470KTable S1. Number of species analyzed after removing missing life-history traits, selecting only time series supported by Gompertz growth, and removing time series supported for depensatory density feedback. Time series showing Gompertz growth were split by core and redundant species (see Methods), and all analyses were replicated for all species' time series, and only high-quality times series (Table S2). Colored numbers in table match with colored text below relative to the four types of analyses undertaken.
ece3298-sup-0005-TableS2.docxWord document4470KTable S2. Summary of stages used to obtain the high-quality time series with extreme departures from the assumptions of the Ricker and Gompertz growth models.
ece3298-sup-0006-TableS3.docxWord document4470KTable S3. Summary of content of figures and tables for model sets accounting for life-history effects on strength of compensatory density feedback. Control variables are time-series length (q) and number of generations monitored (q/age at first reproduction).
ece3298-sup-0007-TableS4.docxWord document4470KTable S4. Bayesian information criterion (BIC) support for the top-ranked modelsa derived from GLMM relating life history to strength of compensatory density feedback for all time series supported for Gompertz growth, including all major groups (aquatic invertebrates, amphibians, birds, fish, insects, mammals, plants, reptiles), and the subsets of mammal and bird species. wBIC, %DE, and Δ%DE are medians (in bold) from 100 bootstrapped samples (95% confidence intervals)b. Models included number of generations monitored (G) and body size (Body) as controls. We show effect sizes in Figure S1, model sets in Table 1, and sample sizes in Table S1.
ece3298-sup-0008-TableS5.docxWord document4470KTable S5. Bayesian information criterion (BIC) support for the top-ranked modelsa derived from generalized linear mixed modeling relating life history to strength of compensatory density feedback for high-quality time series supported for Gompertz growth, including all major groups (aquatic invertebrates, amphibians, birds, fish, insects, mammals, plants, reptiles), and the subsets of mammal and bird species. wBIC, %DE, and Δ%DE are medians (in bold) from 100 bootstrapped samples (95% confidence intervals)b. Models included time-series length (q) and body size (Body) as controls. We show effect sizes in Figure S2, model sets in Table 1, and sample sizes in Table S1.
ece3298-sup-0009-TableS6.docxWord document4470KTable S6. Bayesian information criterion (BIC) support for the top-ranked modelsa derived from generalized linear mixed models relating life history to strength of compensatory density feedback for high-quality time series supported for Gompertz growth, including all major groups (aquatic invertebrates, amphibians, birds, fish, insects, mammals, plants, reptiles), and the subsets of mammal and bird species. wBIC, %DE and Δ%DE are medians (in bold) from 100 bootstrapped samples (95% confidence intervals)b. Models included number of generations monitored (G) and body size (Body) as controls. We show effect sizes in Figure S3, model sets in Table 1, and sample sizes in Table S1.
ece3298-sup-0010-TableS7.docxWord document4470KTable S7. Bayesian information criterion (BIC) support for the top-ranked models derived from GLMM (all taxa: aquatic invertebrates, amphibians, birds, fish, insects, mammals, plants, reptiles) and GLM (birds and mammals) relating life history to strength of compensatory density feedback for time series supported for Gompertz growth. wBIC, %DE, and Δ%DE are medians (in bold) from 100 bootstrapped samples (95% percentile ranges)a. Models included time-series length (q, years) and body size (Body, mm) as controls and three further life-history traitsb, and we present results for observed (as in Table 2) and simulated (below) time series with incorporation of 5%, 10%, and 15% of measurement error. Samples sizes are given in Table S1.

Please note: Wiley Blackwell 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.