| title | subtitle | author | header-includes | |||
|---|---|---|---|---|---|---|
Supplementary ecoregion-level and site-level figures |
Supplementary material to *Spatially explicit predictions of food web structure from regional level data* |
|
\renewcommand{\thefigure}{S\arabic{figure}}
\newpage
{#fig:ecoregion_bivariates height=60%}
\newpage
{#fig:sampling_options}
\newpage
{#fig:mean_richness height=40%}
{#fig:mean_links height=40%}
Site-level examples of a community measure (a, expected species richness) and a network one (b, expected number of links). Both measures are assembled from the predicted probabilistic communities and networks, respectively. Values are measured separately for all sites and represented directly unlike in the main text where they are later summarized by ecoregion.
{#fig:bivariate_estimates height=40%}
{#fig:bivariate_uncertainty height=40%}
Bivariate representation of the estimates (a) and uncertainty (b) of species richness and the number of links. Values are grouped into five quantiles separately for each variable. The colour combinations represent the nine possible combinations of quantiles. The richness (horizontal axis) goes left to right from low richness (light grey, bottom left) to high richness (green, bottom right). The number of links goes bottom-up from low (light grey, bottom left) to high (blue, top left). Uncertainty values (b) are computed for every site as the sum of the standard deviations returned for each species by the Gradient Boosted Trees with a Gaussian maximum likelihood estimator (for richness uncertainty) and as link variance measured on the local probabilistic networks (for links uncertainty).
\newpage
{#fig:lcbd_bivariate}
\newpage
{#fig:motifs_ndti height=40%}
{#fig:motifs_ndci height=40%}
Comparison of the dominant ecological motifs across ecoregions. a) Normalized Difference Index (NDTI) comparing the trophic motifs S1 (tri-trophic food chains) and S2 (omnivory). Positive values indicate a dominance of S1 while negative values indicate a dominance of S2. Values equal or superior to |0.5| are shown with the same color as they indicate a high dominance of one motif. b) Normalized Difference Index (NDCI) comparing the competition motifs S4 (exploitative competition) and S5 (apparent competition). Positive values indicate a dominance of S4 while negative values indicate a dominance of S5.