Fig. 1: Goodness of fit of SAD models in 13,819 animal and plant communities. | Nature Communications

Fig. 1: Goodness of fit of SAD models in 13,819 animal and plant communities.

From: The powerbend distribution provides a unified model for the species abundance distribution across animals, plants and microbes

Fig. 1

The predicted values are plotted against observed values for (a) species abundance. b SAD dominance. c SAD evenness, and d SAD rareness. Each dot in (a) represents a species, and each dot in (b–d) represents a community. The color represents the density of the dots: red represents the densest dots, and dark blue represents the least dense ones. The black diagonal line is the 1:1 line that represents a perfect fit. Goodness of fit for each SAD model is determined by the modified coefficient of determination against the 1:1 line (\({r}_{m}^{2}\)) and its mean value, weighted by the sample sizes of the datasets is shown. Because \({r}_{m}^{2}\) is calculated against the 1:1 line with a fixed intercept of 0, it is possible that its value drops below zero, which indicates a poor fit. Source data is provided as a Source Data file.

Back to article page