Fig. 4: Probability of black skin occurrence estimated by MAXENT56 using solar radiation (kJm-2day-1), average precipitation (mm), and average maximum temperature (°C) as predictor values in passerines (A) and non-passerines (B).
From: Exposure to UV radiance predicts repeated evolution of concealed black skin in birds
Additional panes reflect the difference of the predicted black skin probabilities in relation to UV radiation in passerines (c) and non-passerines (d); The difference of predicted probabilities and relative maximum temperature (maximum temperature/maximal maximum temperature value) in passerines (e) and non-passerines (f); The difference of predicted probabilities and relative precipitation (precipitation/maximal precipitation value) in passerines (g) and non-passerines (h); The relative bird biodiversity (bird biodiversity/maximal bird biodiversity) (i) and the distribution of black skin colour in humans (j) (Adapted from Chaplin40). The darkness gradients in a, b and j indicate the probability of black skin being present: high probability regions are dark, low probability regions are white. In b–d values near 0 (white) indicate a good fit between predictor value and black skin probability while non-zero values indicate increasing tendencies to a mismatch between predictor and predictor indicating that another variable contributes more to the black skin prediction model. All climatic variables were yearly averaged over a span of 30 years.
