Figure 3: Human influence on six fish community metrics for six different categories of areas.
From: Marine reserves lag behind wilderness in the conservation of key functional roles

Boxplot distributions of (a) total biomass (B), (b) species richness (S), (c) herbivores biomass (B Herb.), (d) functional richness (FRic), (e) apex predator biomass (B APEX) and (f) biomass-weighed functional diversity (Rao Entropy) for fish communities in fished areas at <3 h from the main market (red), in the small no-take marine reserves at <1 h from the main market, in the large, old and no-entry marine reserve at <2 h from market, in traditionally managed areas at <5 h travel time, isolated traditionally managed areas (19 h travel time) and in areas located at >20 h from the main market (blue) (Supplementary Fig. 4). Biomasses were log transformed. The median values across fish communities sampled in wilderness areas (>20 h) for these metrics were considered as benchmarks and were set at the maximum possible value (100%) (right y axis). Coloured horizontal bars indicate the clusters of areas after a post-hoc Kruskal–Wallis test, the red bars indicate areas with fish community metrics similar to those in exploited areas (<3 h), blue bars indicate areas with fish community metrics similar to those in wilderness areas (>20 h) and grey bars indicate areas with similar intermediate values for fish community metrics (Supplementary Table 3).