The global macroecology of penguin distribution and conservation

Current global biodiversity crises have reinforced the urgency to accurately quantify macroecological patterns of species distributions with the aim to develop efficient evidence-based protection strategies. The establishment of protected areas (PAs) is one of the major approaches to mitigate biodiversity declines during the Anthropocene. However, the distribution of biodiversity (e.g., hotspots, threatened species) is spatially asymmetric among regions and lineages, and the extent to which PAs offer an effective method for species and ecosystem protection remains uncertain. Penguins, broadly regarded as prime indicators of environmental health, are emerging as priority organisms for such quantitative assessments. The vast majority of all 18 penguin species are suffering population declines, and 11 of them are considered vulnerable/endangered given environmental and anthropogenic threats. However, how protected are penguins globally? Using spatial ecology approaches, we create novel distribution, hotspot, and endemism maps for all penguin species on Earth. We use these data to quantitatively assess the protection offered by PAs validated by the International Union for Conservation of Nature (IUCN) and other national/international organizations. We then create a universal target protection model which predicts the appropriate protection level required for each individual species based on population and range size (both predominant factors underlying population stability). For six species, less than 20% of their range is protected. PAs are only related to range size, whereas other parameters critically linked to population stability such as population size and IUCN conservation status are insignificant. Half of the nine global hotspots of penguin diversity are underprotected, and protection for 10 species is less than the calculated target protection. Consequently, global protection of penguins is not satisfactory. We suggest that more comprehensive ways to account for the multi-dimensional interactions between areas and organisms are needed to further the effectiveness of PA networks