Which turtle should I study? Uneven distribution of research effort across Testudines species

Abstract

Understanding why some species are more often studied than others can provide valuable information on potential biases that research effort discrepancies can cause and help to direct future research to mitigate these biases. Turtles form an order of reptiles in which the majority of species (67%) are threatened with extinction, but they show high variation in the amount of research devoted to each species. We aimed to quantify this variation across all recognized extant species, and to decipher how the distribution of this research effort varied according to species taxonomy, phylogeny, distribution, ecology, life history and extinction risk. Using the number of references listed on the Web of Science for each species as a proxy for species research effort, we first show that this index is both robust to variation in search methods within WoS, and consistent with an index extracted from another online library (Scopus). The number of articles per species varied substantially, with 3441 articles for the green sea turtle alone, but only 475 total articles for all 50% least studied species taken together. This heterogeneity was consistent across time since the 1970s. Phylogenetic non-independence explained 66% of the variance in research effort (vs 58% for taxonomy). In addition, marine species, a group that only includes seven species, were particularly highly studied: five of the eight most studied Testudines are marine turtles. In contrast, freshwater and terrestrial species shared similar research effort values. Species occurring in Europe and North America were also more studied than species from, e.g., Central America or Africa, as were species that occur in the most scientifically active countries. In addition, species with a larger distribution range or introduced outside of their native range, as well as species with certain life history characteristics like larger clutches, smaller eggs or a longer longevity were more frequently investigated. In contrast, population trend and extinction risk did not predict research effort, although species that have not been assessed by the IUCN were characterized by a low research effort. This study underlines the importance of accounting for the heterogeneity in research effort across turtle species in global analyses and helps to identify taxa, regions, and life history strategies that should be the focus of more research.