Low sensitiveness of taxonomic distinctness indices to human impacts: Evidences across marine benthic organisms and habitat types

Ecological indicators likely constitute the mainstream tools in assessing the quality of aquatic ecosystems as they condense composite biological information into single measures, easier to handle for environmental managers and more understandable for non-scientific users. However, sampling settings can influence the performance of most indices, and their use is often constrained to specific habitat types. Average taxonomic distinctness (Delta+) and variation in taxonomic distinctness (Lambda+) may represent promising tools in overcoming sampling bias, and potentially applicable to a wide range of environmental contexts. In marine systems, such metrics showed higher sensitivity than classical indices in discriminating among perturbed and unperturbed conditions, though a number of studies found Delta+ and Lambda+ also varying along natural gradients, suggesting a lower ability to discern human-driven variations from natural variability than what expected. Here, analyzing existing data sets from previous impact assessment studies, we test the response of Delta+ and Lambda+ in detecting the effects of different sources of anthropogenic disturbance on marine mollusks and polychaetes from Mediterranean soft sediments and hard substrates. Our results showed that neither classical univariate analyses on Delta+ and Lambda+ values nor their associated statistical framework were able to discern among perturbed and unperturbed sites, highlighting a low sensitiveness of such metrics in detecting assemblage variations related to anthropogenic disturbance. Delta+ and Lambda+ are based on presence/absence data, and assume impacts being likely to induce variations in taxonomic structures of assemblages. As a consequence, they could experience reduced discrimination power when impacts mostly affect relative abundances of organisms rather than assemblage composition, or drive species replacement within higher taxa. Our results also showed that habitat type could strongly affect taxonomic relatedness of species within assemblages, and that this effect can vary among different organisms, suggesting that the departures from expectation of Delta+ and Lambda+ values might not be always univocally attributable to human perturbations, since possibly depending on habitat effects. Taxonomic distinctness indices can provide useful additional information on taxonomic diversity of assemblages, crucial to better address the wide concept of biological diversity. However, the effectiveness of such measures in disclosing the effects of human disturbance is still unclear, requiring further investigations especially on their potential application in defining the ecological status of coastal rocky systems.