Evaluating the response of current biotic indices and functional metrics to natural and anthropogenic predictors in disconnected pools of temporary rivers.

Temporary rivers, forming the majority of river networks worldwide, are key biodiversity hotspots. Despite their great value for maintaining biodiversity and ecosystem functioning, they are often neglected in biomonitoring programs due to several challenges, such as their variable hydromorphology and the difficulty of establishing reference conditions given their dynamic nature, resulting in highly variable communities. Disconnected pools often form in temporary rivers when flow ceases, providing refuge for aquatic taxa. Given their importance for biodiversity conservation, revising and adapting biotic indices are needed. Here, we evaluate the performance of current biological indices designed for perennial rivers (macroinvertebrates, diatoms) and functional metrics (macroinvertebrates) in assessing biological quality of disconnected pools. We sampled 55 disconnected pools in Catalonia, NE Spain, covering local (e.g., physico-chemical variables, water chemistry) and regional (e.g., human influence, hydrological variables at the water body level) natural and anthropogenic gradients. Only a few macroinvertebrate biotic indices (e.g., family richness, EPT/EPT + OCH and OCH) showed strong responses to anthropogenic predictors and were unaffected by natural predictors at both local and regional scales, making them suitable for biomonitoring. Of the newly adopted functional metrics of macroinvertebrate communities tested, only two (i.e., functional redundancy of predators and response diversity based on the total community) responded strongly to anthropogenic predictors. The rest showed varying responses to the interactive effect of anthropogenic and natural predictors, requiring calibration efforts. Models assessing these metrics explained <40 % of the total variation, likely due to the interplay of colonization/extinction dynamics and density-dependent trophic interactions governing community assemblages in disconnected pools. Although some existing biological metrics could potentially be used to monitor the ecological status of disconnected pools, we call for further development of biomonitoring tools specifically designed for these habitats since they will become more widespread with global change.