Biotic indices of hydrological variability as tools to inform dynamic ecological status assessments in river ecosystems.

River biomonitoring uses biotic indices that assess human-induced degradation, including pollution, by comparison with type-specific static biological reference conditions. However, index scores that indicate pollution may reflect biological responses to natural hydrological variability associated with low flows and drying, leading biomonitoring schemes to misclassify sites as degraded. To address this, dynamic, site-specific adjustments of static biological reference conditions have been proposed, but current biomonitoring tools cannot facilitate implementation of these adjustments. We analyzed 329 samples from mediterranean-climate rivers in Greece, to evaluate the use of six stressor-specific macroinvertebrate-based indices of hydrological variability (CEFI, DEHLI, ELF, LIFE, LIFENZ, MIS-index) as tools to facilitate dynamic adjustments of static biological reference conditions. We examined macroinvertebrate assemblage responses to physicochemical and land use drivers in relation to each sample's hydrological conditions, as assessed by the six indices. We evaluated index performance beyond the region of development by exploring correlations among indices, including correlations with the region-specific Greek ELF index, for which 100% of taxa were represented. We also examined the influence of inorganic nutrient pollution on index performance by comparing index scores from samples with and without nutrient pollution. Season, water temperature, agricultural land use and nutrient pollution were major drivers of macroinvertebrate assemblage composition. Indices were positively correlated but correlation strength varied considerably, driven primarily by taxonomic representation (the proportion of sampled taxa included in each index's calculation), and potentially also by differences in river types, taxonomic resolution and sampling strategies. All indices identified site-specific hydrological conditions both in the presence and absence of nutrient pollution. We recommend the development of region-specific biotic indices of hydrological variability, or regional adaptation of existing indices, to represent 100% of the regional taxa pool and thus to enable acceptable performance beyond their region of development. Such indices could inform dynamic adaptation of static biological reference conditions by assessing site-specific hydrological conditions based on a macroinvertebrate assemblage, without the collection of additional, abiotic field data. Application of our proposed approach could prevent misclassification of ecological status, thus avoiding time-demanding and costly mismanagement of rivers and streams.