WFD ecological status indicator shows poor correlation with flow parameters in a large Alpine catchment

Since the implementation of the Water Framework Directive, the ecological status of European running waters has been evaluated using a set of harmonised ecological indicators that should guide conservation and restoration actions. Among these, the restoration of the natural flow regime (ecological flows) is considered indispensable for the achievement of the good ecological status, and yet the sensitivity of the current biological indicators to hydrologic parameters remains understudied. The Italian Star_ICMi well represents similar WFD indicators; it is a macroinvertebrate-based multimetric index officially adopted to assess the ecological status of running waters at the national level. Recent legislation has also included the Star_ICMi as one of the indicators used to assess and prescribe ecological flows in river reaches regulated by water abstraction. However, the relationship between river hydrology and the Star_ICMi index is so far virtually unknown. Using data from the Trentino – Alto Adige Alpine region, we first assessed the relationship between the Star_ICMi and synthetic descriptors of the physico-chemical (LIMeco) and morphological (MQI) status of respectively 280 and 184 river reaches. Then, we examined the relation between the Star_ICMi and a set of ecologically-relevant hydrologic parameters derived from discharge time-series measured at 21 hydrometric stations, representing both natural and regulated river reaches. Although the Star_ICMi showed significant and linear relationships with the physico- chemical character and, slightly, with the morphological quality of the reaches, its response to flow parameters appeared weak or non-existent when examined with linear models. Mixed quantile regressions allowed the identification of flow parameters that represented limiting factors for macroinvertebrate communities and the associated Star_ICMi scores. In particular, the index showed ‘negative floors’ where lower values were observed in reaches with large temporal variation in flow magnitude as well as frequent low and high flow events. The modelled quantiles also tracked the transition of the index from acceptable to unacceptable conditions. The results suggest that while the central tendency of the Star_ICMi index is not strongly influenced by river flow character, some key flow parameters represent limiting factors that allow the index to reach its lowest values, eventually ‘pushing’ the site towards unacceptable ecological conditions. The identification of limiting flow parameters can aid the setting of hydrologic thresholds over which ecological impairment is likely to occur. Overall, however, results imply caution is needed in using biological indicator like the Star_ICMi for the quantitative assessment and design of ecological flows