Predicting river diatom removal after shear stress induced by ice melting

The augmentation of fine particles and water velocity may have important effects on river biota, which have not been fully clarified. The aim of this study is to relate the removal of diatoms to changes in environmental variables during high‐stage flow in alpine streams. To achieve this goal, we adopted a path analysis approach. We hypothesized a causal model relating diatom presence to environmental predictors, and we tested it against data collected in two alpine rivers in NW Italy (Aosta Valley): Dora di Veny, directly fed by glaciers of the M. Bianco massif, and Savara, fed by springs. Total suspended solids (TSS) and velocity values were much higher in Dora di Veny, with a maximum velocity of 2.9ms−1 and a maximum TSS of 2180 mg L−1. Absence of epilithic diatoms was recorded only three times in Savara, whereas it occurred in roughly half the samples in Dora di Veny. Diatom recolonization after natural removal generally occurred in early autumn, with a dominance of tightly attached forms. According to our path model, the explanatory variables causally related to the presence/absence of diatoms were water velocity, TSS and river feeding. The model estimated both direct and indirect effect weights of water velocity, thus discriminating the respective roles of shear stress (velocity) and abrasion of the substrates/increased turbidity (TSS, partly mediated by velocity). Our model allowed estimation of TSS and velocity threshold values and therefore inferences on the impact of physical alterations induced by natural or human causes. From a practical point of view, this may represent an applied outcome in the environmental impact assessment of engineering works and other human activities that could increase the TSS in rivers.