Anthropic impacts adversely affect the productivity, integrity, connectivity, and resilience of riverine ecosystems, with widespread cumulative effects on the biota and biodiversity. The natural flow regime is a fundamental driver of physical and chemical processes, determining the morphological profile of the river systems and sustaining the complex network of ecological interactions and biological patterns. Therefore, in order to reach the environmental goals required by the binding legislation, and achieve a sustainable use of water resources, it is urgent to understand the mechanisms behind changes in the structure of biological communities along gradients of human disturbances which affect the flow regime. Indicators based on macroinvertebrates are widely used to assess the ecological status of water bodies, given their sensitivity/tolerance to pollution. However, in Alpine running waters, where chemical quality is less impacted than in lowland rivers, it is particularly important and valuable to detect the hydro-morphological alterations, and to discriminate them from chemical degradation, based on the responses of benthic macroinvertebrates to such multiple stressors. Therefore, this thesis aims at: i) examining the taxonomic and functional responses of macroinvertebrate communities to the different anthropogenic pressures acting on river systems; ii) evaluating the taxa/functional traits which mostly discriminate between hydrological and morphological alterations, and chemical degradation, to support effective bioindication methods. Focus of the research is to assess the macroinvertebrate community responses to the alterations caused by flow regulation and morphological alterations, which include water abstraction, diversion, stocking and the intermittent release of water from hydropower plants, banks artificialization and construction of weirs, dams, and other structures, each of these with environmental consequences of different scale and magnitude, such as the interruption of the longitudinal continuity, residual flow release and hydropeaking. The first part of this thesis is based on two empirical field studies, following respectively a manipulative and a mensurative approach, and focuses on changes in the taxonomic and functional composition due to river regulation and hydrological alterations. The first study, conducted in a set of seminatural streamside experimental flumes, simulates a residual flow stretch by reducing the discharge of the downstream sections (treatment) to 50% of the discharge of the upstream sections (control). Even within the short-term of our experiment (i.e., 3 weeks from the beginning of the simulation), we successfully simulated a small run-of-the river water abstraction and we recorded substantial changes in the EPT (Ephemeroptera, Plecoptera and Trichoptera) benthic assemblages. In fact, we observed shifts in functional (rather than taxonomic) EPT community composition over time, likely due to the active drift, from a typically rheophile to a more limnophile one as a response to the stress imposed by the flow reduction, related to decrease of flow. In the second study, we investigated the effectiveness of a hydropeaking mitigation measure on flow and biotic components, in a case study of hydropeaking reduction on a 10-km reach of the Noce Stream, a unique approach for Alpine streams to date. The hydrological analysis conducted applying two hydropeaking quantification indices (HP1 and HP2 of Carolli et al., 2015, and the COSH method by Sauterleute & Charmasson. 2014) confirmed a partial mitigation of the hydropeaking in the stretch. As a consequence of the change in hydrological regime, we observed a different taxonomic and functional recovery in the benthic and hyporheic communities. In fact, macrobenthos was negatively affected by the reduced dilution of point and diffuse pollution; conversely, the hyporheic communities showed an increase in diversity and abundance of interstitial taxa, especially those exclusive to the hyporheic zone, likely due to changes in the interstitial space availability, brought by a reduction of clogging caused by fine sediments which were previously released with each hydropeaking wave. The second part of the thesis is based on large dataset analysis where expert knowledge has been integrated with machine learning and data-based approaches: the focus of thesis shifts towards a holistic approach, extending the investigation to the entire watershed of the Trentino Province by including macroinvertebrate field data collected between 2009 and 2019 from 160 sampling sites, distributed over 90 rivers and streams. Based on the expertise of field operators from the local Environment Agency (APPA), and the quality indices currently used according to the Water Framework Directive (WFD), all the APPA stream sites were classified according to the presence of known hydrological, morphological, and chemical alterations, including the co-occurrence of two or more alteration types; sites in pristine conditions were also identified. Seasonality, stream order and type, and other stream characteristics associated with the elevation gradient are important in flow-ecology investigation, and for this reason were included in the analysis. Moreover, these features are proxies for other variables which are closely related with the structure of the benthic community, such as current velocity, organic matter availability and substrate composition, and can also be related to the probability of expecting the presences of small hydropower plants and/or a diffuse or localized pollution sources. This second section of the thesis is divided into two parts: the first part describes the initial overall qualitative and quantitative analysis, which was conducted to determine to which extent a functional diversity-based approach better recognizes patterns in the benthic community compared to the WFD diversity indices. The second part describes the machine learning approach which we used to examine the degree to which a-priori expert classification matched data-driven classification based on the taxonomic and functional composition of benthic macroinvertebrates across different binary classification disturbances. A Random Forest analysis was performed independently on benthic-macroinvertebrate abundance (expressed as number of individuals per m2) and their functional compositions. The majority of stream sites were a-priori classified as impacted by either one or a combination of anthropogenic alterations (80%), with only 16% of sites in reference or pristine conditions. We observed high variability in benthic community assemblages, likely due to complex environmental interactions and caused by the cumulative/synergic effect of different alterations that negatively affect the discrimination between stressor-specific responses. The overall results of these large-dataset based analyses showed relevant outcomes, the main one being the good discrimination of unaltered sites from the altered ones, but a low discriminating power for the types of alteration (hydrological, morphological, pollution pf combination of two or three of them) based on taxonomic and functional composition of the benthic communities. The functional parameters directly related to the stream longitudinal preference, microhabitat preferences, flow velocity, hydrological and thermal regime, and food availability in the river network, well the most suitable to identify any type of river degradation. A further step in the detection of significant indicator taxa/traits was achieved with the machine learning approach, which resulted in robust and dependable predictive models, that identified the specific taxa and traits related to different stressors, thus representing a promising tool to support environmental assessment and water management. Overall, this thesis contributes to the identification of appropriate indicators based on macroinvertebrates taxonomic and functional sensitivity to different specific stressors, to use in the assessment of the Ecological Status of streams in mountain areas, with relevant outcomes for the water management of Alpine running waters, with particular regard to the definition of environmental flows, and to the mitigation of hydropeaking.