Headwaters in Alpine regions represent the large majority of streams in natural or nearly natural conditions, which provide essential ecosystem services. These catchments are particularly sensitive to temperature changes and may suer significant changes because of climate variations. Thus, identifying the main mechanisms controlling streamflow generation and understanding the nature and variability of streamflow in Alpine streams, represent a very important contribution towards a better understanding of these complex systems. Among the multiplicity of streamflow sources (e.g., rain, snow-melt, ice-melt and groundwater), in particular snow and ice-melt play a fundamental role on the hydrological cycle of Alpine catchments and strongly affect streamflow regime. Despite several research efforts over the past decades focused on understanding the complex dynamics of the hydrological processes that characterize these environments, there is still much to disclose. Hence, the interpretation of streamflow sources can become very difficult with water discharge as the sole observed variable. The previous calls for the use of alternative data sources and methods for data analysis and visualization. This doctoral thesis aimed to contribute with new insights into the multifaceted aspects of streamflow generation in Alpine river catchments, exploring the different roles played by hydrological and geochemical information and the use of several techniques, such us tracer-based analysis, continuous wavelet transform, wavelet coherence, cross-correlation and Hovmöller diagrams; in order to investigate the mechanisms controlling streamflow generation on real case studies at different temporal scales. Hence, the present thesis is based on four main elements. In the first part of this work we show how tracer data (i.e., electrical conductivity and stable isotopes of stream water) can be used to separate the contribution of pre-event and event waters applying a two-component mixing analysis on four single rainfall events identied in the Vermigliana catchment, North-Eastern Italian Alps. The separation of streamflow into two different components allowed us to improve the conceptual model of the catchment introducing constraints that are impossible to envision counting only on streamflow measurements. Moreover, we show that the relative contribution of event water with respect to pre-event water does not change only according to the magnitude of the precipitation event and on the variations in air temperature, but it also depends on the presence and thickness of the snowpack present during the event. Second, we explored the correlation between stream water electrical conductivity (EC) and water discharge (Q) using continuous records collected during two melting periods of the Vermigliana catchment. The analysis of the hysteresis relating EC and Q at the annual scale evidenced the limitations of the use of EC measurements as a proxy of Q in these type of catchments. In addition, the combined analysis of the correlation between both signals using wavelet coherence and cross-correlation, evidenced the nature of their relationship (i.e., out of phase) and the existence of relatively constant time lag between both signals. Wavelet coherence proved to be likewise useful to identify specic periods of significant changes in the dynamics controlling streamflow generation. Furthermore, the analysis of EC and Q diurnal cycles allowed us to obtain new insights related to snow-dynamics and were also used to estimate the daily contribution to streamflow from snow-melting processes. The previous contributions may support future research on the different transfer functions that characterize water and solute transport in snow and ice-melting dominated catchments. Third, the need to understand how short and long-term climate variations may influence streamflow variability in Alpine environments lead us to the use of alternative techniques to analyse traditional long-term hydrological time series, i.e., precipitation (P), temperature (T) and streamflow (Q). We compared streamflow variability and explored the relationship between atmospheric forcing and streamflow of two case studies: Vermigliana and Sarca di Genova catchments, both located in the same region and presenting similar features, like the presence of glaciers in their upper part. Hovm öller diagrams and continuous wavelet transform were used to investigate daily and seasonal climate influences on streamflow variability, while wavelet coherence analysis was used to explore the periods on which two time series experienced oscillations at a similar frequency. Moreover, the use of these alternative techniques for data analysis and visualization, provided further insights into the hydrological response and sensitivity of the systems under study to climate changes, leading to the improvement of current conceptual models and allowing us to define a suitable framework for modelling applications, as foreseen within the following research element. The fourth element of this thesis, includes the application of an existing stochastic analytical modelling framework to the two case studies mentioned above, with the aim of characterizing and predicting streamflow distribution in these glacierized catchments. Results evidence that the size of glacier coverage on these type of catchments represents a very important feature of the system that needs to be taken account for, in fact, glaciers store a large amount of water as snow and ice, which can be rapidly released affecting signicantly the magnitude and distribution of streamflow. Overall, the results obtained during this thesis provide new insights into the multi-faceted aspects of streamflow generation in snow and glacier dominated catchments, where geochemical data as an addition to hydrological information on real case studies played an essential role. Likewise, the application of different techniques for data analysis and visualization considering a variability of temporal scales provided valuable information about the sensitivity of Alpine systems to climate changes, which may serve as a support for water resources management in these important environments. Moreover, testing the applicability of an stochastic analytical approach to this complex context allowed us to understand the influence of the presence and size of glaciers on streamflow variability. Thus, the outcomes of this study may contribute to the improvement and development of new modelling structures.

The fate of reactive solutes in groundwater is largely determined by mixing, since dilution and reactions are controlled by mixing rates. By mixing we refer to the overlap of solute bodies with a different composition, which makes possible the encounter between reacting molecules. Therefore the quantification of mixing has an important role in contamination and risk assessment and remediation technology, when they rely on processes of natural attenuation, biodegradation or chemical delivery. As porous formations are ubiquitously heterogeneous, and heterogeneity features, besides being deterministically unknown, belong to a hierarchy of scales, the description of transport processes has to deal with two main issues: epistemic uncertainty and reference scale. While the heterogeneous nature of porous media interferes with physical and chemical processes (which are inherently related to the quantification of mixing and mixing-controlled processes), the choice of the reference scale is related to the means of modeling the phenomena. In order to have an accurate representation of mixing at the continuum scale, we develop a few numerical tools, all belonging to the Lagrangian framework, and compare them with classic Eulerian and Eulerian-Lagrangian schemes. Typical transport scenarios are characterized by highly fingered plumes and sharp fringes, and pose several numerical problems (e.g. artificial diffusion and spurious oscillations). In particular, artificial diffusion can in some cases overcome the actual local dispersion, thereby possibly determining gross overestimations of reaction rates. Our numerical tests provide a set of guidelines for a conscious choice of the numerical scheme according to the objectives of the investigation and to the heterogeneity level, highlighting the drawbacks of the numerical schemes on both the evaluation of dilution and of the overall effect of reactions. Under the assumption of complete mixing at the Darcy scale, we model both instantaneous and kinetically-controlled reactive transport on synthetic bi-dimensional hydraulic conductivity fields in order to investigate the complex interplay among velocity non-uniformities, local dispersion and reaction rates at increasing levels of physical heterogeneity. We also compare the effects of different local dispersion models and injection modes (uniform vs non-uniform), still analyzing the results on a single-realization basis. Realizations share the same log-conductivity structure but are characterized by variances ranging from low (0.2) to high (10). Resorting to single-realization analysis is uncommon in the literature, unless when ergodicity conditions are fulfilled. On the other hand, ensemble analysis is insensitive to local features and does not often offer a reliable representation of actual field phenomena, especially in non-ergodic conditions. Hence single-realization scenarios can be used for understanding the key processes and their interaction, or for grasping aggregated information on the whole solute body behavior. Under simplified conditions, that is, limiting the investigation to low heterogeneity fields, these numerical results are compared to simplified Lagrangian semianalytical relations aiming at reproducing plume-averaged quantities. This Lagrangian theory provides relevant information relying on a limited amount of information, i.e. low-order geostatistical properties of the formation, aquifer's geometry, reactive parameters and problem forcings (e.g. initial and boundary conditions for the flow field and the concentration of the involved species). The match between empirical and theoretical global moments is very good in all tested conditions (two different Peclet numbers, a few heterogeneity levels up to log-transmissivity variance equal to 2 and three different source sizes), and also Beta Cumulative Frequency Distributions (CFDs) with shape parameters obtained by substituting the theoretical global moments compare well with the numerical CFDs. As expected, coherent estimates of peak concentration are not equally good, because of an inherently different nature of this quantity as opposed to plume-scale concentration moments. The a-priori information expressed by statistical analysis both at the global scale and at the local scale for a conservative tracer z can be transferred to reactive species in case of very fast kinetics. Given this useful property of equilibrium reactions, we develop explicit semianalytical relations for the moments and the probability distribution functions of the concentration of chemical species reacting according to a bimolecular equilibrium homogeneous reaction. We assume that the conservative tracer probability distribution function, both at the local scale and at the global scale, can be modeled with a Beta distribution, fully characterized by the mean and the variance of z. Rigorous numerical testing on highly heterogeneous velocity fields confirms that this assumption holds. A few illustrative cases shed some light on the role of the reaction on the time evolution of (local and global) concentration for the different reactive species, and on the different quality of information contained in local statistics as opposed to global statistics. The Beta distribution is a powerful predicting tool for the space and time evolution of passive concentration and, by extension, also for reactive species in particular chemical conditions. Analytical procedures are needed for predicting the z moments, as for example the Lagrangian ones used in the present work, which are limited to weakly heterogeneous formations. Finally we explore, analytically and numerically, the upscaling from the pore scale to the Darcy scale. Via multiple scale analysis we identify a homogenizability region, in terms of the dimensionless numbers regulating a multicomponent precipitation/dissolution reactive problem, where Darcy-scale (upscaled) transport equations can be used, regardless of sub-Darcy scale inhomogeneities.

Remotely-sensed precipitation and soil moisture products are becoming increasingly important sources of information in earth science system. However, there are still high degree of uncertainties inherited in remotely-sensed precipitation and soil moisture products, and limited studies have focused on evaluation of these products. In this study, GEOtop model (Rigon et al. 2006), which is physically-based distributed hydrological model, is used to assess the use of remotely-sensed precipitation and soil moisture products for hydrological applications. The study area is Little Washita watershed (583 km2), Oklahoma, USA. To assess these products, the model has to be first calibrated and validated at different locations in the watershed using extensive ground-based measurements. The Southern Great Plains 1997 (SGP97) and SGP99 Hydrology Experiment are used for model calibration and validation, respectively. The model is reasonably calibrated and validated at watershed scale at different locations in the watershed for: heat fluxes, soil temperature profiles, soil moisture profiles, and streamflows. Regarding soil moisture evolution, we studied the spatial variability of the near-surface soil moisture from GEOtop simulations and estimates from Electronically Scanned Thinned Array Radiometer (ESTAR). Results show that GEOtop simulations and ESTAR estimates show very different magnitude and spatial patterns of near-surface soil moisture. Spatial patterns derived from GEOtop simulations are in agreement with the previous findings obtained from the same study area using ground-based measurements of soil moisture and theoretical model simulations. We conclude that GEOtop simulation results are more accurate and that ESTAR estimates are not a reliable source of data for characterizing the spatial variability of near-surface soil moisture. GEOtop simulations show that the spatial distribution of near-surface soil moisture is highly controlled by soil texture and river network. Furthermore, we investigated the effect of vegetation, surface roughness, and topography on ESTAR. Results show that there are insignificant effects of vegetation except for interception, surface roughness, and topography on ESTAR. In addition, we investigated the scaling properties of near-surface soil moisture. Results show that near-surface soil moisture has multiscaling behaviour. On the other hand, spatial soil moisture patterns are studied using geostatistical techniques: Ordinary kriging, external drift kriging and conditional Gaussian simulations (CGSs). Krigings show that soil moisture patterns in the watershed are highly controlled by gradient and cosine aspect. All CGSs clearly show soil moisture patterns. Spatial soil moisture patterns produced by CGSs are much better than the patterns reproduced by kriging algorithms. Regarding remotely-sensed precipitation products, we have investigated the utility of these products for hydrological simulations during non-winter seasons. Results show that all remotely-sensed precipitation products (Climate Prediction Center’s morphing technique (CMORPH), Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks - Cloud Classification System (PERSIANN-CCS)- and Next Generation Weather Radar (NEXRAD Stage III)) are fairly reproducing the streamflows, but CMORPH often overestimates streamflows. Thus it concluded that all the above mentioned remotely-sensed precipitation products have value for streamflow simulations.

The importance of subsurface flow in hillslope hydrology has been widely demonstrated in the past twenty years. Neal and Rosier (1990), Sklash (1990), McGlynn (2003), McDonnell (2003), Kirchner (2003) and many others demonstrated through field monitored experiments that the most part of the hillslopes runoff production comes from the subsurface, reaching often percentages around 80% of total runoff. Furthermore they found that the subsurface flow is mainly made of old water previously stored within the catchment slopes. Torres (1998), Pierson (1980) and others evidenced that catchment subsurface runoff response time could be very fast especially for wetter ground moisture contents. Nevertheless Fiori (2006), Kumar (2004), Montanari et al. (2004), Bertoldi et al. (2006) and many others highlighted that the capability of the actual modelling techniques of predicting the temporal scales of subsurface hillslope hydrology response is very poor and in order to get good results they require the use of overestimated soil hydraulic parameters. Furthermore the uncertainties and the variability of such parameters exert indeed a crucial role and require then a deep analysis in order to highlight their effects in modelling accuracy. Nonetheless the equations for saturated and unsaturated flow modelling sometimes do not describe with great precision the physical processes that are instead highlighted in many experimental works, and this is to be imputed to all the constitutive laws employed within them (Hassanizadeh, 1993; Shultze, 1999; Germann, 1999; Torres, 2001; and many others). Therefore the modelling of subsurface processes requires great care and attention and the work done in this thesis focuses on some aspects related to this problem. The objectives of this work are to investigate some features connected to subsurface flow. First of all an extensive analysis of the most used numerical schemes on convergence and on the influence of heterogeneous hydraulic parameter fields in their behaviour in both flow and transport is achieved. We have seen that the most classical method have a more robust structure than new postprocessing schemes that are aimed at improving the classical solutions. The performed studies revealed the importance of being aware of how we are solving the equations and how they deal with the domain features, that are hydraulic parameters, in order to be conscious that the numerical solution might fail to predict correctly the processes we want to model and to give the correct weight to modelling uncertainties. The second important point regards the use of constitutive laws in the governing equations which can have a great drawback on the modelling problems exposed above. The analysis on the validity of such physical constitutive laws employed in saturated and in non saturated flow equations is done. In particular the validity of Darcy law in non stationary flow fields in both saturated and unsaturated media is done as well as the comparison with solutions achieved with the more general momentum balance equation. This is done specifically to investigate some strange soil behaviour revealed in field and laboratory based works in which the unsaturated flow showed unexpected responses Germann, 1999; Torres, 2001; Torres, 2002). On the basis of this study we could make our own speculations on what happens at the base of strange flow behaviour in unsaturated flow. In our opinion a kind of capillary barrier formation in non localized areas of the flow field might explain many of the evidences arisen experimentally. We decided then to design and realize a laboratory column experiment. The work is actually ongoing and preliminary work is reported. Then discussion on our expectations and speculations around the physical processes that are poorly described by the unsaturated flow equations are then exposed.

Conceptual and physically based environmental simulation models as products of research environments efforts became complex software over time in order to allow describing the behaviour of natural phenomena more accurately. Results from these models are considered accurate but often require to operate an entire system of modeling resources with dedicated knowledge, an extensive set up, and sometimes significant computational time. Model complexity limits wide model adaptation among consultants because of lower available technical resources and capabilities. However, models should be ubiquitous to use in both research and consulting environments. This dissertation aims to address and alleviate two aspects of research model complexity: 1) for researchers, the model design complexity with respect to its internal software structure and 2) for consultants, the model application complexity with respect to data and parameter setup, runtime requirements, and proper model infrastructure setup. The first contribution provides modeling design and implementation support by managing interacting modeling solutions as “Directed Acyclic Graph”, while the second one helps to create surrogate models of complex physical models as a streamlined process. Both contributions are implemented within the OMS/CSIP modeling framework and infrastructure and were applied in various studies. First, a machine learning (ML)-based surrogate model approach is presented to respond to field application requirements to get quick but “accurate enough” model results with limited input and limited a-priori knowledge of the internal physical processes involved. The surrogate model aims to capture the behaviour of a physical model as an ensemble system of artificial neural networks (ANN). Here, the NeuroEvolution of Augmenting Topology (NEAT) technique has been leveraged because of its integration of a genetic approach to build and evolve its ANNs during supervised training. Throughout this phase, the thorough design of the services facilitate seamless monitoring of structural mutations of the artificial neural network and its performances with respect to behavioural emulation of the original model response. This results in a streamlined surrogate model generation. Furthermore, the stochasticity inherent to the evolutionary genetic algorithm combined with a specially designed cross-validation approach allows for straightforward use of the ensemble application. Several, slightly different artificial neural networks are concurrently trained. The ensemble system is built upon the selection of the utmost performant surrogate models and is used collectively to provide uncertainty quantified results when applied against new data. Secondly, a Directed Acyclic Graph (DAG) modeling structure NET3 was developed. NET3 provides appropriate data structures to represent modeling states interactions as relationships based on network topologies. The inherent structure of the DAG commands the execution of modeling tasks. NET3 implicitly manages the parallel computation depending on the network topology. A node of a NET3 modeling structure encapsulates any sort of modeling solution such as a system of ordinary differential equations, a set of statistical rules, or a system of partial differential equations. Each link connects these modeling solutions by handling their data flow. As a result, NET3 simplifies 1) the translation of physical mathematical concepts into model components, and 2) the management of complex interactions of modeling solutions. NET3 also pushes forward the idea of separating concerns between software architecture and scientific model codebase. It manages aspects that relate to the architectural design of the graph modeling structure and lets research scientist focus on their model’s domain. NET3 improves encapsulation and reusability of scientific/mathematical concepts. It avoids code duplication by allowing the same modeling solution to be adopted in different nodes and finely adapted to specific requirements. In summary, NET3 enables a new level of modeling flexibility by allowing to quickly change model representations to explore new modeling solutions. The two presented contributions were integrated into the Object Modeling System/Cloud Services Integrated Platform (OMS/CSIP) environmental modeling framework (EMF). EMFs are standard practice in environmental modeling because they represent a software solution of separating the burden of software architectural design management from scientific research. Here, OMS/CSIP has been identified “advanced” in terms of EMFs design. It offers high flexibility, low language invasiveness, fine and thorough architectural design, and innovative cloud computing deployment infrastructure. These aspects make OMS/CSIP infrastructure the suitable platform to host NEAT based surrogate modeling and NET3 extensions. Framework-enabled NEAT based Surrogate modeling (FeNS) results from the full integration of NEAT based surrogate modeling approach with OMS/CSIP platform. Here, the surrogate model approach was developed as CSIP services to help transitioning from research models to “field models” by enabling the modeling framework to interact with CSIP services, ML libraries, and a NoSQL database to emerge model surrogates for a(ny) modelling solution. OMS/CSIP was extended to harvest data from each model run and automatically derive the surrogate model at the modeling framework level. NET3 extends OMS modeling simulations to run as a graph network of interconnected modeling solutions. Furthermore, it enhances available OMS calibration algorithms to become multi-site calibration procedures. OMS already provided implicit parallel computation of independent components in a modeling solution. NET3 now adds a further layer of implicit parallelism by concurrently running independent modeling solutions. Two studies were carried out to develop and test FeSN while three applications supported the development and testing of NET3. Surrogate models of the Revised Universal Soil Loss Equation, Version 2 (R2) were generated to scale up from simple test cases with a constrained input space to more generic applications including a larger variety of input parameters. The main goal of the surrogate model was to streamline and simplify access to the R2 model behaviour. We performed sensitivity analysis of R2 to limit the input space to only relevant parameters (e.g. soil properties, climate parameter, field geometries, crop rotation description). The main study area was the State of Iowa starting from a single county (Clay county) ending up to four counties (Buena Vista, Cherokee, Clay, and Wright). Clustering methodologies were applied to improve surrogate model accuracy and to accelerate the training process by reducing the dataset size. The overall “goodness-of-fit” against the testing dataset estimated on the median of the uncertainty quantified result of the surrogate models ensemble was always above 0.95 Nash-Sutcliffe (NS), root mean squared error (RMSE) between 0.13 and 0.36, and bias between -0.07 and 0.02. In many cases, accuracy of the surrogate model with respect to testing dataset was above 0.98 NS. Surrogate models of the AgroEcoSystem (AgES) were generated to apply and test FeNS methodology to a semi-distributed hydrologic model. The main goal of the surrogate model was to streamline and simplify access to the AgES model behaviour. Only relevant lumped parameters on watershed centroid were used to train the surrogate models and limit the input space to only relevant parameters (e.g. precipitation, groundwater level, LAI, and potential evapotranspiration). The main study area was the South Fork Iowa River (SFIR) watershed in the State of Iowa across Wright, Franklin, Hamilton, and Hardin counties. The overall “goodness-of-fit” against the testing dataset estimated on the median of the uncertainty quantified result of the surrogate models ensemble was above 0.97 Nash-Sutcliffe (NS), root mean squared error (RMSE) of 2.24, and bias of -0.0794. With respect to NET3, the first application is the real-time modeling of flood forecasting through GEOframe system for the Civil Protection of Regione Basilicata implemented by PhD Bancheri. To scale the computation and finely tune calibration parameters, the Basilicata river basins were split into subcatchments where each was represented by a different NET3 node. The second application was part of Mr. Dalla Torre’s master thesis where the computational core of the rainfall-runoff model of Storm Water Management Model (SWMM by EPA) was componentized. NET3 now allows for reimplementing a concise and lightweight SWMM modeling core and highly parallel model runs. Software architectural design of rainfall-runoff, routing and sewer pipe design components targeted separation of concerns, single responsibility, and encapsulation principles. It resulted in clean and minimized code base. NET3 manages component connections and scalable computation by hosting rainfall-runoff modeling solution into separated nodes from routing and sewer pipe design modeling solution. It also enables each node of the modeling structure to 1) access a shared data structure to fetch input data from and push results to (SWMMobject), and 2) internally analyze the upstream subtree in order to adjust sewer pipe design parameters. The third test case is the application of a “system of systems” of urban models where each node of the graph modeling structure encapsulates a single responsibility system of models. Because of the stochasticity involved in each system of models, the entire graph modeling solution was required to run several times and generate independent realizations. Hence, NET3 was enabled to run a “graph of graphs” modeling structure.

In hydrological studies, it is very important to properly analyze the relationship among the different components of the water cycle, due to the complex feedback mechanisms typical of this system. The analysis of available time series is hence a fundamental step, which has to be performed before any modeling activity. Moreover, time series analysis can shed light over the spatial and temporal dynamics of correlated hydrological and climatological processes. In this work, we focus on three tools applied for time series analysis: homogeneity tests, wavelet analysis and copula analysis. Homogeneity tests allow to identify a first important kind of variability in the time series, which is not due to climate nor seasonal variability. Testing for inhomogeneities is therefore an important step that should be always performed on a time series before using it for any application. The homogenization of snow depth data, in particular, is a challenging task. Up to now, it has been performed analyzing available metadata, which often present contradictions and are rarely complete. In this work, we present a procedure to test the homogeneity of snow depth time series based on the Standard Normal Homogeneity Test (SNHT). The performance of the SNHT for the detection of inhomogeneities in snow depth data is further investigated with a comparison experiment, in which a dataset of snow depth time series relative to Austrian stations has been analyzed with both the SNHT and the HOMOP algorithm. The intercomparison study indicates that the two algorithms show comparable performance. The wavelet transform analysis allows to obtain a different kind of information about the variability of a time series. In fact, it determines the different frequency content of a signal in different time intervals. Moreover, the wavelet coherence analysis allows to identify periods where two time series are correlated and their phase shift. We apply the wavelet transform to a dataset of snow depth time series of stations distributed in the Adige catchment and on a dataset of 16 discharge time series located in the Adige and in the Inn catchments. The same datasets are used to perform a wavelet coherence analysis considering the Mediterranean Oscillation Index (MOI) and the North Atlantic Oscillation Index (NAOI). This analysis highlights a difference in the behavior of the snow time series collected below and above 1650 m a.s.l.. We also observe a difference between low and high elevation sites in the amount of mean seasonal snow depth and snow cover duration. More interestingly, snow time series collected at different elevations respond differently to temperature and more in general to climate changes. The wavelet analysis allows us also to distinguish between gauging stations belonging to different catchments, while the wavelet coherence analysis revealed non-stationary correlations with the MOI and NAOI, indicating a very complex relation between the measured quantities and climatic indexes. Finally the application of copulas allows modeling the marginal of each variable and their dependence structure independently. We apply this technique to two relevant cases. First we study snow related variables in relation with temperature, the NAOI and the MOI, which we already investigated with the wavelet coherence analysis. Then we model flood events registered at two stations of the Inn river: Wasserburg and Passau. This last analysis is performed with the goal of predicting future flood events and derive construction parameters for retention basins. We test three different combinations of variables (direct peak discharge-direct volume, direct peak discharge-direct volume-rising time-base flow, direct peak discharge-direct volume-rising time-moving threshold) describing the flood events and compare the results. The consistency in the results indicates that the proposed methodology is robust and reliable. This study shows the importance of approaching the analysis to hydrological time series from several points of view: quality of the data, variability of the time series and relation between different variables. Moreover, it shows that integrating the use of various time series analysis methods can greatly improve our understanding of the system behavior.

The fate and transport of solutes introduced into a watershed and sampled at the catchment outlet depends on many environmental, chemical and hydro-climatological forces.Moreover, if the solutes are micro and emerging pollutants (i.e. pharmaceuticals), which are non-regulated contaminants not routinely monitored but often-detected in fresh waters, the description of the transport sources and routes becomes an interesting and challenging topic to investigate and describe, especially in conjunction with the well-known travel time transport approach at the catchment-scale. In fact, with the travel time approach to pharmaceuticals represents a framework that allows dealing in a unitary and simple way the main two mitigation mechanisms controlling them, which are dilution and biological decay. Moreover, possible consequences on the health of humans and of aquatic organisms have become issue of increasing concern by the scientific community worldwide. The topics have been extensively studied in the last decades, with some recent benchmark contributions. Nevertheless, there is still room for further development for emerging contaminant models and there is still the necessity of complementing the applications with measured data. This doctoral thesis aimed at contributing with new insights into the multi-faceted aspects of solute transport at catchment-scale, proposing novel solutions, with applications to real-world case studies and including a detailed description of the major aspects that influence the water quality dynamics in rivers. The thesis is divided into three interconnected and chronological subsequent parts. In the first part, a detailed description of three large European river basins are presented (i.e. Adige, Ebro and Sava), believing that an accurate analysis of existing information is therefore useful and necessary to identify stressors that may act in synergy and to design new field campaigns. In addition, a detailed data analysis of the main water quality variables is presented: advanced statistical analyses (i.e. Spearman rank correlation, Principal Component Analysis, andMann-Kendall trend tests) were applied to long-term time series of water quality data both in the Adige River Basin and in the Ebro and Sava catchments, aiming at providing an integrated and comparative analysis of recent trends, in order to investigate the relationships between water quality parameters and the main factors controlling them (i.e. climate change, streamflow, land use, population) in the Mediterranean region. These catchments are included into the EU project “Globaqua ”, dealing with the analysis of the combined effect of several stressors on the freshwater ecosystems inMediterranean rivers. In fact, little attention has been paid to linkages between long-term trends in climate, streamflow and water quality in European basins; nevertheless, such analysis can represent, complementary to a deep knowledge of the investigated systems, a reliable tool for decision makers in river basin planning by providing a reliable estimate of the impacts on the aquatic ecosystem of the studied basins. In the second part, sampling campaigns performed in our study basin, the Adige catchment, are presented in detail. Special attention is also given to emerging pollutants, whose study on the occurrence patterns and spatiotemporal variability in the Adige River Basin has been conducted in conjunction with population patterns and touristic fluxes. In the third and last part, novel theoretical solutions of the well-known advection-dispersion-reaction (ADR) equation are presented. The theory was developed for both general water quality variables and pharmaceuticals, evidencing differences and analysing the main factors that influence water quality dynamics. An application is also proposed to the Adige catchment.

The impact of changing climate on the hydrological cycle in Alpine regions has attracted in the last decades a wealth of attention by the scientific community and decision makers. Indeed, the implications of changes in the intensity and in the temporal and spatial patterns of precipitation, temperature and other climatic forcing have been widely observed accompanied with an increased frequency of drought and flood events, and a general degradation of water quality and health of aquatic ecosystems. Accordingly, in the present thesis, the effect of changes in hydro-climatic variables on the hydrological cycle is investigated over a range of temporal and spatial scales. In particular, the research moves along two main directions: 1) changes in historical time series of streamflow, precipitation and temperature, recorded in the Adige River Basin (i.e., Northeastern Italy), are analyzed with a water balance approach and compared to those of other large European river basins (i.e., Ebro and Sava) in order to quantify alterations of the main hydrological fluxes due to climate change and water uses and to disentangle their reciprocal effects; 2) a framework for evaluating the hydrological coherence of available gridded meteorological datasets, including one developed in the first part of the thesis, is introduced and tested. Regarding the first line of research, hydro-climatic and water quality variables of some important European river basins have been analyzed in order to quantify the main alterations of streamflow and to understand the most important factors controlling them. Particular attention is drawn to the Adige River Basin (an Alpine catchment located in the North-East of Italy), for which in depth studies, data measures and analyses have been performed. At this purpose, advanced techniques, besides novel approaches, have been applied. In particular, statistical methods (i.e., Mann-Kendall trend tests, Senâ€TMs slope estimates, multivariate data analyses and Kriging algorithms) have been used to assess the water budgets and the variations in time and space of the aforementioned variables. Disentangling climatic and human impacts on the hydrological fluxes is a difficult task and it has not been fully explored yet, since concurring drivers of hydrological alterations (e.g., climate and land use changes, hydropower and agricultural developments and increasing population) are intimately intertwined one to each other and combined in a complex nonlinear manner. At this purpose, spatial and temporal patterns of change in the hydrological cycle of the Adige River Basin have been identified by comparing annual and seasonal water budgets performed in four representative sub-basins (sized from 207 to 9,852 km2) characterized by different climatic and water uses conditions. A significant downward trend of streamflow is found in the lower part of the Adige since the â€TM70s , which can be attributed to the intense development of irrigated agriculture in the drainage area of the Noce River (one of the main tributaries of the Adige River). Conversely, headwater catchments showed a significant positive trend in streamflow due to a shift in the seasonal distribution of precipitation. These results suggest that climate change is the main driver only in headwater basins, while water uses overcome its effect along the main stream and the lower end of the tributaries. Therefore, a comparative analysis of recent trends in hydro-climatic parameters in three climatologically different European watersheds (i.e., the Adige, Ebro and Sava River Basins) has been performed. The main results suggest that the highest risk of increasing water scarcity refers to the Ebro, whereas the Adige shows better resilience to a changing climate. In the second part, this thesis deals with the uncertainty associated with climate datasets, that typically represents the largest part of the total uncertainty in hydrological modeling and, more in general, in climate change impact studies. In particular, this thesis describes a new framework for assessing the coherence of gridded meteorological datasets with streamflow observations (i.e., HyCoT - Hydrological Coherence Test). Application to the Adige catchment reveals that using inverse hydrological modeling allows testing the accuracy of gridded temperature and precipitation datasets and it may represent a tool for excluding those that are inconsistent with the hydrological response.

The increasing impacts of climate changes on water related sectors are leading the scientists’ attentions to the development of comprehensive models, allowing better descriptions of the water and solute transport processes. "Getting the right answers for the right reasons", in terms of hydrological response, is one of the main goals of most of the recent literature. Semi-distributed hydrological models, based on the partition of basins in hydrological response units (HRUs) to be connected, eventually, to describe a whole catchment, proved to be robust in the reproduction of observed catchment dynamics. ’Embedded reservoirs’ are often used for each HRU, to allow a consistent representation of the processes. In this work, a new semi-distributed model for runoff and evapotranspiration is presented: five different reservoirs are inter-connected in order to capture the dynamics of snow, canopy, surface flow, root-zone and groundwater compartments. The knowledge of the mass of water and solute stored and released through different outputs (e.g. discharge, evapotranspiration) allows the analysis of the hydrological travel times and solute transport in catchments. The latter have been studied extensively, with some recent benchmark contributions in the last decade. However, the literature remains obscured by different terminologies and notations, as well as model assumptions are not fully explained. The thesis presents a detailed description of a new theoretical approach that reworks the theory from the point of view of the hydrological storages and fluxes involved.Major aspects of the new theory are the ’age-ranked’ definition of the hydrological variables, the explicit treatment of evaporative fluxes and of their influence on the transport, the analysis of the outflows partitioning coefficients and the explicit formulation of the ’age-ranked’ equations for solutes.Moreover, the work presents concepts in a new systematic and clarified way, helping the application of the theory. To give substance to the theory, a small catchment in the prealpine area was chosen as an example and the results illustrated. The new semi-distributed model for runoff and evapotranspiration and the travel time theory were implemented and integrated in the semi-distributed hydrological system JGrass-NewAge. Thanks to the environmentalmodeling framework OMS3, each part of the hydrological cycle is implemented as a component that can be selected, adopted, and connected at run-time to obtain a user-customized hydrologicalmodel. The system is flexible, expandable and applicable in a variety of modeling solutions. In this work, the model code underwent to an extensive revision: new components were added (coupled storages water budget, travel times components); old components were enhanced (Kriging, shortwave, longwave, evapotranspiration, rain-snow separation, SWE and melting components); documentation was standardized and deployed. Since the Thesis regards in wide sense the building of a collaborative system, a discussion of some general purpose tools that were implemented or improved for supporting the present research is also presented. They include the description and the verification of a software component dealing with the long-wave radiation budget and another component dealing with an implementation of some Kriging procedure.

NewAge-JGrass system for forecasting and modelling of water resources in general at the basin scale. As a modern hydrological modelling, it is composed of two parts: (i) the system for data and results visualization based on the Geographic Information System uDig and (ii) the component based modelling system. All the system is based on Java because of its portability. Java is a modern and mature language aware of the web and has features such as multithreading that are essential to build scalable modelling platform. There are a few open source frameworks available that allow adaptation for our task, such as the GeoTools project by the Open GIS Consortium, representing a solid foundation for spatial analysis. OMS was chosen for facilitating model connectivity because of it low invasiveness in code practice and capability in production of leaner and more descriptive modelling code . uDig as visualization/GIS platform, including GIS services, and its integration with the JGrass GIS, developed by http://udig.refractions.net/, offers a spatial toolbox which contains the features previously offered by JGrass. Compared to traditional hydrological models, which are built upon monolithic code, JGrass-NewAge allows for multiple modelling solutions for the same physical process, provided they share similar input and outputs constraints. Modeling components are connected by means of a concise scripting language NewAge-JGrass components can be grouped in several categories. The geomorphic and DEM analyses which solves the problem of basin delineation; the tools for making spatial extrapolation/interpolation of the meteorological data; the estimation of the radiation forcing; the estimation of evapotranspiration; the estimation of the runoff production; the channel routing and tools for automatic model parameter calibration such as DREAM, Particle Swarm and LUCA. NewAge requires interpolated meteorological variables (such as air temperature, precipitation, and relative humidity) as input data for each hillslope. They can be computed by a deterministic or geostatistic approaches. The energy model includes both, shortwave and longwave radiation calculation components for each hillslope. The first implements algorithms that take into account shade and complex topography and cloud cover. Evapotraspiration can be modelled using two different solutions: the Fao-Evapotraspiration model and the Priestley-Taylor model. A snow melting and snow water equivalent model is also part of the system. Duffy's model and Hymod model are the runoff production models implemented in NewAge. In both cases the model is applied for each hillslope. Finally, the discharge generated at each hillslope is routed to each associated stream link. Modeling solutions (connections of different components) are applied in three different river basin and verifications against measured data (discharge, radiation fluxes, snow water equivalent) are presented by using traditional goodness of fitting indices.