The objective of this research was to determine the impact of climate change on the Churchill River basin and perform analysis on uncertainty related to this impact. Three hydrological models were used to determine this impact and were calibrated to approximately equivalent levels of efficiency. These include WATFLOODTM, a semi-physically based, distributed model; HBV-EC, a semidistributed, conceptual model; and HMETS, a lumped, conceptual model. These models achieved Nash-Sutcliffe calibration values ranging from 0.51 to 0.71. Climate change simulations indicated that the average of simulations predict a small increase in flow for the 2050s and a slight decrease for the 2080s. Each hydrological model predicted earlier freshets and a shift in timing of low flow events. Uncertainty analysis indicated that the chief contributor of uncertainty was the selection of GCM followed by hydrological model with less significant sources of uncertainty being parameterization of the hydrological model and selection of emissions scenario.

The objective of this research was to determine the impact of climate change on the Churchill River basin and perform analysis on uncertainty related to this impact. Three hydrological models were used to determine this impact and were calibrated to approximately equivalent levels of efficiency. These include WATFLOODTM, a semi-physically based, distributed model; HBV-EC, a semidistributed, conceptual model; and HMETS, a lumped, conceptual model. These models achieved Nash-Sutcliffe calibration values ranging from 0.51 to 0.71. Climate change simulations indicated that the average of simulations predict a small increase in flow for the 2050s and a slight decrease for the 2080s. Each hydrological model predicted earlier freshets and a shift in timing of low flow events. Uncertainty analysis indicated that the chief contributor of uncertainty was the selection of GCM followed by hydrological model with less significant sources of uncertainty being parameterization of the hydrological model and selection of emissions scenario.

Delineating spatial patterns of precipitation isotopes (“isoscapes”) is important for studies including the hydrology of terrestrial systems, present and past interpretations of climate, and tracer-aided hydrological modelling, among others. However, the extent to which precipitation isoscapes can be predicted across Canada has not been fully articulated. This thesis combines isotopes in precipitation (δ18Oppt) observations from two regional and one global network to create long term and time series precipitation isoscapes for Canada and the northern United States. Multi-linear regressions of a small suite of geographic and climate variables generate the best performing long-term and seasonal models of δ18Oppt. These models are used to develop long term isoscapes for Canada, which capture the general spatial and seasonal trends in δ18Oppt, showing an improvement upon results from previous studies using global models. Building upon long-term δ18Oppt prediction, δ18Oppt observations alongside climatological and geographic predictors are used to create empirical time series prediction models. Five regionalization approaches are used to separate the study domain into isotope zones to explore the effect of spatial grouping on simulations. Generally, the models capture the timing and magnitude of intra-annual (seasonal) δ18Oppt cycles across the study domain while simulating moderate inter-annual variation; however often fail to capture the anomalies in observed δ18Oppt. Uncertainty in predictions is quantified spatially and temporally, and the Köppen-Geiger (Kpn) regionalization is selected as the preferred regionalization scheme for future applications due to adequate model performance and lack of border issues at regional boundaries. Finally, estimates of monthly δ18Oppt from Kpn models, long term annual averages, and daily REMOiso output are used to force an isotope-enabled hydrological model, isoWATFLOOD, in the Fort Simpson Basin, NWT, Canada. Results show streamflow simulations are not significantly impacted by choice of δ18Oppt input; however, oxygen-18 in streamflow and the internal apportionment of water (and model parameterizations) are impacted, particularly during large precipitation and snowmelt events. This work shows how isoWATFLOOD can be used in regions with limited δ18Oppt observations, and that the model can be of value in such regions. This study reinforces that a tracer-aided modelling approach works towards diagnosing issues surrounding model equifinality. / February 2017

Desequilíbrios ambientais provocados pela combinação de queimadas, desmatamentos e os fenômenos de ENSO (Oscilação Sul de El Nino) podem ser os responsáveis pelo aumento de períodos de seca na região Amazônica. Com o propósito de compreender as conseqüências que longos períodos de seca podem causar na Floresta Nacional de Tapajós, foi desenvolvido um modelo numérico que simula a dinâmica da água no solo e o Balanço Hídrico para um latossolo (Haplustox, na taxonomia Americana), muito comum na região Amazônica. As simulações foram realizadas para o período de 1999 a 2003, utilizando dados de precipitação, evapotranspiração, umidade do solo, curvas de retenção, propriedades físicas do solo coletadas no local de estudo. Este estudo integra o Projeto Seca-Floresta do grupo de pesquisa do LBA (Experimento de Grande Escala da Biosfera-Atmosfera na Amazônia) para o experimento de exclusão parcial da chuva no projeto na Floresta Nacional Tapajós no que diz respeito a componente de modelagem hidrológica. Os resultados mostraram que, mesmo com a diminuição da quantidade de água disponível para a Floresta, não houve mudanças significativas em relação ao balanço hídrico da floresta, mostrando que a floresta provavelmente se adaptou, para sobreviver a longos períodos de seca / Environmental instability caused by the combination of fire, deforestation and the ENSO phenomena (El Nino South Oscillation) can be the responsible for increases of dry periods in the Amazonian region. With the purpose of understanding the consequences that long dry periods can cause on the National Forest of Tapajós, a mathematical model that determines the water dynamics in soil and the hydrological balance was developed for a typical soil of the Amazonian region the Amazonianlatossol (Haplustox). The simulation were performed for the period from 1999 to 2003, using precipitation, evapotranspiration, soil moisture, retention curves and soil physical properties data obtained in the study area. The present study integrates the Dry-Forest Project of the LBA (Large Scale Biosphere-Atmosphere experiment in the Amazon) research group for the experiment of rainfall exclusion in the National Forest of Tapajos and concerns the hydrological modeling component of the project. The results showed that even with a decrease in the amount of water available to the Forest, significant changes in the hydrological balance of the forest did not occur, showing that the forest had probably adapted itself to survive to longer periods of drought

L'objet de la présente thèse est de caractériser, quantifier et modéliser les flux d'eau au sein de la Réserve Naturelle Nationale de la Tourbière des Dauges, située en Limousin (Massif Central, France) et qui inclue une tourbière acide de fond de vallon et son bassin versant. Un ensemble de techniques, incluant la description de coupes superficielles existantes, la réinterprétation de sondages géologiques profonds, la tomographie de résistivité électrique et une modélisation de la distribution spatiale des formations affleurantes, ont été utilisées pour caractériser la nature et la géométrie des formations d'altération du granite. Les dépôts alluviaux et tourbeux ont été caractérisés et cartographiés par sondage à la tarière et à la tige filetée, et leur conductivité hydraulique estimée par choc hydraulique. Les précipitations, les paramètres météorologiques nécessaires au calcul de l'évapotranspiration potentielle, les débits et niveaux dans les ruisseaux, et les niveaux piézométriques dans la tourbe et les formations minérales sous jacentes ont été mesurés en continu pendant trois ans. Le modèle hydrologique distribué à base physique MIKE SHE / MIKE 11 a été utilisé pour modéliser les écoulements et les niveaux piézométriques au sein de la tourbière et de son bassin versant avec un pas de temps quotidien et une résolution spatiale de IO m. Il est montré que les apports souterrains issus de la zone fissurée du granite et suintant au travers du dépôt tourbeux constituent une part quantitativement importante et fonctionnellement essentielle de la balance hydrique de la zone humide. La présence d'une nappe affleurante entraîne une évacuation rapide vers les cours d'eau des apports par ruissellement ou par précipitation directe du fait de la saturation des histosols. Toutefois, il est montré que le fonctionnement hydrologique à l'échelle locale peut s'éloigner de ce schéma général du fait d'une grande hétérogénéité du taux d'humification et de la conductivité hydraulique de la tourbe, de la présence de dépôts alluviaux très perméables sous ou au sein du dépôt tourbeux et de perturbations anthropiques passées. Une fois calibré, le modèle hydrologique, qui représente la zone fissurée du socle granitique comme un milieu poreux équivalent, donne des résultats satisfaisants à très bons selon les indicateurs de performance utilisés: il est capable de reproduire les débits dans les cours d'eau au niveau des quatre stations de jaugeage disponibles, et le niveau de la nappe dans la plupart des piézomètres installés. A l'échelle du bassin versant étudié, le niveau moyen de la nappe simulé par le modèle montre une très bonne concordance avec la distribution observée des végétations de zone humide, cartographiée de manière indépendante. Les analyses de sensibilité ont montré que la porosité efficace et la conductivité hydraulique horizontale de la zone fissurée du granite sont les paramètres auxquels les débits et les niveaux de nappe (y compris dans la tourbe) simulés par le modèle sont les plus sensibles, ce qui démontre l'importance d'une meilleure caractérisation des formations d'altération du granite dans tout le bassin versant pour la compréhension et la modélisation du fonctionnement hydrologique de ce type de zone humide. Le modèle a été utilisé pour simuler l'impact potentiel d'un changement d'occupation des sols au sein du bassin versant sur la balance hydrique et les niveaux de nappe dans la zone humide, ainsi que sur les débits dans les cours d'eau. Le modèle suggère que le remplacement des végétations conduirait à une réduction substantielle des apports de surface et souterrains à la tourbière et à un abaissement conséquent des niveaux de nappe dans les histosols en période estivale. / This thesis identifies, quantifies and models water fluxes within the Dauges National Nature Reserve, an acidic valley mire in the French Massif Central. A range of techniques were used to investigate the nature and geometry of granite weathering formations and of peat deposits. Rainfall, reference evapotranspiration, stream discharge, stream stage, groundwater table depths and piezometric heads were monitored over a three-year period. The distributed, physics-based hydrological model MIKE SHE / MIKE 11 was used to model water flow within the mire and its catchment. lt was shown that the mire is mostly fed by groundwater flowing within the densely fissured granite zone and upwelling through the peat deposits. Upwelling to the peat layer and see page to overland flow were highest along the mire boundaries. However hydrological functioning differs from this general conceptual model in some locations due to the high variability of the peat hydraulic characteristics, the presence of highly permeable alluvial deposits of past human interference including drainage. The equivalent porous medium approach used to mode groundwater flow within the fissured granite zone gave satisfactory results : the model was able to reproduce discharge at several locations within the high-relief catchment and groundwater table depth in most monitoring points. Sensitivity analyses showed that the specific yield and horizontal hydraulic conductivity of the fissured zone are the parameter to which simulated stream discharge and groundwater table depth, including in peat, are most sensitive. The model was forced with new vegetation pararneters to assess the potential impacts of changes in catchment land use on the mire hydrological conditions. Replacement of the broad leaf woodlands that currently cover most of the catchment with conifer plantations would lead to a substantial reduction in surface and groundwater intlows to the mire and to a substantial drop in summer groundwater table depths, particularly along the mire margins.

Desequilíbrios ambientais provocados pela combinação de queimadas, desmatamentos e os fenômenos de ENSO (Oscilação Sul de El Nino) podem ser os responsáveis pelo aumento de períodos de seca na região Amazônica. Com o propósito de compreender as conseqüências que longos períodos de seca podem causar na Floresta Nacional de Tapajós, foi desenvolvido um modelo numérico que simula a dinâmica da água no solo e o Balanço Hídrico para um latossolo (Haplustox, na taxonomia Americana), muito comum na região Amazônica. As simulações foram realizadas para o período de 1999 a 2003, utilizando dados de precipitação, evapotranspiração, umidade do solo, curvas de retenção, propriedades físicas do solo coletadas no local de estudo. Este estudo integra o Projeto Seca-Floresta do grupo de pesquisa do LBA (Experimento de Grande Escala da Biosfera-Atmosfera na Amazônia) para o experimento de exclusão parcial da chuva no projeto na Floresta Nacional Tapajós no que diz respeito a componente de modelagem hidrológica. Os resultados mostraram que, mesmo com a diminuição da quantidade de água disponível para a Floresta, não houve mudanças significativas em relação ao balanço hídrico da floresta, mostrando que a floresta provavelmente se adaptou, para sobreviver a longos períodos de seca / Environmental instability caused by the combination of fire, deforestation and the ENSO phenomena (El Nino South Oscillation) can be the responsible for increases of dry periods in the Amazonian region. With the purpose of understanding the consequences that long dry periods can cause on the National Forest of Tapajós, a mathematical model that determines the water dynamics in soil and the hydrological balance was developed for a typical soil of the Amazonian region the Amazonianlatossol (Haplustox). The simulation were performed for the period from 1999 to 2003, using precipitation, evapotranspiration, soil moisture, retention curves and soil physical properties data obtained in the study area. The present study integrates the Dry-Forest Project of the LBA (Large Scale Biosphere-Atmosphere experiment in the Amazon) research group for the experiment of rainfall exclusion in the National Forest of Tapajos and concerns the hydrological modeling component of the project. The results showed that even with a decrease in the amount of water available to the Forest, significant changes in the hydrological balance of the forest did not occur, showing that the forest had probably adapted itself to survive to longer periods of drought

The cryosphere is defined as the portions of the earth where water is in solid form. It represents a very important part of the hydrologic cycle, affecting ecological, human and climate systems. A number of component models describing the energy and mass balances of a snowpack have been developed and these component models are finding their way into watershed models and land surface schemes. The purpose of this thesis is to examine the incorporation of a number of snow processes in the coupled land-surface-hydrological model WATCLASS. The processes under consideration were mixed precipitation, variable fresh snow density, maximum snowpack density, canopy interception and snow-covered area (SCA). The first four of these processes were based on similar work done by Fassnacht (2000) on a watershed in Southern Ontario. In the case of this thesis, the work was completed on a basin in Northern Manitoba. A theory of the relationship between snow-covered area and average snow depth was developed and an algorithm was developed to implement this theory in WATCLASS. Of the five snow processes considered, mixed precipitation was found to have the greatest impact on streamflow while the new canopy interception algorithm was found to have the greatest impact on sensible and latent heat fluxes. The development of a new relationship between SCA and average snow depth was found to have a minimal impact in one study case, but a significant impact on the sensible and latent heat fluxes when snow fell on a pack that had begun to melt and was partially free of snow. Further study of these snow processes in land-surface-hydrologic models is recommended.

Weather radar has long been recognized as a potentially powerful tool for hydrological modelling. A single radar station is able to provide detailed precipitation information over entire watersheds. The operational use of radar in water resources applications, however, has been limited. Interpretation of raw radar data requires several rigorous analytical steps and a solid understanding of the technology. In general, hydrologists’ lack of meteorological background and the persistence of systematic errors within the data, has led to a common mistrust of radar-estimated precipitation values. As part of the Enhanced Nowcasting of Extreme Weather project, researchers at McGill University’s J.S. Marshall Radar Observatory in Montreal have been working to improve real-time quantitative precipitation estimates (QPEs). The aim is to create real-time radar precipitation products for the water resource community that are reliable and properly validated. The validation of QPEs is traditionally based on how well observed measurements agree with data from a precipitation gauge network. Comparisons between radar and precipitation gauge quantities, however, can be misleading. Data from a precipitation gauge network represents a series of single-point observations taken near ground surface. Radar, however, estimates the average rate of precipitation over a given area (i.e. a 1-km grid cell) based on the intensity of reflected microwaves at altitudes exceeding 1 km. Additionally, both measurement techniques are susceptible to a number of sources of error that further confound efforts to compare the two. One of the greatest challenges facing radar meteorologists is the variation in the vertical profile of reflectivity (VPR). A radar unit creates a volumetric scan of the atmosphere by emitting microwave beams at several elevation angles. As a beam travels away from the radar, its distance from ground surface increases. Different precipitation types are sampled at a number of heights (i.e. snow above the 0º C elevation and rain below it) that vary with range. The difficulty lies in estimating the intensity of precipitation at the Earth’s surface, based on measurements taken aloft. Scientists at McGill University have incorporated VPR correction techniques into algorithms used to automatically convert raw radar data into quantitative hydrological products. This thesis evaluates three real-time radar precipitation products from McGill University’s J.S. Marshall Radar Observatory in the context of hydrological modelling. The C0 radar product consists of radar precipitation estimates that are filtered for erroneous data, such as ground clutter and anomalous precipitation. The C2 and C3 radar products use different VPR correction techniques to improve upon the C0 product. The WATFLOOD hydrological model is used to assess the ability of each radar product to estimate precipitation over several watersheds within the McGill radar domain. It is proposed that using a watershed as sample area can reduce the error associated with sampling differences between radar and precipitation gauges and allow for the evaluation of a precipitation product over space and time. The WATFLOOD model is run continuously over a four-year period, using each radar product as precipitation input. Streamflow hydrographs are generated for 39 gauging stations within the radar domain, which includes parts of eastern Ontario, south-western Quebec and northern New York and Vermont, and compared to observed measurements. Streamflows are also modelled using distributed precipitation gauge data from 44 meteorological stations concentrated around the Montreal region. Analysis of select streamflow events reveals that despite the non-ideal placement of precipitation gauges throughout the study area, distributed precipitation gauge data are able to reproduce hydrological events with greater accuracy and consistency than any of the provided radar products. Precipitation estimates within the McGill radar domain are found to only be useful in areas within the Doppler range (120-km) where the radar beam is unobstructed by physiographic or man-made features. Among radar products, the C2 VPR-corrected product performed best during the greatest number of the flood events throughout the study area.

The cryosphere is defined as the portions of the earth where water is in solid form. It represents a very important part of the hydrologic cycle, affecting ecological, human and climate systems. A number of component models describing the energy and mass balances of a snowpack have been developed and these component models are finding their way into watershed models and land surface schemes. The purpose of this thesis is to examine the incorporation of a number of snow processes in the coupled land-surface-hydrological model WATCLASS. The processes under consideration were mixed precipitation, variable fresh snow density, maximum snowpack density, canopy interception and snow-covered area (SCA). The first four of these processes were based on similar work done by Fassnacht (2000) on a watershed in Southern Ontario. In the case of this thesis, the work was completed on a basin in Northern Manitoba. A theory of the relationship between snow-covered area and average snow depth was developed and an algorithm was developed to implement this theory in WATCLASS. Of the five snow processes considered, mixed precipitation was found to have the greatest impact on streamflow while the new canopy interception algorithm was found to have the greatest impact on sensible and latent heat fluxes. The development of a new relationship between SCA and average snow depth was found to have a minimal impact in one study case, but a significant impact on the sensible and latent heat fluxes when snow fell on a pack that had begun to melt and was partially free of snow. Further study of these snow processes in land-surface-hydrologic models is recommended.

Weather radar has long been recognized as a potentially powerful tool for hydrological modelling. A single radar station is able to provide detailed precipitation information over entire watersheds. The operational use of radar in water resources applications, however, has been limited. Interpretation of raw radar data requires several rigorous analytical steps and a solid understanding of the technology. In general, hydrologists’ lack of meteorological background and the persistence of systematic errors within the data, has led to a common mistrust of radar-estimated precipitation values. As part of the Enhanced Nowcasting of Extreme Weather project, researchers at McGill University’s J.S. Marshall Radar Observatory in Montreal have been working to improve real-time quantitative precipitation estimates (QPEs). The aim is to create real-time radar precipitation products for the water resource community that are reliable and properly validated. The validation of QPEs is traditionally based on how well observed measurements agree with data from a precipitation gauge network. Comparisons between radar and precipitation gauge quantities, however, can be misleading. Data from a precipitation gauge network represents a series of single-point observations taken near ground surface. Radar, however, estimates the average rate of precipitation over a given area (i.e. a 1-km grid cell) based on the intensity of reflected microwaves at altitudes exceeding 1 km. Additionally, both measurement techniques are susceptible to a number of sources of error that further confound efforts to compare the two. One of the greatest challenges facing radar meteorologists is the variation in the vertical profile of reflectivity (VPR). A radar unit creates a volumetric scan of the atmosphere by emitting microwave beams at several elevation angles. As a beam travels away from the radar, its distance from ground surface increases. Different precipitation types are sampled at a number of heights (i.e. snow above the 0º C elevation and rain below it) that vary with range. The difficulty lies in estimating the intensity of precipitation at the Earth’s surface, based on measurements taken aloft. Scientists at McGill University have incorporated VPR correction techniques into algorithms used to automatically convert raw radar data into quantitative hydrological products. This thesis evaluates three real-time radar precipitation products from McGill University’s J.S. Marshall Radar Observatory in the context of hydrological modelling. The C0 radar product consists of radar precipitation estimates that are filtered for erroneous data, such as ground clutter and anomalous precipitation. The C2 and C3 radar products use different VPR correction techniques to improve upon the C0 product. The WATFLOOD hydrological model is used to assess the ability of each radar product to estimate precipitation over several watersheds within the McGill radar domain. It is proposed that using a watershed as sample area can reduce the error associated with sampling differences between radar and precipitation gauges and allow for the evaluation of a precipitation product over space and time. The WATFLOOD model is run continuously over a four-year period, using each radar product as precipitation input. Streamflow hydrographs are generated for 39 gauging stations within the radar domain, which includes parts of eastern Ontario, south-western Quebec and northern New York and Vermont, and compared to observed measurements. Streamflows are also modelled using distributed precipitation gauge data from 44 meteorological stations concentrated around the Montreal region. Analysis of select streamflow events reveals that despite the non-ideal placement of precipitation gauges throughout the study area, distributed precipitation gauge data are able to reproduce hydrological events with greater accuracy and consistency than any of the provided radar products. Precipitation estimates within the McGill radar domain are found to only be useful in areas within the Doppler range (120-km) where the radar beam is unobstructed by physiographic or man-made features. Among radar products, the C2 VPR-corrected product performed best during the greatest number of the flood events throughout the study area.