Water availability and management issues related to the supply of drinking water in northern communities are problematic in Canada. While rivers and lakes are abundant, they are vulnerable to contamination and may become dry in winter due to freezing. Groundwater can often provide a more secure and sustainable water source, however its availability is limited in northern Canada due to the presence of permafrost. Moreover, the exploitation of northern aquifers poses a dual challenge of identifying not only permafrost-free areas, but also permeable areas which will allow groundwater recharge and exploitation. Suitable aquifers are not as common in northern Canada since the shallow subsurface is mostly composed of low-permeability crystalline rocks or unconsolidated sediments of glacial origin that are highly heterogeneous. In order to investigate groundwater occurrence and associated geological contexts in Nunavik (northern Quebec, Canada), along with exploring how these resources will evolve in response to climate change, field and compilation work were conducted in the surroundings of the four villages of Salluit, Kuujjuaq, Umiujaq and Whapmagoostui-Kuujjuarapik. These villages are located in different permafrost zones, ranging from continuous to discontinuous, as well as in different geological environments. It was found that despite the ubiquitous presence of permafrost, unfrozen aquifers could be identified, which suggests that groundwater may be available as a source of drinking water for small communities. Expected climate change, with predicted permafrost thawing and increases in temperature and precipitation, should enhance groundwater availability and may contribute to a more secure source of drinking water for northern communities.

As part of a province-wide groundwater characterization program, a detailed groundwater geochemistry survey was undertaken in the Outaouais Region (Québec, Canada) in order to identify the primary processes responsible for groundwater quality and to develop a conceptual model for groundwater flow and geochemical evolution. During the summers of 2011 and 2012, 139 samples were collected from municipal and private wells which were analysed for major ions, nutrients, trace elements and sulphides. About 70 % of the samples were obtained from bedrock wells, mainly in the silicate rocks of the Canadian Shield and the remainder from wells screened in Quaternary deposit aquifers. Hydrogeochemical facies distributions were determined from 127 of these samples which had anion-cation charge balance errors within ±10 %. The classification by facies was also supported by a multivariate statistical analysis, namely factor analysis combined with hierarchical cluster analysis. The study identified Champlain Sea invasion, cation exchange and freshwater recharge as the main geochemical processes affecting groundwater chemistry in this region. Secondary processes, related to the bedrock geology, are responsible for exceedances of Canadian drinking-water standards, namely for fluoride, uranium, iron and manganese.

Groundwater distribution and flow dynamics were studied in a small watershed located in the discontinuous permafrost zone near Umiujaq in Nunavik (Québec), Canada, to assess the seasonal variations and perform a quantitative analysis of the water cycle in a subarctic watershed. Due to the complexity of the subsurface geology within the watershed, an integrated investigation was instrumental to provide a detailed understanding of the hydrogeological context as a basis for the water balance. Based on this water balance, for the two studied hydrological years of 2015 and 2016, the average values are 828 mm for precipitation, 337 mm for evapotranspiration, 46 mm for snow sublimation, 263 mm for runoff, 183 mm for groundwater exchange (losses with other aquifers outside the watershed), and 0 mm for change in water storage. Although these values likely have significant uncertainty and spatial variability, this water balance is shown to be plausible. It was also found that permafrost influences surface water and groundwater interaction, even if located in low-permeability sediments. It is expected that permafrost degradation will likely increase stream baseflow, especially in winter.

Groundwater recharge (GR) is a complex process that is difficult to quantify. Increasing attention has been given to unsaturated zone modeling to estimate GR and better understand the processes controlling it. Continuous soil-moisture time series have been shown to provide valuable information in this regard. The objectives of this study were to (i) analyze the processes and factors controlling GR in an unconfined granular aquifer in a cold and humid environment and (ii) assess the uncertainties associated with the use of data from different sources. Soil moisture data monitored over three years at three experimental sites in southern Quebec (Canada) were used to calibrate the HYDRUS-1D model and to estimate ranges of possible GR in a region where groundwater is increasingly used as a source of fresh water. The simulations identified and quantified important factors responsible for the near-surface water balance that leads to GR. The resulting GR estimates from 2016 to 2018 showed marked differences between the three sites, with values ranging from 347 to 735 mm/y. Mean GR for the three sites was 517 mm/y for 2016–2018 and 455 mm/y for the previous 12-year period. GR was shown to depend on monthly variations in precipitation and on soil textural parameters in the root zone, both controlling soil-water retention and evapotranspiration. Monthly recharge patterns showed distinct preferential GR periods during the spring snowmelt (38–45% of precipitation) and in the fall (29% of precipitation). The use of different meteorological datasets was shown to influence the GR estimates.

Estimation of groundwater recharge is of critical importance for effective management of freshwater resources. Three common and distinct approaches for calculating recharge rely on techniques of baseflow separation, well hydrograph analysis, and numerical modeling. In this study, these three methods are assessed for a watershed in southwestern Quebec, Canada. A physically based surface–subsurface model provides estimates of spatially distributed recharge; two baseflow separation filters estimate recharge from measured streamflow; and a well hydrograph master recession curve technique calculates recharge from water-table elevation records. The recharge results obtained are in good agreement over the entire catchment, producing an annual aquifer recharge of 10–30 % of rainfall. The annual average estimated across all methods is 200 mm/year. High variability is obtained for the monthly and seasonal recharge patterns (e.g. respectively, 0–30 mm for September and 0–95 mm for the summer), in particular between the baseflow filters and the well hydrograph technique and between the hydrograph technique and the simulated estimates at the observation wells. Recharge occurs predominantly in the spring months for the different approaches, except for the master recession curve method for which the highest recharge estimates are obtained during the summer. The recharge distribution obtained with the model shows that the main recharge area of the aquifer is the Covey Hill region. The use of a fully integrated physically based model enables the construction of an arbitrary number of well hydrographs to enhance the representativity of the master recession curve technique.

Dissolved methane concentrations in shallow groundwater are known to vary both spatially and temporally. The extent of these variations is poorly documented although this knowledge is critical for distinguishing natural fluctuations from anthropogenic impacts stemming from oil and gas activities. This issue was addressed as part of a groundwater research project aiming to assess the risk of shale gas development for groundwater quality over a 500-km² area in the St. Lawrence Lowlands (Quebec, Canada). A specific study was carried out to define the natural variability of methane concentrations and carbon and hydrogen isotope ratios in groundwater, as dissolved methane is naturally ubiquitous in aquifers of this area. Monitoring was carried out over a period of up to 2.5 years in seven monitoring wells. Results showed that for a given well, using the same sampling depth and technique, methane concentrations can vary over time from 2.5 to 6 times relative to the lowest recorded value. Methane isotopic composition, which is a useful tool to distinguish gas origin, was found to be stable for most wells, but varied significantly over time in the two wells where methane concentrations are the lowest. The use of concentration ratios, as well as isotopic composition of methane and dissolved inorganic carbon (DIC), helped unravel the processes responsible for these variations. This study indicates that both methane concentrations and isotopic composition, as well as DIC isotopes, should be regularly monitored over at least 1 year to establish their potential natural variations prior to hydrocarbon development.

Several groundwater-level forecasting studies have shown that data-driven models are simpler, faster to develop, and provide more accurate and precise results than physical or numerical-based models. Five data-driven models were examined for the forecasting of groundwater levels as a result of recharge via tailings from an abandoned mine in Quebec, Canada, for lead times of 1 day, 1 week and 1 month. The five models are: a multiple linear regression (MLR); an artificial neural network (ANN); two models that are based on de-noising the model predictors using the wavelet-transform (W-MLR, W-ANN); and a W-ensemble ANN (W-ENN) model. The tailing recharge, total precipitation, and mean air temperature were used as predictors. The ANN models performed better than the MLR models, and both MLR and ANN models performed significantly better after de-noising the predictors using wavelet-transforms. Overall, the W-ENN model performed best for each of the three lead times. These results highlight the ability of wavelet-transforms to decompose non-stationary data into discrete wavelet-components, highlighting cyclic patterns and trends in the time-series at varying temporal scales, rendering the data readily usable in forecasting. The good performance of the W-ENN model highlights the usefulness of ensemble modeling, which ensures model robustness along with improved reliability by reducing variance.

Permafrost thaw is a complex process resulting from interactions between the atmosphere, soil, water and vegetation. Although advective heat transport by groundwater at depth likely plays a significant role in permafrost dynamics at many sites, there is lack of direct measurements of groundwater flow patterns and fluxes in such cold-region environments. Here, the finite volume point dilution method (FVPDM) is used to measure in-situ groundwater fluxes in two sandy aquifers in the discontinuous permafrost zone, within a small watershed near Umiujaq, Nunavik (Quebec), Canada. The FVPDM theory is first reviewed, then results from four FVPDM tests are presented: one test in a shallow supra-permafrost aquifer, and three in a deeper subpermafrost aquifer. Apparent Darcy fluxes derived from the FVPDM tests varied from 0.5 × 10⁻⁵ to 1.0 × 10⁻⁵ m/s, implying that advective heat transport from groundwater flow could be contributing to rapid permafrost thaw at this site. In providing estimates of the Darcy fluxes at the local scale of the well screens, the approach offers more accurate and direct measurements over indirect estimates using Darcy’s law. The tests show that this method can be successfully used in remote areas and with limited resources. Recommendations for optimizing the test protocol are proposed.