Past Research Projects

Climate Change Impacts on Water

Rathay (MSc, 2016) investigated how a shift in the frequency and magnitude of rainfall could influence the amount and timing of aquifer recharge by characterizing the response of a fractured bedrock aquifer to heavy rainfall events. The study site was the Gulf Islands, British Columbia. Both the recharge and discharge environment (using the stable isotopes of water and thermal imagery) were examined. Historical heavy rainfall events were correlated to the groundwater level fluctuations in order to determine a threshold rainfall amount that results in a distinguishable response in groundwater level. The historic climate record were then used to estimate mean annual recharge and model analytically how the occurrence of more heavy rainfall events in future could influence groundwater recharge over longer periods of time.

Burgess (MSc, 2017) characterized recharge to a fractured bedrock aquifer in a temperate climate. The study area was Gabriola Island located in the Gulf Islands region of British Columbia. The study aimed to assess how low storage fractured bedrock and seasonal rainfall patterns affect rainfall-runoff-recharge processes, taking a holistic approach. The research involved using a coupled surface water-groundwater numerical model (MIKE-SHE) to simulate the recharge and discharge processes. The effect of climate change stressors on recharge was also modeled.

Van Pelt (M.Sc, 2016) investigated paleorecharge to Yucatan Peninsula in Mexico. The goals of this study were: (a) to compare multiple sources of paleoproxy data and with paleo-model simulations of the past1000 years (from the Community Climate System Model Version 4, part of the Coupled Model Intercomparison Project Phase 5 past1000 experiment) to analyze changes in precipitation and temperature, and (b) to use this comparison to generate a daily climate time series that is representative of the Terminal Classic Period (T.C.P.) 800-1000 A.D. for use in a groundwater recharge model.  A coupled surface water-groundwater model (MIKE SHE) was used to assess how groundwater recharge has changed between the T.C.P. and the present climate.  This study aimed to increase our understanding of climate variability in this region, which has the potential to cause extreme weather conditions in the future climate.

Secondary Impacts of Climate Change on Human and Ecosystem Health: A Risk-based Approach (Project Leader, Diana Allen). Climate change is predicted to have significant direct impacts on air temperature and precipitation in terms of long-term trends, seasonality, and occurrence of extreme events. Such changes are anticipated to have primary (i.e., direct) impacts on hydrologic regimes (e.g., melting glaciers, sea-level change) and air quality, and consequent secondary (i.e., indirect) impacts on health of humans and ecosystems. Primary climate-change impacts have been investigated by researchers around the world, but relatively few studies have considered secondary effects of these impacts or adaptation responses, particularly in terms of the potential spread of infectious diseases, quantity of water available, degradation of water quality due to contamination, and loss of biodiversity, which is important for maintaining resilient and functioning ecosystems in the presence of changes. Among the many secondary changes anticipated are major shifts in species distributions and ecosystems, and potentially widespread (human) emigration. These secondary effects raise questions about how existing human infrastructure (e.g., water storage and delivery systems, health-care systems) will be able to support growing urban populations. Research on the secondary impacts of climate change was conducted by the Climate Change Impacts Research Consortium. The research spanned the physical, biological, health and social sciences, resource and environmental management, communication, and computing science, and brought together a group of researchers with expertise in climate, water, air quality, disease, ecology, human health, risk analysis, emergency preparedness, and visualization. Building on previous research in British Columbia (BC) and abroad, the research team investigated the secondary effects of climate change using risk-assessment approaches to evaluate various risk-management options for dealing with these problems arising from climate change. Novel computer visualization techniques were developed and applied to support knowledge translation and enabled use of our results by policy-makers and other stakeholders.

An Integrated Approach to Estimating Groundwater Recharge and Storage Variability in Southern Mali, Africa (Chris Henry M.Sc., 2011) This research was carried out as part of a broader iniative in partnership with Global Aquifer Development Foundation (now partnered with Hydrogeologists without Borders) to assist the Mali (Africa) government with the monitoring and management of their groundwater resource. Groundwater recharge in southern Mali was investigated using a variety of methods. The aquifer system comprises a surficial unconfined aquifer in laterite that is hydraulically connected by vertical fractures through a sedimentary rock layer to a deep fractured semi-confined aquifer. Observed groundwater storage fluctuations from historical water level data correlate with GRACE satellite terrestrial water storage (TWS) variations, with peaks in September and lows in May; however, soil-moisture corrected GRACE data peaked in November due to the GLDAS model poorly predicting the timing of soil-water storage changes. Recharge modeling using HELP gave an average annual net recharge of 132.2 mm (12.6% of rainfall), comparing well with estimates from historical water level (149.1 mm; 16.4%) and GRACE (149.7 mm; 14.8%) data. Major ion chemistry suggests groundwater is fresh (average TDS of 205 mg/L) and rapidly recharged. d18O and d2H concentrations in groundwater and precipitation indicate July-September rainfall as the recharge source.

A Basin Approach to Groundwater Recharge in the Okanagan: Bridging the Gap Between Science and Policy (Project Leader, D. Allen) The Okanagan is one of the driest regions of Canada , and rapid development in the region due to both population and agricultural growth has significantly increased demands on both surface and groundwater resources. While exploitation of surface water is regulated, there is no current legislation governing the development and use of groundwater. This unregulated use of groundwater has the potential to have negative impacts on the sustainable development of the resource, and consequent negative impacts on long term social, economic, and agricultural activity in the watershed that rely on it. This Canadian Water Network project (comprising studies by Liggett, Toews, and Voeckler) aimed to enhance scientific understanding of groundwater recharge variability and mechanisms, and directly feed this understanding back to stakeholders through a series of targeted decision support tools.

Modeling Climate Change Impacts on Groundwater Recharge in Semi-Arid Region, southern Okanagan , British Columbia (Mike Toews, M.Sc. 2007) The impacts of future predicted climate change on groundwater recharge resources are modelled for the arid to semi-arid south Okanagan region, British Columbia. The hydrostratigraphy of the region consists of Pleistocene-aged glaciolacustrine silt overlain by glaciofluvial sand and gravel. Spatial recharge is modelled using available soil and climate data with the HELP 3.80D hydrology model. Climate change effects on recharge are investigated using stochastically-generated climate from three GCMs. Recharge is estimated to be ~45 mm/year, with minor increases expected with climate change. However, growing season and crop water demands will increase, posing additional stresses on water use in the region. A transient MODFLOW groundwater model simulates increases of water table in future time periods, which is largely driven by irrigation application increases. Spatial recharge is also used in a groundwater model to define capture zones around eight municipal water wells. These capture zones will be used for community planning.

Climate Change and Low Flows: Influences of Groundwater and Glaciers (Moore, Allen Stahl, Werner, Hutchinson, Cannon and Whitfield) This collaborative project, funded by the Climate Change Action Fund, explored the linkages between low flows in streams, glacier melt and groundwater under climate change. The groundwater component of the study involved comparing the relationships between groundwater, climate, and surface water within and between groups of well records from the two major hydro-climatic zones in BC. We developed a system for detecting the influence of climate change and variability on groundwater in the absence of long term records; suitable correlation coefficients were applied to evaluate the strength of these interactions. Different aquifer types are assessed with respect to vulnerability to climate change influences.

Modelling the Impacts of Climate Change on Groundwater: A Comparative Study of Two Unconfined Aquifers in Southern British Columbia and Northern Washington State (Scibek, M.Sc. 2005). A methodology was developed for linking climate and groundwater models to investigate future impacts of climate change on groundwater resources using two case study sites of unconfined aquifers in southern British Columbia and northern Washington State . One semi-arid site is compared with one wet coastal site. The two groundwater systems differ in river-aquifer interactions, recharge, aquifer heterogeneity, scale, and groundwater use. Climate change scenarios from the Canadian Global Coupled Model 1 model runs for 1961-2000, 2010-2039, 2040-2069 and 2070-2099 are downscaled to local conditions, modelled at daily time scales using a stochastic weather generator, and applied to the spatially-distributed infiltration model. At one site the basin-scale runoff is also downscaled to predict river discharge and river-aquifer interactions in future climates. The impacts of predicted climate change on the groundwater system for each site are modelled in three-dimensions using Visual MODFLOW. Grand Forks Aquifer Study. (Research Funded by Climate Change Action Fund, BC Ministry of Environment, and Environment Canada ).

Impacts of Predicted Climate Change on Groundwater Recharge, Gulf Islands , British Columbia , Canada (Appaih-Adjei, M.Sc. University of Lund, Sweden, 2006). This research investigated the potential impact of climate change on groundwater recharge to the fractured bedrock aquifers, which serve as the main source of potable water supply to the inhabitants of Gulf Islands in BC, Canada . Using Statistical DownScaling Model (SDSM) in combination with the LARS-WG stochastic weather generator, daily current and future (i.e., 2010-2039, 2040-2069, and 2070-2099) climate data were generated from CGCM1 predictions of the study location. These predictions were used as input to the HELP hydrologic model for estimation of recharge for the different climate change periods. The main properties of the aquifer – soil permeability, aquifer permeability and water table depth - used for recharge modeling were linked to ArcGIS for generating recharge zones, which allowed spatial and temporal integration of the recharge results. The combination of SDSM and LARS-WG in downscaling and predicting both the observed monthly temperature and precipitation was very successful. Mean annual precipitation downscaling with SDSM is predicted to increase by 52%, 65% and 88% relative to the observed for 2020's, 2050's and 2080's, respectively. On the other hand, the mean monthly temperature is predicted to rise by 1.14­­ o C in 2020's, 2.05 o C in the next 30 years, and up to 3.5 o C by the end of the century. According to HELP, the current mean annual recharge is about 44 % of the annual precipitation and is predicted to increase progressively by 7%, 8% and 9% in the 2020's, 2050's and 2080's, respectively, from the current.

Water Security and Risk Assessment

Groundwater Resource Vulnerability for Small Island Developing States (Lead Researcher: Allen; Students: Holding, Foster, Hsieh, Larocque, Klassen, Van Pelt, 2014) Groundwater constitutes the majority of naturally-occurring freshwater supply on many small islands. However, in addition to potential impacts of climate change and population growth that may compromise water supply and demand on small islands, the Small Island Developing States (SIDS) in particular lack of hydrogeological data for characterizing aquifers and have limited institutional means for managing the groundwater resource. As part of UNESCO-IHP’s Transboundary Water Assessment Programme (TWAP), an assessment of groundwater systems on 43 SIDS was undertaken. This SIDS assessment represents the first baseline global assessment of the status of groundwater in SIDS, covering all major islands states of the world below a surface threshold of 50,000 km2. As many SIDS comprise multiple islands with different hydrogeologic settings, a representative island from each SIDS was chosen to represent the majority of the population. Data were compiled from publications, accessible datasets and a questionnaire distributed to collect local expertise. A total of 74 variables were defined for each SIDS, according to the level of confidence in the data (i.e. high confidence when based on specific studies and low confidence when inferred). Although all islands are vulnerable to saltwater intrusion, SIDS reliant on small coastal aquifers are at higher risk of saltwater contamination from sea level rise, pumping, and wave overwash events. Despite limitations and uncertainty in the data used in the assessment, the results provide a broad assessment of current and future groundwater resource vulnereability for SIDS throughout the world.

Risk Assessment Framework for Coastal Bedrock Aquifers. This project aimed to develop a risk assessment methodology suitable for coastal bedrock aquifers that is both straightforward and standardized, based on common risk assessment principles. A risk framework, spatial risk maps and the set of indicators that can be used to establish guidelines for drilling, monitoring and well operation (groundwater development and use) will be developed. Ultimately, these deliverables can be used as tools for land use planning, direct monitoring efforts, and to build community preparedness (i.e. risk management). The study was funded by Natural Resources Canada (NRCan), the Pacific Institute for Climate Solutions (PICS), and was conducted in partnership with the BC Ministry of Environment, the BC Ministry of Forests, Lands and Natural Resource Operations, and the Salt Spring Island Water Council. Two SFU students were involved in this project (Larocque, M.Sc. 2013 and Klassen, M.Sc. 2015). Glenna Erlandson (Vancouver Island University) helped to refine aquifer susceptibility maps that were previously developed for the Gulf Islands region (Denny et al., 2007; Hydrogeology Journal). See detailed descriptions of the student projects below.

Assessing the Risk of Saltwater Intrusion on the Gulf Islands, BC (MSc, Klassen, 2015) In coastal regions, the quality of groundwater can be compromised due to saltwater intrusion (SWI) caused by various natural (sea level rise and storm surge) and anthropogenic (pumping) hazards. The goal of this research was to distinguish groundwaters impacted by SWI in the bedrock aquifers of the Gulf Islands, BC and identify thresholds for select chemical parameters that can be used for monitoring purposes, as well as to develop and test an approach for assessing risk to groundwater quality in coastal aquifers. The most reliable indicators were Cl/(HCO3 + CO3), BEX (base exchange index), Cl vs. EC, depth vs TDS, and a quantile analysis, resulting in 138 well samples (out of 795) that appear to be impacted by SWI. Based the 95th percentiles, for which 100% of the samples graphically showed strong evidence of SWI, the recommended threshold for Cl is 480 mg/L, 2,090 µS/cm for EC, and 970 mg/L for TDS. These samples were collected from wells that predominantly fall along the coastline. The vulnerability of the bedrock aquifers to SWI was assessed spatially by mapping hazards in combination with the aquifer susceptibility. Hazards due to pumping have the greatest influence on the vulnerability. Risk was evaluated spatially using an economic valuation of loss – here replacement of a water supply. The combination of chemical indicators and risk assessment maps are useful tools for identifying areas vulnerable to SWI, and these tools can be used to improve decision-making related to monitoring and community development for coastal areas.

Risk to Water Security on Small Islands: A Numerical Modeling Approach (PhD, Holding, 2014) The aim of this research is to characterise risk to water security for small islands. This is achieved by modeling the spatial and temporal impact from major stressors affecting water resources on small islands, and then evaluating the risk to water security through an integrated assessment framework. Numerical density-dependent flow and transport modeling is used to evaluate the response of the freshwater lens on Andros Island in The Bahamas to various climate change and human stressors including: sea level rise, changes in recharge, and increased pumping.  SEAWAT models showed a reduction of freshwater lens volume by up to 24% under projected sea level rise and reduced recharge. The response time of the freshwater lens increased with stressor magnitude, resulting in a longer lens adjustment period.  In addition, greater upconing was observed for pumping scenarios simulated under projected climate change conditions than under current conditions. The impact of a 2004 storm overwash event on Andros Island was simulated using HydroGeosphere. Results show that potable water is restored one month sooner when timely remedial actions are implemented; however, if delayed by four days or more, there is no improvement in recovery time. To extend the research more broadly, simulations of overwash for various island types observed worldwide were conducted. Dominant factors affecting freshwater lens response include vadose zone thickness and geologic heterogeneity, such as low or high permeability zones, whereas the dominant factor affecting freshwater lens recovery is recharge rate.   A framework to characterise risk to water security was developed specific to an island hydrogeological setting. A freshwater lens susceptibility map was generated using the results of the numerical modeling. Hazard threats from climate change and human stressors (derived from numerical modeling and a land-use survey) were overlaid on the susceptibility map to represent vulnerability. Combining vulnerability with loss (or consequence) yielded a risk to water security map. High risk areas are largely concentrated within the developed areas near high chemical hazard activities, as well as along portions of the coastline. These maps were provided to local partners to inform water management policies and raise awareness about factors impacting water security.

Developing a Canadian Water Security Framework as a Tool for Assessing Cumulative Impacts and Improving Watershed Governance Two student research projects (Simpson and Cavalcanti de Albuquerque) formed part of a collaborative research project funded by the Canadian Water Network (CWN) to develop a Water Security Framework for Canada. The overall objective of the CWN project was to develop a framework for assessing the security of water. In collaboration with Natural Resources Canada, we developed a set of decision support tools that include both groundwater and surface water, and that address issues of quantity as well as quality. The tools aim to support existing methodologies, such as vulnerability mapping methods, numerical / analytical models, be flexible enough to be used in different jurisdictions (e.g., Ontario with a rigid legislative framework and BC without), and provide output in a format that is understandable to water managers.

Assessing Risk to Groundwater Quality Using an Integrated Risk Framework (M.Sc., Simpson, 2012) Risk to groundwater quality is defined as a function of susceptibility, hazard and consequence. Aquifer susceptibility combines the intrinsic susceptibility of the physical system with potential preferential pathways. Hazard threats are assessed based on the potential impact and probability of release. The consequence is the financial cost of the loss of the resource. The risk assessment methodology is applied to the Township of Langley, BC. The results define vulnerable areas as those where susceptible aquifers coincide with chemical and biological threats. The risk is greatest where vulnerable areas coincide with high potential financial loss: within municipal well capture zones and where wells serve high value agriculture. A regional numerical model was constructed and used to outline capture zones for wells that may be at risk. The model was also used to model contaminant transport and highlight the need to consider horizontal groundwater flow when assessing vulnerability.

Comparison of Approaches for Aquifer Vulnerability Mapping and Recharge Modelling at Regional and Local Scales, Okanagan Basin, British Columbia (Liggett, M.Sc., 2008) Aquifer vulnerability and direct recharge from precipitation were modelled in Okanagan Basin, British Columbia. The vulnerability study evaluated mapping approaches for regional and local scales using the DRASTIC method. Original rating tables provide sufficient detail for mapping at the regional scale, where broad ranges of geologic material are present. However, modified rating tables improved spatial representation of input parameters at local scales, which is useful for local planning. Spatially-distributed recharge throughout the valley bottom was modelled using the HELP code. Average annual recharge is 65 mm/yr, with 109 mm/yr near Vernon, and 37 mm/yr near Oliver. The regional recharge map adequately captured the magnitude and distribution compared to a local map constructed using HELP (42 mm/yr); However, regional recharge results were higher compared to a local map constructed using the MIKE-SHE code (6 mm/yr). Compared to measured evapotranspiration data, HELP appears to under-estimate evapotranspiration, therefore over-estimating recharge within semi-arid regions.

Aquifer-Stream Hydraulic Connectivity

Aquifer - Stream Connectivity at Various Scales: Application of Sediment – Water Interface Temperature and Vulnerability Assessments of Groundwater Dependent Streams (PhD, Middleton, 2016) Streams with greater connectivity to an aquifer are potentially more sensitive to changes in groundwater levels and fluxes. Aquifer-stream connectivity during the summer low flow period is of particular concern because this is a period of maximum groundwater contribution to stream flow volumes, which coincides with periods of peak water demands and critical aquatic habitat needs. This research focused on the use of streambed-interface temperature in combination with a range of field methods to characterize aquifer-stream connectivity and evaluate factors influencing the groundwater flux to streams at different scales. Field methods included streambed-interface temperatures as well as manual stream discharge measurements, seepage meters, and in-stream piezometers.  A simplified heat budget was used in combination with groundwater flux measurements in two case study streams, Fishtrap and Bertrand Creeks, British Columbia, to demonstrate that the groundwater flux during the summer periods was higher in Fishtrap Creek than in Bertrand Creek, due to a higher vertical hydraulic gradient into the stream and higher aquifer sediment permeability. The results also demonstrated that a combination of field measurements improved the estimation of the groundwater flux due to measurement uncertainties. Independent component analysis (ICA), combined with cross-correlation was a novel approach to temperature signal separation that directly linked the extracted signals to factors in the heat budget that influence streambed-interface temperatures within a stream reach. Surface heating from solar radiation was the dominant factor influencing the interface temperature in most years, but there is evidence that thermal exchanges took place at the water-sediment interface, and the correlation with groundwater levels indicated these heat exchanges were associated with groundwater influx. Overall, the combined approaches were able to attribute temporal and spatial variability in streamflow and streambed-interface temperatures to relative contributions of groundwater to streams. The understanding of aquifer-stream connectivity at different scales was applied in the development of a vulnerability framework for assessing groundwater-dependent streams, and to determine stream vulnerability to changes in groundwater conditions. This framework can be used in support of decision making surrounding Sensitive Stream Designation in British Columbia and water allocation under the Water Sustainability Act. This research project was funded by the Pacific Institute for Climate Solutions. The project also contributed to research on the secondary impacts of climate change on the health of aquatic ecosystems carried out by the former Climate Change Impacts Research Consortium (CCIRC).

Hydrogeochemistry

Hydrochemical Evolution and Arsenic Mobilization in Confined Aquifers Formed Within Glaciomarine Sediments (R. Cavalcanti de Albuquerque, M.Sc. 2011). The hydrogeochemical evolution and arsenic mobilization mechanisms in groundwater occurring in confined aquifers formed within glaciomarine sediments in the Lower Fraser Valley, British Columbia, were addressed. Methodology included analysis of chemical and isotopic composition of groundwater, and mineralogical and chemical analysis of sediment samples sourced from core. Groundwater in confined aquifers is Na-HCO3 or Na-Cl type, basic and reduced; whereas groundwater in unconfined aquifers is Ca-Mg-HCO3 type, near neutral and oxidized. The chemistry of groundwater in confined aquifers is controlled by cation exchange, dissolution of carbonate minerals, silicate mineral weathering, and mixing with saline connate water suggesting freshening conditions.  Arsenic release occurs as groundwater flows through glaciomarine sediments; its mobility is favoured by basic pH and reducing groundwater conditions. Possible arsenic release mechanisms are iron (hydr)oxide reduction and sulphide oxidation. A method of spatially representing likelihood of arsenic occurrence in groundwater based on geochemical interpretation and available data was developed.

Nitrate Contamination: Abbotsford-Sumas Aquifer Nitrate contamination of the trans-national Abbotsford-Sumas aquifer in the central Fraser Valley has become a significant problem over the last 30 years. Nitrate concentrations above the maximum allowable concentration (10 mg/L NO3-N) have been recorded in many of the aquifer's wells since the 1970's. Agricultural land-use above the aquifer is primarily raspberries, and although application practices for fertilizer have improved in recent years, nitrate concentrations in the aquifer have not dropped. Vadose zone transport simulations have been undertaken to determine the potential loading to the aquifer by synthetic fertilzer and the loading concentrations have been applied to a 3D numerical groundwater model to investigate the transport of nitrate within the aquifer, both from historical and future perspectives (Chesnaux, post-doctoral research, 2008). Recent research focused on how climate variability (related to Pacific Decadal Oscillation and El Nino Southern Oscillation) affect nitrate concentrations measured in observation wells. This work was conducted in collaboration with Environment Canada (Graham et al. 2014; Canadian Water Resources Journal 39(1)).

A Study of Aquifer Heterogeneity and its Effects on Nitrate Transport and Distribution Using Geophysics and Numerical Groundwater Modelling in the Abbotsford-Sumas Aquifer, British Columbia, Canada and Washington State, USA (S. McArthur, M.Sc. 2006). Heterogeneity within the sand and gravel glacial outwash deposits of the Abbotsford-Sumas aquifer was investigated using ground penetrating radar (GPR) and borehole geophysical logging. Layering consists of fining upward sequences up to 5 m thick that are continuous over 10's of metres. Smaller heterogeneities were identified visually at a local gravel pit. Heterogeneity is best represented in a local scale groundwater model using vertical anisotropy based on the relative comparison of model travel times and groundwater ages. Model ages, however, are consistently underestimated. The spatial distribution of nitrate provides initial and calibration concentrations for the transport model. Observed concentrations are considerably higher than those predicted by the model, suggesting that either the source of nitrate used in the model is too low and that other sources should be considered (such as mobilization during summer when berries are irrigated), or that the current concentrations reflect a much longer history of contamination.

Stream Discharge and Hydrochemical Variation Over the Low Flow Season in the Abbotsford Aquifer, British Columbia (B.Sc. Honours, M.A. Berg, 2005). The Canadian portion of the Abbotsford-Sumas aquifer is located in the Lower Fraser Valley in southwest British Columbia . Understanding the interaction between the surface water and the groundwater in this aquifer is important in order to model the aquifer system response to potential climate change. Groundwater-surface water interactions are also important to understand for the protection of the endangered Nooksack Dace and Salish Sucker fish populations in the vicinity. This study was conducted on Fishtrap Creek, Pepin Brook, and Bertrand Creek . Regional flow measurements and chemistry sampling were conducted along the length of all three streams, and data were collected monthly. The water was sampled for physical parameters and dissolved chemical constituents. The study found groundwater discharging into Fishtrap Creek and Pepin Brook, and that both these streams have dissolved oxygen levels that fall below the accepted level for aquatic health. Nitrate levels also fluctuate in these streams and exceed background levels, indicating possible contamination. The low flow repeatability component of the study was conducted at two sites; one on Fishtrap Creek, and one on Bertrand Creek . The flow measurements were repeated across the channel, and down the length of a 7.5m section, and the results analyzed to determine the relation between the standard deviation, the percent error, and the mean discharge. At Bertrand Creek , there was a strong negative correlation (R 2 =0.9) between the percent error and the mean discharge. At Fishtrap Creek, the correlation was weak, and the results for that stream are inconclusive.

Determining the Origin of Groundwater Using Stable Isotopes of 18O, 2H and 34S (Allen, 2004) Stable isotopes of 18 O and 2 H in water and 34 S and 18 O in dissolved SO 4 are used to verify the interpretation of the chemical evolution and proposed sources of salinity for two islands located in southwest British Columbia. Results for d 18 O and d 34 S in SO 4 suggest a three component mixing between 1) atmospheric SO 4 derived largely from recharge of meteoric origin, 2) modern marine SO 4 associated with either modern day saltwater intrusion or Pleistocene-age seawater, and 3) terrestrial SO 4 . The age of the marine SO 4 is uncertain based on the geochemistry and SO 4 isotopes alone. Two options for mixing of saline groundwaters are proposed; either between current day marine SO 4 and atmospheric SO 4 , or between older (Pleistocene age) marine SO 4 and atmospheric SO 4 . d 18 O and d 2 H compositions are relatively consistent between both islands with a few samples showing evidence of mixing with water that is a hybrid mixture of Fraser River water and ocean water. The isotopic composition of this hydrid water is approximately d 18 O = 10 o / oo . d 18 O and d 2 H values for many saline groundwaters plot close to the global meteoric water line, which is distinctly different from the local meteoric water line. This suggests a meteoric origin for groundwaters that is different from the current isotopic composition of meteoric waters. It is proposed that these waters may be late Pleistocene in age and were recharged when the island was submerged below sea level and prior to rebound at the end of the last glaciation.

Chemical Evolution of Groundwater on the Gulf Islands (Suchy, B.Sc. 1998; Matsuo, B.Sc. 2001) These benchmark hydrochemical studies on Saturna Island and Hornby Island, B.C. were completed as undergraduate B.Sc. Honours theses and involved a large-scale sampling programs (funded by the Islands Trust) to investigate the chemical character of groundwaters and surface waters on the islands. The analyses were subsequently used to look at the evolution of groundwater in the Gulf Islands and to describe salinity variations on the islands.

Use of Stable Isotopes ( 206 Pb, 18 O and 2 H) in Delineating Plumes for Acid Rock Drainage Problems (Lepitre, M.Sc. 2001) This research project was a collaborative effort with Jim Mortinsen, University of British Columbia. The project involved sampling groundwaters and spring waters in the mine area of the Sullivan Mine in Kimberly, B.C. The mine is currently being decommissioned. The research demonstrated that stable isotopes of lead in combination with those of water can be used delineate or fingerprint mine effluent from tailings pond in acid rock drainage problem.

Integrated Geochemical and Stable Isotope Analysis of Tailings Effluent at the Seepage Collection System at the Sullivan Mine BC (Voormeij, B.Sc. 2001) This integrated study explored the combined use of hydrochemical data and stable isotopes, 18 O and 2 H, to characterize and quantify the origin and percentage of mixing between the tailings pond effluents, seepage collection and background waters at the Sullivan Mine, BC.

Hydrogeology of Heterogeneous and Fractured Rock Aquifers

Geostatistical Modeling and Upscaling Permeability for Reservoir Scale Modeling in Bioturbated, Heterogeneous Tight Reservoir Rock: Viking Fm, Provost Field, Alberta (M.Sc., Hsieh, 2015) While burrow-affected permeability must be considered for characterizing reservoir flow, the marked variability generated at the bed/bedset scale makes bioturbated media difficult to model.  Study of 28 cored wells of the Lower Cretaceous Viking Formation in the Provost Field, Alberta, Canada integrated sedimentologic and ichnologic features to define recurring hydrofacies possessing distinct permeability grades.  Transition probability analysis was employed to model spatial variations in biogenically enhanced permeability at the bed/bedset scale.  Results suggest that variations in permeability are strongly related to variations in hydrofacies rather than grain size.  The variability in permeability at the bed/bedset scale was simplified by calculating an equivalent permeability that represents the thickness-weighted sum of permeability at the bed/bedset scale using expressions for layered media.  Numerical block models were then generated for both the bed/bedset hydrofacies and the upscaled hydrofacies.  Vertical and horizontal flows were simulated at both scales, and the volumetric flows in each direction were compared to verify the representativeness of the equivalent permeability.  Vertical and horizontal flows simulated for bed/bedset scale and composite hydrofacies differ by less than ±5%, suggesting that permeabilities at the bed/bedset scale can be simplified through upscaling.  Reservoir-scale groundwater flow was simulated along a hydrogeological cross section comprised of the composite hydrofacies.  The resulting flow regime was consistent with those simulated using permeability estimates from tight reservoir units of the Viking Formation.  This approach may lead to improved reserve calculations, estimates of resource deliverability, and understanding of reservoir responses during recovery.

The Hydrogeology of Salt Spring Island, British Columbia (M.Sc., Laroque, 2014) Groundwater on Salt Spring Island, British Columbia, flows through fractures in  sedimentary and igneous rock aquifers. Recharge is dominantly by rainfall infiltration. At a local scale, groundwater discharges into lakes and streams; regional flow is toward the coast where the groundwater discharges to the ocean. Groundwater evolves from a Na-Cl rainwater to a Ca-HCO3 type through calcite dissolution. Cation exchange (Ca exchanges for Na) is a dominant process in the sedimentary rocks (but not in the igneous rocks), resulting in a Na-HCO3 type water. Mixing with a Cl-rich end-member is also a dominant process in both rock types. Some wells near the coast are known to be impacted by saltwater intrusion. Despite the differences in scale for testing, and rock type, pumping and tidal response tests yield similar averages across the Gulf Islands. Transmissivity is estimated on the order of 10-5 to 10-4 m2/s and hydraulic conductivity on the order of 10-7 to 10-6 m/s. Using Visual MODFLOW, a steady-state fresh groundwater simulation for the Swan Point area generated a representative flow system and established a reasonable range for the aquifer properties. A tidally-forced transient model simulated the tidal response and further constrained the aquifer properties. Next, a density-dependent flow and transport model was constructed in SEAWAT to simulate the current position of the saltwater interface, which was found to be near vertical at the coast. The lack of a wedge likely reflects the relatively steep topography at this site and suggests that submarine groundwater discharge may occur. However, the small model domain and imposed boundary conditions may over-estimate the amount of inflowing water. The SEAWAT model was then used to determine the sensitivity of the aquifer to climate change, encompassing an increase in mean annual recharge by ~1.5 % and SLR of 1.17 m by the end of this century, as well as pumping. Model results showed no significant impacts to the salinity distribution or saltwater wedge geometry at this particular site due to the steep topography. Areas with less steep topography may have greater impacts and merit further research.

Constraining a Density-Dependent Groundwater Flow Model Using Multiple Calibration Time Periods (Trapp, University of Stuttgart, M.Sc., 2011). Calibrating a density-dependent groundwater flow model can be a challenging task and is often afflicted with great uncertainty. Data are usually limited spatially and temporally; however, projections are made well beyond those limitations. To test the validity of such models, a study area was chosen where extensive data were available. These included steady water level measurements, pumping tests, aquifer response to ocean tides, long-term water level measurements and geochemical data. A finite element model was set up using FEFLOW, calibrated for each data set individually and the results were then checked for consistency. With increasing temporal extent of the data, lower hydraulic conductivity values were required for calibration. Furthermore, the study area comprises a fractured bedrock aquifer, which highlights difficulties using the equivalent porous medium approach when describing such systems.

Vulnerability Mapping Method for Fractured Rocks: DRASTICFm (in collaboration with S. Denny and M. Journeay, Geological Survey of Canada). DRASTIC, the commonly-used methodology for mapping the intrinsic vulnerability of aquifers, is modified to incorporate the structural characteristics of fractured bedrock aquifers. In these aquifers, groundwater flow is predominantly through fractures, with large-scale fracture zones and faults acting as primary conduits for flow at the regional scale. The methodology is applied to the southern Gulf Islands region of southwestern British Columbia , Canada . Bedrock geology maps, soil maps, structural measurements, mapped lineaments, water well information, and topographic and bathymetric data, assembled within a comprehensive GIS database, form the basis for assigning traditional DRASTIC indices, while adding the structural indices necessary for capturing the importance of regional structural elements in recharge and well capture zone determinations.

A Hydrostructural Domain Approach to Quantifying Heterogeneity in Fractured Rock Aquifers, Gulf Islands, BC (M. Surrette, M.Sc., 2006). A hydrostructural domain approach was used to model regional scale groundwater flow in the fractured bedrock aquifers of the Gulf Islands , British Columbia , Canada . The domains were defined using fracture intensity and modeled using a stochastic, discrete fracture network-equivalent porous medium (DFN-EPM) approach. Results showed that the “highly” fractured interbedded sandstone and mudstone (<10 cm spacing) and fault and fracture domains had greater potential porosity than the “less” fractured sandstone (>1.0 m spacing) domain. The two highly fractured domains had an average permeability of 10 -13 m 2 compared to 10 -14 m 2 for the less fractured domain. The model results also showed a westward decrease in transmissivity, porosity and permeability. This decrease appears to be associated with the hinge line of a large anticline. Independently collected pump test analysis confirmed this interpretation. The DFN-EPM approach used in this thesis may have applications to other areas where groundwater resources in fractured rock aquifers are of interest.

The Role of Aquifer Heterogeneity in Saltwater Intrusion Modelling, Saturna Island , British Columbia ( E. Liteanu , M.Sc. 2003). Density-dependent flow and solute transport simulations were carried out using USGS SUTRA. Models simulations indicate that the magnitude of the permeability and the nature of layering exercise a major control of the magnitude and appearance of the freshwater-saltwater interface. The Pleistocene sea level history for the Gulf Islands , BC and chemical analyses for groundwater together suggest that saltwater intrusion on Saturna Island has two different origins: direct intrusion and older entrapped groundwater. A number of steady–state simulations were first undertaken to calibrate the model using groundwater geochemical data. To verify the model calibration, a transient simulation was conducted to simulate the behaviour of the freshwater-saltwater interface over the last 12,000 years. Over that time period, the island was submerged for a period of about 1,000 years, and rebounded, in a near instantaneous fashion with sea level at its current position. The transient simulations were undertaken to test if the period of 1,000 years of submergence was long enough to saturate the island with saltwater, and to test if the period following rebound is sufficient to result in the current observed saltwater-freshwater interface position. These simulations show that, considering the coarse approximations used in this research, the conceptual model is consistent with the Pleistocene sea level history for the area.

An Integrated Structural and Hydrogeological Investigation of the Fracture System in the Upper Cretaceous Nanaimo Group, Southern Gulf Islands , BC . (Mackie, M.Sc. 2002) The purpose of this collaborative study (with P. Mustard) was to identify differences in fracture distribution and character with respect to lithology and different generations of geologic structures, and to apply this fracture distribution and characterization to the development of a conceptual model for fractured controlled groundwater. A total of over 8000 fracture measurements were made at 157 stations on 8 islands using scanline, grid mapping and random data collection techniques. Results indicate that fracturing related to the Late Cretaceous to Neogene tectonic history is heterogeneously distributed and can be separated into groups. Four primary groups were identified: deformation bands, bedding-perpendicular fractures, faults and fracture zones, and non-bedding-perpendicular fractures. Relationships between structures, lithology and fracture spacing are used to define hydrostructural domains, areas of differing bulk permeability. Four domains are defined and supported by geochemical, geophysical and well yield data. The four domains are: discrete fault and fracture, fracture zone, bedding-perpendicular fracture, and fault zone. The bedding-perpendicular fracture domain includes two subdomains, the mudstone-dominant and sandstone-dominant domains. The fault zone and mudstone-dominant domain are the highest relative permeability. A methodology is proposed for delineating domains using a combination of lineament analysis and geologic mapping. The net effect has implications for flow system controls, the amount and location of recharge, and potential for saltwater intrusion along shorelines. Regional structural history is determined to have a direct and significant effect on groundwater resources via the distribution of brittle fractures.

The Applicability and Scale Dependence of Aquifer Testing Methods: An Integrated Geophysical and Hydrogeological Characterization of Two Fractured Systems (Abbey, M.Sc. 2000). In this study, geological, horizontal loop electromagnetic (HLEM) and borehole geophysical surveys were used to characterize the lithology and structure of two fractured bedrock aquifers of low primary porosity (limestone/argillite and sandstone/mudstone), and to identify hydrostratigraphic and hydrostructural units and the associated boundary conditions. The applicability of constant-discharge aquifer testing and slug testing for determining hydraulic parameters of fractured aquifers was investigated by evaluating quantitatively each testing method and its associated analytical models (radial, linear, double porosity, unconfined). Pressure derivative analysis of the hydraulic test data aided in identifying boundary conditions and component flow regimes, thereby enhancing the analytical procedures.

Evaluation of the Analytical Methods Currently Used in B.C. to Analyze Hydraulic Test Data in Bedrock Aquifers This research project involved analyzing hydraulic test data from bedrock wells in British Columbia using various analytical techniques (both radial and linear flow models) to calculate the hydraulic parameters of the aquifers. The objectives of the study were to determine the range of values that are calculated using each method, to identify the most appropriate method of analysis, to estimate the possible error in using radial flow models when linear flow models are more suitable, and to recommend a strategy for analyzing test data from fractured rocks.

Groundwater Flow in Mountainous Regions

Characterizing Groundwater - Surface Water Interactions within a Mountain to Ocean Watershed, Lake Cowichan, British Columbia ( M.Sc., Foster, 2014) Watersheds located within a mountain to coast physiographic setting have been described as having a highly inter-connected surface water and groundwater environment. The quantification of groundwater-surface water interactions at the watershed scale requires upscaling. This study uses MIKE SHE, a coupled numerical model, to explore the seasonally and spatially dynamic nature of these interactions in the Cowichan Watershed on Vancouver Island, British Columbia, Canada. The hydrostratigraphy of the watershed is constructed using several datasets, including electrical resistivity tomography data. The calibrated model simulates a transition of the Cowichan River from mostly gaining within the valley, to losing stream near the coast where groundwater extraction is focused. Losing and gaining sections correlate with geological substrate. Recharge across the watershed accounts for 17% of precipitation. Climate change is projected to lessen snowpack accumulation in the high alpine and alter timing of snowmelt, resulting in higher spring river discharge and lower summer flows.

Modeling Topographically-Driven Groundwater Flow in Mountains (PhD, Neilson-Welch, 2012) The role of groundwater as a component of mountain water systems is poorly understood due to a lack of data for these often remote and inaccessible high relief regions. Yet, two of the main processes that act to replenish water resources for populated valleys, baseflow (BF) and mountain block recharge (MBR), are a direct result of mountain groundwater flow. The research conducted for this thesis involved numerical modelling of 2-D and 3-D domains representing simple generic systems, as well as real topographic scenarios, to investigate the influence of topography on deep groundwater (DG) flow patterns, and the sensitivity of these patterns to uncertainty in recharge (R) and bedrock hydraulic conductivity (K). Regional-scale topographically-driven DG flow patterns that develop in mountain bedrock are reflective of prominent topographic features comprising the mountain block which, for linear-ridge mountain ranges, consist of watershed-bounding ridges, deeply incised primary stream valleys, and triangular facet areas at the mountain front. DG contributing areas for BF at primary stream valleys are generally close to watershed boundaries, and contributing areas for MBR are generally coincident with triangular facet areas. At the regional (~ 10 – 100 km) and subcatchment (~ 500 m to 2 km) scales, the fundamental characteristics of topographicallydriven DG flow patterns are dominant despite smaller-scale variations in topography, and uncertainty in recharge and sub-catchment K heterogeneity. At the local scale, topographically-driven DG flow patterns vary due to uniqueness and complexity of topography, recharge, spatial and/or temporal variability, and K heterogeneities, indicating the importance of understanding these factors for a detailed understanding of DG flow systems. DG flow patterns and boundaries are influenced by the topography of a given catchment as well as surrounding topography and regional topographic setting, and therefore, analysis of DG flow systems and boundaries must consider areas outside the catchment of interest. The results of this study support the development of topography-based predictive tools for conceptualizing DG flow patterns and BF/MBR contributing areas. Numerical modelling using topographic data, which are often the only available data for mountainous areas, provides a convenient approach for preliminary characterization of topographically-driven DG flow patterns in mountains on a site-specific basis.

Modeling Deep Groundwater Flow Through Fractured Bedrock in a Mountainous Headwater Catchment using a Coupled Surface Water - Groundwater Model, Okanagan Basin, British Columbia (PhD, Voeckler, 2012) Quantifying recharge to the mountain block from headwater catchments in snowmelt dominated upland mountainous regions is an important aspect of hydrologic studies. This study contributes to understanding of the interaction between surface water, soil water and deep groundwater flow in headwater catchments. A novel approach was developed for estimating the bedrock hydraulic conductivity of a regional-scale fractured bedrock aquifer using discrete fracture network (DFN) modeling. The methodology was tested in the mountainous Okanagan Basin, British Columbia, Canada. Discrete fractures were mapped in outcrops, and larger-scale fracture zones (corresponding to lineaments) were mapped from orthophotos and LANDSAT imagery. Outcrop fracture data were used to generate DFN models for estimating hydraulic conductivity for the fractured matrix (Km). The mountain block hydraulic conductivity (Kmb) was estimated using larger-scale DFN models. Simulated Km and Kmb values range from 10-8 to 10-7 m/s, are consistent with estimates from regional modeling studies, and are greatest in a N-S direction, coinciding with the main strike direction of Okanagan Valley Fault Zone. Kmb values also decrease away from the fault, consistent with the decrease in lineament density. Simulated hydraulic conductivity values also compare well with those estimated from pumping tests. The estimates of Kmb were then used to represent the deep bedrock in a coupled surface water - groundwater model using MIKE SHE for the Upper Penticton Creek 241 headwater catchment in the Okanagan Basin. Although highly uncertain due to parameter uncertainty and calibration error, recharge to deep groundwater was ~4% of the annual water budget. An specified outward flux from the catchment boundary, representing ~6% of annual water budget, did not significantly impact streamflow calibration, indicating that such deep groundwater losses from the catchment can be accommodated in a model. This outflow may contribute to cross-catchment flow and, ultimately, to groundwater inflow to lower elevation catchments in the mountain block. The modeling exercise is one of the first in catchment hydrology modeling within steep mountainous terrain in which the lower boundary of the model is not treated as impermeable, and in which recharge to the deep bedrock and discharge to the surrounding mountain block were estimated.

Regional-Scale Groundwater Flow Models for the Kelowna Area and the Mission Creek Watershed, and the BX Creek Watershed near Vernon, Okanagan Basin (Brian Smerdon, Post-doc, 2009) Two regional-scale groundwater flow models were constructed for the Mission Creek and BX Creek Watersheds to assess the interconnection and related water balance between broad-scale hydrogeologic units, including upland bedrock and major aquifers. The approach provides a first-order approximation of the groundwater flow system between the upland recharge areas and valley-bottom aquifers, including average groundwater flow rates through bedrock and alluvial aquifers.

CO2 Sequestration

Predicting the geochemical effects of SO2 impurities during carbon storage: Batch experiments and reaction path geochemical modelling (Anja Frank, M.Sc. 2015) The objective of this study was to improve our ability to predict CO2-SO2 geologic storage. SO2 is an impurity of industrial CO2 gas streams which is expected to intensify brine acidification resulting in enhanced mineral reaction. Short-term H2SO4-brine-rock experiments were combined with reaction path modelling to identify reactions and evaluate the pH and temperature dependency of reaction rates. In addition, available reactive surface area was investigated to enhance our ability to upscale to reservoir scale. Kinetically controlled reaction path models that included CO2, SO2 and O2 were generated and then run at reservoir conditions for 100 y. The models predicted a rapid buffering of the SO2 induced acidification. Compared to pure CO2 storage the CO2-SO2-O2 reservoir models resulted in enhanced carbonate reaction extents and a greater porosity increase, which have significant ramifications for the safety of the seal and the storage capacity of the storage formation.

Multiphase flow and reactive transport modelling of CO2 storage in heterogeneous reservoirs (Hermanson, M.Sc. 2013) This study addresses how physical heterogeneity, representing different sedimentary rock layers and the relationships between those layers, impacts the distribution of CO2, and thus the type and extent of mineral dissolution and precipitation reactions during CO2 geologic storage in deep saline aquifers. Numerical multiphase flow (TOUGH2) and reactive transport codes (TOUGHREACT) were used to construct a series of reservoir scale simulations to investigate how the flow controlling parameter values, distribution, and grid refinement of various hydrostratigraphic units (HSUs) affect the distribution of CO2, pH and mineral reactions. Physical heterogeneity is critical for controlling the distribution of supercritical and dissolved CO2, the redistribution of ions from geochemically reactive materials to more stable portions of the reservoir, mixing and dilution of CO2-rich waters, and the extent of mineral dissolution and precipitation reactions. The highest magnitude of carbonate mineral precipitation occurs at the sandstone-siltstone interface and along the extent of the CO2-water contact.

Geothermal Energy

Influence of Aquifer Heterogeneity on the Design and Modelling of Aquifer Thermal Energy Storage (ATES) Systems (Bridger, M.Sc. 2006). A modelling study was carried out to evaluate the influence of aquifer heterogeneity, as represented by geologic layering, on heat transport and storage in aquifer thermal energy storage (ATES) systems. An existing ATES system installed within a heterogeneous aquifer system in Agassiz , British Columbia , Canada was used as a case study. Two 3D heat transport models of the study site were developed and calibrated using the heat transport code FEFLOW including: a “simple” model domain with uniform hydraulic and thermal properties (no layering); and, a “complex” aquifer domain with variable hydraulic and thermal properties assigned to discrete layers to represent aquifer heterogeneity. Comparison of simulation results indicated heat transport in higher permeability layers was significant. Effects of heterogeneity on thermal energy storage and recoverability were not observed. Heat transport in the aquifer was determined to be more sensitive to properties and boundary conditions which influence convective heat transport.

Other Projects

1. Snowmelt and Soil Thaw Energy in Sub-alpine Tundra, Wolf Creek , Yukon Territory , Canada (Shirazi, M.Sc., Geography, SFU, 2006).
2. Observation Well Testing and Recharge Characterization of the Okanagan Basin , BC (Liskop, B.Sc. Honours, 2004).
3. Hydrogeological Assessment of the Belcarra Aquifer (Holt, B.Sc. Honours, 2004).
4. Development of a Hydrostratigraphic Model and Data Integration Strategies for Groundwater Management in the Abbotsford - Sumas Aquifer, B.C. / Washington State (Deshpande, M.Sc, 2004.)
5. Constraining Aquifer Architecture with Electrical Resistivity Imaging in a Fractured Hydrogeological Setting (Rayner, M.Sc., University of Calgary , 2004).
6. Modelling Fluid Flow and Drug Diffusion Through the Stratum Corneum (upper skin layer). (Marquez-Lago, M.Sc. Applied Mathematics, SFU, 2002)
7. Investigation of the Shallow Groundwater Regime in a Small Alluvial Valley, Cheakamus River, BC (Jordan-Knox, M.Sc.)
8. The Characterization of Gentle over Steep Slopes in BC Forest Terrain (Paddington, M.Sc. with D. Stead)
9. Integrating high resolution sequence stratigraphy and ichnology with petrophysical data (Hobbs , M.Sc.; Lerette, M.Sc. with J. MacEachern)