NWT RESEARCH DATABASE

Regions: Dehcho Region, North Slave Region

Tags: climate change, discontinuous permafrost, groundwater, stream flow, Scotty Creek

Objective(s): To assess the spatial and temporal patterns of hydrologic change (groundwater and surface water) in the La Martre River and Trout River catchments and to identify the drivers, processes, and feedbacks responsible for the hydrologic changes identified and how they differ between the study catchments.

Project Description: This licence has been issued for the scientific research application No. 5846. Objective 1 (O1): Assess the spatial and temporal patterns of hydrologic change (groundwater and surface water) in the La Martre River and Trout River catchments. Objective 2 (O2): Identify the drivers, processes, and feedbacks responsible for the hydrologic changes identified in O1 and how they differ between the study catchments. Objective 3 (O3): Predict land cover types where hydrologic changes are most likely to occur in the Study Region. Objective 4 (O4): Provide initial estimates of the rate and magnitude of change in streamflow for the Trout River and La Martre Rivers under different climate warming scenarios until 2100. The primary research question is: How is permafrost thaw changing landscape runoff and groundwater interactions with streamflow at the catchment scale in discontinuous permafrost regions? Sub-questions will be answered in three phases that tie to each objective: 1) What are the spatial and temporal patterns of hydrologic change in the Study Region? (O1) 2) What are the drivers, processes, and feedbacks responsible for the changing hydrologic patterns? (O2) 3) Where, when, and how should we expect future hydrological changes to occur? (O3 and O4) Overview: The project will be carried out in three phases as guided by the objectives outlined in Section 4: 1) a region-wide assessment of hydrological change through timeseries analysis and remote sensing for the La Martre and Trout River catchments; 2) field studies near Whatì and at SCRS to improve process understanding (O2); and 3) hydrologic model development to improve prediction (O3 and O4). O1 will be conducted during year 1 (June 2022–2023), with updated maps following ground-truthing during O2 field investigations. Community mapping workshops/meetings and preliminary O2 field investigations will begin summer 2022. Most of the field work will occur in year 2 with the goal of sustaining long-term data collection beyond the 3-year project. O3 modelling will be initiated early in year 2 using findings from the mapping work and existing insights from previous studies at SCRS. Site-specific data collected will then be used to refine HRU models and build catchment models for early projections of streamflow under different climate change scenarios. Findings will be discussed with community collaborators before the submission of final results. O1: Masters Student (MSc) #1 will lead a time-series analysis using existing streamflow (hydrometric) and climate data for the La Martre and Trout River catchments to identify temporal trends and assess how each basin hydrograph has changed over the last half century. This will build on the hydrometric analysis conducted in the Groundwater Supply Vulnerability Assessment for Whatì. Daily streamflow has been recorded by the Water Survey of Canada since 1969 for the Trout River and since 1975 for the La Martre River. Historical weather data for the catchments will be estimated from the Government of Canada’s Statistically Downscaled Climate Scenarios (10 km2 resolution) and Global Climate Model Scenarios (~10,000 km2 resolution, or roughly catchment size). To assess spatial landscape changes over time, historical air photos and satellite imagery dating back to the 1950s will be compiled and sourced from open access databases, the NWT Centre for Geomatics (NWTCG), and from the NASA ABoVE project. This work will leverage existing data products developed by NWTCG, such as the Landsat Long-term Change Detection dataset and the Inventory of Landscape Change. For example, the Inventory of Landscape Change can be compared to periods of time in the hydrometric record to determine if disturbances such as large wildfires have a statistical influence on large-scale basin response. Insight from previous plot-scale field hydrology studies at SCRS will be used to interpret results from the Landsat Long-term Change Detection map and apply this information to new locations in the study catchments. Key indicators for groundwater recharge and discharge will be mapped, including groundwater springs, icings, and possibly indicator plant species like young tamarack trees (Larch). Since tamaracks require mineral-rich waters, new growth of this tree (as observed in thawed bogs at SCRS) may indicate relatively recent groundwater activation. Various statistical and geospatial analyses will be conducted with the spatial data and streamflow data to assess possible relationships. Changes in groundwater indicators and interpretations of the Long-term Change Detection map will be supported by Traditional Knowledge and community land user observations through mapping workshops and meetings described in Section 7. O2: The spatial datasets and analysis from O1 will aid in developing an inventory of HRUs, which are land areas that have similar land cover, topography, subsurface geology, etc. and therefore have similar hydrological behaviour. These units can then be represented in hydrologic modelling platforms like Raven and form the building blocks for larger catchment- and regional-scale models. A catchment may be modelled as a single, “lumped” HRU, or could consist of tens, hundreds, or thousands of HRUs depending on watershed heterogeneity and model complexity. HRUs are commonly divided up based on land cover (e.g. wetlands, grasslands, forest, open water etc), elevation, and/or aspect. The first step in developing a model for this project is to delineate representative HRUs and ensure that our understanding of their hydrologic functioning and processes are accurate. To do this, satellite imagery, spatial analysis, and community workshops from O1 will help identify which HRUs are most common in the Study Region, and/or which are most sensitive to change. MSc #1 and MSc #2 will work to identify and instrument sub-catchments that represent HRUs of interest. Representation of hydrological processes for some HRUs in the Trout River catchment will be obtained from fieldwork at SCRS since the HRUs found in the Scotty Creek catchment are similar to the nearby Trout River. It also allows for the maximum utilization of an extensive network of pre-existing scientific instrumentation, equipment, and knowledge. Such knowledge is particularly important for applying the NWT’s Long-term Change Detection dataset/maps to the two study catchments. Sub-catchments that represent key HRUs (2-3) will be gauged and instrumented with water level sensors to monitor stream discharge. Groundwater characterization will be guided by the techniques outlined in the recent report prepared for GNWT department of Environment and Natural Resource (ENR) entitled “Hydrogeological Site Characterization Methods for Discontinuous Permafrost Terrain.” Thermal imagery will help identify zones of groundwater discharge for targeted shallow well installation and groundwater sampling. Where the ground is suitable for hand augering and the water table is shallow enough, slotted PVC piezometers will be hand-installed for groundwater level monitoring. For harder ground and/or deeper monitoring (up to 7 m), drive-point piezometers will be installed using a manual slide hammer. Temperature will be monitored in surface water, groundwater, subsurface soils, and at the ground surface. Subsurface characterization and permafrost detection will be completed using geophysical methods such as electrical resistivity tomography (ERT). Surface water and groundwater will be sampled by following the Standardized Water Sampling Instructions developed by GNWT Taiga laboratory and following relevant Community-Based Water Monitoring (CBM) Protocols. Water samples will be analyzed for stable water isotopes at Queen’s University, Tritium at the University of Waterloo, and major ions will be analyzed in Taiga Laboratories, Yellowknife. Standard water quality parameters (e.g., temperature., pH, conductivity, etc.) will be measured at the time of sampling with a multi-parameter sonde following GNWT-ENR and CBM protocols. Deep groundwater samples will be sourced from the community well in Whatì prior to water treatment. A small number of groundwater and surface water samples have been analyzed for water isotopes and major ions as part of the Groundwater Supply Vulnerability Assessment in Whatì. This project will build on this to extend the period of observation and include seasonal sampling. Geochemical data will support hydrometric data in determining relative landscape runoff and groundwater contributions to streamflow and refining the water balance. This will also build our conceptual understanding of the hydrologic system which can be tested using hydrologic models built in O3. For O3 and O4, MSc #2 will use spatial data from O1 and field data collected in O2 to calibrate model structure and parameters in the study catchments using existing physically-based numerical modelling platforms (e.g., Raven, MESH, HydroGeoSphere, FEFLOW). Raven will primarily be used as it is a robust and physically-based hydrological modelling platform and is currently used by project collaborators and by WMMD personnel to model hydrologic changes associated with permafrost thaw. A similar model structure used by Brown and Craig (2020) for the nearby Liard River catchment will be applied here. Some HRUs from the Liard River may be transferrable to this project which will leverage existing work and allow for funds to be allocated toward understanding understudied HRUs. Since MSc #2 will participate in the field component of this project, an iterative modelling approach can be undertaken. This involves building simple models of HRUs forced with and compared to available field data. If there is a gap in field data or in our understanding of underlying hydrologic processes, additional data will be collected to improve model performance and prediction. For HRUs where physical conditions of the subsurface (e.g. porosity, hydraulic conductivity, etc) are unknown, Ostrich, a model-independent optimization and parameter estimation tool will be used. Once HRUs have been tested, they will be built into catchment-scale models of the Trout and La Martre Rivers, which will be calibrated with historic data, and forced with climate data from future climate scenarios. Climate projections will consider high, medium, and low emission scenarios until 2100. This data will be sourced from the Government of Canada’s Statistically Downscaled Climate Scenarios (10 km2 resolution) and Global Climate Model Scenarios (~10,000 km2 resolution, or roughly catchment size). Model results will provide a first attempt at estimating changes to streamflow in the study catchments. However, the iterative modelling process described here will be necessary for the catchment models beyond the 3-year project as new process understanding evolves and data collection expands. Nevertheless, the field investigation and modelling of HRUs will lay the foundation for future work to refine and improve model projections of streamflow for enhanced water resources management. Field data and numerical simulations will be built into a user-friendly data visualization tool that communities can use to view project data and model simulation results. Throughout the project, results will be presented in workshops, training sessions, and meetings. All data collected during the project will be co-owned by the communities. Project leads and community collaborators will coordinate the co-presentation of results at local meetings including the NWT Environmental Monitoring Results Workshop. Community collaborators will be fully supported with the necessary materials and information to present results at relevant local and regional meetings. Results will also be synthesized in annual reports and plain language summaries that will be made available in English, Dene Zhatie´, and Tli?cho Yatiì. The SCRS will be used as a land-based location for community members to gather and to facilitate cross-community discussions. The co-hosted on-the-land camp and activities will be key mechanisms for knowledge mobilization. Data and digital maps will be available through the NWT Discovery Portal and the Mackenzie Datastream and will also be sent directly to community collaborators. GIS data files will be transferred directly to Tli?cho Government GIS specialists in the Department of Culture and Lands Protection. Social media platforms will be used to disseminate co-produced materials such as videos, images, updates, and reports to communities and public audiences. This project will work towards the development of a hydro-climate dashboard for data visualization. The dashboard will be developed using the R Shiny package to provide visualization of hydrometric and climate data being collected in this study. This product will be developed using the R statistical programming environment but produces an interactive output with clean data visualization. The GNWT’s COVID-19 Dashboard is an example of a product developed using the R Shiny package. The initial interface will include access to all data collected in this project but could be expanded to view data from Environment and Climate Change Canada (ECCC) that are collected in real time. The project team will also explore the possibility of building in results from the Raven streamflow simulations and climate scenarios into the user-friendly data visualization tool. Training on the tool will be provided as part of the community workshops and supported beyond the 3-year project. The fieldwork for this study will be conducted from: February 22 - October 31, 2024