Scientific Team:
Cat. No.: En14-42/2011E-PDF
ISBN 978-1-100-18471-5
Information contained in this publication or product may be reproduced, in part or in whole, and by any means, for personal or public non-commercial purposes, without charge or further permission, unless otherwise specified.
You are asked to:
Commercial reproduction and distribution is prohibited except with written permission from the Government of Canada's copyright administrator, Public Works and Government Services of Canada (PWGSC). For more information, please contact PWGSC at 613-996-6886 or at droitdauteur.copyright@tpsgc-pwgsc.gc.ca.
Cover photo credits: © Photos.com – 2011, Environment Canada
© Her Majesty the Queen in Right of Canada, represented by the Minister of the Environment, 2011
On the 16th December, 2010 the Federal Oil Sands Advisory Panel presented its report to the federal Environment Minister. They reviewed current monitoring activities in the lower Athabasca River system, identified key shortcomings, and provided recommendations on what would constitute a world-class monitoring program for the oil sands region (http://www.ec.gc.ca/pollution/default.asp?lang=En&n=E9ABC93B-1).
In response to this report and other concerns, the Federal Minister of the Environment committed Environment Canada to lead, in collaboration with the Government of Alberta, the development of a preliminary surface water quality monitoring plan for the lower Athabasca River and tributaries, and to complete this task within 90 days. Subsequent steps would expand the plan to other environmental media like air and biodiversity and would ensure the media-specific plans are integrated into a single, holistic, ecosystem-based approach.
The monitoring plan presented in this document is a first step towards a comprehensive, integrated monitoring program for the oil sands region. This Phase 1 plan deals with surface water quality monitoring in the mainstem of the Athabasca River, and its major tributaries, between Fort McMurray to Wood Buffalo National Park Boundary. It focuses on the physical and chemical attributes of water quality. Subsequent phases in the development of the water quality monitoring program will include biological endpoints and “effects-based” monitoring and assessments, and the expansion of the geographic scope to include environments further downstream in the Athabasca River system where appropriate, and upland lakes in the air shed that could be affected by aerial contaminant deposition from oil sands development.
Contributors to Phase 1 plan development had expertise in: surface and groundwater quality and quantity, hydrology, climatology, environmental chemistry, paleo-limnology, aerial deposition and contaminants, oil sands related contaminant chemistry, ecosystem-based monitoring program design and cumulative effects assessment, and statistical design. Each contributed significantly to the development of the Phase 1 plan and the production of this document.
This document presents a technical monitoring plan about when, where, why and how to monitor surface water quality. It does not present a complete monitoring system that includes specific elements like data management, and reporting. In the implementation of this plan, key principles articulated by the Federal Oil Sands Advisory Panel will be followed. The full plan will be holistic and comprehensive. It will be adaptive and robust. All data will be publically available in a standardized, accessible format. Stakeholders will be involved and their interests and concerns fully taken into account as data are analyzed and the meaning and impacts of conclusions are determined.
As with all responsible science plans, and in keeping with the guidance of the Federal Panel, this plan will be continuously updated and refined to ensure its relevance and technical competence.
This plan presents a marked change in approach to oil sands monitoring. Although the individuals involved in developing this Plan had solid technical backgrounds and most have experience in the oil sands region, this plan can be improved by incorporation of traditional knowledge.
The proposed water quality monitoring program is ecosystem-based, and focuses on the Lower Athabasca mainstem and its tributaries.
It is designed to be adaptable and has the ability to expand and contract its spatial and temporal coverage, and revise its procedures and protocols to address new and emerging issues as they arise.
The monitoring plan presented in this report is based on the principles recommended by the Federal Oil Sands Advisory Panel and is the first of several phases in the design of an integrated, comprehensive and holistic monitoring program for the Lower Athabasca system.
The purpose of the monitoring program is to obtain scientifically credible information that will allow:
Environmental monitoring and assessment programs must be designed to provide data that can be used to answer specific questions.
Over the years, a number of consultative processes have occurred within the lower Athabasca River basin (e.g., the Northern River Basins Study (1991-1996), Northern Rivers Ecosystem Initiative (1998-2004)). Key stakeholder questions have emerged and been identified as pertinent to understanding the impacts of the oil sands and other industrial developments on water quantity, quality and ecosystem health in the Athabasca basin. Box 1 provides examples of the nature and scope of environmental concerns and key questions that have been raised by First Nations, basin residents, government agencies and NGOs.
This monitoring plan is designed to provide comprehensive data that can be used to answer a full suite of key monitoring questions. Specific questions can be answered, or hypotheses tested, by performing targeted analyses on sub-sets of data collected under this monitoring plan.
Stakeholder issues and concerns must be revisited on a regular basis to ensure appropriate focus.
Box 1: Examples of key questions to be answered by the monitoring program:
The Federal Minister of the Environment committed to have a preliminary water quality monitoring plan within 90 days after the tabling of the Federal Oil Sands Advisory Panel report in December 2010. Given the scientific complexities involved with adequately addressing the geographic scope, and associated environmental stressor- and effects-based monitoring program design components, the water quality monitoring program is being developed in several integrated phases that chronologically and progressively build upon one another (Figure 1).
Figure 1. Components of a monitoring program design for the Lower Athabasca system.
This report (Phase 1) is concerned with the physical (hydrological and climate) and chemical components and stressors of the system. The primary goal of this technical plan is to present a comprehensive and integrated approach that quantifies and assesses the sources, transport, loadings, fate, and types of oil sands and other industrial and municipal contaminants into the Athabasca River system. Phase 1 is targeted at obtaining a better spatial and temporal quantitative understanding of the key physical/chemical “stressors” affecting the system and also improving knowledge on historical baseline conditions.
Phase 2 will address the “so what” question by identifying key biological and ecological endpoints that will be monitored and used to assess local and regional impacts, including cumulative effects. The development of this phase will provide data that can be used to answer overarching questions such as, whether fish have changed in ways that make them undesirable (e.g. malformations), whether fish and/or other aquatic organism health has been affected, whether key species have been lost or ecosystem function impaired, and whether current guidelines for individual contaminants are useful for detecting ecosystem impairment?
The development and validation of new integrated modelling and environmental predictions systems that allow for improved assessment and projection of site-specific and regional impacts is also an important component of the design. The “cause-effect” relationships that will be developed through the concurrent implementation of Phases 1 and 2 will be used ultimately to predict cumulative environmental impacts and inform the requirements for subsequent cumulative effects monitoring. This component will be an ongoing and iterative scientific process that continues to improve and advance the modelling frameworks through incorporation of new scientific knowledge obtained from the monitoring programs themselves and other sources.
The Athabasca River system originates from the eastern slopes of the Rocky Mountains in Alberta, flows northeast and drains an area of approximately 160,000 km2, eventually flowing through the Peace-Athabasca Delta and discharging into Lake Athabasca and the Slave River, which crosses into the Northwest Territories. The Peace-Athabasca Delta within the Wood Buffalo National Park is also recognized by the RAMSAR convention on Wetlands and is a UNESCO World-Heritage Site (Wrona et al. 2010).
Figure 2. Alberta Oil Sands Region (source: Wikipedia, created by Norman Einstein, 2006)
Primary environmental concerns regarding water quality, quantity and ecosystem health related to oil sands development are within the lower portions of the Athabasca River, primarily downstream of Fort McMurray, Alberta, although they are potentially farther reaching. This lower portion of the Athabasca River drains an area of approximately 58,000 km2 and includes numerous river tributaries and sub-catchments (Figure 2). In this area, the Athabasca River and tributaries flow through the Athabasca oil sands deposits, eroding the oil-containing sands and interacting with groundwater systems also in contact with the oil formations.
The Athabasca River Basin is located in a constantly changing and dynamic landscape that has seen significant rates and magnitudes of change in cumulative land use and industrial development in recent years (e.g., Suncor in 1967, Syncrude in 1978, major expansion in 1998 with Suncor Steepbank, the early 2000s with Shell MRM, Syncrude Aurora, Suncor Millennium and, proposed developments such as the Joslyn mine.
Similar to other existing and emerging energy sources, oil sands development has multiple potential environmental impacts on the air, water and land related to stack and particulate emissions (e.g., mine dust, coke stockpile, etc.), mine fleet and greenhouse gas emissions, water use (both surface and groundwater), production of waste streams including tailings ponds and associated contaminants, potential groundwater contamination, land disturbance and associated alterations in hydrological connectivity, and habitat loss and fragmentation.
The monitoring program must also be designed to inform pertinent legislative and regulatory action. In this context, Box 2 summarizes the primary federal and provincial legislative authorities that the water quality monitoring system can inform. Government representatives were queried for their respective surface water quality monitoring data needs. These data needs are accommodated in this monitoring plan.
BOX 2: Key Federal and Provincial Legislative Authorities
Government of Canada - Legislative Authorities
Province of Alberta - Legislative Authorities
There is monitoring currently conducted in the lower Athabasca River region. The Province of Alberta and the Government of Canada have monitoring stations, industry monitors as a condition of their license or approval to operate, and additional monitoring is performed by the industry-funded Regional Aquatic Monitoring Program (RAMP). However, there is a lack of integration of these activities. This lack of integration has been reflected in a number of independent science reviews or journal papers (e.g., Timoney and Lee 2009; Kelly et al. 2009, 2010; Giesy et al. 2010; Schindler 2010) and expert reviews conducted by the Royal Society of Canada (RSC 2010), the RAMP program (RAMP 2004, 2011), the Federal Oil sands Advisory Panel (2010); and, the Alberta Water Monitoring Data Review Committee – Final Report (2011)). Key criticisms of the RAMP monitoring design are that it:
Collectively, the above reviews found that the current monitoring system did not deliver data of sufficient quantity or quality to detect or quantify the effects of oil sands development. In addition, current oil sands monitoring of air, groundwater and surface water have not been integrated in a source, transport and fate construct.
It is important to note that, although widely criticized and in need of redesign, past monitoring activities have yielded data at some sites and during some time periods that will be critical to the optimal functioning of the new monitoring plan. The goal of this new plan is to incorporate the sound components of existing monitoring and improve data collection activities to facilitate holistic assessment of contaminant sources, their transport in the environment and their ultimate fate. With this, risk-based assessments to aquatic and human health can be made.
Phase 1 deals with surface water quality monitoring in the main stem of the Athabasca River, and its major tributaries, between Fort McMurray to the Wood Buffalo National Park Boundary. Phase 1 targets sampling designed to quantify fluxes of contaminants from point and non-point sources.
Phase 1 deals with surface water quality monitoring in the mainstem of the Athabasca River, and its major tributaries, between Fort McMurray to Wood Buffalo National Park Boundary. Phase 2 will expand the geographic scope to include the Peace-Athabasca Delta, Lake Athabasca, the Slave River systems, and upland lakes in the region that could be affected by aerial deposition.
Current and projected environmental pressures stemming from oil sands development on the lower Athabasca River are a culmination of historical land use alterations, and enhanced levels of industrialization, coupled with broader environmental changes linked to altered hydrological conditions arising from a warming climate and enhanced climatic variability.
Figure 3 summarizes various point and non-point source contaminant pathways entering the Lower Athabasca river system. Each of these pressures has direct and indirect consequences on water resources and there are many interactions among them at local and regional scales.
Thus, an appropriately designed and adaptive integrated, multi-media monitoring and research program is required to understand, predict and report on the status and trends of water quality and quantity, accumulated state, changes in ecosystem structure, function and health and ultimately determine cumulative impacts. This document gives the monitoring plan for a portion of the Athabasca mainstem and its tributaries as a first step in the development of such an integrated monitoring plan.
Figure 3. Key Direct and Indirect Contaminant Pathways into the Lower Athabasca River
The design and successful implementation of an effective, water monitoring program requires several core elements, including:
The design and successful implementation of an effective, monitoring program requires several core elements. Below we describe each of these elements and how they are incorporated in the new monitoring program design.
Data collection and data management are currently fragmented. Strategic decisions for environmental protection (including water quality) and industry sustainability cannot be made under such conditions. An integrated regional monitoring framework needs to be implemented where the various types and sources of monitoring are linked to provide the information required to inform decision-making.
The new integrated regional oil sands monitoring program will consist of a series of inter-related types of monitoring that have specific objectives. There is a plethora of monitoring definitions, many of which are interchangeable. The definitions used in this document are:
Baseline Monitoring quantifies background levels of physical and chemical parameters at locations that are least-developed or ideally “non-impacted” by anthropogenic disturbance. Baseline monitoring is critical as it establishes the benchmark to compare to sites potentially impacted by development. In previous monitoring in the oil sands region, very few fixed-location, long-term monitoring sites existed (those that do, are operated by government agencies and primarily located outside of the active oil sands mining region), and many sites that did have pre-development information were either discontinued or gradually absorbed by an expanding industrial development footprint. The new plan will increase the number of long-term monitoring stations at appropriate spatial and temporal scales in the catchment to assess regional baseline.
Still, defining “baseline” or “reference” conditions in a region that has undergone, and is continuing to undergo, change in its environmental regime poses significant challenges. Very few regions of the basin, if any, are now untouched by some form of direct or indirect anthropogenic influence (e.g., aerial deposition of contaminants). In addition, the richness of the bitumen ore body is heterogeneous, as are the areas of natural exposure or contact with streams and rivers, further complicating a “control/impact” approach to impact assessment.
To address these challenges, the new monitoring framework will employ new approaches that will integrate synoptic information from long-term monitoring locations, with modelling and other predictive tools to assess changes in the environmental state. A key requirement to the enhancement of our understanding of regional and localized “baseline” conditions, will be to ensure that there will be adequate pre-development, site-specific monitoring data, against which environmental change can be measured, and that post-development monitoring will continue for the long-term. Consideration could also given to establishing additional, long-term reference areas/sites within the oil sands region that could be used to further define “baseline” or “reference” environmental conditions. A complicating factor is that remaining “reference” watersheds within the bitumen-bearing McMurray formation are either larger or smaller than impacted streams, which would limit their ability to function as a “baseline”.
The proposed use of sediment coring and related paleo-analyses may also further help discriminate medium-term (i.e., decadal) from long-term (i.e., multi-decadal) environmental change and variability. Also, existing hard copy and electronic reports will be mined for historic data and archived in the monitoring system database.
Accumulated State Monitoring measures changes in the aquatic environment that are outside of both local and regional baseline. It identifies “hot spots” of change in space and “hot moments” of change over time. It is conducted at local and regional scales relative to local and regional baseline. This monitoring requires use of baseline monitoring data and identifies changes outside of baseline, their magnitude and direction. It is fundamental monitoring for management decisions. The first component of accumulated state monitoring is also commonly called effects-based monitoring. The effects monitoring program operates consistent with the baseline monitoring suite of parameters.
Accumulated state monitoring also measures changes on the landscape occurring over space and time. This includes quantification of changes in land use as well as synthesis of point and non-point source loadings. The second component of accumulated state monitoring is also commonly called stressor-based monitoring.
Compliance Monitoring involves facilities monitoring of specific parameters to determine whether the site is in compliance to its operating approvals and related licensing/permitting conditions. This monitoring is currently conducted in the oil sands, although not integrated into a regional monitoring program. This information potentially contributes to cumulative effects and must become available, transparent, and integrated with electronic reporting mechanisms in a timeframe that aligns with the reporting requirements of the integrated regional program.
Performance Monitoring is site/facility-specific and would be conducted after development has occurred. This type of monitoring would be used, for instance, to verify and/or validate whether predictions made through the Environmental Impact Assessment (EIA) process were accurate. Currently, little performance monitoring is conducted and the indicators and parameters used during the EIA process have little to no connection to the local accumulated state and effects monitoring discussed above. It is critical that performance monitoring be conducted or there will be no mechanism to improve ability to predict impacts of specific developments or to identify whether EIA predictions were accurate.
Cumulative Effects Monitoring is the largest and most integrated of monitoring components. It is fundamentally the most critical for managing regional development-related change. Cumulative effects monitoring involves comparison between the predicted ecosystem trajectory towards the future state of the system against the observed conditions and trajectory. It is implemented at a regional scale following a cumulative effects assessment, but is not currently conducted in an effective manner in the oil sands region.
Surveillance Monitoring is generally hypothesis driven and addresses specific knowledge gaps about system processes or attempts to ascertain a mechanistic understanding of causality. This type of monitoring is often both time (short-duration) and spatially limited (sites and attributes being measured follow an explicit experimental design). An example of surveillance monitoring is the use of “synoptic surveys”.
Currently there is no consistent reporting in a format needed to track environmental change, track industrial development on the landscape, and understand existing or future threats to environmental protection (including water quality). Regional monitoring must produce three Core Results on a consistent and ongoing basis:
In the absence of these Core Results, cumulative change cannot be detected, predicted, managed, or mitigated.
Oil sands monitoring to date has functioned as a series of independent monitoring programs conducted for specific and independent reasons. Where it has failed, is in integrating and converting the sum total of the data being generated to a knowledge base to support approaches to dealing with cumulative effects.
Figure 4. A “world-class” monitoring program measures how much environmental change has already occurred (Assessment of accumulated environmental state), as well as helps define what caused the change (Relationships between system drivers and environmental response), and uses this information to plan the desired landscape for the future by predicting the outcomes of different development trajectories (Cumulative Effects Predictive Modelling).
A regional monitoring framework provides a consistent regional approach in terms of sampling strategy, endpoints, and protocols, such that monitoring efforts become more coordinated, approaches become standardized for related components, and data become regionally available.
Thus, in addition to meeting program-specific objectives, the monitoring programs must be integrated to achieve the three Core Results given above and shown in Figure 4.
Ultimately the goal of having an integrated monitoring framework is to move towards the implementation of an integrated cumulative effects assessment approach. Such an approach reduces the focus from assessing effects (both stressor-specific and cumulative) on a project-specific basis, and instead provides a regional basis for addressing key questions of concern. This provides opportunities for cost-efficiencies, develops synergies by focusing questions, and reduces duplication of effort in existing monitoring requirements. The result is a framework that underpins the ecological assessment components required for Environmental Impact Assessment (EIA), focuses research, provides regional baselines, develops thresholds for responses, and detects cumulative effects.
Cumulative Effects Monitoring then follows cumulative effects assessment and essentially determines if the observed future state followed the expected trajectory. This is a comparison of the expected (or predicted) to the actual response (Figure 4).
Adaptive program design - alterations in spatial coverage and temporal sampling frequency are linked to science-based (and policy-relevant) decision-triggers. A key objective is to maintain required statistical rigour to detect and predict effects and assess trends in environmental parameters. The program design is continually assessed and refined based on stakeholder feedback and new science findings.
It is not necessary to monitor everything, everywhere, all the time. The monitoring system must incorporate a decision-support system that will provide clear direction on when and where monitoring intensity should change based on monitoring results and what actions should be taken if monitoring reveals significant effects to ecosystem health. The decision-triggers will allow opportunities to increase or decrease monitoring frequency, ensuring the best possible information is available for making management decisions and optimizing the efficiency of the monitoring network. In addition, triggers are defined that when a monitored parameter reaches a predetermined threshold of concern, it is inspected to determine if and what types of management actions are warranted.
Within each monitoring activity, a “decision framework” will be developed and incorporated so monitoring efforts and intensities (i.e., tiers) can be adjusted or triggered “on or off” for specific monitoring stations and related data acquisition activities depending on the decision threshold. Triggers will be used to both increase and decrease monitoring intensity and would link back to factors such as the rate of development and restrictions to it, such as reducing airborne emissions.
Conceptually these triggers would occur when there is an observed effect (e.g., a significant statistical change), a warning sign (e.g., a critical effects size is exceeded and is confirmed for an environmental stressor), or a response sign (e.g., a critical threshold or effects size is exceeded and is getting worse). Hence the monitoring program will be adaptive with tiers and triggers that move, focus and optimize monitoring efforts throughout the basin, according to needs (Table 1).
Table 1. Examples of decision “Triggers” that would be place to adjust monitoring program design, sampling intensity, in a tiered, hierarchical manner. In this manner the monitoring programs would adapt to information needs.
Level
Trigger
Consequence
Effect
Significant statistical change
Seek confirmation
Warning Sign
Exceeds critical effect size, and is confirmed
Increase monitoring frequency to define extent and magnitude of change
Response Sign
Exceeds critical threshold effect size and is getting worse
Investigate cause
Action Level
Passes probable effects level or water quality criterion
Change in management strategy warranted
It is expected that all sites for baseline monitoring can be triggered to lower frequencies or tiers of monitoring, and reduced parameter lists, once the ability to predict relevant levels becomes possible (i.e., if the status at monitoring station 3 can be adequately estimated from data collected at stations 2 and 4, then monitoring at station 3 can be reduced). Such monitoring would be triggered back into operation if appropriate threshold levels were surpassed in subsequent monitoring. This triggering “on and off”, and changes in intensity of monitoring, are essential for the proper operation of an adaptive monitoring program, and function as a first level of defining the extent and magnitude of changes, and as a first step towards localizing the potential impact and identification of causative factors.
Once a predictive model for assessing cumulative effects is developed, it will be possible to track the system’s performance to compare the predicted and observed outcomes. When there is a sufficient signal difference between the predicted and observed state, increased monitoring frequency can be triggered (Table 1).
A comprehensive, integrated monitoring framework and open and transparent data management system are necessary to produce and archive data/information that is used to predict the trajectory(ies) of changes in water quality, quantity and ecosystem health. Comparison of predicted modeled trajectories with reality increases insight into system function and ultimately improves management decisions.
The proposed monitoring program, along with subsequent monitoring plans for other environmental media, must be supported by a highly efficient, integrated database and a systems-based, decision-support framework. Database and decision-support software must be consistent with the format of, and output from, each type of monitoring program in the framework, as well as the integration products (relational analysis, cumulative effects assessment, cumulative effects monitoring). As mentioned above, this aspect of regional monitoring and assessment has been largely ignored, resulting in inefficiencies and a lack of coordination and responsive action to detected change. In the absence of such a facilitation system, this program will not meet its target of being “world-class”.
In addition, there is recognition for the need of strong and explicit relationships between pertinent research, monitoring and modelling activities. Linkages among these activities will ensure better interpretation of monitoring data and maintenance of an optimal monitoring system design. Hence, the production of timely, peer-reviewed scientific publications of key findings must also be an ongoing required deliverable of the program.
Testing and the incorporation of new and emerging monitoring technologies (e.g., new sensor-based integrative samplers, gene-array chips as biomarkers of exposure, remote-sensing and satellite-linked automated sampling networks) should be continually assessed for their applicability to improve data acquisition efficiencies and interpretability in the program.
Proposed design addresses identified scientific and design shortcomings of previous monitoring programs – the monitoring program design builds upon existing programs/sites where deemed appropriate and expands the range of core water chemistry parameters, the spatial coverage and sampling frequency.
Water quality monitoring in large and complex freshwater systems presents many challenges, particularly in cold-regions such as the Athabasca River and its tributaries. Climatic and hydrological conditions along with chemical concentrations can vary substantially over scales of space and time. A plethora of monitoring programs has been implemented in the lower Athabasca river region, and many active sites currently exist (Figure 5).
Figure 5. Summary of existing water and Hydromet monitoring programs/sampling sites in the oil sands region
These sites were selected for program-specific purposes such as, to assess “upstream-downstream” impacts, quantify areas of significant contaminant loading sources, and to assess medium- to long-term trends in water quality at focused locations (e.g., RAMP 2010; Alberta Environment 2010 - Contaminant Load Study); AENV 2009 - Muskeg River Watershed Integrated Water Quality Monitoring Program). However, many of these locations might not have been adequately assessed from the standpoint of biological relevance. In all large systems there will be “hotspots” of damage where fish spawn or feed at the same locations where contaminants are deposited or habitat otherwise destroyed (Choy et al. 2008; Fowlie et al. 2008). Sampling site selection may need refinement based on validation that will occur in Phase 2 which will ensure sites are important not only from a water quality perspective, but also from a biological perspective.
A comprehensive overview of past and current monitoring activities in the oil sands regions is provided by Lindeman et al. (2011).
Recent 2010-11 external science reviews and scientific publications have been unanimous in highlighting that existing monitoring programs have generated a significant amount of information over the past four decades. However, they have also identified primary shortcomings and concerns related to:
The proposed program is designed to improve both the spatial and temporal resolution of the data being collected. This will improve our ability to detect change and predict effects, and adaptively manage for changing environmental conditions. Not only will this increase our power to detect change, but we will be in better position to understand the variability and responses in the system in relation to point and non-point sources, natural versus mined bitumen deposits, and cumulative effects.
Ultimately, the monitoring program must be able to detect contaminant loading and/or concentration trends that exceed key biological thresholds. Since biological thresholds will likely always be more sensitive than chemical guidelines (e.g., Palmer et al. 2010), it will be important to ensure the final design of the program utilizes biological/ecological endpoints as key indicators of stress and impacts. The integration of such endpoints into the current water quality design will occur in Phase 2.
A mass-balance approach was used to define the network of sites to be monitored for key water quality, hydrometric and aerial deposition variables. The approach allows for the quantification and modelling of sources, fate and loadings of contaminants in the Lower Athabasca system.
A mass-balance approach has been used in designing the monitoring network assessing the quantity, movement, and cycling of materials in the watershed. Applying this approach within the lower Athabasca River system requires an expansion of the current water quality and quantity monitoring sites to include a more systematic and comprehensive quantification of the sources, transport, flux, and fate of materials and contaminants entering the watersheds. The major contaminant input sources that must be considered are:
Quantification of oil sands related contaminant loadings and fluxes by necessity requires a more comprehensive network of hydrometric and suspended and bed sediment measurements, better quantification of historical background or “baseline” conditions, and improved estimates of atmospheric contributions. In addition, standardized QA/QC analytical procedures and Standard Operating Protocols (SOPs) must be implemented, along with proper data archiving and reporting.
The proposed monitoring program expands an existing suite of lower Athabasca River monitoring sites of “state” to a formal integrated network. This will allow quantification of the reach-specific and regional-scale accumulated state, and the spatial and temporal status and trends in contaminant loadings. The design builds on Alberta Environment sites that since 2008 have been the foundation for a regional Contaminant Load Study (CLS) and more focused reach-specific efforts in the Muskeg River watershed.
In addition, RAMP sites that have been deemed to have relevant site-specific and spatial and temporal coverage characteristics have also been incorporated. New sites have been proposed to enhance spatial coverage where necessary. The program proposes increased sampling frequency, for both existing and new sites, to address issues of past designs, with respect to temporal variability and statistical power.
Consideration will also be made of adding additional tributary locations as they are important habitat and support 17-24 species of fish, including thousands of migrating fish each year, and have a high diversity of invertebrate taxa (Barton and Wallace 1980; Bond 1980; Machniak et al. 1980). These areas will likely be hotspots for assessing impacts in Phase 2 of Plan design and would complement the physical/chemical network design.
The geographic and geological settings, along with the scope, both in the rate and magnitude of the oil sands development, poses significant challenges in evaluating and attributing observed change in environmental conditions and related impacts on ecological conditions. An ongoing constraint and challenge to understanding and predicting the long-term causes of water quality changes in the oil sands region has been the difficulty identifying appropriate “reference” sites or “baseline” conditions because of the lack of reliable historical records. To address this challenge, the proposed water quality monitoring program incorporates a combination of existing medium- to long-term sites, and establishes an enhanced network of core locations that will serve to better define regional “baseline” or “reference” conditions. Hind-casting approaches will be used to define historical “baseline” conditions. An assessment will be performed of the applicability of all potential hind-casting techniques including paleo-sediment coring and tree ring analyses. Appropriate techniques will be used at strategically selected sites to assess the historical levels and trends in hydrological conditions, general water chemistry and contaminants where applicable.
Within the Lower Athabasca Basin, and in particular in the oil sands region, both “pulse” and “press” environmental stressors, and related impacts, are occurring (Table 2). These affect the temporal variability in observed responses and will influence the spatial and temporal replication necessary to detect impacts with sufficient statistical power (Stewart-Oaten, et al., 1986). The proposed increase in the number of long-term, standardized monitoring sites (each having a common suite of measured environmental parameters along with an increased temporal frequency of sampling) will enhance the statistical power to assess status, trends and impacts. Detecting “press” related environmental impacts (e.g., a regime shift in parameters such as climate, flow or sediment loads) requires a large number of spatially distributed “baseline” or reference sites, while, shorter-term “pulse” impacts (e.g., detection of enhanced contaminant loadings from point source discharges) are best detected when the number of periods sampled is maximized.
TABLE 2. Planned sampling design to detect different types of impacts (after Underwood and Chapman 2001; Bulleri et al. 2008)
Potential Impact to Detect
Design should
Long-term ‘‘press’’ response to disturbance (e.g., slow and long-term changes in contaminant concentrations in water and/or sediments resulting from diffuse, non-point source discharges)
Incorporate numerous control locations (i.e., maximize the number of “baseline” or reference sites)
Shorter-term ‘‘pulse’’ response to disturbance (e.g., rapid or short-term changes in contaminant concentrations in water and/or sediments resulting from site-specific, point-source discharges)
Incorporate numerous periods in sampling (i.e., maximize the number of periods being sampled - seasonality)
Changes in temporal variation or very short-term ‘‘pulse’’ responses to disturbance (e.g., severe storm event and related contaminant flux)
Incorporate as many times of sampling in each period sampled, or increase the number of periods sampled (i.e., maximize periods or frequency of sampling within a period)
The monitoring program is designed around three integrated environmental components (hydrology and sediments, water quality, air deposition), allowing for an improved capability to address hypothesis-driven questions related to current and projected reach-specific and regional impacts of oil sands developments on water quality.
This water quality monitoring plan is structured around three integrated core components:
For each of these components and related monitoring system types, we describe the proposed new spatial and temporal coverage of sampling locations within the overall Athabasca catchment, the Lower Athabasca mainstem reach (LMR) in the vicinity of the oil sands developments, and for key tributaries (Sections 4 to 6). We further provide the rationale behind the proposed design addressing What, Why, Where, When and How, and discuss how the design is an improvement over the existing system(s) (e.g., What information is gained?) (Box 3).
The proposed design utilizes the following schematic representation of the lower Athabasca River system and its major tributaries (Figure 6). The design follows a statistical framework that allows for both reach-specific and cumulative loadings and water quality/quantity to be quantified and characterized with sufficient power. Notably, approximately 55% of the Athabasca River flow exiting the LMR at M9 originates from flow entering the LMR at M0, reinforcing the need to also understand upstream hydro-climatic drivers.
Box 3. Rationale for Monitoring Program Design:
What water quality/quantity parameters/processes need to be sampled
Why - are the chosen parameters important to quantify (linked to data use and issues/concerns to be addressed), and why is the chosen sampling design being used (statistical considerations)
Where - spatial resolution & replication, by sub-catchment/drainage basin, by media (surface water, pore water, sediment)
When - temporal resolution (discrete time vs. event-based vs. continuous)
How - what sampling and analytical methodologies should be used
Figures 6 and 7 illustrate the locations of the proposed water quality monitoring sites in the Lower Athabasca system and Table 3 (see page 57 ) provides detailed descriptions of each site. Appendix A also contains additional related information.
This stratified sampling is a fundamental shift in monitoring approach. By dividing the Athabasca River and tributaries into specific reaches or segments and monitoring in those reaches, quantification and assessment of site-specific loadings and therefore isolation of potential sources on contaminants is possible.
A standardized and enhanced suite of water quality and quantity variables are integrated into the monitoring program design, focusing on environmental parameters relevant to assessing oilsands development impacts.
Figure 6. Schematic representation of proposed sampling sites on the Athabasca River mainstem and major tributaries
Figure 7. Proposed monitoring sites on the Athabasca River mainstem and major tributaries
The hydrological and climatic conditions affecting the overall Athabasca catchment need to be understood, because of the strong control the upper part of the catchment and associated tributaries have on downstream river flow.
Sediment dynamics play an important role in the fate and transport of contaminants in the system.
An understanding of the hydrological and climatic conditions affecting the overall Athabasca catchment is required, because of the strong control the upper part of the catchment has on downstream river flow. In addition, river-ice dynamics play a major role in physical and chemical processes (e.g., sediment transport, mixing regimes, flushing flows, background chemistry) in the LMR (e.g., AENV 2010). Furthermore, knowledge of the controlling hydro-climatic systems (e.g., synoptic climatic patterns and atmospheric teleconnection patterns) is also key to understanding the hydrologic processes affecting the LMR and its tributary catchments (e.g., snow accumulation/runoff, extreme events, etc.).
Information obtained from the hydrometric/sediment monitoring program will address the following questions:
The hydrology and sediment monitoring program, while described separately, is linked to the surface water quality monitoring program (section 5.0 below). Sediment dynamics (i.e. erosion, transport and deposition) can be a significant driver of contaminant dynamics since many contaminants partition strongly to fine sediments, with resulting influences on benthic communities. This linkage allows quantification of sediment/contaminant loadings to downstream environments and permits better understanding and management of sources, fates and effects of these contaminants.
To quantify and characterize the hydrologic fluxes entering the LMR, ongoing analyses need to be conducted of the upstream major source-water hydrologic regimes and types (e.g., nival, pluvial and glacial regimes). Specifically addressed will be the importance of each regime to seasonal flow contributions at M0, their sensitivity to variations in current and future climatic drivers, and potential to generate extreme events within the LMR (e.g., low flows, open-water floods, river-ice breakup surges). The latter type of extreme has previously been identified as a transport mechanism for major fluxes of material beyond that possible under open-water conditions (e.g. Milburn and Prowse 2002; Prowse and Culp 2003; Beltaos et al. 2011). For this, a companion focus on the advancement of, and effects from, ice breakup through the LMR will also be required.
To properly quantify and characterize the hydrological and sediment regimes/budgets within the LMR, the following core components are needed:
Climate parameters will be monitored by an enhanced network of regional meteorological stations. These stations will provide necessary site-specific and synoptic regional information on the climatic drivers of hydrological and landscape-related processes. The meteorological stations will provide finer spatial resolution data for process-based studies (i.e., focused monitoring), and contribute to hydrological and sediment transport model development and validation. In addition, an enhanced network of precipitation/snow gauges will be installed for evaluating meso-scale variability during events such as intense convective rainfall events and those affecting winter snow dynamics (distribution, sublimation, etc.). These stations are part of the meteorological network identified in Section 6.0 for the aerial deposition monitoring program.
Flow (hydrology) has been reasonably quantified over the past 40 years in the Lower Athabaska River (LAR) at Fort McMurray. Hydrologic, hydraulic and river-ice modelling has been performed elsewhere in the Mackenzie basin that could be applied to the LAR (e.g., Beltaos and Prowse 2009; Hicks and Beltaos 2008; Hicks 2009). The RAMP program has provided an expanded spatial hydrometric network in the region; however, enhanced continuous time series (particularly improvements on winter monitoring) are required. The requirements of the Instream Flow Framework for the LAR (AENV 2007) have enhanced winter monitoring of flow at existing Water Survey of Canada sites, however, daily representative monitoring of major tributaries, their primary sub-catchments and the LAR at strategic locations between Fort McMurray and Old Fort are required. This monitoring program requires daily discharge estimates for all major tributaries and the LAR upstream (M0 and M2) and downstream of Fort McMurray (M3), and downstream of the Firebag River near Embarrass airport (M9) on a continuous basis. We further specify a design where flow uncertainty does not exceed 5% during ice cover, or explicitly report uncertainty with flow estimates where a 5% uncertainty cannot be achieved.
Low flows in the LAR occur during late winter and have been identified as potentially having adverse impacts to the aquatic ecosystem (CEMA 2009). These impacts may occur at flows near 100 m3/s, and, with industrial withdrawals managed between 4 and 15 m3/s – accuracy of river flow estimates needs to be within this range. Achieving this level of accuracy in flow estimates under ice is a challenge but can be achieved with advanced monitoring systems. EC and AENV have committed to developing a number of advanced monitoring technologies through space-based telemetry and through-ice Doppler systems. Space-based systems will evaluate the entire LAR at critical junctures (evaluating feasibility of monthly estimates), thereby help identifying access restrictions into critical side-channels and tributary habitats. Doppler systems could be deployed during low flow winter conditions (e.g., when flows are below 120 m3/s) when yellow conditions/thresholds are reached (see AENV 2007).
In order to quantify contaminant mass balance, there is a need for accurate hydrologic mass- balance calculations. This will be achieved through hydrologic modelling using changes in stable isotope ratio (Deuturium & 18O) mass balances, and using improved regional data on relevant hydrological input and outputs such as evaporative flux, precipitation, source ratios. These data will be collected at suitable meteorological stations in the region.
The behavior of waters as they mix in the lower Athabasca river controls the site-specific characteristics and the transport of contaminants downstream. Hydraulics are described by a number of reasonable models and the impact on transport are understood generally through lateral mixing of dissolved or suspended species and Shields equations for transport of sediment-bound compounds. Hydraulic behavior, in the context of understanding contaminant fate and transport, will be better quantified and assessed in this program through monitoring of cohesive and non-cohesive sediment transport. Research programs have already been identified within EC and AENV to address unknowns in the science of cohesive and non-cohesive sediment transport. These programs are essential to the long-term success of the monitoring program.
Mainstem sites provide upstream and downstream boundary conditions for the LMR through the oil sands development region. Where possible, the sites were selected to include (active) hydrometric stations with historical hydrologic data (river flow; suspended sediment data). In addition, sites were co-located where possible with other existing medium to long-term monitoring activities (e.g., AENV LTRM Water Quality Monitoring Sites; AENV Contaminant Load Monitoring Sites, RAMP sites) to build upon additional historical and present-day water quality data.
Four long-term integrated hydrological/suspended sediment monitoring stations are recommended for the LMR. Sampling stations are proposed for sites M0-at Athabasca, M2-upstream of Fort McMurray, M3-downstream of Fort McMurray and the Clearwater River, and M9-downstream of Firebag River and near Embarras Airport (Figures 6,7; Table 3). At each of these locations, in addition to the water quality parameters identified in Section 5 (below), suspended sediment-related parameters to be measured include Total Suspended Sediment (TSS), Turbidity, Sediment particle size distribution, % Organic Content and standard Water Survey Canada hydrometric data (e.g., flow, discharge, water levels, temperature).
Hydrometric and site-specific meteorological data will be collected on a continuous time scale. Sampling for other physical suspended sediment parameter such as TSS, particle size determination, etc., will be conducted at a minimum monthly, but the frequency will be adjusted to weekly or bi-weekly as necessary to capture variation through hydrograph rise, peak and recession. Bulk-time integrated suspended sediment sampling will be conducted following Philips et al. (2000) and Macdonald et al. (2010).
Estimation of the magnitudes and variability in tributary hydrology and related sediment dynamics is necessary to estimate the transport and relative contributions of materials from upper catchments into the LMR. These include sites located near the confluence with the Athabasca mainstem to provide necessary information on the water and sediment inputs (quantity, quality) to the mainstem and their variability over various time-scales (individual storm events, seasonal, annual regimes, inter-annual variability, trends, etc.). Tributary sites also provide reach-specific information on the impact of oil sands development on tributary water and sediment regimes in addition to providing mass-balance loading estimates to the Athabasca River.
Eight primary tributary sampling stations have been selected for continuous hydrometric and suspended sediment sampling. Proposed sites include: CH1, CL2, ST1, MU1, MA1, EL1, CA1 and FI1 (Figures 6,7). The mouth of the Tar River was also originally considered, however, this river is now actively being “relocated” as part of an oil sands development project. Information on similar hydrometric and suspended sediment parameters will be collected as described above for the mainstem sites. Tributary sites provide important information for closing the water and sediment budgets for the Lower Athabasca (using a mass balance approach).
Tributary sites have been spatially distributed in the Athabasca basin to include catchments that are highly, moderately and minimally impacted by oil sands development. Tributary catchments minimally impacted by development may be the closest approximation to a “baseline” or “reference” condition before approved oil sands development commence, although defining a true, unimpacted “reference” system is probably no longer possible to find.
Tributary sites also provide information on the possible effects of different oil sands developments/technologies on water and sediment delivery to the Athabasca. Hence sites have been identified that include catchments minimally affected by oil sands development, affected by mining, affected by SAGD, and affected by a combination of mining and SAGD activities. To improve our understanding and modelling/prediction capabilities in unmonitored tributaries, the selected tributaries and sites also cover a range of catchment size and runoff regimes.
Table 3 provides specific details on site descriptions, sampling regimes, the rationale for the chosen spatial and temporal resolution. A summary of other design and implementation considerations/issues is also provided in Appendix B.
Given challenges in year-round accessibility of sites and in quantifying river flow and sediment loads during dynamic periods of ice conditions, new and developing technologies will be incorporated into the monitoring program to address these types of issues. These include:
Figure 8. Schematic diagram of the fixed (left) and variable (right) model traps in cross-section. For comparison, a schematic diagram of the Phillips et al. (2000) trap design is shown (bottom figure). Note that the figure is not to scale (from McDonald et al. 2010).
An integrated and standardized surface and groundwater quality monitoring component has been designed to ensure to provide the necessary data/information to assess and predict current and future impacts of oil sands development on water quality (including sediments) in the Lower Athabasca mainstem and its tributaries.
The ability to assess changes in reach-specific and regional water quality conditions is directly dependent on the availability of standardized data obtained at sites located at relevant spatial and temporal scales. The objectives of the surface and groundwater quality monitoring components are to provide the necessary data/information that address the questions:
Although this component of the monitoring program deals specifically with surface and related groundwater quality parameters in the Lower Athabasca and its tributaries, it is directly linked to the hydrometric/sediment monitoring component (above) from which contaminant fluxes will be calculated, the air deposition-water quality component (Section 6) which provides information on atmospheric contaminant deposition directly to the aquatic environment and to the adjacent landscape (i.e., onto snow in winter; on the landscape); and, to future biomonitoring and effects-based monitoring programs that will be added in the upcoming Phase 2 of the Lower Athabasca Water Quality Monitoring Program.
To adequately quantify and characterize the water quality in the Lower Athabasca and its tributaries, the following core components are needed:
State-of-the-art sampling/environmental data acquisition systems will be implemented and validated to improve spatial/temporal coverage and data interpretation.
Concentrations of chemicals, both naturally occurring and industrial inputs can vary substantially over time and space within any system. A confounding challenge also arises when compounds of concern are at concentrations at or below analytical detection limits or are at levels where the concentrations are at or close to the threshold of concentrations known to have an observable biological and/or ecological effect. For example, in the RAMP program, many temporal trends in PAC concentrations in the Athabasca River could not be statistically assessed with sufficient confidence because levels were at or below analytical detection limits (RAMP 2010). In such cases, standard point-in-time bulk water sampling may not adequately measure potentially ecologically relevant exposure concentrations to biota. Hence, with the exception of spring melt pulses (where metals and acids in the first few days of melt can be higher than average), the implementation of integrative passive sampling and related analytical procedures could improve the quantification of such compounds, which includes metals, PACs, NAs, and other oil sands or industrial related contaminants.
Bulk water sampling frequency at these sites will be monthly, although provisions will be made for enhanced daily, weekly or bi-weekly sampling to capture variations through hydrograph rise, peak and recession and to capture key contaminant loading events such as aerial deposition through spring snow melt. In addition to conventional grab water sampling, a series of integrated passive samplers will be deployed to better quantify environmental concentrations of oil sands related organics contaminants, trace metals and naturally occurring petroleum compounds.
Building on the experience gained in the AENV Contaminant Load Study (AENV 2010), the Muskeg River Water Quality Program (AENV 2009), and Environment Canada’s integrated oil sands research program, the proposed design will utilize and assess the efficacy of new passive sampling systems at each major tributary mouth and at key main stem locations (Table 3) to better integrate and quantify contaminants/compounds of concern. Passive sampler types will include:
SPMD and POCIS samplers also enable in situ extraction and isolation of trace organic contaminant mixtures for future toxicity assessments and toxicity identification evaluation (TIE) approaches as necessary.
The core Lower Athabasca River mainstem sampling stations are proposed for locations M1-M9 (Figures 6,7; Table 3), which would quantify key tributary-specific loadings. In addition to the hydrometric/meteorological and sediment data being collected at these locations (see Section 5.0), bulk water samples will be taken to measure the following chemical parameters: major Ions, nutrients, dissolved and particulate organic carbon, metals, and organic contaminants (e.g., PACs, Naphthenic Acids). Measurements of trace metals will include both total and filtered forms, and methyl mercury. Sample types for organic contaminants and metals include:
Appendix B provides summary of the preliminary core chemical analyte list for the suite of chemical elements and compounds to be tested.
Water sampling will be done at fixed depths with Nislin samplers or with portable battery-powered submersible pumps, at multiple points at a given site and subsamples may be pooled to obtained cross-sectionally averaged samples.
Monthly (final frequency to be confirmed from preliminary sampling) cross-sectional composite bulk samples using Ponar grabs of surficial sediments will also be taken at each location (frequency is to be determined). Analysis of the samples will provide information on the physical and chemical characterization of surficial sediments (e.g., lithology, grain size, and trace-metal distribution). The final selection of sites will need to consider the biological relevance and contaminant deposition.
Certain physical parameters such as lithology can be subsequently compared between locations to define spatial and temporal change in status and trends. Sediments will also be analyzed for oil sands organic contaminants and trace metal, providing additional information on possible exposure concentrations and pathways to aquatic benthic biota and fish early life stages.
Figures 6,7 and Table 3 summarize the proposed sites for tributary sampling. In addition to these, sites located in smaller catchments such as Clarke Creek, McLean Creek and Fort Creek on the east side of the lower Athabasca mainstem and Emundson Creek on the west side will be considered.
Surface water chemical parameters to be measured at all tributary stations are the same as for the mainstem stations indicated above (section 5.2.2) and outlined in Appendix B. Passive sampling of contaminants will only be conducted initially at tributary mouths as discussed above.
Estimation of groundwater seepage and possible surface water interactions requires a broader synoptic approach to spatial sample placement and related temporal frequency. A series of distinct geographic (mainstem and tributary) sampling regimes will be conducted to characterize the groundwater quality (and its variability) discharging to the reach under examination (with an appropriate number of replicates taken for QA/QCs purposes). For example, in areas more distant from oil sands operations and tailings ponds, spatial sampling will be less dense. Sample densities proximate to tailings ponds (e.g., Suncor Pond 1) will be approximately 50-150 m apart (i.e., based on preliminary Environment Canada studies in the vicinity of Suncor’s Pond 1; about 70 samples along the 4 km shoreline). Adjustments to site spacing will be conducted if deemed necessary after preliminary analyses. Within the Athabasca mainstem, riverine groundwater sampling will build on existing tailings pond studies and sites currently being used by Environment Canada and Alberta Environment. Riverine groundwater samples will be collected during four time intervals (summer, spring, autumn, winter), with the initial proposed levels of site replication focused on operations closest to surface water and new areas scheduled for development.
Groundwater chemical analytes to be measured will be the same as described for the surface water stations (Appendix B). Table 4 summarizes groundwater quality and quantity indicators that can be used to describe the potential impacts of oil sands developments on local and regional groundwater resources.
| Indicators | Condition indicator | Development indicator | |
|---|---|---|---|
| Quality | Primary | pH, conductivity, dissolved oxygen, redox; major anions/cations, trace metals (e.g., As, B, V, Se. U), ammonia/ammonium, naphthenic acids, TDS | Density of mine-related seepage sites for relevant aquifers. Density of disposal operations in a given area. |
| Secondary | BTEX (benzene, toluene, ethybenzene, xylenes) - (near ponds and upgrading operations); stable isotopes (deuterium and O18) | ||
| Tertiary | PACs, other stable & radiogenic isotopes (e.g., C14, S36, N15,) | ||
| Quantity | Primary | Temporal change in groundwater surface elevation in an aquifer management unit at an established monitoring location. Changes in measured baseflow (tributaries) in receiving waters. (determined from hydrometric gauging) | Density of dewatering activity in overburden and basal aquifer for mine development. |
| Secondary | Impact to sensitive water body or wetland as demonstrated by water level changes. Accuracy of modelled versus measured conditions in established monitoring wells. |
An assessment of potential paleo techniques that could be used to provide historical contextual information to aid in monitoring data interpretation will be performed. If appropriate, historical levels and trends in trace metals, mercury, and organic contaminants will be assessed using sediment core samples obtained from suitable depositional areas. The cores will be age dated using standardized methods (e.g., 210Pb and/or geochronological methods) and analyzed for relevant contaminants. Profiles of stable metal isotopes, metal ratios, and changes in various organic compounds will provide information on historical deposition to the region and help quantify the influence of emissions from oil sands activities on contaminant burdens in sediments. The potential for spatial variability in sediment contaminant loading necessitates collection of cores from multiple locations (e.g., lakes) within the Athabasca region.
Particular focus will be placed on quantifying the types, sources, transport and fate of contaminants of concern such as Naphthenic Acids (NAs). NAs are a large class of cyclic, organic compounds that are associated with oil sands operations and have been determined to be a significant source of toxicity of oil sands process water (OSPW) (Giesy et al. 2010). NAs are a complex mixture of compounds that have been difficult to characterize, but progress is being made by Environment Canada’s oil sands fingerprinting research program using state-of-the art analytical procedures to identify and characterize this group of contaminants. Monitoring and analyses of NAs will be tightly linked to these research efforts.
Many of the PACs that will be measured in the proposed water quality program (see Appendix B for analyte list) have at present, no established effects thresholds. Hence, limitations exist in taking water chemistry information and using concentration values to predict toxicological effects. For example, for compounds such as alkylated PACs, it is not yet possible to conduct a proper risk assessment linking measured ambient concentrations with potential levels of toxicological response. Such issues will need to be further clarified in relation to how the water chemistry information can be subsequently used and interpreted to assess biological and ecological impacts.
Water quality criteria for metals that are designed to be protective of aquatic life, can at times be over-protective as a result of natural, site-specific differences in water chemistry. Such differences can affect metal speciation and bioavailability, which are fundamental to toxicity assessment. Many studies over the last two decades have strengthened our understanding of metal chemistry in water, and their toxicity to organisms. Our awareness of processes that can mitigate toxicity through formation of organic and inorganic metal complexes, and sorption of metals to organic matter has also increased considerably.
The biotic ligand model provides a quantitative framework for assessing metal toxicity over a range of hardness, pH and dissolved organic carbon (DOC) levels, and allows for the prediction of the degree to which site-specific water chemistry will affect metal toxicity. As a result, it becomes important for aquatic monitoring programs to track changes in these characteristics in time and space as proposed in this program.
Using this approach the toxicity potentially associated with a given measured metal concentration can be successfully predicted. Recent studies such as Paquin et al. (2000) have outlined the potential utility of this approach to metal monitoring, for establishing site-specific water quality criteria, that take into account local differences in water chemistry parameters that affect the partitioning and speciation of heavy metals. Such an integrated perspective will need to be considered during the Phase 2 design of the water quality program.
Given the spatial heterogeneity and difficulties associated with accessing sites, new sampling technologies should be constantly assessed in terms of their possible implementation in the program. New buoy-based water quality sondes, satellite imagery for monitoring changes in cryospheric conditions (river ice; snow distribution and depth; snow water equivalence), and automated instrumentation that can transmit real-time data via satellite are examples of developing technologies that have potential applicability in the program.
With respect to connectivity with groundwater, remote sensing technologies may be applied to further focus and optimize monitoring efforts. In particular, connectivity of basal water (i.e., groundwater in contact with the bitumen ore body) with surface waters is believed to provide a conduit for organic contaminants, whether naturally occurring or anthropogenic. Given that the basal water typically has an elevated saline component relative to surface water, there may be an opportunity to apply remote sensing technologies to identify saline seeps. Alberta Environment has commissioned such preliminary surveys for sections of the Athabasca River using electromagnetic induction to identify areas of higher conductivity in the river bed, thereby inferring connectivity with saline aquifers. In addition, airborne geophysics using electrical of electromagnetic spectrometry, for example, could be applied to the Athabasca River and its tributaries to further constrain the location of saline seeps. When properly validated (e.g., by electromagnetic induction), this information could be used to focus the surface water monitoring program, especially in relation to the identification of areas of greater or lesser impact (e.g. positive controls, reference areas), and to populate surface water quality models with more detailed boundary information.
Aerial deposition of stack emissions, mining operations particulates and emissions from tailings ponds has been identified as a possible significant source of both direct and indirect oil sands contaminant loadings to the Athabasca River basin.
Aerial deposition of stack emissions, mining operations particulates and emissions from tailings ponds has been identified as a possible significant source of both direct and indirect oil sands contaminant loadings to the Athabasca basin (Kelly et al. 2009, 2010). This research also identified some of the key gaps in data and the importance of having better regional estimation of aerial inputs and their relationship to changes in water quality parameters. Moreover, it reinforced the need for integrated air, hydrometric, and water quality sampling at appropriate spatial and temporal scales.
The goal of the air deposition monitoring program is to provide the necessary data to address the following questions:
In order to answer the third question above, the program includes studies within selected catchments to provide the necessary information to determine key processes affecting transport and fate of material deposited aerially to the landscape and to quantify the contribution to surface and groundwater.
Table 5 (see page 63 ) and Appendix C summarize the current air quality stations operating or being installed in the region or adjacent areas. Data collected at these sites are currently being used by four air sampling programs operating in the oils sands region:
The above programs form a solid basis upon which to build a more comprehensive regional atmospheric monitoring program in support of the determination of atmospheric deposition of toxic or potentially toxic organic substances and metals emitted from oil sands development.
To determine atmospheric deposition pathways, there is a need to know both dry and wet deposition. Inputs related to dry deposition to the mainstem Athabasca River may be variable; wet deposition to the river may be elevated for episodic periods during spring snow melt, spring freshets and major rain events. Hence, associated watershed sampling must occur to calculate actual concentrations and loading coefficients.
Snow collections serve to quantify winter deposition but are subject to uncertainties due to melting and loss of chemicals. There has been only limited comprehensive snow collection (Kelly et al. 2009, 2010; Alberta Oil Sands Environmental Research Program (AOSERP) studies in the late 1970s/early 1980s – Barrie and Kovalick 1980; Murray 1981) for pollutant inputs and given the importance of snow melt inputs in this climate zone more systematic transect watershed sampling of snow will be integral part of the air monitoring program.
Since estimation of contaminant flux to surface waters is a core objective, paleo-environmental approaches will be explored to quantify present and historical baseline conditions. The potential for spatial variability in sediment contaminant loading necessitates the collection of cores from multiple lakes. Parallel analysis of surface water samples for a temporal assessment of dissolved and suspended contaminant concentrations is also necessary.
The air deposition monitoring component is designed to address two strategic objectives in relation to quantifying the distribution, fate and magnitude of the deposition of aerial contaminants on the Athabasca mainstem and primary tributary catchments. The two program objectives are:
Figure 9 summarizes the proposed new and expanded sites for the air deposition monitoring program described below.
What is the direct deposition of oil sands related contaminants to the surface of the Athabasca River and its tributaries?
A new synoptic network of monitoring sites will be established. Building on the three sites in the core region established under the current regional surveillance study, two new sites will be added downstream along the Athabasca (one near Firebag River and one near the boundary of the Wood Buffalo National Park) to look at expansion to the north. One site in the existing core industrial region would continue as a long-term station and the two new sites would be both long-term. Two of the existing sites could prove to obtain better estimates of spatial heterogeneity and variability in deposition patterns. The new sites will also be supplemented with passive samplers as indicated by the surveillance study that is underway. The final latitudinal and longitudinal coordinates of the proposed new sites will be determined by modelling scenarios of current and future development and demonstrating the impact of the data from each site to the quality of deposition maps produced. Active, continuous measurement sampling sites will require locally-clean power and an operator. “Passive sites” do not require power and will be visited on a monthly and/or seasonal basis. The preliminary proposed “analyte” list for all active and passive sites is provided in Appendix B.
What is the deposition to the landscape in the Athabasca Basin? How does this deposition affect water quality in the tributaries and main stem?
In addition to the sites outlined above and as informed by modelling, up to eight new sites will be installed. The proposed locations include: one south of Fort McMurray; two to the east and two to the west of the mineable area, along the boundary of the mineable area; one along the western edge of the Athabasca basin; one at the Alberta/NWT border; and one at the Saskatchewan border. Each of the sites will also be supplemented with passive samplers as used by the current surveillance study. The sites will be co-located where possible with sites monitoring air for criteria air contaminants (particulate matter, sulphur and nitrogen compounds and volatile organic substances) as directed by the forth-coming development of the monitoring plan for air in Phase 2. The amount and timing of precipitation to the landscape and the contribution of snowmelt to the aquatic system will be determined through measurements made in support of Section 6 above, remote sensing and modelling with loadings determined using measured concentrations from the monitoring sites.
The proposed synoptic sampling design includes a large number of sites within a relatively small geographic area, when compared to monitoring in other areas of Canada. Many spatially-distributed sites are necessary because of:
Both wet and dry deposition need to be measured over a far larger geographic area than currently done by WBEA. For atmospheric deposition the key issue is having background sites for gas phase, particles and wet precipitation outside of the immediate or proposed development area – N, S, E and W. Also atmospheric deposition sites are needed at the outer edge of development to assess more direct regional emissions and deposition including dust and pollutants from mining itself (e.g., vehicles). Currently the atmospheric deposition from WBEA is only along the developments near the river.
Since the region is expected to undergo rapid expansion of industrial activity and the information will be needed on short time lines (e.g., changes over 1-year compared to decadal for more spatially distributed networks), the additional sampling sites will be needed and sampling frequency will be high (as frequent as possible for detection of deposition pathways and magnitudes).
Precipitation sampling will be event-based where access to the site permits and every two weeks or monthly at sites difficult to access. If insufficient volume for analysis is available, samples will be added for analysis.
We will attempt to model dry deposition based on observed gas and particle phase concentrations, since it is difficult to measure directly. Additional air particle measurements will be made to provide the input needed for modelling. A large number of hydrocarbons co-exist in gas and particle phase. The dry deposition for gases is different than for particles; furthermore, the size of the particles affects their deposition rates. To avoid large uncertainties in dry deposition calculation, it is recommended that the total ambient concentrations, gas-particle partitioning and the particle size distributions should be measured.
In all cases, data analysis would be performed taking back-trajectory into account. By this means, measurements at a single site will be used to characterize different sources in the area. Sampling frequency will need to be high for this purpose. Key parameters to quantify must include core meteorological attributes (e.g., wind speed and direction, temperature, relative humidity) and fast-response plume indicators such as PM2.5 number concentration and NO2 and SO2 mixing ratios.
To what extent are atmospherically deposited contaminants from oil sands mining and upgrading and in-situ extraction exported from select catchments in the Athabasca River basin?
To better identify and model transport pathways, efficiency, transformation, and export of atmospherically deposited contaminants within small catchments in the Athabasca oil sands area, two sub-catchment monitoring networks will be established. One will be situated in an ”undisturbed” landscape to assess baseline or background conditions and the other will be in an industrially modified landscape. Information from these catchment-specific networks will be used to estimate aerial deposition related contaminant transport to surface water at a landscape scale, based on deposition estimates from atmospheric transport models and from directly measured deposition rates.
The “undisturbed” catchment must be located within a physically undisturbed area of relatively high deposition of organic contaminants and heavy metals from sources within the mineable oil sands area. Ideally the site must have a defined surface water flow pathway (defined stream channel or lake) and must contain hydrologic response areas that are representative of the larger study region, although this might be difficult to attain except in headwater areas.
Alternatively, if an appropriate fluvial system cannot be identified, monitoring could focus on wetland-dominated catchments on the assumption that precipitation on upland areas goes to storage and that there is no export of contaminants. A lake-based catchment may be the ideal study location in that contaminants that are exported to surface waters are not immediately exported from the catchment. An analysis of contaminant flux using lake sediment cores or sediment traps may also be possible.
Wet deposition of key contaminants (hydrocarbons, metals and their degradation products) will be measured throughout the catchment. Deposition may be further characterized by type (rain, snow, adsorptive, dustfall, etc.) and composition (particle size, major ions). Analysis of contaminants of concern in peatland vegetation may provide a robust assessment of actual deposition rates and composition. Consideration however, must be given to the transformation of metals in the highly reducing peatland environment. Providing that metals deposited in peatlands are initially in relatively insoluble forms, they may be mobilized as they transition into a reducing or acidic environment with continuing biomass accumulation at surface. In this context, export of metals from peatland-dominated catchments may be delayed. Analysis of metal concentration and speciation in peat cores may be appropriate.
Figure 9. Map showing the proposed locations for an expanded sampling network. Red dots are WBEA and EC (2010) passive air samplers, and Green dots are WBEA sampling sites, currently in use for the surveillance study. + signs are proposed additional powered hi-volume air and precipitation monitoring.
What is the historical deposition of selected oil sands contaminants in the Athabasca River basin region and are there chemical markers of anthropogenic emissions?
An assessment of the potential for paleo-limnological reconstruction of historical contaminant trends will be performed. There is concern that variable sediment deposition rates may bias results, with a consequent poor ability to use paleo data as contextual information with which to evaluate contaminant trends over time. If appropriate, collection of lake and pond sediment cores from the region within 50 km of major developments would support the atmospheric deposition study as well as provide information on transformation and mobilization of contaminants. This work has begun (e.g. Curtis et al. 2010; Hazewinkel et al. 2008) and can possibly be extended to provide geographical coverage of the region. Careful consideration will be made regarding the final selection of sites to ensure areas are representative of good historical records.
Sediment cores would be collected from deep areas of the lakes and ponds and dated (using 210Pb, 137Cs) and sedimentation rate estimated with dating models. Selected cores would be analyzed for the same contaminant suite as for atmospheric deposition and a historical profile constructed. Interpretation of the data would be aided by ancillary measurements such as organic carbon, diatoms and other microfossils and algal pigments. The potential for spatial variability in sediment contaminant loading, due to factors such as size of the lake catchment, changing water levels particularly in smaller water bodies, necessitates the collection of cores from multiple lake basins.
Open, transparent analytical standards and data access: QA/QC protocols, Standard Operating Practices will be clearly defined and implemented.
State-of-the-art analytical procedures will be utilized and developed through qualified laboratories.
Many good QA/QC methods exist for organic and inorganic contaminants in water, biota and atmospheric samples and the QA/QC approach in this program will adopt the best available practices. Examples are the National Air Pollution Surveillance Network and the integrated Atmospheric Deposition Network (PACs, heavy metals), the Northern Contaminants Program (mercury, heavy metals), and QUASIMEME (selected metals biota and sediment), Alberta Environment (2007) and Environment Canada’s EOALRSD Laboratory Services for the Oil Sands Project (Environment Canada 2011).
In general, a robust quality assurance program for chemical analysis will:
Sample analyses will be carried out by government or private sector labs qualified (e.g., accredited by CALA to the standards of ISO/IEC 17025) for the analysis of metals and PACs in environmental samples. It is anticipated that multiple analytical laboratories will be involved in the program due to the large numbers of samples to be analyzed and wide range of analytes.
To encourage continuous innovation in method development no single analytical method for given analytes is recommended. However for PACs and multi-element analysis well established methodology exists for sample extraction/digestion to isolate analytes and for instrumental quantification. Method performance will be evaluated by the QA an inter-lab comparison program
PACs:
A large suite of polycyclic aromatic compounds (PACs) have been determined in river water, snow and sediments (e.g. Kelly et al 2009; RAMP database 2011). The methodology generally follows USEPA 8100 for determination of PACs by GC-MS. However more analytes are possible. For example, alkylated phenanthrenes and dibenzothiophenes have been found to comprise more than 70% of the alkylated PACs in the oil-sands extractable material but have not yet been measured routinely (Yang et al. 2010). RAMP surveys (2002-2004) showed non-detectable PACs in river water (Detection Level = ~ 1 ug/L) for all 50 PACs but most of the same suite of PACs are reported in Athabasca River sediments (RAMP database 2011). While approximately 46 PACs are listed in the RAMP and 42 were determined by Kelly et al (2009), a larger suite could be determined if suitable analytical standards were available. There is also a lack of reference materials certified for a large number of PACs, particularly alkylated isomers, and for S and N- containing heterocyclics. This can be solved by repeated analysis of suitable laboratory control samples and inter-laboratory comparisons with the eventual goal of certifying one or more existing commercial reference materials for the full suite of organic analytes. The monitoring and identification of alkylated PACs will also be useful for fingerprinting or identification of main sources (e.g., when the ratio of alkylated to non-alkylated PACs is high, this identifies a petrogenic source. In contrast for low ratios, a pyrogenic source is suggested (i.e., forest fires, combustion processes)).
Multi-elements:
Multi-element analysis should follow USEPA Method 200.8 and EPA 1669 for determination of trace elements in water and wastewaters. Most recent studies (e.g. RAMP 2010; Kelly et al. 2010 and unpublished; Headley et al. 2005), have utilized this methodology and determined up to 32 elements. Additional elements are desirable in order to develop the best multi-element profile possible. Limitations to this are the need for reference materials certified for the large suites of elements. This can be solved by repeated analysis of suitable laboratory control samples and inter-laboratory comparisons with the eventual goal of certifying one or more existing commercial reference materials for up to 50 elements.
Naphthenic Acids (NAs):
RAMP (2011 database) has routinely determined “naphthenic acids” in river water and concentrations in the μg/L range (< ~10 – 3000 μg/L) have been reported. However the naphthenics and related heterocyclic acids are a very complex mixture for which only a limited number of analytical standards exist.
A tiered approach is recommended for the naphthenic acids as follows: (a) Total naphthenic acids reported as total extractable oil sands organic acids - low resolution MS is sufficient for this measurement using electrospray ionization, and the instrumentation is widely available in most laboratories. (b) Specific classes of oil sands acids reported as their relative concentrations.
High resolution Mass Spectroscopy is needed for this measurement and is available in limited number of laboratories. The species should include organic acids containing the O2 class (often assumed to be naphthenic acids) along with other species: containing On where n = 1-16; NOn and N2On where n = 1-13; and OnS and OnS2, where n = 1-10; and 1-8 respectively.
Information from above will be required for fingerprinting purposes. As the Program progresses there will likely be a need to fine tune the classes in (b) above.
The detection limit for total naphthenic acids in water is 1 mg/L based on 100mL sample.
However, reporting the NAs according to compound classes is difficult. There are no commercially available authentic standards for these classes and thus the analysis at the present time is qualitative and at best semi-quantitative based on the relative abundances.
Total and methyl mercury:
Total and methyl mercury sampling and analysis procedures will follow US EPA methods 1631 and 1630, respectively. Briefly, environmental samples are collected into acid-cleaned glass, polyethylene, or fluoropolymer jars using clean hands-dirty hands ultra trace sampling techniques. Total mercury analysis is carried out by bromine oxidation digestion, tin chloride reduction, purge and trap, thermal desorption, and cold vapor atomic fluorescence spectrometry. Methyl mercury analysis in water is carried out by distillation, aqueous phase ethylation, purge and trap, thermal desorption, pyrolysis, and cold vapor atomic fluorescence spectrometry. Methyl mercury analysis of solids such as biological samples, sediments, and vegetation is similar but includes digestion rather than distillation. Detection limits for total and methyl mercury in water are 0.2 ng/L and 0.02 ng/g, respectively and for solids are 0.1 ng/g.
Other organic analytes:
RAMP (2011 database) has determined selected phenolics (nitrophenols, cresols) at multiple sites and N- containing polycyclics (Acridine, quinoline) in river water at a single site (Steepbank River (mouth, STR-1), in 1997. Phenolics were detectable (1-2 μg/L) in the Athabasca River at Donald Creek ATR-DC-CC and STR-1. Further surveillance for these contaminants could be conducted. Emissions from the wastewater treatment plant at Fort McMurray could include pharmaceutical and personal care products, which are found in receiving waters downstream of all WWTPs. Selected chemicals may be useful tracers of that effluent (e.g. artificial sweeteners) as well as the additional effluents associated with all mining operations. These efforts will support biological effects studies on fish in the Athabasca River and associated tributaries. No recommendations for methods are made here but it should be a subject of future discussion.
Analytical:
Standard operating procedures for core analytes will be determined by the participating laboratories tasked with the monitoring. A chemical monitoring analysis management group with representatives from each participating laboratory should jointly decide on which procedures are to be used, and if current SOPs are satisfactory for the new program. Where SOPs do not exist these should be developed among this group or external assistance sought if required. It is expected that SOPs for chemicals emerging from continuous research will be integrated as necessary. SOPs should conform to those developed by international bodies such as ASTM, OECD and ISO. Protocols exist both federally and provincially for many routinely collected water quality parameters (e.g. dissolved organic carbon), however protocols for other methods exist in research laboratories only (e.g. C1-C4 PACs, naphthenics and other acid extractable compounds) and will require additional validation for full implementation as discussed above.
Field Sampling and Sample Handling, Processing:
A variety of field sampling protocols are being used in the monitoring program (e.g., AENV 2006). Documented ASTM, USEPA, USGS and Environment Canada SOPs methods will be followed and noted. Any alterations of methods will be approved through an expert consultation process and include a period of inter-calibration with older methods.
Example SOPs to be listed include:
ASTM, 2002. Standard Practice for Low-Flow Purging and Sampling for Wells and Devices Used for Ground-Water Quality Investigations, ASTM Standard D6771-02, ASTM International, West Conshohocken, PA.
ASTM, 2005. Standard Guide for Direct-Push Ground Water Sampling for Environmental Site Characterization, ASTM Standard D6001-05, ASTM International, West Conshohocken, PA.
ASTM, 2005. Standard Guide for Field Filtration of Ground-Water Samples, ASTM Standard D6564-05, ASTM International, West Conshohocken, PA.
ASTM, 2005. Standard Guide for Field Preservation of Ground-Water Samples, ASTM Standard D6517-05, ASTM International, West Conshohocken, PA.
ASTM, 2009. Standard Practice for Detecting Hot Spots Using Point-Net (Grid) Search Patterns, ASTM Standard D6982-09, ASTM International, West Conshohocken, PA.
Field sampling for mercury/methyl mercury and multi-element analysis will follow the clean-hands dirty hands protocol described in USEPA Method 1669 (US EPA 1996)
Power analyses have been used conventionally in monitoring programs to determine the levels of spatial and temporal replication necessary to detect change or impacts. In the context of a monitoring assessment, the main issue is the establishment of effect size (i.e., the deviation from baseline or “reference” condition that needs to be detected) and the relative importance of associated Type I and Type II statistical errors. The proposed spatial network of long-term, standardized sites in this program, along with associated increased temporal sampling frequency, allow for more rigourous development and application of suitable power analyses methodologies. These analyses will include conventional approaches to statistical inference where applicable (i.e., the power associated with the detection of water quality threshold exceedances), but also new and emerging approaches dealing with environmental gradient design and prediction (i.e., the power associated with detecting rates of change and trends).
Effects-based Monitoring, Multiple-stressor and Ecological Impacts –
Phase 1 of the water quality monitoring plan concentrates on the physical/chemical attributes in the Lower Athabasca river system. Ultimately, the Athabasca River Water Quality Monitoring plan must address not only surface water quality, but also assess the aquatic and relevant terrestrial ecosystem components holistically. It would also involve monitoring and understanding of all components of the aquatic ecosystem, such as benthic biota, fish and other aquatic species.
A key requirement will be to expand the existing program to include consideration and implementation of appropriate effects-based biological and ecological monitoring that directly addresses questions related to the assessment of oil sands impacts, and other point- and non-point source contaminants, on the structure (biodiversity), function and health of relevant aquatic and terrestrial environments.
Determination of observed and projected biological and ecological effects on aquatic biota such as fish and wildlife are of utmost importance to residents in the region, particularly First Nation communities. It will be important to ensure that appropriate validation and consultation occurs to properly inform the design and implementation of Phase 2 to ensure that valued ecosystem components are appropriately addressed in the monitoring plan.
Integrated Process-based Modelling and Environmental Prediction –
Modelling past, current and projected environmental conditions and impacts is a necessary complementary activity of the monitoring program (Figure 1). Modelling serves two main functions: 1) data integration and testing our process-based understanding of ecosystem structure and function; and, 2) provide a predictive, decision support tool such as cumulative effects assessment. Process-based modelling can also be used to identify where and when we need to improve the monitoring efforts and focused surveys. The monitoring activities should therefore be constructed and integrated in such a way as to support the development, calibration, and operation of hydrological and water quality models for the lower Athabasca River system. Currently several models are being used by Alberta Environment and Environment Canada in the Lower Athabasca River (e.g., EFDC, river-ice dynamics and 2-D flow). Ultimately an improved integrated environmental prediction framework should result.
Other Monitoring Locations - Altered or reclaimed aquatic habitats -
Although this document focuses on surface waters, in particular particularly the Athabasca mainstem and tributaries, future phases of the sampling program design must consider other types of water bodies in the region. Given the requirement for the oil sands industry to reclaim existing and future open pit mines and tailings ponds, a systematic monitoring program of these altered or reclaimed systems must also occur. The objective of these monitoring efforts would be to assess the compliance and performance of these reclaimed systems and to evaluate their conditions and success. Examples of such aquatic systems include: re-constructed stream courses; impoundments; fisheries compensation waters; and constructed wetlands. A prime example of such systems is the Beaver Creek Diversion system (implemented in 1977) and the upcoming construction and use of End Pit Lakes.
The water quality monitoring program framework and proposed sampling design fully meet the key principles that were identified by the Federal Oilsands Advisory Panel for the design and implementation of a “world-class” oilsands monitoring program. Although the overall monitoring plan is being designed in phases given the geographic and sampling complexities involved, assurances are being taken that the program will be one that is: holistic and comprehensive, scientifically rigorous, adaptive and robust, inclusive and collaborative, and transparent and accessible.
An integrated, ecosystem-based approach was used that incorporated multiple essential components of the system (hydrology, surface and groundwater quality and quantify, climatology, sediment dynamics and quality, snow and river-ice, regional air deposition) as well as the relationships among the components. Sites were chosen to integrate multi-scale, spatial measurements (tributary and mainstem catchments, impacted vs. baseline), recognizing the importance of addressing the temporal variability, and improving the ability to define “baseline” or historical environmental conditions.
A best available science-based approach was used to select the chemical, hydrological and atmospheric variables to be measured, methodologies for field sampling and laboratory analyses, and field and laboratory quality assurance and quality control. Standardized reporting, including peer-reviewed and plain language publications, was identified as a core required outcome of the program.
The new water quality monitoring framework builds upon and integrates existing monitoring efforts into an adaptive, and holistic system. Decisions regarding the sampling locations and the types of associated analyses will be evaluated using science-based and policy-relevant decision triggers, and revised as new knowledge and technologies become available and needs and circumstances change. The proposed design moves the current monitoring program paradigm from being “retrospective” to one that will be “prospective” in assessing current and projected multiple stressor impacts and ultimately cumulative effects. In addition the program design is based on a mass-balance conceptual framework, allowing for hypothesis-driven analyses to be conducted and models to be developed to assess the fate, distribution, transport and loadings of oil sands related and other contaminants in the lower Athabasca River system. This stressor-based program forms the foundation for a comprehensive, a geographically expanded effects-based monitoring plan to be developed in Phase 2. Air quality and terrestrial biodiversity monitoring plans, and their integration, will follow.
Although phased in its design, the approach recognizes the importance of ensuring the monitoring program addresses the key issues, questions and concerns of stakeholder groups, First Nations and governments at all levels. This includes recognition of the need for appropriate validation with concerned parties in the design and execution, including the prioritization of issues and setting of monitoring objectives.
The concepts presented in this plan necessitate a dramatic change and improvement from the current monitoring regime, and the actions needed to implement this plan require a significant increase in monitoring effort. The plan will therefore need a suitable transition period in which to implement the proposed actions.
D. Baird, Research Scientist
C. Banic, Atmospheric Scientist
G. Bickerton, Research Scientist
D. Burn, Hydrologist
P. di Cenzo (Ed.) Physical Scientist
P. Dillon, Watershed Biogeochemist
I. Droppo, Research Scientist
M. Dube, Aquatic Ecotoxicologist
R. Hazewinkel, Water Quality Specialist
M. Hewitt, Research Scientist
E. Kelly, Environmental Scientist
D. Lindeman, Biologist
P. Marriott, Civil Engineer
E. McCauley, Population Ecologist
P. McEachern, Limnologist
D. Muir, Senior Research Scientist
K. Munkitrrick, Fisheries Ecologist
L. Noton, Water Quality Specialist
T . Prowse, Senior Research Scientist
J. Rasmussen, Aquatic Ecologist
J. Smol, Paleolimnologist
F. Wrona (Ed.), Senior Science Strategist
US EPA. 1996. Method 1669 Sampling Ambient Water for Trace Metals at EPA Water Quality Criteria Levels. July 1996. U.S. Environmental Protection Agency
Office of Water, Washington DC 39 pp.
Alberta Environment. 2006. Aquatic Ecosystems Field Sampling Protocols, March 2006. Edmonton, Alberta.
Alberta Environment. 2007. Alberta Environment Data Quality Assurance Policy Procedures and Guidelines, July 30, 2004. Edmonton, Alberta.
Alberta Environment. 2008. Muskeg River Interim Management Framework for Water Quantity and Quality Summary Report, June 2008. Edmonton, Alberta.
Alberta Environment. 2009. Muskeg River Watershed Integrated Water Quality Monitoring Program Annual Report, September 2009. Edmonton, Alberta.
Alberta Environment. 2010. Comprehensive Contaminant Load Study: Water Quality Monitoring, October 12, 2010. Edmonton, Alberta
Alberta Environment - Water Monitoring Data Review Committee. 2011. Water Monitoring Data Review Committee Final Report. Prepared for the Government of Alberta. 46 pp.
Beltaos, S., Cater, T. and Prowse, T.D. 2011. Morphology of deep scour holes in the Mackenzie Delta. Can. J. Civil Eng. (in press)
Beltaos, S. and Prowse, T. 2009. River-ice hydrology in a shrinking cryosphere. Hydrological Processes, 29, 122-144. Invited review.
Barrie, L. A. and Kovalick, J. 1980. A Wintertime Investigation of the Deposition of Pollutants
Around an Isolated Power Plant in Northern Alberta. Prepared for the Alberta Oil Sands
Environ Res Program by Atmos Environ Service. AOSERP Report 90 (Alberta Oil Sands
Environmental Research Program).
Barton, D.R., and R.R. Wallace. 1980. Ecological studies of the aquatic invertebrates of the Alberta Oil Sands Environmental Research Program study area of northeastern Alberta. Prep. for the Alberta Oil Sands Environmental Research Program by Fisheries and Environment
Canada, Freshwater Institute. AOSERP Report 88, 215 pp.
Bond, W.A. 1980. Fishery resources of the Athabasca River downstream of Fort McMurray,
Alberta. Volume 1. Prep. for the Alberta Oil Sands Environmental Research Program by Department of Fisheries and Oceans, Freshwater Institute. AOSERP Report 89, 81 pp.
Bulleri, F., Underwood, A. J., and Benedetti-Cecchi, L. 2007. The assessment and inter-predations of ecological impacts in human-dominated environments. Environmental Conservation, 34, 181-182.
Choy, E.S., Hodson, P.V., Campbell, L.M., Fowlie, A.R., Ridal, J. 2008. Spatial and temporal trends of mercury concentrations in young-of-the-year spottail shiners (Notropis hudsonius) in the St. Lawrence River at Cornwall, ON. Archives of Environmental Contamination and Toxicology, 54(3), 473-488.
Curtis, C. J., Flower, R. W., Rose, N., Shilland, J., Simpson , G.L., Turner, S., Yang, H., and Pla, S. 2010. Palaeolimnological assessment of Lake Acidification and environmental change in the Athabasca Oil Sands Region, Alberta. Journal of Limnology, 69(Suppl. 1), 92-104.
Droppo, I. and Jaskot, C. 1995. Impact of river transport characteristics on contaminant sampling error and design. Environmental Science and Technology, 29(1): 161-170.
Environment Canada. 2011. EOALRSD Analytical Chemistry Schedule of Services – Oil Sands Project, 30 pp..
Federal Oil Sands Advisory Panel Report 2010
(http://www.ec.gc.ca/pollution/default.asp?lang=En&n=E9ABC93B-1)
Fowlie, A.R., Hodson, P.V., Hickey, M.B.C. 2008. Spatial and seasonal patterns of mercury concentrations in fish from the St. Lawrence River at Cornwall, Ontario: Implications for monitoring. Journal of Great Lakes Research, 34(1) 72-85.
Giesy, J. P., Anderson, J. C., and Wiseman, S. B. 2010. Alberta oil sands development, Proceedings of the National Academy of Sciences, 107(37), 915-952.
Headley, J. V., Crosley, B., Conly, F. M., and Quagraine, E. K. 2005. The characterization and distribution of inorganic chemicals in tributary waters of the lower Athabasca River oil sands region, Canada, Journal of Environmental Science and Health, A40, 1-27.
Hazelwinkel, R. R. O., Wolfe, A. P., Pla, S., Curtis, C. and Hadly, K. 2008. Have atmospheric emissions from the Athabasca oil sands impacted lakes in northeastern Alberta Canada? Canadian Journal of Fisheries and Aquatic Sciences, 65, 1154-1567.
Hicks, F., and Beltaos, S. 2008. River Ice. In M. K. Woo (ed), Cold Regions and Hydrolgoic Studies: The Mackenzie EWWEX Experience, Hydrologic Processes, Springer-Verlag, Berlin, 281-205.
Hicks, F. 2009. “An overview of river ice problems” invited paper, Journal of Cold Regions Science and Technology, Special Issue on River Ice, 55, 175-185.
Horowitz, A.J. 1991. A primer on Sediment-Trace Element Chemistry. Lewis Publishers Inc., Chelsea, Michigan, USA, 136.
Kelly, E. N., Short, J. W., Schindler, D. W., Hodson, P. V., Ma, M., Kwan, A. K., and Fortin, B. L. 2009. Oil sands development contributes polycyclic aromatic compounds to the Athabasca River and its tributaries. Proceedings of the National Academy of Sciences, 106(52), 22346-22351.
Kelly, E. N., Schindler, D. W., Hodson, P. V., Radmanovich, R., and Nielsen, C. C. 2010. Oil sands development contributes elements toxic at low concentrations to the Athabasca River and its tributaries. Proceedings of the National Academy of Sciences, 107(37), 16178-16183.
Lindeman, D. et al. 2011. Analysis of Current of Historical Water Quality Monitoring Programs/Activities, Environment Canada (see Appendix A)
Machniak, K., W.A. Bond, M.R. Orr, D. Rudy, and D. Miller. 1980. Fisheries and aquatic habitat investigations in the MacKay River watershed of northeastern Alberta. Prep. for the Alberta Oil Sands Environmental Research Program and Sync rude Canada Ltd. by Department of Fisheries and Oceans. AOSERP Report 93, 173 pp.
McDonald, D. M., Lamoureux, S. F. and Warburton, J. 2010. Assessment of a time-integrated fluvial suspended sediment sampler in a high arctic setting, Swedish Society for Anthropology and Geography, 92(2), 225-235.
Milburn, D. and Prowse, T.D. 2002. Under-Ice Movement of Cohesive Sediments Before River-Ice Breakup. Hydrological Processes, 16(4), 823-834.
Murray, W.A. 1981. The 1981 snowpack survey in the AOSERP study area. Prepared for the
Alberta Oil Sands Environ Res Program by Promet Enviromental Group. AOSERP Report
125 (Alberta Oil Sands Environmental Research Program).
Northern River Basins Study Board (Canada). 1996. Northern Rivers Basins Study – Report to the ministers 1996. Edmonton, Alberta, Canada, 287.
Phillips, J. M., Russel, M. A. and Walling, D. E. 2000. Time-integrated sampling of fluvial suspended sediment: a simple methodology for small catchments, Hydrological Processes, 14(14), 2589-2602.
Palmer, M.A., E.S. Bernhardt, W.H. Schlesinger, K.N. Eshleman, E. Foufoula-Georgiou, M.S. Hendryx, A.D. Lemly, G.E. Likens, O.L. Loucks, M.E. Power, P.S. White, and P.R. Wilcock, 2010. Environmental and Human Health Consequences of Mountaintop Removal Mining, Science, 327, 148-149.
Paquin, P.R., Santore, R.C., Wu, K.B. Kavvadas, C.D. and Di Toro D.M. 2000. The biotic ligand model: a model of the acute toxicity of metals to aquatic life. Environmental Science and Policy, 3, 175-182.
Prowse, T.D. and Culp, J.M. 2003. Ice break-up: a neglected factor in river ecology. Can. J. Civil Eng., 30, 128-144.
RAMP (Regional Aquatics Monitoring Program) 2010. RAMP: 2009 Technical Report. Prepared by RAMP 2009 Implementation Team: RAMP.
RAMP (Regional Aquatics Monitoring Program) 2009. RAMP: Technical Design and Rationale. Report prepared for RAMP steering committee by Hatfield Consultants: RAMP.
Roy, J.W. and Bickerton, G. 2010. Proactive screening approach for detecting groundwater contaminants along urban streams at the reach-scale, Environmental Science and Technology, 44, 6088-6094.
Royal Society of Canada (RSC). 2010. Environmental and health impacts of Canada’s oil sands industry. Royal Society of Canada Expert Panel report, 414.
Schindler, D. 2010. Tar sands need solid science. Nature, 468 (7323), 499-501.
Stewart-Oaten, A., Murdoch, W.W., Parker, K.R. 1986. Environmental impact assessment: pseudo-replication in time? Ecology, 67, 929-940.
Timoney, K. P., and Lee, P. 2009. Does the Alberta tar sands industry pollute? The scientific evidence. Open Conservation Biology Journal, 3, 65-81.
Underwood, A.J., M.G. Chapman. 2003. Power, precaution, Type II error and sampling design in assessment of environmental impacts. Journal of Experimental Marine Biology and Ecology, 296, 49-70.
Wrona, F. J., Culp, J. M., and Prowse, T. D. 2010. Basin Management Approaches used in a High-latitude Northern Catchment - the Mackenzie River Basin. In. Handbook of Catchment Management. West Sussex, Blackwell Publishing Ltd., pp. 476-499.
Yang C, Wang ZD, Hollebone B, Brown CE, Yang ZY, Landriault M, Fieldhouse B. 2010. Preliminary fingerprinting analysis of Alberta oil sands and related petroleum products. 33rd Arctic and Marine Oilspill Program (AMOP) Technical Seminar on Environmental Contamination and Response. Halifax NS, June 7-9, 2010.
http://www.ec.gc.ca/scitech/default.asp?lang=En&n=A0477462-1
Table 3. Specific details on selected monitoring sites for the Water Quality Monitoring program. (Green shading = “Baseline/Reference” Sites; Red = Sites affected by development; Dark Yellow = New Sites; Light Yellow = proposed requirements for the new program)
Table 5. Locations and sampling details on current and proposed monitoring sites for atmospheric deposition studies. Further details on measurements by WBEA at their sites are provided in Appendix D.
Additional Site-Specific Details for the Lower Athabasca River and Major Tributary Catchments
M0 – Athabasca River at Athabasca
M2 – Athabasca River upstream of Fort McMurray
M3 – Athabasca River downstream of Fort McMurray and confluence with the Clearwater River
M9 – Athabasca River downstream of Firebag River and near Embarras
CH1 – Christina River (Athabasca sub-tributary near confluence with the Clearwater River)
CL1 – Clearwater River
PO1 – Poplar Creek
ST1 – Steepbank River
MU1 – Muskeg River
MA1 – MacKay River
EL1 – Ells River
CA1 – Calumet River
FI1 – Firebag River
Similar to the hydrology/sediment program, tributary monitoring sites were identified to address the following criteria:
M4 – Athabasca River downstream of Steepbank River and upstream of Muskeg River
M5 – Athabasca River downstream of Muskeg River and upstream of MacKay River
M6 – Athabasca River downstream of MacKay River and upstream of Ells River
M7 – Athabasca River downstream of Ells River and upstream of Tar River
Three monitoring sites on the Clearwater River – to separate the Clearwater into two reaches to look at the effect of input from the Christina and Hangingstone on the Clearwater.
Proposed tributary sites CH1, CL1, ST1, PO1, MU1, MA1, EL1, CA1 and FI1 are the same as for the proposed hydrometric/sediment monitoring.
CL2 – Clearwater upstream of Hangingstone River
CL3 – Clearwater upstream of Christina River
HA1 – Hangingstone River
HO1 – Horse River
MU2 – Muskeg R upstream of Canterra Road Crossing
MU3 – Muskeg River downstream of Alsands Drain
MU4 – Muskeg River upstream of Jackpine Creek
MU5 – Muskeg River upstream of of Muskeg Creek and downstream of Stanley Creek
MU6 – Muskeg River downstream of Wapasu Creek
MU7 – Muskeg River upstream of Wapasu Creek
JA1 – Mouth of Jackpine Creek
JA1 – Mouth of Jackpine Creek
JA2 – Upper Jackpine Creek
NM1 – Mouth of North Muskeg Creek
PE1 – Mouth of Wapasu Creek
KL1 – Kearl Lake
MA2 – MacKay River above Oil Sands development
DO1 – Upper Dover River Catchment
EL2 – Upper Ells River catchment
TA1 – Upper Tar River Catchment
CA2 – Calumet River upstream of oil sands development
FI2 – Firebag River upstream of Oil Sands Development
UN1 – Unnamed Creek near mouth
1 Since 202, total mercury (Hg) has been measured to ultra-trace levels (i.e., method detections limit of 0.000012 mg/L, or 1.2 mg/L).
1 CCME 4-fraction test added in 2004; AEP Tier-1 variables (i.e., TVH, THE, TRH) eliminated in 2005
2 10-day L. variegatus test eliminated in 2004
Table 6. Summary of Current Atmospheric Monitoring Stations/Programs in the Oil sands Region
The sampling frequencies for all the samplers are:
PACs/metals – 24 hr integrated sample, midnight to midnight, 1day in 6
Precip samples – monthly samples, event based
WBEA monitors – 1 hour
Weather sites – 1hour
Passive samplers (gaseous PACs +WBEA): 1 or 2 months based on helicopter schedule and weather conditions