ISSUE 20

Synthesis of a hydrological, water chemistry and contaminants research program in the Peace-Athabasca Delta

Informing the long-term monitoring of shallow, flood-prone lakes

The Water Institute

Contact

Roland Hall

Roland Hall
Department of Biology

Laura Neary

Laura Neary
Department of Biology

Introduction

Evidence-based stewardship of aquatic ecosystems requires long-term monitoring data to understand the timing and magnitude of environmental change and potential causes. The Peace-Athabasca Delta in northeastern Alberta, Canada, is a Ramsar Wetland of International Importance within Wood Buffalo National Park where concern for aquatic ecosystem degradation has elicited requests by Indigenous, national and international governance bodies for implementation of a long-term lake monitoring program to track changes in hydrology and water quality attributable to upstream energy projects, climate change and or other processes. Challenges imposed by the delta's size, hydrological complexity, and inaccessibility have long impeded implementation of a delta-wide lake monitoring program.

This study reviews results from seven years of intensive, multifaceted research at rivers and shallow lakes spanning the deltas broad hydro-ecological gradients to inform an integrated monitoring program. The study synthesizes research findings to identify ecosystem changes and probable causes, and provides recommendations for implementing a lake monitoring program in the delta or other hydrologically complex landscapes where shallow lakes are exposed to multiple stressors.

Methodology

The study reviewed research that included systematic collection of water depth measurements and water, sediment and biota samples during the period 2015 to 2021 at 60 spatially dispersed lakes spanning the hydrological gradient from open- to closed-drainage and nine rivers (Figure 1). Key evidence from hydrological measurements (water isotope composition, water depth variation), water chemistry and concentrations of oil-sands-indicator metal(loid)s in sediment and periphytic biofilm was synthesized and integrated.

Figure 1 - Map of Peace-Athabasca Delta

Figure 1. Lower map shows locations of the Peace-Athabasca Delta within the Peace River and Athabasca River watersheds and Wood Buffalo National Park, the upstream hydroelectric dams on the Peace River (solid black diamonds), and oil sands mining project boundaries (1985–2015) in the Lower Athabasca River watershed (orange). Upper map shows locations of the lakes (solid black circles) and river (solid black triangles) sites sampled during 2015–2021.

The absence of historic concentration data for substances of concern was identified as a key limitation challenging the ability to determine the extent of contamination by oil sands development. Paleolimnological analyses was used to address this limitation, define pre-disturbance baseline concentrations and quantify the enrichment of contaminants since onset of oil sands development.

Outcomes

The study noted that approaches capable of tracking the influence of floodwaters (eg, ice-jam floodwaters, open-water floods) and meteorological conditions (eg, rainfall, snowmelt, evaporation) over space and time are essential for monitoring shifts in lake water balances and depth and informing water resource management decisions. The study demonstrated that geospatial interpolation of information obtained from systematic, repeated spatial surveys of water isotope composition in the form of maps called ‘isoscapes’ effectively tracks changes over space and time in lake water balance, and integration with hourly water-depth measurements helped to discern the influence of floodwaters, precipitation and evaporation on lake water balance during the study period.

Systematic measurement of lake water clarity and chemistry, and analysis by principal components analysis (PCA) ordination, provided a means to evaluate ecosystem changes mediated by hydrological processes. Geospatial interpolation of PCA results as “limnoscapes” provided a complementary data visualization to isoscapes in illustrating spatiotemporal variation in lake water chemistry and its strong association with lake water balance (Figure 2).

Figure 2 - Isoscapes of lake water balances

Figure 2. (a) Isoscapes of lake water balances showing interpolated evaporation-to-inflow (E/I) ratios that are based on water isotope composition. White contour lines identify regions where E/I exceed 1.0, whereas black lines identify regions where lakes received river floodwater. (b) Pie charts showing the proportion of lakes that fell by lake level category based on hourly water depth measurements during open-water seasons of 2018–2021. (c) “Limnoscapes” showing interpolated principal component analysis scores of the water chemistry data. Blue hues indicate higher scores typical of flowing rivers, open-drainage lakes and recently flooded perched lakes; green hues indicate lower PCA scores typical of non-flooded perched lakes.

The study synthesized results of oil-sands-indicator metals nickel (Ni) and vanadium (V) in surface sediment and periphytic biofilm and compared them with pre-disturbance baseline concentrations derived from analyses of radiometrically dated sediment cores. Ni and V concentrations were close to the pre-disturbance baselines and mostly within the 95% prediction intervals, indicative of little to no enrichment (Figure 3).

Scatterplots showing linear relations between (a) pre-disturbance concentrations of oil sands indicator metals (nickel and vanadium) and geochemical normalizer aluminum.

Figure 3. Scatterplots showing linear relations between (a) pre-disturbance concentrations of oil sands indicator metals (nickel and vanadium) and geochemical normalizer aluminum. Peace River and Athabasca River regression lines (blue and orange lines, respectively) and 95% prediction intervals (blue and orange shaded area) are based on concentrations in lake sediment (closed circles) and river levee samples (open circles) deposited before ∼1920. (b) Metal concentrations in lake surface sediment (brown) and periphyton (green) from the Peace (circles) and Athabasca (triangles) sectors collected in 2017 and 2018.

Indices were used to assess and communicate relationships among water balance, water chemistry and contaminant enrichment and their relationship with hydrometeorological conditions. The average and range of environmental conditions can serve as reference points for comparison against future monitoring data to quantify the extent of change caused by contaminant emissions and hydrological processes (Figure 4).

Figure 4. Distribution of index scores

Figure 4. Distribution of index scores at each study lakes for (a) water balance (based on lake E/I ratios), (b) water chemistry (PCA axis 2 scores) and (c) contaminant enrichment (Ni and V EFs). The median is shown as a horizontal black line and mean as open circles in (a) and (b). The Contaminant Enrichment Index in (c) is based on enrichment factors (EF) computed for surface sediment collected from all lakes in 2017 (solid circles) and 20 recently flooded lakes in 2018 (open circles) relative to pre-disturbance baseline (orange = Athabasca sector, blue = Peace sector).

The research revealed associations between climatic conditions and lake water balance in the delta. Generalized additive model (GAM) trendlines of time series of water balance and water chemistry indices for the study lakes displayed common patterns with climate indices for the Pacific Decadal Oscillation and El Niño Southern Oscillation, demonstrating the sensitivity and predictability of lake ecosystems to shifts in weather (Figure 5).

Figure 5 Timeseries

Figure 5. Timeseries for (a) climate indices for the Pacific Decadal Oscillation Index (positive values in pink, negative in teal) and the Oceanic Niño Index (black dots) and (b) water balance index and (c) water chemistry index for lakes in the Peace (blue) and Athabasca (orange) sectors. Trendlines in the plots (solid black lines) were generated using generalized additive models (GAMs).

Conclusions

The study reviewed and synthesized seven years of research results to guide design of a long-term aquatic ecosystem monitoring program that addresses concerns related to upstream industrial development for lakes of the internationally recognized, hydrologically complex Peace-Athabasca Delta. As the research coincided with a period of marked hydroclimatic variation, it provided an opportunity to evaluate the effectiveness of methodologies and metrics responsive to changes in hydrometric, meteorological and large-scale climatic drivers. Results revealed that geospatial interpolation as isoscapes and limnoscapes effectively captured the considerable spatial and temporal variation in lake water balances and water chemistry that closely tracked shifts in hydroclimatic conditions. Results also confirmed that concentrations of key indicator metals of oil sands development remain within the range of natural variation that existed before 1920.

Four key recommendations were offered to guide development of a long-term aquatic monitoring program in the Peace-Athabasca Delta other hydrologically dynamic and complex landscapes where abundant shallow lakes are threatened by multiple potential stressors:

Monitor a sufficient number of well-dispersed lakes that span the delta's hydrological gradients to support geospatial interpolation of results.
Collect samples for water isotope composition at seasonal intervals and deploy water depth data loggers every year.
Perform systematic water chemistry surveys at three-year intervals and targeted sampling shortly after widespread flooding of lakes.
Collect surface sediment annually at large, centrally-located, river-connected Mamawi Lake and undertake targeted collection at other lakes after widespread flooding.

Windy river . Laura Neary . Delta