Using water isotope tracers to delineate river flooding: Results from the Peace-Athabasca Delta

Introduction

Challenges to landscape-scale hydrological monitoring have long prevailed in the Peace-Athabasca Delta (PAD), a complex floodplain in northern Alberta, Canada. Spanning about 6,000 km², the PAD is one of the world's largest inland boreal freshwater deltas that receives input from two major rivers, the Peace River and the Athabasca River. The hundreds of shallow lakes and wetlands that characterize the PAD provide important habitat for a variety of biota and natural resources for Indigenous communities. In recognition of its ecological, historical, and cultural significance, the PAD is a Ramsar Wetland of International Importance.

Lakes and wetlands of the PAD are influenced by river floodwater, snowmelt, rainfall, and evaporation, resulting in a broad spectrum of spatially varying hydrological regimes. As a low-relief system, small changes in water levels can alter hydrological connectivity and set forth temporal patterns of change in limnology, habitats, biodiversity and ecosystem productivity. As freshwater availability has declined over large portions of the PAD, research has focused on ice- jam flood events because of the importance of this hydrological process as a recharge mechanism for lakes and wetlands in the Delta.

The objective of this study was to delineate the extent and magnitude of a spring ice-jam flood event in the PAD that occurred in late April and early May 2018 using water isotope tracers, supplemented by in situ measurements and observations. Building upon previous approaches, the study compared the water isotope data with an isotope framework established using 16 years of measured meteorological conditions and on-line resources and then developed a set of landscape-specific binary mixing models that allow estimation of the proportion of input to flooded lakes attributable to river floodwater and precipitation.

Methodology

During May 15-17, 2018, a set of 68 lakes and nine river sites spanning a range of hydrological conditions across the PAD were sampled with the aid of a helicopter (Figure 1). Water samples were sent for oxygen and hydrogen isotope analysis, while in situ measurements of conductivity were taken and additional field observations were recorded.

FIGURE 1: Locations of the Peace-Athabasca Delta, Alberta, and the 68 lake (black circles) and nine river (black triangles) sites sampled from May 15–17, 2018. Approximate locations of river ice‐jams are indicated in red.

To determine the flood status for each sampled lake, water isotope compositions were compared with river water isotope compositions, considering both values of conductivity and field observations. Lakes were designated as flooded if (a) water isotope values were close to or overlapping with river water isotope compositions, (b) specific conductivity values were close to the range of the river water values, and (c) there was visible evidence of flooding.

A previously developed isotope framework representing average conditions during the period 2000-2015 and a coupled-isotope tracer approach were used to calculate the isotope composition of input water for each flooded lake. A set of four Meteoric Water Line Segments were developed capturing the range of lake input water isotope compositions that result from mixing of either rain or snow with floodwater from the Peace or Athabasca river. Values generated by the equations represent estimates of the proportion of input water attributable to river floodwaters and do not necessarily represent the total amount of river floodwater in the lake at the time of sampling because nearly all lakes contained pre-existing water.

To visualize flood magnitude and explore spatial patterns, the proportion of input water attributed to river floodwater was interpolated between lakes across the surface of the Delta. A sensitivity analysis was performed that incorporated uncertainty in the precipitation end-member isotope compositions to evaluate their influence on flood magnitude estimates.

Outcomes

In total, 44 per cent (30 out of 68) of the sampled lakes received river floodwaters based on consideration of the water isotope compositions, specific conductivity, and field observations, including 28 and 58 per cent of the lakes sampled in the Peace and Athabasca basins, respectively (Figure 2a). The isotope compositions of non-flooded lakes are generally higher than for flooded lakes due to the influence of evaporative isotopic enrichment in the absence of flooding (Figure 2b).

Figure 2: δ¹⁸O‐δ²H graphs showing the water isotope compositions of (a) flooded lakes and (b) not flooded lakes sampled in May 2018. Isotope compositions of rain, snow and the Peace and Athabasca rivers are also shown.

Based on lake specific estimates of the isotope composition of input waters, the proportion of input to lakes attributable to river floodwaters was calculated. Input water for five of the nine flooded lakes in the Peace basin consisted mainly of a mixture of river floodwater and rainfall (Figure 3a), while input water for the other four lakes consisted mainly of a mixture of river floodwater and snowmelt (Figure 3b). In the Athabasca basin, two of the 21 flooded lakes received input water mainly comprised of river floodwater and rainfall (Figure 3c), while the other 19 flooded lakes mainly received input water from river floodwater and snowmelt (Figure 3d).

Figure 3: Isotope composition of input water (δ_I) for lakes flooded in May 2018 assigned to one of four Meteoric Water Line Segments: (a) Peace River—rain, (b) Peace River—snow, (c) Athabasca River—rain, and (d) Athabasca River—snow.

In the Peace basin, the proportion of input water attributable to river floodwater in lakes typically dominated by precipitation in the form of rain was 81.8 per cent, while the proportion of input water attributable to river floodwater in lakes dominated by precipitation in the form of snow was 93.9 per cent. In the Athabasca sector, theses shares are 95.5 per cent and 86.3 per cent, respectively.

Calculations of the proportion of input to lakes attributable to river floodwater were interpolated across the PAD using a radial basis function to visualize flood extent and magnitude. The analysis resulted in different patterns of flooding between the Peace and Athabasca rivers, with the proportion of input water attributed to floodwater decreasing with increasing distance from the rivers and their distributaries (Figure 4a). A sensitivity analysis incorporating uncertainties in the rain and snow end-member isotope compositions in the binary mixing model equations generated small differences in the flood map magnitude estimates (Figure 4b, 4c).

Figure 4: (a) Map of the Peace‐Athabasca Delta showing spatial interpolation of per cent of input water to lakes attributed to river floodwaters in May 2018. Approximate locations of river ice‐jams are indicated in red. (b and c) Maps of the Peace‐Athabasca Delta showing spatial interpolation of per cent of input water to lakes attributed to river floodwaters in May 2018 incorporating sensitivity analyses (Figure 4b = rain δ¹⁸O − 1.08‰ and snow δ¹⁸O − 1.88‰; Figure 4c = rain δ¹⁸O + 1.08‰ and snow δ¹⁸O + 1.88‰).

Conclusions

Ice-jams in late April and early May 2018 resulted in the flooding of lakes in both sub-basins of the PAD, although flood extent and magnitude differed between the Peace and Athabasca rivers. In the Athabasca, river floodwaters originated primarily from the Athabasca and Embarras Rivers and entered lakes along the western portion of these rivers. Given the importance of the Athabasca Delta terminus region to land users, and their concerns about decreasing water levels, confirmation that floodwater is not reaching this area, even during the large ice-jam flood of 2018, is an important contribution to conservation dialogues.

Monitoring data are an essential component of science-based policy. In complex and dynamic floodplain landscapes such as the PAD, hydrological monitoring approaches must be sensitive, timely, flexible and cost-effective. The use of water isotope tracers meets these requirements and as such provides an important complementary approach to capture signals of recent hydrological events and processes. The most important delineation in this study was whether the lake flooded or not, which was effectively identified by the water isotope tracers, supplemented by measurements of conductivity and field observations. Sensitivity analysis demonstrated that incorporating reasonable uncertainty of both the snow and rain end-member isotope compositions does little to change the outcome at the landscape scale, which is most appropriate in this study for ecosystem management purposes. This approach hence provides an effective means to track the extent and magnitude of river flood events in the PAD.