Water quality parameters and constituent concentrations measured in the Peel and Arctic Red Rivers, 2007–2010

Although river floodplains in all biomes are complex interfaces between the terrestrial and marine environments (Bianchi and Allison 2009), those located in the circumpolar Arctic are especially complex because of the extreme seasonal variations in temperature, solar exposure, ice cover, and river discharge they experience. The Mackenzie River Delta, which sits at the confluence of the north-flowing Mackenzie River and the Beaufort Sea basin of the Arctic Ocean, is the second largest delta by surface area in the circumpolar Arctic and contains over 45 000 lakes that collectively cover almost half of its 13 000 km² surface area (Emmerton et al. 2007). The quantity and quality of river-borne constituents that are ultimately discharged from the Mackenzie River Delta to the Beaufort Sea coastal zone are strongly affected by processes that occur during passage through its floodplain (Emmerton et al. 2008; Gareis 2018), with downstream effects on the food webs and ecology of the nearshore Beaufort.

From 2007 to 2010, a team of researchers from across Canada conducted an International Polar Year project titled “The Study of Canadian Arctic River-delta Fluxes” to better characterize the seasonal hydrology and biogeochemistry of the Mackenzie River and its delta. The overall objectives of the IPY-SCARF project were to (1) fully characterize the seasonal hydrology of the Mackenzie River and the Mackenzie River Delta, (2) assess seasonal variations in the flux and quality of nutrients and sediment carried in the Mackenzie River, and (3) determine the fluxes of nutrients and sediment that are ultimately discharged to the Beaufort Sea basin of the Arctic Ocean. To assess the seasonal variations in constituent fluxes and quality, water chemistry samples were taken from 14 sites within and upstream of the Mackenzie River Delta between May 2007 and September 2010 (Figs. 1 and 2).

Here, we present a water quality data set that was collected from the Peel and Arctic Red Rivers during the IPY-SCARF project. The Peel and Arctic Red are tributaries of the Mackenzie River and its delta, and drain mountainous basins that are ecologically and culturally important to the region (Teetł’it Gwich'in First Nation 2003; Canadian Heritage Rivers System 2017). While much of the data that were collected during IPY-SCARF have been presented in other publications (e.g., Nafziger et al. 2009; Hopkinson et al. 2011; Emmerton et al. 2013; Lesack et al. 2013, 2014; Gareis and Lesack 2017, 2020; Gareis 2018), most data collected from the Peel and Arctic Red Rivers have remained unpublished to date. As such, we provide information on the data that were collected from both the Peel and Arctic Red Rivers during the IPY-SCARF project, how they were collected, and their significance. The complete data set is available from the Federated Research Data Repository (Gareis and Lesack 2022).

To date, no publications have directly drawn on the Peel and Arctic Red River data set presented herein, although the following were derived from complementary IPY-SCARF data:

Objective 1: seasonal hydrology of the Mackenzie Delta

Objective 2: nutrient and sediment flux and quantity

Objective 3: fluxes of nutrients and sediment to the Beaufort Sea

In support of the IPY-SCARF project objectives, water chemistry data were collected from the Peel and Arctic Red Rivers upstream of the Mackenzie River Delta (Fig. 1) at irregular intervals between May 2007 and September 2010 (Fig. 2).

Both the Peel River and Arctic Red River are situated in largely mountainous basins underlain by permafrost (Heginbottom 2000; Kokelj 2001), which strongly influence the flow of water in both rivers and their tributaries (Environment Canada 1985; Canadian Heritage Rivers System 2017). The Peel River is formed at the confluence of the Blackstone and Ogilvie Rivers in the Ogilvie Mountains in the central Yukon (Juurand 1986), and drains a 74 000 km² basin (MRBB 2003) before discharging into the southwest end of the Mackenzie River Delta near the community of Fort McPherson, Northwest Territories (NT). The Peel River has an average annual flow rate of 691 m³ s⁻¹ (HYDAT n.d.), and was sampled upstream of Fort McPherson at the location where the Dempster Highway crosses the river via either a commuter ferry (during open water periods) or an ice road (during ice covered periods). The Arctic Red River flows northwest for 430 km from its headwaters in the north Mackenzie Mountains, draining a 18 600 km² basin before reaching its confluence with the Mackenzie River approximately 25 km south of the Mackenzie River Delta (Canadian Heritage Rivers System 2017). The Arctic Red River has an average annual flow rate of 161 m³ s⁻¹ (HYDAT n.d.), and was sampled approximately 0.5 km upstream of its confluence with the Mackenzie River near the community of Tsiigehtchic, NT. In both rivers, the annual maximum flow rates typically occur during the spring freshet (Kokelj 2001; Canadian Heritage Rivers System 2017); however, since both basins are mountainous with steep terrain underlain by permafrost that limits infiltration, high rates of discharge may also occur later in the open water season in response to heavy precipitation events (Kokelj 2001).

A total of 57 samples were taken from the Peel River, and 32 from the Arctic Red River, between May 2007 and September 2010 (Fig. 2). Measurements for each sample included water temperature, specific conductivity, pH, chlorophyll a, total suspended sediments (TSS), particulate nutrients (carbon, nitrogen, and phosphorus), soluble reactive silica, major cations (calcium, magnesium, potassium, and sodium), major anions (chloride and sulfate), dissolved carbon (organic and inorganic), and dissolved nutrients (total dissolved nitrogen, ammonium, nitrate, total dissolved phosphorus, and soluble reactive phosphorus). Due to sampling constraints, analytical constraints, and sample loss, measurements are not available for all parameters on all sampling dates (i.e., there are missing data points).

Our goal was to sample most frequently during periods of high discharge and periods when discharge was changing most rapidly. Our priority was therefore to sample both the Peel and Arctic Red Rivers as frequently as possible during the freshet (May), while samples were taken less frequently during the open water season (June through September). Only the Peel River was sampled during the ice-covered winter season (October through April), and at sporadic intervals, because both constituent concentrations and flow rates were far less variable during winter than during other hydrological seasons.

Sampling methods differed throughout the year based on prevailing weather and hydrological conditions, as described in Gareis and Lesack (2017). During ice-covered periods, samples were taken through a hole in the river ice using a slow-filling integrated sampler that was continuously lowered and raised through the water column (Shearer 1978). During the rising freshet period, while it was still possible to land a helicopter on the river ice sheet, water samples were collected through drilled ice holes using an integrated sampler. When the ice sheet broke up to the point when landing a helicopter on it was no longer possible, near-surface grab samples were taken from the pontoons of a helicopter after landing in an opening in the ice in the middle of the river channel. During the falling freshet and open water periods, the sampling method differed between the two rivers. In the Peel River, samples were taken from the deck of the commuter cable ferry MV Abraham Francis. A 20 L sampling bucket was lowered over the side of the ferry at midriver, rinsed with Peel River water, filled, and hauled back on to the deck. All measurements and samples were taken from the sampling bucket while the water was vigorously mixed to prevent settling of particulates. In the Arctic Red River, measurements and samples were taken after wading as far into the river channel as was safely possible, and samples were taken as near-surface grab samples.

All water samples were collected in precleaned high-density polyethylene (HDPE) water bottles that were stored in cool, dark conditions for less than 8 h before processing at the Aurora Research Institute in Inuvik, NT. Sample water was well mixed and a known quantity was filtered through a Whatman GF/C filter (1.2 µm pore size) for the fluorometric determination of chlorophyll a concentrations following the methods of Wetzel and Likens (2000). Aliquots of sample water were also filtered onto precombusted (16 h at 550 °C) Whatman GF/C filters and oven-dried at 60 °C for later analysis of TSS and particulate carbon (PC), nitrogen, and phosphorus concentrations following the methods of Stainton et al. (1977). GF/C filtration was chosen for the determination of particulate nutrient concentrations to be consistent with historical Mackenzie River and Delta data sets. Testing of GF/C filtration versus filters with smaller pore sizes showed negligible differences in nutrient and particulate concentrations in this system (data not shown), as has been observed in prior studies of the Mackenzie River Delta (Emmerton et al. 2008; Tank 2009). Filtrate that passed through the precombusted GF/C filters was partitioned into precleaned and sample-rinsed bottles for the analysis of dissolved constituents. A subsample of GF/C filtrate was passed through a Millipore membrane filter (GPWP, 0.2 µm pore size) and stored in precleaned and precombusted (16 h at 550 °C) borosilicate vials for analysis of dissolved organic carbon and dissolved inorganic carbon (DIC). Potential loss of CO₂ from DIC samples was negligible, with CO₂ never comprising more than 3% of total DIC during testing on 10 occasions (data not shown). All water quality measurements and constituent concentrations were analyzed following standard analytical protocols in one of four laboratories (see footnotes in Table 1). Additional methodological details on the processing, preservation, storage, and analysis of all constituents can be found in Table 1.

Although a total of 57 water samples were taken from the Peel River during this study, and 32 from the Arctic Red River, some samples were not analyzed for all constituents due to sampling constraints, analytical constraints, and sample loss. All missing data points are recorded in the data set for this study, which is available for download from the Federated Research Data Repository (Gareis and Lesack 2022).

All analytical methods included the following to validate data and ensure its reliability:

This water quality data set for the Peel and Arctic Red Rivers provides information on two waterways that have been historically under-sampled for water chemistry parameters. Sampling of rivers in the region has generally focused on the Mackenzie River, with sampling in the Peel and Arctic Red generally restricted to parameters of concern several times per year. This data set therefore expands the long-term water quality records for both the Peel and Arctic Red Rivers, and contributes information that complements prior and ongoing studies. Some examples of prior and ongoing studies include the Peel River Basin Water Quality Program that measured several basic water quality parameters, as well as contaminants of concern (metals, hydrocarbons, and organochlorines), in the Peel River between 1980–1998 and 2002–2007 (INAC 2008), and the Government of the Northwest Territories Department of Environment and Natural Resources Community-Based Water Quality Monitoring Program, which has collected data from more than 40 sites on 24 rivers and lakes in the territory in partnership with 21 communities since 2012 (GNWT-ENR).

Additionally, this data set includes samples that were taken during historically under-sampled hydrological periods. In 2007 and 2008, samples were taken from the Peel and Arctic Red Rivers during the spring freshet period, when both rivers transition from fully ice-covered to open-water conditions. This transition period is very short and is characterized by rapidly increasing discharge as winter snowpacks melt and flow overland into drainage channels. The freshet is a historically under-sampled part of the hydrological cycle in north-flowing circumpolar rivers, and therefore this data set contributes critical information about freshet water quality in both the Peel and Arctic Red Rivers.

Both the Peel and Arctic Red Rivers flow through the Gwich'in Settlement Area. The rivers themselves, and their surrounding watersheds, are socially and culturally important to the Gwich'in people and are used by community members from both Fort McPherson and Tsiigehtchic, NT. Additionally, the Arctic Red River has been a designated Canadian Heritage River since 1993 in recognition of its rich natural and cultural heritage (Canadian Heritage Rivers System 2017).

The water quality data collected from the Peel and Arctic Red Rivers also provides context and information related to one of the world's great circumpolar river deltas. The Mackenzie River Delta is the second largest delta by surface area in the circumpolar Arctic and contains more than 45 000 floodplain lakes (Emmerton et al. 2007). The delta is far more biodiverse and productive than the surrounding tundra due to the annual flooding it experiences during the spring freshet (Squires et al. 2009), and the water that flows out of the delta supports a productive food web in the nearshore Beaufort Sea (Hsiao 1976; Carmack et al. 2004) that is both culturally and economically important to those living in the Inuvialuit Settlement Region. Although the Mackenzie River accounts for approximately 90% of the total water flow into the delta in all seasons (Morley 2012), the Peel and Arctic Red Rivers exert a local influence in the southern reaches of the delta where both rivers nourish large populations of fish, wildlife, and waterfowl.

Finally, water quality data for the Peel and Arctic Red Rivers provide baseline information in support of climate change research in the region. Climate change is occurring most rapidly at high latitudes, with Arctic regions warming at a rate over four times faster than the global average since 2000 (Chylek et al. 2022). In the western Canadian Arctic, mean annual temperatures have increased by more than 3 °C since 1948 (Zhang et al. 2019). Both the Peel and Arctic Red Rivers are situated in basins that are underlain by permafrost and vulnerable to rapid thaw as temperatures increase (Kokelj et al. 2017). This data set therefore provides useful baseline information for future research that assesses how water chemistry in the Peel and Arctic Red Rivers responds to continued changes in temperature, hydrology (e.g., the timing of thaw/freeze-up, length of thaw season, and increases in rainfall), and permafrost extent (e.g., increased thaw slumps and channel bank erosion).

The research reported in this data paper was conducted in the Gwich'in Settlement Region under NWT Research Licenses 14153, 14512, and 14687 issued by the Aurora Research Institute. Funding to support this research was received from the International Polar Year Canadian Federal Grant Program, the Natural Sciences and Engineering Research Council of Canada, the Polar Continental Shelf Program, and the Northern Scientific Training Program. Field data collection and laboratory analyses were assisted by Larissa Duma, Sean Magee, and Adam McPherson. External laboratory analyses were facilitated by the lab of Dr. Christopher L. Osburn at North Carolina State University and the Freshwater Institute, Fisheries and Oceans Canada. Critical logistical support was provided by staff at the Aurora Research Institute, Gwich'in Renewable Resources Board, Aklak Canadian Helicopters, and the Water Survey of Canada Inuvik office. The authors would also like to thank the two anonymous reviewers, whose comments greatly improved the quality of this data paper.