Effect of seasonally-varying hydrology and circulation on transport of trace elements through a moderately-sized reservoir
Richard A. Wildman, Jr.1,2,3,* and Noelani A. Forde2,4
1 Harvard University Center for the Environment
2 Harvard University School of Public Health
3 present address: Quest University Canada, 3200 University Boulevard, Squamish, BC V8B 0N8, Canada
4 present address: Department of Earth, Ocean and Atmospheric Sciences, University of British Columbia, 2020-2207 Main Mall, Vancouver, BC V6T 1Z4, Canada
* e-mail: ; Tel. 604 898 8047
Abbreviated title: Trace Element Transport through a Reservoir
In preparation for Lake and Reservoir Management, last updated 12 June 2015
Abstract
We assessed the effect of Grand Lake, Oklahoma on the transport of Fe, Mn, P, As, Zn, Pb, and Cd through the watershed of the Grand River. We measured filtered and suspended sediment samples collected upstream of, within, and downstream of this moderately-sized reservoir and then used water flow to estimate instantaneous, seasonal, elemental fluxes. In winter and spring, when storms brought high flows to the reservoirs, Grand Lake modified flood water minimally; trace element distributions were determined by the passage of storm inflows through the reservoir. In summer, Fe, Mn, P, and As were enriched in anoxic bottom water, and they were exported out the dam, which draws water from below the surface mixed layer. Fall overturn appeared to decrease water-column concentrations of aqueous elements due to precipitation and settling. Zinc did not track Fe and was instead sequestered in Grand Lake during all seasons. Logistic regression indicated that Zn predicts Cd concentrations, and so Grand Lake probably sequesters Cd as well. Lead was less clear. This study shows that watershed hydrology determines the transport of trace elements through a reservoir during times of high flow but that vertical circulation and biogeochemistry dominate during summertime and autumn low flows. It thus can aid reservoir managers in understanding when and how upstream pollution may be retained by or passed through reservoirs to river systems downstream.
Key words: anoxia, flood, hydrology, reservoir, stratification, trace metals, watershed
Introduction
[1] Reservoirs of moderate or large size influence chemical transport in their watersheds. They are often located on large rivers to maximize the ratio of watershed area to lake area, and so they often receive large loads of dissolved and particulate chemicals relative to lakes of similar size (Kalff 2002). As reservoirs trap suspended sediment, they trap a particulate chemical load (Horowitz et al. 2001, Lee et al. 2001, Castelle et al. 2007, Wildman et al. 2011). Dissolved, inorganic, non-nutrient chemicals (e.g., trace metals) have been characterized in rivers (e.g., Horowitz et al. 2001, Müller et al. 2008), but less attention has been devoted to the modification of the downstream transport of such chemicals by reservoirs.
[2] The hydrology of sizeable reservoirs and their major tributaries can affect chemical transport downstream in two important ways. First, increased flow in tributaries brings substantially increased chemical loads to a reservoir (Horowitz 2008), and so reservoirs might have the largest effect on chemical transport in their watersheds during times of high flow. Second, sizeable reservoirs contain lacustrine zones when tributaries are at base flow (Kalff 2002). In temperate latitudes, these contain stratification in summer and the potential for hypolimnetic anoxia and subsequent reaeration during overturn or release out the dam. This can lead to reductive dissolution of iron-oxide minerals, which can leave the reservoir in dam releases (Ashby et al. 2004). Elements that sorb to metal-oxides (e.g., arsenic and phosphorus) can behave similarly to Fe in reservoirs (e.g., Kneebone and Hering 2000). The overlaid, seasonally-varying influences of watershed hydrology and biogeochemistry suggest that reservoir managers cannot predict with confidence the effect of a specific reservoir on elemental transport in a watershed despite the importance of trace-element cycling for recreational water quality and ecology.
[3] The purpose of this study was to explore the effect of seasonal variation in flow and vertical circulation on the transport of a suite of trace elements through Grand Lake, Oklahoma. We studied metals (Fe, Mn, Pb, Zn, Cd), a metalloid (As), and a nutrient (P) that represent most of the primary abiotic threats to water quality in this watershed. Grand Lake is particularly interesting because it lies at the upstream end of a chain of reservoirs and downstream of multiple sources of contamination (see below). Thus, the transport of trace elements through this reservoir is important for management of water quality for some distance through its watershed.
Study Site
[4] Grand Lake (sometimes known as Grand Lake O’ the Cherokees) is a moderately-sized (36 m maximum depth, >80 km long) reservoir in northeast Oklahoma (Figure 1). Its watershed is 26,600 km2, and <30% falls within another flood control project well upstream, so most of the flow into Grand Lake is unregulated. This implies considerable variation in inflow rates during a year. The median of recorded daily mean inflows in the 10 years preceding this study was 94.1 m3/s (3,322 cfs), although, during our study (December 2010 through November 2011, see below) drought conditions induced median inflows of 34.7 m3/s (1,225 cfs; Wildman, submitted). Small storms in late winter induced inflows from 85–485 m3/s (3,002–17,128 cfs), and two large storms in spring increased inflows above 550 m3/s (19,423 cfs) for less than a week each (Wildman, submitted). Despite these variations, the water level of Grand Lake generally varies <1 m throughout the year because, except when it retains floodwater, outflows match inflows to satisfy a management objective of maintaining consistent water levels for months at a time (D. Townsend, Grand River Dam Authority, personal communication). Consequently, the reservoir volume was near a neutral water balance in December, increasing by 40 megaliters/day (ML/day) in February during storm inflows, decreasing by 16 ML/day in May as water was released following a large storm, and decreasing slightly in August and November as part of minor water-level drawdowns (Wildman, submitted).
[5] Inflows to Grand Lake come primarily from the Neosho River in the northwest, Spring River in the north, and Elk River in the east, which contributed 75% of the water released from the dam during this study. The former two of these drain predominantly agricultural land. The Elk River drains forested uplands and receives permitted discharge from poultry plants. In the northwest, Tar Creek drains the Tar Creek Superfund Site, a historic Pb and Zn mining district with abundant metal-rich mine tailings (Schaider et al. 2007, Andrews et al. 2009). In the southeast, several industrial-scale chicken farms, which might release arsenic to the environment (Hilleman 2007), operate in the Honey Creek watershed.
[6] The confluence of the Neosho and Spring Rivers, which has been submerged by Grand Lake, was the historical origination of the Grand River; now, the Grand River originates from Pensacola Dam. Penstocks of the dam, which are used for the entirety of dam releases except during floods, are 4 m tall screened openings centered ~16 m below the water surface and located at the southwestern end of the dam.
[7] Grand Lake is a warm, monomictic reservoir. During our study, it turned over in October, reached a vertically-mixed minimum temperature of 5 C in winter, and exhibited stratification by early May. In mid-summer, stratification was pronounced; the surface mixed layer was 8–11 m thick (Wildman, submitted). Grand Lake is eutrophic (OWRB 2009). Its metalimnion and hypolimnion are anoxic in summer, and the entire water column readily becomes oxic upon overturn in autumn (Wildman, submitted).
Methods
[8] Water samples were collected during excursions in December 2010, February/March, May, August, and November 2011 from 4 (December and November) or 5 (February-August) locations spaced roughly evenly across the thalweg of Grand Lake, the tributaries described above, and the Grand River <500 m below the dam. Based on circulation parameters that were measured concurrently, 1–4 depths at each site were sampled. One sample was always collected from the surface mixed layer (SML); additional samples were collected from below the SML and from water ≤1 m from the sediment-water interface. Samples were pumped through silicone tubing, filtered through pre-weighed 0.45 μm polyethersulfone membranes either within 24 h in a class-100 clean bench (December-May) or immediately inline (August-November), and acidified with ultrapure nitric acid within 12 h. Field blanks and occasional field duplicates verified the low background concentrations of analytes and reproducibility of field protocols.
[9] Aqueous-phase elements were quantified by inductively-coupled plasma (ICP) mass spectrometry (Mn, Pb, Zn, Cd, As) and ICP optical emission spectrometry (Fe, P). Measurements were calibrated with standard solutions made by diluting commercially-available stock solutions. Detection limits were 0.1 μg/L for Mn, Pb, Zn, Cd, and As and 1 μg/L for Fe and P.
[10] Suspended-sediment-associated elements were quantified after extraction from filter membranes, which were first oven-dried (at <65 C until mass was constant) and weighed to determine sediment mass. Membranes were subsequently microwave-digested in concentrated ultrapure nitric acid. Digestates were diluted 1:10 with water and analyzed like the water samples described above. Further 100-fold or 1000-fold dilutions were usually necessary; 5% ultrapure nitric acid was used. Digestion efficiency was verified by digesting NIST 2709 and 2711 certified reference materials several times along with batches of samples; filter and liner blanks verified trace-metal-clean techniques. Detection limits of these measurements varied between samples because, although the ICP detection limits were constant, the digestate volumes, sediment masses on filters, and volumes of water filtered were not. Detection limits varied between 0.0–71.9 mg/kg with a median of 0.5 mg/kg and an IQR of 0.2–1.7 mg/kg when expressed as mass of element per mass of bulk solid.
[11] Elemental fluxes were calculated for each sampling excursion by multiplying the average flow of the days when sampling occurred by the concentrations of elements in samples collected from tributaries and the dam tailrace and then subtracting the latter from the sum of the former. When concentrations were below detection limit (BDL), half the detection limit was used for this calculation.
[12] Contour plots were created from measurements at discrete depths and distances from the dam using Tecplot 360 (Tecplot, Inc.; Bellevue, Wash.). It was necessary to interpolate vertically based on circulation parameters before allowing Tecplot to interpolate longitudinally between sampling locations. Relationships between trace elements and qualitative predictor variables (e.g., depth, location, season) were explored using principal component analysis (PCA). This statistical technique expresses the variance of a many-dimensional dataset not on axes that correspond to individual variables but on new, orthogonal axes called principal components (PCs) that are aligned with successively decreasing fractions of the variance of the dataset (see Shine et al. 1995 for excellent background and a previous implementation of this technique). This allows visualizations of groupings of both variables and samples based on the concentrations of the trace elements measured in this study. In filtered samples, when some elements were too often BDL to permit useful inclusion in PCA, relationships with other elements were instead explored with multivariable logistic regression. These analyses treated trace elements for which many above-BDL concentrations were available as continuous predictor variables for low-concentration elements that were evaluated as bimodally-distributed (i.e., above or below the detection limit) response variables.
Results
Concentrations of Trace Elements
[13] Iron in filtered samples (henceforth, “Fef” and similar) ranged from BDL to 220 μg/L with a median of 15 μg/L and an interquartile range (IQR) of BDL–33 μg/L (Figure 2). Across all sampling excursions, concentrations in tributaries did not differ meaningfully from those in the reservoir. Concentrations were insignificantly lower below the dam. Elevated concentrations of Fef occurred in anoxic summertime bottom water and in the large inflow captured in the February/March sampling, indicating that, during different seasons, aqueous-phase Fe enters the reservoir from both internal loading and the watershed upstream. Otherwise, spatial variation was minor (Figure S1, where “S” denotes the Supplement). Oxic waters were not devoid of Fef; although concentrations in oxic water were BDL in autumn and winter, they were 10–35 μg/L in warmer months.
[14] We observed a general resemblance between Mnf, Pf, Asf, and Fef. Anoxic bottom water contained 1–4 mg Mnf/L, which far exceeded Fef concentrations in the same samples; the IQR of Mn for all samples was 2.2–56.9 μg/L (Figure 2). Hypolimnetic anoxia in summer led to concentrations 2–3 orders of magnitude higher than in tributaries and downstream. Dam releases were 93 μg/L and 22 μg/L in summer and autumn, respectively; the former value exceeds the United States secondary drinking water standard (50 µg/L) (US EPA 2013), which provides a useful comparison for this concentration. The IQR of Pf concentrations was 22–140 μg/L (Figure 2). It was higher in the reservoir than in tributaries during much of the year and especially in summertime bottom water. Concentrations of Asf were <5 μg/L (Figure 2). Anoxic hypolimnetic waters reached 4.7 μg/L of Asf; concentrations were low (1–2 μg/L) and invariant otherwise. Aside from variation with depth as described here, spatial variation of these elements was minor (Figures S2–4).
[15] Elements derived from mines upstream behaved differently than Fef, Mnf, Pf, and Asf. Anoic water did not show an enrichment of Znf, unlike Fef, Mnf, Pf, and Asf were (Figure 2). Its concentrations were elevated well above 1000 μg/L in Tar Creek and >100 μg/L in the main tributaries of Grand Lake and in the upper part of the reservoir. Concentrations of Znf were consistently higher in the tributaries feeding the northern end of Grand Lake than in our sampling location >30 km downstream (99 vs. 24 µg/L in December, 157 vs. 51 µg/L in February, 42 vs. 8 µg/L in May, and 5 vs. 0.9 µg/L in August). Concentrations in the rest of the reservoir were <15 μg/L with no clear spatial pattern (Figure S5). The highest concentrations of Pbf and Cdf occurred in Tar Creek in February; values were 1.9 µg/L and 1.7 µg/L, respectively. Otherwise, these elements were usually BDL in filtered samples.
[16] The IQR of Fe in suspended sediment (henceforth, “Fess” and similar) was 7,800–24,600 mg/kg with extrema an order of magnitude lower and higher, respectively (Figure 3). Highest values occurred in bottom water during autumn, and these high concentrations also occurred downstream in autumn (Figure 3). Otherwise, suspended-sediment-associated concentrations in the reservoir and below the dam resembled those in the tributaries, which did not vary appreciably throughout the year (Figure S6).