Dataset for the assessment of metallic pollution in the Saint-Charles River sediments (Québec City, QC, Canada)

This Data in Brief article presents sedimentological and geochemical parameters from a set of sedimentary samples collected in the Saint-Charles River, a tributary of the Saint-Lawrence River flowing in Québec City (QC, Canada). It details the experimental design, methods, materials and results of destructive analyses related to a multi-proxy study of polymetallic contamination in sediments collected within an urban reservoir (Spatial and temporal patterns of metallic pollution in Québec City, Canada: Sources and hazard assessment from reservoir sediment records, https://doi.org/10.1016/j.scitotenv.2019.04.021, (Chassiot et al., 2019)). The present article summarizes the results of relevant parameters on a set of 68 samples: total organic carbon (TOC), sulfur content, grain-size, and concentrations of heavy and trace metals. It also presents the calculation of enrichment factors, geoaccumulation indexes, and metallic pollution index.


Data
Data presented in this article are related to a multi-proxy study of pollution in the sediments of the Saint-Charles River, a tributary of the Saint-Lawrence River flowing in Qu ebec City [1]. The present article focuses on destructive analyses used to acquire sedimentological and geochemical data, in complement to non-destructive analyses and age-depth model presented in Chassiot et al. [1]. Sedimentological and geochemical data include total organic carbon (TOC), sulfur (S), grain-size, and heavy and trace metals content for silver (Ag), arsenic (As), cadmium (Cd), cobalt (Co), chromium (Cr), copper (Cu), mercury (Hg), manganese (Mn), molybdenum (Mo), nickel (Ni), tin (Sn), lead (Pb), vanadium (V), and zinc (Zn).
A total of 68 samples is presented. Among them, a first dataset of 39 samples (Table 1) includes 6 surface samples collected at the intersection between the Saint-Charles River and its tributaries (JAU, NEL, LOR, BER, CAR, and LAI), 3 surface samples collected in the downstream section (VER, DRA, and FLE), and 30 samples (A, B, and C) extracted from a series of short-cores collected in the river channel (RSC16-01 to À08, BER16, and FLE17). The second dataset consists in 29 samples extracted from longcore RSC17 (Table 2) to document the historical distribution since the creation of the reservoir in the early 1970s [1].
This article also includes the calculation of three pollution indexes: enrichment factors (EF), geoaccumulation indexes (Igeo), and the metallic pollution index (MPI) for the two datasets displayed in Tables 1 and 2, respectively. Contamination categories for EF and Igeo are listed in Table 3. Results and interpretations of EF, Igeo, and MPI are presented in two Excel sheets in supplementary data.

CHNS analyzer
Total Carbon (TC) and Total Organic Carbon (TOC) contents were determined using a CHNS analyzer TruSpec® Leco 932 (catalytic combustion method and infrared detection), with a Limit of Detection Value of the data A geochemical and sedimentological dataset to document metallic pollution and associated environmental hazards in Qu ebec City. A dataset to be considered for local restoration plans and urban management policies, as well as pollution issues within the Saint-Lawrence Estuary.
A benchmark for future studies dedicated to pollutants in urbanized environments across Canada. A support for multi-disciplinary research in urban centers and urban reservoirs.

Table 1
Geochemical and sedimentological parameters in surface sediment and short-core samples, including heavy and trace metal content in mg/kg, TOC (%), S (mg/kg), and fine fraction (silts and clays) in %. The reference sample represents the background geochemistry of the studied area [1]. Data are listed following an upstream-downstream transect. A, B, and C refer to top, middle, and bottom-core samples, respectively [1]. Limits of Detection (LOD) include analytical precision and dilution factors. n.a. ¼ not analyzed. Int. Fe ¼ analytical interference with iron.  Table 2 Geochemical and sedimentological parameters in RSC17 sediment samples (see Table 1 for the signification of abbreviations, and [1] for age-depth model). (LOD) of 0.05% and a Limit of Quantification (LOQ) of 0.17%, respectively. The sample set was first dried during 24h at 50 C and then analyzed for the assessment of TC content. The same set was used for TOC measurements by using silver capsules. They were placed on a plastic plate with small numbered wells. The samples were then moistened with about 20mL of ELGA water which allowed acidification. The plate was then placed in a sealed glass desiccator in the presence of a small beaker containing about 25mL of concentrated HCl. The samples were exposed to HCl steam for 4 hours at room temperature. They were then removed and placed in the oven for 1 hour at 50 C to remove HCl and water residues. The capsules were then closed and placed in the CHNS analyzer without reweighing. Analyses were performed in duplicates using PACS-2 (Marine sediment) and OAS as standard reference materials for control. For additional information about certified reference values, the reader is referred to supplementary data.

Grain-size analyses
Grain-size analyses were performed by sieving the coarse fraction using apertures of 16, 11.3, 8, 5.6, 4, 2.8, 2, 1.4, and 1 mm. Laser diffraction was performed without pretreatment to characterize the fraction under 2 mm in duplicate or triplicate using a Horiba® LA-950 Laser Particle Size Analyzer. Data were then combined and interpreted using the Folk and Ward method [2] in the GRADISTAT Excel spreadsheet [3] to extract parameters such as silt and clay contents and d50.

ICP-AES
Total acid attacks were performed on ca. 0.1 g of crushed sediment by mixing 4 ml of HNO 3 with 1.6 ml of HClO 4 , and 2 ml of HF in Teflon tubes completed to 15 ml with ultrapure water. The quantification of major elements and trace metals, except for mercury, have been performed using an Inductively Coupled Plasma Atomic Emission Spectrometry (ICP-AES) Varian X® with multi-elements solutions, reference materials and sample replicates. LSKD-2, LSKD-4 (Lake sediments) and Buffalo RM8704 (River sediment) were used as certified reference materials (SI).

AAS
Mercury (Hg) content was analyzed on ca. 50 mg of dried powders following thermal decomposition, amalgamation, and Atomic Absorption Spectroscopy (AAS) analyses using a DMA-80 with an instrumental LOD of 0.005 ng/g of sediment. Different certified control masses of known concentrations were analyzed to make a calibration curve ranging from 1 to 25 ng of Hg. For each analysis, the sample is heated to 200 C for 1min, then the temperature increases for 1min30s to reach 650 C. This temperature is maintained for another 1min30s. During this time, the Hg steam is captured in the "amalgamator" containing gold, which captures Hg. After 1min30sec. At 650 C, the "amalgamator" is heated to 900 C for 12s, which releases the Hg that goes into the detection cell. Hg is then detected by AAS at 253.65nm. This method allowed to determine a mean LOD of 0.03 ng/g of sediment for the whole dataset, which varies according to the mass and the Hg concentration of each sample.

Pollution indexes
The assessment of pollution was made by calculating Enrichments Factors (EF, equation 1) [4e6] and Geoaccumulation Indexes (Igeo, equation 2) [7,8]. EF and Igeo are both seven classes indexes used to assess a pollution by a single metal (Table 3).
(1) EF (X) ¼ ðX=ðTiÞsampleÞ ðX=ðTiÞref Þ where X and Ti represent the metal and titanium concentrations, respectively, in sample or reference sample in mg kg À1 .
The calculation of EFs requires a reference sample for background geochemical values and a conservative element to normalize geochemical data that can be affected by grain-size effect. The reference sample was provided by sampling a deep layer in a core (LSC17) collected in the lake feeding the Saint-Charles 30 km upstream [1]. According to the age-depth model presented in Tremblay et al. [9], the layer sampled at 85e86 cm depth in core LSC17 predates the European settlement in Canada and was thus targeted to evaluate natural background concentrations for metals [1]. The affinity of Ti for fine sediments was first suggested from Itrax® data [1], and then confirmed when plotting Ti inferred from ICP-AES analyses versus grain-size. Fig. 1 shows the relationship between Ti and d50 is negative (y ¼ -0.1883x þ 863.54; r ¼ 0.79) and significant (p < 10 À4 ). This relationship is even stronger (r ¼ 0.83) when two outliers are removed from the dataset.
We inferred the extent of polymetallic contamination for each sample by calculating Metallic Pollution Index (MPI, equation 3) [10]. MPI values > 1 indicate pollution whereas MPI values < 1 indicate no pollution.
where M represents the metal concentration whereas n indicates the number of metals considered. acknowledged for assistance in analytical procedures: St ephane Pr emont, Lise Rancourt, Anissa Bensadoune, Brigitte Patry, Philippe Girard and Jean-François Dutil.