Chemical data of contaminants in water and sediments from the Doce River four years after the mining dam collapse disaster

Chemical datasets describing the occurrence of both inorganic and organic contaminants along the Doce River Basin (DRB) could provide a better understanding of the potential impacts of a major mining dam collapse disaster combined to additional chronic sources of contamination. This data article presents datasets of main contaminants detected in the water and sediments sampled four years after the mining dam collapse in the DRB. A summary table of data obtained in the literature is also provided to allow a comparison of the variation of chemicals before, right after in 2015/2016 and after the event (current data). In addition, there are also provided physical-chemical parameters of water and sediments of different sampling sites, which could support the investigation of chemicals distribution. For this purpose, triplicate samples of water and sediment were obtained in 8 sampling sites along the DRB during wet and dry seasons of 2019, totalizing 48 samples of each environmental matrix. The sampling sites were strategically selected according to their different main sources of pollution along the river. Concentrations of trace elements and organic contaminants (polycyclic aromatic hydrocarbons, and pyrethroids) were determined in samples of water and sediments by inductively coupled plasma mass spectrometry (ICP-MS) and gas chromatography - mass spectrometry GC-MS, respectively. Main data obtained in the literature consisted in published reports from environmental agencies (IGAM) and private companies (RENOVA) as well as journal articles. The datasets provided may be useful to the stakeholders, which include scientific community, authorities and public agencies, and private companies interested to understand the impacts of the contaminants introduced along the River Basin four years after the environmental disaster.

of water and sediment were obtained in 8 sampling sites along the DRB during wet and dry seasons of 2019, totalizing 48 samples of each environmental matrix. The sampling sites were strategically selected according to their different main sources of pollution along the river. Concentrations of trace elements and organic contaminants (polycyclic aromatic hydrocarbons, and pyrethroids) were determined in samples of water and sediments by inductively coupled plasma mass spectrometry (ICP-MS) and gas chromatography -mass spectrometry GC-MS, respectively. Main data obtained in the literature consisted in published reports from environmental agencies (IGAM) and private companies (RENOVA) as well as journal articles. The datasets provided may be useful to the stakeholders, which include scientific community, authorities and public agencies, and private companies interested to understand the impacts of the contaminants introduced along the River Basin four years after the environmental disaster.  Table   Subject Environmental Sciences -Pollution Specific subject area Environmental contaminants -Detection and quantification of inorganic and organic contaminants in field samples of water and sediment by analytical chemistry. Type of data Tables  How the data were acquired Physical-chemical parameters of water Water -Data available in the literature from RENOVA ( https://www.fundacaorenova.org/arquivos-e-relatorios ) and IGAM websites ( http://www.igam.mg.gov.br/monitoramento-da-qualidadedas-aguas2/monitoramento-da-qualidade-das-aguas-superficiais-do-riodoce-no-estado-de-minas-gerais ).
Sediment properties Sediment -Data generated from sediment samples obtained from the Doce River (MG/ES, Brazil). The analysis of composition and grain size distribution of sediment was performed following the wet and dry sieving method. The quantity of Organic Matter (OM) was determined through the method of wet organic matter oxidation. Chemical analyses Inorganic chemicals Data were acquired by an inductively coupled plasma mass spectrometry (ICP-MS) Organic chemicals Data were acquired by GC-MS and HPLC-DAD Data format Raw Analyzed Description of data collection Data was generated from the analysis of samples of water and sediment obtained in two field surveys of the Doce River, located in Southeast Brazil, during wet and dry seasons of 2019. In each survey, eight sampling sites were established across the DRB, and three replicates of water and sediment samples were collected, in both wet and dry seasons.

Value of the Data
• Relevant datasets are provided for screening assessment of major inorganic and organic contaminants occurring along the Doce River basin, and their distribution, three to four years after the mining dam collapse disaster • Data exploration may be useful to distinguish the impacts from multiple sources of pollutants present along the river that have been affecting the river basin before and after the event, and thus better clarifying the distribution of these contaminants four years after the event. • The datasets are relevant for performing a more appropriate environmental diagnostic to better predict further ecological and human risks of exposure, including the cumulative effects due to the disaster and the contribution of other pollution sources. • The data will be useful for the assessment of public authorities (environmental agencies), riverine communities, private companies affected by the mining dam disaster, as well as scientific community, which are concerned by the impacts of the mining dam rupture and are still investigating such impacts. • The data can be further explored by the application of environmental diagnostic indexes and multiple statistical tools to evidence the major contaminants of concern, the most impacted regions and identify the main sources of pollution (causes) that are affecting the local water quality. • Chemical data can be associated and integrated with biological data to understand potential impacts of these chemicals in exposed organisms [2] . Finally, these data can be used in comparisons focusing the contamination levels before and after the dam rupture.

Review of Literature Data
Table S1 -Water: summarizes the data obtained in the literature by Pubmed database, describing the levels of inorganic chemicals present in the water along different sites of the Doce River Basin before and after the dam rupture event in 2015.
Table S2 -Sediments: summarizes the data obtained in the literature by Pubmed database, describing the levels of inorganic chemicals present in sediments along different sites of the Doce River Basin before and after the dam rupture event in 2015. Table 1. Physical-chemical parameters of the water obtained from RENOVA, described for six different sites along the DRB during wet and dry seasons of 2019. Table 2. Physical-chemical parameters of water obtained from IGAM, described for five different sites along the DRB during wet and dry seasons of 2019. Table 3. Granulometry and Organic Matter content of sediment samples surveyed along the DRB during wet and dry seasons of 2019. Table 4. Concentrations of main inorganic chemicals detected in the water (in mg.L −1 ) of the DRB during wet and dry seasons of 2019, by Procedure 1. Table 5. Concentrations of main inorganic chemicals detected in the water (in mg.L −1 ) of the DRB during wet and dry seasons of 2019, by Procedure 2.

Materials and methods
The The seasonal variation was also assessed in two sampling surveys performed during the wet season (Jan -Feb) and dry season (Jul-Aug) of 2019. The water samples (3 L per sampling site and season, divided in three field replicates of 1 L each) were directly collected with the own bottle for storage, at a depth approximately below 30 cm from the surface, in three different sites of each location. For the analysis of inorganic contaminants, the water samples (1 L per sample in triplicate) were acidified with HNO 3 0,5% within 24 h after sampling, and stored in new polypropylene bottles at 4 °C, in the absence of light. For the organics, the water was stored in new amber glass bottles, and kept at 4 °C in the absence of light. The surface sediment samples were obtained with plastic shovels, more than 5 meters distant from the riverbank and in a maximum depth of 10 cm, and stored in white polyethylene bottles for inorganic analysis, and in aluminum containers in the case of aliquots for organic compounds analyses, kept at -20 °C until analysis. Samples of approximately 1 kg were collected in three different sites for each location along the Doce River (total of 3 kg per location, for triplicate analysis).

Physicochemical parameters of water
Data of physicochemical parameters were obtained by weekly and monthly reports available in the literature, from RENOVA Foundation and from the environmental agency IGAM, for the same periods of the field surveys of Wet (01/14/2019 to 01/23/2019 for sampling sites in MG state and 02/10/2019 to 02/18/2019 in ES state) and Dry (07/10/2019 to 07/19/2019 for sampling sites in MG state and 12/08/2019 to 18/08/2019 in ES state) of 2019. A total of six weekly reports for the wet season and three-monthly reports for the dry season published by REN-OVA described Clorophyll a (mg/L), Cyanobacteria (mg/L), Conductivity (mS/cm), temperature ( °C), turbidity (NTU), Dissolved oxygen (mg/L), pH and average precipitation (mm) in water. The search for physicochemical data considered six sampling sites described in the reports, which could correspond spatially to at least six out of the eight sites described in the field surveys. Regarding the IGAM data, four reports from Wet and four for the Dry described the levels of: (dissolved Al, As, Cd, Pb, dissolved Cu, Cr, dissolved Fe, Mn, Hg, Ni, NO 3 , Zn, total Coliforms, Clorophyll a (μg.L −1 ), Cyanobacteria (μg.L −1 ), Conductivity (μS.cm −1 ) Temperature ( °C) Turbidity (UNT), Disolved O 2 (mg.L −1 ) pH) at 5 different sites (GUA, CAN, NAQ, GOV, AIM) across the DRB. The quali-quantitative parameters of the water (QQPM) were obtained through an automatic monitoring of fixed equipments distributed along the Doce River Basin.
References for database for physicochemical parameters:

. Sediment grain size distribution
The composition and grain size distribution of the sediment was analyzed following the wet and dry sieving method (CETESB, L6.160). A fraction of the lyophilized sediment sample was weighed (100-120 g per sample) and wet sieved (distilled water) in a 0.062 mm mesh to remove muddy fraction, being subsequently dried for 2 days at 60 °C. Then, the material was sieved for 20 min in an electromagnetic sieve shaker (Bertel), using a 6-piece sieve for 20 minutes, to separate the sandy fractions. The fraction removed from each sieve was weighed and recorded, and the contents of sand, silt and clay were determined, according to the classification of [6] .

Organic matter
The quantity of Organic Matter (OM) was determined through the method of wet organic matter oxidation [4] . One gram of each lyophilized sediment was oxidized with 10 mL of K 2 Cr 2 O 7 1N within 20 mL of H 2 SO 4 PA. After light agitation and resting for 30 min, 200 mL of distilled water, 10 mL of orthophosphoric acid PA and 1 mL of 1% diphenylamine were added. The excess of remaining dichromate from the oxidation was titrated with 0.5N ammoniacal ferrous sulfate (AFS) solution and the organic matter content was calculated based on the volume of AFS added in the titration. The same procedure was performed without sediment sample to obtain a blank value for comparison with the samples.

Procedure 1 -inorganics in water
The concentrations of the inorganic chemicals in the water were determined by two different procedures. For the procedure 1, performed by the IO-USP Institution, the samples were filtered with PTFE membrane of 0.45 μm pore size (Merck Millipore TM ), and analyzed in triplicate for determination of Al, As, Ca, Cd, Cr, Cu, Fe, K, Mn, Ni, Pb, Sc, Sr, Ti, V, Zn. The results were expressed as mg L −1 and compared to CONAMA Resolution 357/2005, according to its respective class of water (for Doce River, Class II). The quantification was performed in Inductively Coupled Plasma Mass Spectromer (ICP-MS) according to USEPA 6020 (2014).
For the procedure 2, performed by the UFPR Institution, the 48 river water samples were centrifuged for 10 minutes at 3500 rpm to precipitate the suspended material. Subsequently, the solutions were analyzed directly by conventional nebulization in ICP OES for elemental quantification of Al, As, Cd, Co, Cu, Fe, K, Li, Mg, Mn, Na, Ni, Pb, Ti, V, Zn. For quantification of Hg was used the hydride generation technique coupled to the ICP OES.
For instrumental determination it was used an Inductively Coupled Plasma Optical Emission Spectrometer (ICP OES, Thermo Scientific, model iCAP 6500) with axial view, software Thermo   Certified value (mg kg −1 ) and result of mean ± standard deviation (SD), recovery and relative standard deviation (RSD) of SS-2 (n = 3) for inorganic chemicals measured by ICP IOS.

Organic chemicals in water
The organic contaminants (Polycyclic aromatic hydrocarbons -PAHs) in the water were determined according to the method of Solid-Phase Extraction-SPE (Method 3535A) proposed by USEPA (2007), validated by [ 7 ] for the PAHs and according to [ 8 ] for the Pyrethroids . The compounds were extracted from filtered water samples (PTFE membrane of 0.45 μm pore size, Merck Millipore TM ) through solid phase extraction with cartridge SPE C18 (Discovery Ò DSC-18 SPE, 0,5g/6 mL) coupled to a vacuum manifold system (Visiprep TM ). The columns were preconditioned with Miliq water, in a flux of approximately 5 ml.min −1 , and the compounds were eluted with 0.5 mL of acetone, 0.5 mL of ethyl acetate, 0.5 mL of hexane and lastly 0.5 mL of a mixture of the three solvents in equal proportions.
The determination of the organic contaminants was performed with a gas chromatography coupled to a mass spectrometer (GCMS-QP2010, Shimadzu) in the splitless mode. The analytes of interest were separated in a capillary column DB-5 Agilent J&W (5% diphenyl-95% dimethyl polysiloxane) of 30 m x 0.25 mm i.d with 0.25 μm film thickness, with helium (99.999%) as the carrier gas. The initial oven temperature was 60 °C, that was maintained for 3 min at a rate of 15 °C min -1 , after it was increased to 200 °C at a rate of 6 °C min -1 , until reaching the final temperature of 300 °C that was maintained for 7 min. For identification of the 21 PAH compounds, the retention time of the analytical standards were used to match with the injected samples, considering the confirmations of the ion mass fragments and the NIST library. Quantification of analytes was performed based on a 5-point calibration curve, using the internal standard method. Accepted linearity was obtained in all calibrations (r 2 > 0.99).

Inorganic chemicals
The trace elements Al, Cd, Cr, Cu, Fe, Mn, Ni, Pb, Sc, Ti, V and Zn were quantified in the whole sediments as triplicate samples for each of the eight sites and sampling season. All glassware was previously washed with 10% nitric acid. Metals were extracted from 1 g of sediment digested in 5 mL of nitric acid 50%, followed by two successive additions of 2.5 mL of HNO 3 PA (Merck) in the digestive block at 95 °C. Peroxide hydrogen (6 mL progressively added, for 270 min), HCl PA (5ml for 15 min) and water were added to complete 50 ml of solution. The detection was performed in an inductively coupled plasma mass spectrometry (ICP-MS). The metal concentrations in the sediment were expressed as μg.g −1 or dry (weighing after dehydration).
To ensure the reliability of the test, an SS-2 certified reference material (EnviroMAT Contaminated Soil, SCP Science) was subjected to the same analytical procedure. The certified values of each analyzed element are presented in Table 11 , with recovery data and relative standard deviation. All recoveries were within the values recommended by USEPA (1996) showing accuracy and precision. Detection limits were calculated (m = 1.0 g; v = 50 ml) and with the exception of Cd (0.38 mg kg-¹) all results were above.

Total Hg in the sediments
The experimental procedure to determine total Hg in sediment samples was performed according to Santos et al. 2005. Sediment samples were firstly grounded in agate mortar and sieved with a 100 μm stainless steel granulometric sieve. In approximately 30 mg of the sieved sample it was added 1.5 mL of aqua regia (3HCl:1HNO 3 ) in 15 mL polypropylene tubes. The mixture was submerged in an ultrasound bath for 30 minutes. After this period, the suspension was incubated for 24 hours with occasional manual agitation and was subjected to another ultrasound for 30 minutes. At the end the volume was adjusted to 15 mL with deionized water. The final concentration of aqua regia in the suspension was 10% v v −1 . The procedure was performed in duplicate and analyzed according to the ICP OES conditions described for the Procedure 2.
Blank reagents were included in the analysis system to ensure the absence of contamination in all the samples. The recovery test was performed with the analysis of certified material RS-3 (river sediment) from a round-robin test, the recovery rate was 94% for Hg (Santos, et al. 2005).

Organic chemicals in the sediments
Differently than inorganic analysis, sediments sampled for organic analysis were stored in aluminum containers as composite samples, formed by a mixture of three subsamples, in equal proportions (total of 8 composite samples per season).
The extraction of organic contaminants was carried out according to [ 9 ], consisting of the extraction of total organic contaminants followed by clean-up for separation into three fractions with different polarities. All glassware was previously washed with 25% Detertec and rinsed with acetone PA. For each 15 g of lyophilized sediment sample, it was spiked deuterated surrogate standards at final concentrations of 100 ng g −1 (p-terphenyl-d 14 , purchased from Supelco/ Aldrich). The total organic chemicals from the spiked samples of sediment were extracted with a solid-liquid extraction consisting of the addition of a sequence of solvents with decreasing polarity indexes ( P' ): 25 ml of each solvent Acetone, Ethyl acetate, Dichloromethane, Hexane, and the mixture of the four solvents in same proportions, all purchased from MERCK. Samples remained on ultrasound for 20 min for each solvent, followed by centrifugation at 40 0 0 rpm for 15 min. The supernatant was concentrated in a rotary evaporator in a 250 mL flask to a volume of approximately 1 ml.
The clean-up procedure consisted of separating the compounds through chromatographic affinity in an open preparative column (1 × 50 cm) containing 8.0 g of Silica gel (MERCK), 4.0 g of alumina (MERCK), 0.5 g of Cu (MERCK) and 1.0 g of anhydrous sodium sulfate (VETEC). The elution sequence was based on the increasing range of eluotropic strength by adding 40 ml of Hexane (Fraction 1), 60 ml of Hexane: Dichloromethane: Ethyl Acetate (3: 3: 1 -Fraction 2) and 50 ml of Dichloromethane: Methanol (9: 1 -Fraction 3). The extracts were concentrated in a rotary evaporator and stored in amber vials kept at -20 °C until analysis. The determination of the 21 PAHs in the sediment extract (Fraction 2) ( Table 8) and Pyrethroids (Fraction 3) (Table 9) was performed with a gas chromatography coupled to a mass spectrometer (GC-MS) as previously described for the water samples.

QA/QC
Method blanks were included in the analysis system to ensure the absence of contamination in all the samples. The method blank consists of clean glassware included in each batch under the same processing method than environmental samples. This quality control sample is used for sample concentration correction of environmental samples via background subtraction. Besides instrument blanks (n-hexane) were analyzed to evaluate any instrument noise.
Furthermore, the deuterated surrogate standard with known concentration, was spiked at in the beginning of the extraction and quantified at the end of all the process, to assess the efficiency of the analyses (Cavalcante et al., 2008. The average recovery from our environmental samples was 63% ± 29%. The values in the Table 8 were not modified by the recovery values.