Incomplete Mixing in the Fate and Transport of Arsenic at a River Affected by Acid Drainage

Incomplete mixing in the fate and transport of arsenic at a river affected by acid drainage

SupportingInformation

Paula Guerra, Christian Gonzalez, Cristian Escauriaza, Gonzalo Pizarro and Pablo Pasten

Sections

Section 6.Description of cross-sectional mass balances of iron, aluminum and arsenic.

Section 7.Other Supplementary Information.

Figures

Fig. 10Caracarani River cross-section scheme

Fig. 11Arsenic sequential extractions

Fig. 12FTIR spectra of suspended solids obtained downstream the Azufre River-Caracarani River confluence.

Fig. 13Volume density versus particle size for field measurements

Fig. 14Volume density versus particle size for laboratory experiments

Tables

Table 4.Detail of cross-section measurements. Total flowrate: 979 L s-1

Table 5.Detail of water quality cross-section measurements at 0 m downstream the confluence

Table 6.Detail of water quality cross-section measurements at 35 m downstream the confluence

Table 7.Detail of water quality cross-section measurements at 110 m downstream the confluence

Table 8. Correlation indexes between metals found in the suspended solids downstream the Azufre River-Caracarani River confluence.

Table 9. Correlation indexes between metals found in the suspended solids downstream the Azufre River-Caracarani River confluence.

Table 10.Main parameters used for the ANOVA test

6. Description of cross-sectional mass balances of iron, aluminum and arsenic

Introduction

The integrated mass balances of iron, aluminum and arsenic were obtained by multiplying the flowrate of one section of the stream by the total concentration of metals in the same section.

The following assumptions were made:

Constant cross-sectional area downstream. This means that the cross-section where the flowrate was measured was projected downstream to the different points at which metals were measured.
The concentration of metals within a determined area is a representative average within the section.

Methodology

Flowrate was estimated by the velocity-area method. This method consists of dividing the cross-section of the stream into sub-areas and measuring the velocity in each of these sub-areas. The flowrate is then estimated as:

where Q: Total flowrate in m3 s-1, vi: velocity in subsection in m s-1, Ai: area of subsection i in m2.

The total concentrations of metals were determined by sampling water at several points across different sections of the stream.

The total mass of metals passing through a determined cross-section was estimated as:

whereFm: Total flow of metal “m” in mg s-1, [m]i: average concentration of metal m in subsection i in mg L-1, Qi: flowrate in subsection i in L s-1.

Results

Fig. 10 shows the cross-section at which the flowrate was measured during the campaign of March 2013. Table 4 shows the values of each sub area and velocity. Tables 5, 6 and 7 show cross-sectional measurements of total metals. Cross-sectional distance was normalized.

Fig. 10.Cross-section scheme.

Table 4.Details of cross-sectional measurements. Total flowrate: 979 L s-1

Table 5.Details of water quality cross-sectional measurements 0 m downstream from the confluence

Table 6.Details of water quality cross-sectional measurements 35 m downstream from the confluence

Table 7.Details of water quality cross-sectional measurements 110 m downstream from the confluence

7. Other Supplementary Information

Table 8.The concentrations of several metals in the post-confluence suspended solids > 0.45 μm. High concentrations of arsenic are observed. The concentrations of the elements are highly variable in a range of pH from 3.5 to 6.5.

Table 9.Summary of the correlation coefficients calculated for several metals measured in the suspended solids > 0.45 μm. The highest correlation coefficient was the As-Fe pair (ρAs,Fe= 0.975).

Table 10. Main parameters used for the ANOVA test

Fig. 11.FTIR spectra of suspended solids collected by the time integrated suspended sediment sampler (TISS). Samples were filtered and washed with Milli-Q water. The comparison of our sample with literature spectra suggests the presence of Goethite (Gt), Schwertmanite (Schw) and Akageneite (Aka) (peaks found between 250 and 100 cm-1) according to Schwertmann, U., & Cornell, R. M. (2008). Iron oxides in the laboratory.Wiley-VCh.

Fig. 12.Arsenic sequential extractions suggest how arsenic is associated to mineral phases.

Fig. 13.Volume density versus particle size for mixing experiments type 3 for the following mixing ratios: (a) 0.07, (b) 0.19 and (c) 0.24. The mode (µm) represents the mean particle size at which each percentage of volume density was estimated. Distributions were selected arbitrarily for reference in the Supplementary Information.

Fig. 14.Volume density versus particle size for field measurements (selected arbitrarily) corresponding to the following points downstream (a) 25 m, (b) 92 m and (c) 173 m. The mode (µm) represents the mean particle size at which each percentage of volume density was estimated.