The Past Is Never Dead: Soil Pollution from Mining in the Copiapó River Basin (Northern Chile)

Abstract

This short paper analyses the concentrations of two major components (Fe, S) and eight trace elements (As, Au, Co, Cu, Hg, Ni, Mn, Zn) in soils and tailings from Tierra Amarilla (northern Chile) using ICP-Ms analysis. The levels of As, Au, Cu, Fe and S are very high and come from polymetallic sulphides from nearby mines, together with minor contributions of Co, Ni and Mn. Hg has its origin in the extreme seasonal flows of the Copiapó River, which erodes the dumps of old precious metal mines. These high concentrations require further analysis of possible metal immobilisation techniques, bioavailability or analysis in cultivated plants.

1. Introduction

‘The past is never dead. It’s not even past’, said the Nobel Prize in literature 1949 winner William Faulkner. This reflection is particularly applicable to certain regions of our planet whose current environmental scenario is a direct consequence of past actions, especially in mining regions. Waste dumps and slag heaps of past centuries have changed the original landscapes and caused heavy contamination of both soils and sediments [1,2]. Their removal by river currents, rainfall or agriculture can cause a redistribution of heavy metals and a new affectation of current environments, with notable consequences on the environment or on the food chain [3,4]. This is particularly dangerous in the case of some elements such as Hg, which is potentially poisonous to humans [5,6].

The Chilean Iron Belt is one of the most important mineralogenetic provinces in South America, as it encompasses the world’s largest copper ore deposits, which have been exploited since pre-Hispanic times [7,8]. Copper is extracted from more than 40 ore deposits, as well as smaller but significant amounts of Ag, Au, Fe, Hg or Zn, among other ores [9,10]. This has caused significant environmental deterioration and the contamination of soils and sediments near the waste dumps or the redistribution of pollutants through rivers or occasional rainfall [11,12].

This communication analyses the geochemical content of soils and mining tailings located in the Copiapó river basin, around the town of Tierra Amarilla (northern Chile). Correlation analysis is applied to determine chemical associations and the possible sources of the detected elements.

2. Materials and Methods

2.1. Study Area

The Copiapó river flows through the region of Atacama (northern Chile) (Figure 1A,B) and its basin covers some 18,400 km². This area is characterised by a semi-arid climate and the presence of deserts, with an annual rainfall of less than 20 mm [13]. However, this river suffers periodic flooding episodes, as can be seen from historical information [14,15,16]. These floods can inundate areas subject to intense degradation due to heavy extensive mining activities in this region [17].

Wastes from mining operations are very abundant in the basin of this river, with very high Cu contents in most cases [18,19]. In addition, some of them come from the cyanidation and amalgamation of precious metals such as Ag or Au [20] and are in close proximity to this river. This location can lead to a result in the partial redistribution of Hg-contaminated tailings due to periodic fluvial action and the presence of alluvial sediments with high Hg contents in the catchment, as already established for other metals such as Cr, Ni, Pb or Fe [21].

This paper analyses the geochemical content of soils and mining waste adjacent to two mining areas located in Tierra Amarilla (northern Chile) (Figure 1A,B). Minera Candelaria is a large open-pit and underground Cu and Au mine, while Minera Carola is an underground copper mine (Figure 1C).

2.2. Chemical Analysis, Geoaccumulation Index and Statistical Procedures

Sixteen soil samples were taken from two agricultural plots located near the mines of Minera Candelaria (A-1 to A-6) and Minera Carola (B-1 to B-10), as well as three samples from mine waste near Minera Carola (T1 to T3) (Figure 1C). The samples were manually pulverised in an agate mortar at the sedimentology laboratory of the Geology Department of the University of Chile. They were then placed in polypropylene jars. An ICP-MS analysis of 0.5 g of each sampling point was performed via acid digestion with aqua regia (method AQ250) at ACME Analytical Laboratories S.A. (Bureau Veritas Mineral Laboratories Chile). The concentrations of major components (Fe and S, in %) and trace elements (As, Co, Cu, Mn, Ni and Zn in ppm; Au and Hg in ppb) were obtained. The quality control report (number SAN18000095.1) includes a check with three reference materials (STD DS11, STD OREAS45EA and BLK). The geoaccumulation index I_geo [22] was applied to the eight minor elements, based on the equation

I_geo = log₂C_n/1.5B_n where C_n is the concentration of an element in a sample and B_n is the average concentration of that element in the earth’s crust. Factor 1.5 is used to reflect the possible fluctuation of the element in the background value as well as minimal anthropogenic influences or inputs. In this paper, samples were classified as unpolluted (I_geo < 0), very low polluted (0 < I_geo < 1), low polluted (1 < I_geo < 2), moderately polluted (2 < I_geo < 3), highly polluted (3 < I_geo < 4), very highly polluted (4 < I_geo < 5) and extremely polluted (I_geo ˃ 5). This index is also used in conjunction with the enrichment factor (EF) to assess the degree of soil contamination [23].

Linear correlation analysis has been used to test the relationship between the ten elements analysed, as well as to differentiate possible associations between them and different sources of origin, in coordination with their raw values. Positive correlations denote that one variable increases when the other also increases and negative correlations mean that one variable tends to increase when the other decreases. Values closer to 0 indicate an absence of correlation.

3. Results and Discussion

3.1. Chemistry

The results of the chemical analyses clearly differentiate mining waste from agricultural plots (

Table 1. Geochemical analysis of soils (A-1 to A-6 close to Minera Candelaria; B-1 to B-10 close to Minera Carola tailings) and tailings (T1 to T3 near Minera Carola).

SAMPLE	As (ppm)	Au (ppb)	Co (ppm)	Cu (ppm)	Fe (%)	Hg (ppb)	Mn (ppm)	Ni (ppm)	S (%)	Zn (ppm)
A-1	23.7	74.4	14.6	447	3.33	337	688	44.4	0.1	147.4
A-2	28.5	8.1	15.1	333	3.71	861	760	18	0.34	144.3
A-3	58.9	33.5	21.1	408	3.34	5480	946	25.8	0.21	173
A-4	92.2	12.5	14.7	535	3.03	8570	768	22.5	0.28	151.3
A-5	21.7	8.9	13.2	187	3.02	660	633	17.4	0.21	99.2
A-6	69.9	21.2	14.1	287	2.98	6077	761	21.7	0.66	189.4
B-1	59.1	25.5	18.7	658	3.49	2668	990	27.1	0.21	205.7
B-2	47.9	21.9	15.9	676	3.06	1545	799	23.7	0.22	146.2
B-3	52	26.3	16.7	731	3.16	3172	781	22.3	0.26	169.6
B-4	56.4	30	18.3	858	3.47	2749	972	26.7	0.22	188.6
B-5	48.8	16.6	17	473	3.39	889	990	21.2	0.11	159.2
B-6	75.5	48.2	16.7	720	3.16	3220	839	22.3	0.18	166.9
B-7	90	105.5	25	1447	4.14	3007	1205	25.2	0.27	226.8
B-8	67.5	36.5	17.3	571	3.19	4681	846	22.6	0.15	164.9
B-9	79.6	55.3	20.2	896	3.6	6848	1005	29.5	0.13	213.8
B-10	101.4	116	18.2	739	3.22	6017	941	23.3	0.11	205.5
T1	93.7	99.3	74.1	3279	11.53	746	1214	47.1	1.6	125
T2	159.8	1881.7	87.2	5658	12.55	576	1804	46.7	1.65	186.7
T3	163.2	318.4	197.5	4725	15.7	618	1016	120.2	4.5	131.6

). Tailings have the highest concentrations of As (93.7–163.2 ppm), Au (up to 1881.7 ppb), Co (74.1–197.5 ppm), Cu (3279–5658 ppm), Fe (11.53–15.7%), Mn (1016–1804 ppm), Ni (46.1–120.2 ppm) and S (1.6–4.5%), while their Hg levels (576–746 ppb) are lower or similar on average than those of agricultural soils (˃1500 ppb in most samples). Average Zn contents remain in similar ranges in both groups (tailings: 146.2–186.7 ppm; soils: 99.2–226.8 ppm). Although mineralogical studies of these dumps have not been carried out, this chemical composition points to the presence of high percentages of polymetallic Fe and Cu sulphides with significant Au contents, e.g., the ores mined at Mina Carola [24].

The geochemical content of the soils is markedly different between the northwest area, close to Minera Candelaria (Zone 1-Z1-: samples A-1 to A-6), and the southeastern sector, close to the Minera Carola tailings (Zone 2 -Z2-: samples B-1 to B-10). The former parcel includes most of the lowest values of Cu (<450 ppm in most samples), As (21.7–28.5 ppm), Au (8.1–33.5 ppb), Co (13.2–15.1 ppm in most samples) and Ni (17.4–25.8 ppm in most samples), while the Hg values increase significantly from the vicinity of Mina Candelaria (337–861 ppb) towards the Copiapó River (˃5000 ppb in most samples near the river). The absence of high Hg contents near Mina Candelaria (e.g., sample A-1) as well as in the Mina Carola tailings dumps (T1 to T3) would indicate that high levels of Hg in both zones are coming from the Copiapó river. Historically, this region is subject to major hydrometeorological events that cause the Copiapó River to overflow [17] and the erosion of tailings from old Ag, Au and Cu mines, with the consequent contamination of the surrounding floodplain with these metals [17,25].

Z2 shows higher average values than Z1, especially in As (Z1: 49.1 ppm; Z2: 67.8 ppm), Au (Z1: 26.4 ppb; 48.2 ppb), Cu (Z1: 366 ppm; 777 ppm), Mn (Z1: 759.3 ppm; Z2: 936.8 ppm) and, to a lesser extent, Zn (Z1: 150.8 ppm; Z2: 184.7 ppm). These differences are even more pronounced if only the samples closest to the tailings of Minera Carola are taken into account (Figure 1: samples B-7, B-9 and B-10), where the soils that are most contaminated in terms of As (up to 101.4 ppm), Au (up to 116 pp), Cu (up to 1447 ppm), Mn (up to 1205 ppm) or Zn (up to 226.8 ppm) are found. Consequently, the pollution of the Z2 soils would come from the adjoining Mina Carola tailings dumps. This direct relationship between the chemical compositions of mine dumps and adjacent soils has been repeatedly observed in various mining regions of Chile [26,27].

3.2. Geoaccumulation Index

Most of the soil samples are highly to extremely polluted by As (Figure 2A; 2.78 < I_geo < 5.77), while Z2 shows moderate to very high I_geo values for Cu (Figure 2D; 2.81 < I_geo < 4.42), clearly higher than Z1 (1.47 < I_geo < 2.98). Z2 is also heavily contaminated by Au (Figure 2B) and Hg (Figure 2E), especially at its southeastern end. Some samples of Z1 are also heavily contaminated by this metal (e.g., A-4 or A-6). These high values contrast with the very low to low Zn contamination (Figure 2H; 0.02 < I_geo < 1.22) or the negative values of Igeo for Co (Figure 2C), Mn (Figure 2F) or Ni (Figure 2G). These values are similar to those deduced in other polluted soils in Chile, although this index depends on the background considered [11]. The mine tailings from Mina Carola are heavily to extremely polluted by As (4.97 < I_geo < 5.76), Au (3.92 < I_geo < 7.97) and Cu (5.6 < I_geo < 6.39). Also noteworthy is their low to moderate Co contamination (1.1 < I_geo < 2.52) and the negative I_geo values for Hg in these tailings dumps.

3.3. Correlation Analysis

The bivariate analysis reveals a strong positive and significant correlation between As, Au, Co, Cu, Fe, Mn, Ni and S (

Table 2. Correlation coefficient matrix. Bold: r ˃ 0.6; p < 0.01.

	As	Au	Co	Cu	Fe	Hg	Mn	Ni	S	Zn
As	1
Au	0.65	1
Co	0.77	0.44	1
Cu	0.85	0.79	0.86	1
Fe	0.79	0.61	0.94	0.96	1
Hg	0.06	−0.28	−0.37	−0.41	−0.45	1
Mn	0.71	0.82	0.46	0.81	0.65	−0.24	1
Ni	0.66	0.31	0.96	0.75	0.86	−0.36	0.32	1
S	0.73	0.38	0.98	0.8	0.91	−0.33	0.36	0.94	1
Zn	0.22	0.13	−0.24	−0.05	−0.23	0.44	0.37	−0.23	−0.3	1

; r ˃ 0.6 in most cases; p < 0.01). Hg and Zn show negative correlations with most of these elements, which contrasts with a positive and significant correlation between these two elements. The first group would be mainly contributed to by nearby mining facilities and a statistically significant correlation between some of these elements is frequent in soils close to ore deposits due to weathering [28,29]. Nevertheless, Hg would come from the erosion of former small precious metal mines by seasonal flows of the Copiapó river upstream. These small mines are numerous in this region [30].

3.4. Considerations for Future Research

The obtained levels of heavy metals, such as Cu or Hg, require further analysis in the following lines of research: (i) possibilities for the partial immobilisation of these metals in soils; (ii) an analysis of their bioavailability in the plants that grow on them, with the application of the free ion activity model (FIAM) or the correlation between labile metals in soils, determined using simple or sequential extraction methods, and their accumulation in plants; or (iii) a study of metal concentrations in roots, stems and fruits of cultivated plants, in order to prevent possible subsequent negative effects on human health [31,32].

4. Conclusions

The main findings of this research are as follows:

(i)

The studied soils of Tierra Amarilla are highly to extremely polluted by As, Au, Cu, Fe and S from nearby mining operations, mainly from the Mina Carola mine;

(ii)

Hg also shows moderate to high levels and geoaccumulation rates in most of the soils analysed, while the waste dumps are not contaminated;

(iii)

The contents of this metal would come from the erosion of old tailings of precious metal mines, which used it in their production processes;

(iv)

These high concentrations require further analysis to investigate the relationship between these soils and the plants grown in them.