Environmental Geochemistry and Health 26: 383–393, 2004.© 2004 Kluwer Academic Publishers. Printed in the Netherlands.

Impact of mining activities on soils in a semi-arid environment: SierraAlmagrera district, SE Spain

Andres Navarro1,4, Diego Collado2, Montserrat Carbonell3 & Juan A. Sanchez2

1Dto. Mecanica de Fluidos, ETSEIT, Universidad Politecnica de Cataluña (UPC), 08222-Terrassa (Barcelona)2Dto. Edafologıa y Quımica Agrıcola, Universidad de Almerıa (UAL), 04120-La Cañada (Almerıa)3Dto. Mecanica de Fluidos, EUPVG, Universidad Politecnica de Cataluña (UPC),Vilanova i la Geltru(Barcelona), Spain4Author for correspondence (tel.: +93-739-8264; fax: +93-739-8101; e-mail: navarro@mf.upc.es)

Received 3 December 2002 Accepted in revised form 29 October 2003

Key words: column experiments, heavy metals, secondary phases, Sierra Almagrera, tailings

Abstract

Field and laboratory-column studies were undertaken in order to investigate soil contamination derived from pastmining activity in the Sierra Almagrera (SA) district in southeast Spain. The tailings, soil and sediment samplesthat were collected showed high concentrations of Ag, As, Ba, Cu, Pb, Sb and Zn when analyzed. The meanconcentrations of these elements in the tailings were 29.8, 285.4, 54000, 57.7, 2687.5, 179.0 and 2269.0 ppm,respectively. In the soil samples these decreased to 14.3, 96.9, 24700, 37.5, 1859.1, 168.5 and 815.7 ppm, re-spectively. Geochemical analyses demonstrated high levels of As, Pb and Zn which were above the interventionvalues set forth in the Andalusian Regulations for Contaminated Soils for As (>50 ppm), Pb (>500 ppm) andZn (>2000 ppm). Column experiments and mineralogical studies suggest that the dissolution of sulfates and othersecondary phases, accumulated in soils and waste-sites during the dry season, acts to control the mobility of metals.The elution curves obtained from column experiments showed a mobilization of Ba, Cu, Pb and Zn, while a lowmobility was seen for Ag, As and Sb.

Introduction

Mining and smelting related to sulfide ore bodiesexploitation have been the cause of major pollutionproblems in, around and outside mining areas, as theyresult in the mobilization of metals and inorganic sub-stances into the environment (Förstner & Wittmann1979; Azcue 1999).

Exposure of pyrite (FeS2) and other sulfide min-erals to atmospheric oxygen and moisture results inonset of known weathering reactions (Förstner &Wittmann 1979; Appelo & Postma 1996). The con-sequence can be the presence of high amounts ofAs, Cd, Cu, Co, Cr, Fe, Mo, Ni, Pb, Sb, Zn andother minor elements (Stollenwerk 1994; EPA 1996;Herbert 1996) in the soil and/or ground water, as aresult of sulfide oxidation.

As the mobilization of contaminants in the soil iscontrolled by pH and moisture conditions in addition

to other factors, soluble iron-sulfate minerals may pre-cipitate during dry periods, especially at the interfacebetween saturated and unsaturated zones, at whichevaporation leads to an accumulation of dissolvedspecies (Bigham & Nordstrom 2000). Moreover, hy-droxide and oxy-hydroxide of Fe can also precipitate,which can remove many metals by means of adsorp-tion phenomena. Therefore, these secondary phasescontrol the distribution of trace elements in the porewater, by means of reactions such as adsorption,surface-complexation and co-precipitation, which re-sult in heavy metal attenuation (McGregor et al. 1998).The precipitation of Fe compounds and other second-ary phases in the non-saturated zone of contaminatedareas or tailings, occasionally produces precipitatedlayers referred to as ‘hardpans’, which may controlthe movement of O2 and dissolved metals through thetailings and also have significant environmental im-plications (Blowes & Jambor 1990; Ribet et al. 1995).

384 ANDRES NAVARRO ET AL.

Fig. 1. Location map of the study area and volcanic belts of Cartagena-Gata metallogenic arc. (I) Calc-alkaline lavas of Gata; (II) K2O-enrichedcalc-alkaline volcanic domes; (III) Hyper potassic, aluminous volcanic necks; (IV) Basalts of Cartagena. Modified after Lopez Ruiz &Rodrıguez (1980).

The environmental impact of mining activities onarid and semi-arid regions in Spain has received lessattention that in other regions, with the exception ofthe nearby Azogue valley area (Navarro et al. 2000)and the Rodalquilar mining district, where the researchfocused on the contamination of stream sedimentsdownstream of a tailings site with Pb, Cu, Zn, Mo, Sb,Te, Bi and As (Wray 1998). This paper describes theenvironmental impact on a semi-arid region of uncon-fined tailings impoundments caused by recent miningactivities in the Sierra Almagrera (SA) district, and soreflects the significance of obsolescent mining areas asa factor in soil and sediment contamination.

Site description

The SA mining district (Figure 1) is located 90 kmfrom Almeria, the provincial capital of Almeria

province, Spain. It is one of the oldest mining areasin the Iberian peninsula, along with the Iberian PyriticBelt and the Cartagena mining districts (Navarro et al.1994).

The SA mineral deposits are part of the volcano-metallogenetic belt of southeast Spain. The miner-alizations were first mined in the third millenniumB.C. and later again, with greater intensity, un-der Carthaginian and Roman occupation between thefifth century B.C. and the fifth century A.D. Modernproduction began in 1839, with the discovery of theJaroso lode. Over the course of the 19th century5 50 000 t of Pb and 512 t of Ag were extracted fromthe mines.

In the 20th century, the mining activity began todecline slowly (the smelting sites ceased their pro-duction in 1915) due to problems with groundwaterdrainage. Underground selective mining (high-gradelead–silver veins) ceased in 1957, when the exploita-

IMPACT OF MINING ACTIVITIES 385

Fig. 2. Schematic geological map of the SA area. (0) Alluvial Quaternary deposits; (1) Quaternary sedimentary deposits; (2) Upper Miocenevolcanic rocks; (3) Miocene sedimentary deposits; (4) Permian–Triassic metamorphic basement; (5) Main faulting system.

tion level had extended 220 m below sea level. In thelast third of the 20th century (specifically, from 1967to 1991) the mining activity was mainly the processingof low grade stockpiles, which resulted in sand andsilt-size tailings. These constitute one of the largestaccumulations of mining waste in this area (30 ha).At present, the only mining activity in the SA districtis barite extraction in the old native silver mine atHerrerías, which has produced nearly 1 00 000 t year−1

of BaSO4 in an open pit since 1993.

Geological setting

The SA mining district is located at the eastern edgeof the Cordillera Bética, which forms the central partof a wide volcano-tectonic and metallogenetic belt thatextends from Cabo de Gata to the Sierra de Cartagena

(Figure 1). In the study area, the Almanzora riverbasin (1800 km2) overlies a tectonic basin betweentwo metamorphic ranges, the SA and the Sierra deAlmagro, and is filled with Tertiary deposits overlainby the Quaternary alluvial and deltaic deposits of theAlmanzora River (Figure 2).

The main system of faults occurs in the NE–SWdirection parallel to the Palomares fault (Figure 2).This orientation also controls the distribution of vol-canic rocks (K2O-enriched, calc alkaline volcanicshoshonitic domes and breccias) and the hydrothermalsystem that produced the mineral deposits in the studyarea. The Quaternary alluvial and deltaic depositsof the Almanzora river basin cut across these oldermaterials. These deposits form wide terraces and apro-grading delta and are made up of sand, gravel, andclay to reach a thickness of 20–30 m.

386 ANDRES NAVARRO ET AL.

Table 1. Summary of geochemical composition of main epithermalmineralizations in the study area. Values in ppm except Au in ppb.

Element 1 2 3 4

As∗ 5–3300 52.7 393.7 76

Ba∗ 110–39% 27.2% 24820 1700

Sb∗ 4.9–540 364.7 14773 9.4

Hg∗ <1–3.0 3.3 2300–11% <1

Pb∗∗ 0.3%–17% – 67.2 1512

Zn∗∗ 50.0–21.5% 627.4 555.8 180

Ag∗∗ 100–1200 19.8 0.8–160 0.9

Mn∗∗ 245.0 – 456 111

Sr∗∗ <0.05%–1.3% 0.15%–1.3% 0.05% 367

Au∗ 20–20000 <5.0–44.0 5–823 14

1 Sierra Almagrera veins.2 Herrerıas stratabound deposits.3 Siliceous mineralization of Azogue area (Plio-Pleistocene vein).4 Volcanic-shoshonitic dome of Sierra Almagrera.∗ By Neutron Activation.∗∗ By ICP-MS.

The SA mining district consists of three main min-ing areas: Herrerías (stratabound mineralized bodies);Almagrera (N10–40◦W veins) and Azogue (hydro-thermal breccias associated with a major E NE–SWfault). The characteristics of these deposits have beenstudied in recent years (Martínez et al. 1992, 1997;Navarro et al. 1994; Navarro et al. 1998; Viladevallet al. 1999), as have the environmental effects on theAzogue area (Navarro et al. 2000).

The geochemical characteristics of the main min-eral deposits are shown in Table 1. The host rocksof the Herrerías deposit are Upper Miocene marinesediments that belong to the Neogene neotectonicbasin of Vera-Garrucha; the Almagrera mineral de-posit is placed in the phyllites which are part ofthe Permian–Triassic metamorphic basement (SA),and the Azogue deposit is located in the dividebetween the metamorphic basement and the Neogenesediments.

The Almagrera mineral deposits are principallymade up of veins of a large thickness, irregular veins,filling cavities and fissures. A geochemical regionalzonation can be appreciated in this area (Navarro et al.1994), and two basic types of vein structures aredefined (Navarro et al. 1998): veins of Pb–Ag (Fe,Ba) with bournonite, boulangerite, tetrahedrite, ga-lena, sphalerite, chalcopyrite, pyrite, marcasite, barite,siderite and quartz; and veins of Au–As with oxides,sulphates, gold-bearing quartz, arsenopyrite, sphaler-ite, marcasite, galena and chalcopyrite. A combinedsubmarine epithermal and hydrothermal genesis dur-

ing the Upper Miocene has been suggested as anexplanation for the nature of these deposits, associatedwith shoshonitic volcanism.

Hydrogeological setting

The study area is a semi-arid climate that has an an-nual average of 200 mm of precipitation, of whichapproximately 90% is subjected to evapotranspira-tion, according to mean annual balances. Surfacerun-off is very high, since 50% of the annual pre-cipitation can occur in 1–2 days. Therefore, themain mechanism of contaminant mobilization fromprimary sources are run-off and mechanical dispersioncaused by wind. Groundwater pollution and the lossof soil fertility are the main symptoms of the envi-ronmental degradation brought about by uncontrolleddisposal of mining waste in this area (Navarro et al.1998).

The tailings present in the SA are spilled over thenon-confined aquifer formed by the alluvial and del-taic deposits of the Almanzora river. The AlmanzoraDelta is a Quaternary formation developed over animpermeable clayey Miocene platform, and is madeup at the basal level of gravel and fine sand, with sandlayers in the upper part. The Almanzora aquifer hasa thickness of 25–40 m, with a saturated zone of ap-proximately 7 m. The water table is generally situated20–30 m below the surface, but annual and seasonaloscillations are typically observed. In the lateral di-rection, the deltaic aquifer is confined by the Tertiarydeposits of the Las Herrerías depression, except in theeastern sector, where the aquifer is confined by themetamorphic massif of the SA.

Main aquifer recharge are the infiltration of rain-fall, irrigation return, losses due to filtration from theCuevas dam reservoir and the contribution of smallercollateral aquifers. The main outflows of the systemare through direct discharge to the sea and pump-ing for irrigation. About 2–4 hm3 year−1 of water arepumped from the aquifer for agricultural uses. Thecontamination of groundwater associated with min-ing activities show significant amounts of Ba, Fe,Mn, Mo, Pb and Zn, near the obsolescent miningsites, and low concentrations of As, Cd, Cr, Cu,Sb, Se and Ni in the groundwater. Metal attenu-ation is very significant during the transport of metalsthrough the non-saturated zone and probably accountsfor the low concentrations of several contaminants inthe groundwater.

IMPACT OF MINING ACTIVITIES 387

Research methods

Soil and tailings sampling

Tailings and stockpiles are disposed over the non-confined aquifer formed by the alluvial and deltaic de-posits of the Almanzora river, close to the old miningfacilities and near an intensive agricultural productionarea.

Soil and stream sediment samples were collec-ted by driving a small diameter pipe vertically intothe ground, along the lines of a methodology derivedfrom standardized methods for sampling and describ-ing contaminated soils (EPA 1991). Detailed metalcontamination and composition profiles were deter-mined for the surface soils through local geochemicalprofiling. This worked from soil core samples takenfrom areas of high soil contamination near the maintailings disposal area. Tailings samples were obtainedby hand-excavation as deep as 0.4 m.

Soil, stream sediment and tailings samples weresealed to minimize exposure to atmospheric gases, andafter drying at 30◦C for 48 h, they were prepared in thefollowing ways:

(1) Homogenization of tailings samples and determi-nation of grain size, porosity, bulk density, pH andelectric conductivity.

(2) Determination of grain size, porosity, bulk densi-ty, pH and electric conductivity of soil and streamsediment samples.

Chemical analyses were performed on subsamples(<2 mm in diameter) from each soil and sedimentsample core. The results gave the CaCO3 content, thetotal N, P, TOC, the gypsum percentage and the cationexchange capacity (CEC). The chemical analyses oftailing samples were done by grinding homogenizedsamples in a mechanical mortar (Fritsch pulverisette502).

The total content in metals and trace elements wasdetermined by Instrumental Neutron Activation Anal-ysis (INAA) and Inductively Coupled Plasma Emis-sion Spectroscopy (ICP-OES): Au, Ag, As, Ba, andSb were determined by INAA and Cu, Pb, Zn and Agby ICP-OES.

Non-sequential extractions were performed on se-lected tailings, soil and sediment samples to determinethe distilled-water soluble fractions (WDE) of eachsample. The distilled-water fraction was analyzed byICP-MS Hewlett Packard 4500 to determine: the Ag,As, Ba, Cu, Pb, Sb and Zn content and the usual meth-

ods were utilized to determine the major anions andcations (APHA 1989).

The mineralogy of the tailings and soil sampleswas examined under a binocular microscope andby X-ray powder diffraction (XRPD). The re-sults of the mineralogical study (Collado 2002)indicate the most abundant secondary mineralspresent to be jarosite (KFe3(SO4)2(OH)6), natro-jarosite (NaFe3(SO4)2(OH)6), crystalline oxyhydrox-ide of Fe (goethite), amorphous ferric hydroxide(Fe (OH)3), clay minerals, anglesite (PbSO4), alunite(KAl3(SO4)2(OH)6) and gypsum (CaSO4 · 2H2O),which can occur either as a primary or a sec-ondary phase in the SA. Other secondary phasesdetected were plumbojarosite (PbFe6(SO4)4(OH)12),langite (Cu4(SO4)(OH)6 · 2H2O) and uklonskovite(NaMg(SO4)F · 2H2O), in minor quantities.

The chemical analyses of the samples fromthe tailings (Table 2) show high concentrations ofAg (mean: 29.8 ppm), As (mean: 285.4 ppm), Ba(mean: 54000 ppm), Cu (mean: 57.7 ppm), Pb(mean: 2687.5 pp), Sb (mean: 179.0 ppm) and Zn(2269.0 ppm). The chemical analyses of the soils andsediments show high amounts of Ag, As, Ba, Pb, Sb,

Table 2. Summary statistics of tailing and soils samples of SA.Values in ppm, except Au (ppb).

Element Tailings Tailings Soils Soils

(Mean) (S.D.) (Mean) (S.D.)

Au 5.6 3.5 4.4 2.7

Ag 29.8 8.8 14.3 11.8

As 285.4 82.6 96.9 83.0

Ba 54000 15798.7 24700.0 27365.7

Cu 57.7 – 37.5 12.6

Pb 2687.5 – 1859.1 675.6

Sb 179.0 65.4 168.5 240.2

Zn 2269.0 575.6 815.7 696.9

Table 3. Summary of the statistics of stream sediment samples ofSA. Values in ppm, except Au (ppb).

Element Mean S.D.

Au 6.25 2.60

Ag 9.75 8.91

As 91.50 80.41

Ba 11750.00 12416.79

Sb 65.50 57.81

Zn 977.00 709.10

∗ Below the detection limit (0.1 ppm).

388 ANDRES NAVARRO ET AL.

and Zn, though the mean concentrations decrease withrespect to the tailings by approximately 50% in thesoil (Table 2) and stream sediments of the Rambla deCanalejas, the main tributary of the Almanzora river(Table 3).

The As and Zn isoconcentration maps (Figures 3and 4) show that a large contaminated area surroundsthe tailings deposits. Thus, the systematic samplingof waste tailings and contaminated soils show highconcentrations of As (Figure 3) in the range of 31–460 ppm in a large anomalous area, located to thesouth and north of the main tailings area. The Asdistribution is mainly originated by the action and sea-sonal run-off, associated with episodic events. Levelsof Zn contamination (Figure 4) vary from 220 to3620 ppm, which shows a similar shape to As distri-bution.

The distribution of the remaining trace elementssuch as, heavy metals, shows a similar distribution

Fig. 3. Isoconcentration map of As values in ppm.

Fig. 4. Isoconcentration map of Zn values in ppm.

which the greatest amounts are located over the dump-ing area of El Arteal.

Laboratory column experiments

Laboratory column experiments were performed in apacked bed to simulate the processes that are causedby the leaching of water from soils contaminated bymining and to analyze the mobility of contaminantsin similar conditions such as in natural soils, due toinfluence of precipitation.

Moreover, the experimental results were analyzedby comparing the leaching curves obtained by us-ing two analytical models, the perfectly mixed flowreactor model (RQF) and an advective–dispersivemodel, so as to evaluate the influence of hydrodynamicdispersion in contaminant transport.

The column used in this study was 150 mm in in-ternal diameter and 751 mm in length, and had andendpiece equipped with a 0.50 µm filter. The column

IMPACT OF MINING ACTIVITIES 389

Table 4. Geochemical data of contaminated soils used in the lixiviation tests, concentrations inthe lixiviates (L1–L12) and mean concentration of water distilled extractions of soils (WDE).Values in ppm (soils) and µg L−1 (lixiviates and WDE).

Ag As Ba Cu Pb Sb Zn

Soils 27.5 265 49500 27.5 1881 190 2220

L1 <1.0 <5.0 80.0 31.1 298.8 <1.0 902.2

L2 <1.0 <5.0 90.0 24.9 134.3 <1.0 724.7

L3 <1.0 <5.0 110.0 15.3 63.6 <1.0 623.6

L4 <1.0 <5.0 100.0 11.8 52.2 <1.0 529.8

L5 <1.0 <5.0 130.0 9.1 42.1 <1.0 467.8

L6 <1.0 <5.0 120.0 11.5 42.7 <1.0 446.0

L7 <1.0 <5.0 150.0 29.2 48.4 <1.0 353.5

L8 <1.0 <5.0 200.0 22.4 44.9 <1.0 305.5

L9 <1.0 <5.0 190.0 17.5 63.3 <1.0 305.4

L10 <1.0 <5.0 170.0 15.4 57.2 <1.0 275.5

L11 <1.0 <5.0 200.0 15.1 46.9 <1.0 252.6

L12 <1.0 <5.0 240.0 14.5 42.2 <1.0 207.4

WDE 8.0 46.0 117.1 93.4 109.6 38.3 52.3

was uniformly packed with polystyrene particles of2.97 mm in equivalent diameter, which occupied613 mm of the column and had a very low reactiv-ity. Some of the experiments were performed usinga sandy quartz packer to analyze the phenomena ofadsorption onto this material.

Leaching was induced by introducing distilled wa-ter into the column using a ‘rain-simulator’ connectedto a titration pump, that provided a maximum flow rateof 10 L h−1. The contaminated soil to leach was placedin the upper part of the column, after the column waspacked with polystyrene particles.

The soil sample used in the column experimentswas taken from contaminated soil sampled near thedumping area, and was representative of the geochem-ical characteristics of soils and tailings of the SA(Table 4). This soil was subjected to leaching at a sta-tionary flow rate of 2.5 L h−1, until a ‘pore volumes’ of2.5 was obtained. The ‘pore volumes’, or dimensionaltime, is the leached pore volume of flow at the end ofthe column, defined as follows:

T ′ = vt

L(1)

where v is the average linear fluid velocity (pore ve-locity), t is the elapsed time and L the length of thecolumn.

The effluents obtained from the experiments werefiltered using cellulose nitrate films 0.45 µm in di-ameter, and stored at 4◦C. Their pH, temperature,oxidation–reduction potential and electric conduc-tivity had been determined previously. The pH was

measured potentiometrically and the pH meter wascalibrated before each sample was read. Conductivitywas determined by using a conductimeter calibratedwith NaCl for each effluent sample. The oxidation–reduction potential was measured using an ORP meterwith a combined platinum electrode. The acidifiedsamples (ultra trace HCl, pH = 1.9) were analyzedusing ICP-MS methodology at the University of Bar-celona, to determine their Ag, As, Ba, Cu, Pb, Sb andZn content.

The hydrogeochemical data obtained from columnexperiments and obtained from the distilled-water sol-uble fractions of selected samples were interpretedwith the aid of the geochemical model PHREEQCI(Parkhurst 1995), which calculated the saturation in-dices of the main mineral phases.

Results and discussion

Column experiments

The study of tailings and soils from the SA miningarea suggests the potential mobilization of some ele-ments by water, as the distilled water extracts fromthe soils (WDE) indicated the presence of significantamounts of As, Ba, Cu, Pb, Sb and Zn (Table 4).

The column experiments, which can accurately re-produce the transport of contaminants in the soil, showthe mobilization of Ba, Cu, Pb and Zn (Table 4),elements that were also detected in the groundwater

390 ANDRES NAVARRO ET AL.

beneath the contaminated areas, and show the lowmobility of Ag, As and Sb.

The importance of some of the transport contami-nant parameters, such as hydrodynamic dispersion andthe retardation factor (which is directly related to sorp-tion phenomena), becomes clear if one compares theexperimental column curves with the analytical mod-els that were considered: the advection–dispersion andRQF models, by means of respective equations. Thesemodels are used frequently in the study of contaminantleaching into the soil (Brusseau 1996; Shackelford &Glade 1997).

The estimation approach based on the advective–dispersive transport theory is described by solutions tothe advection–dispersion equation (Bear 1988) for a‘non-reactive’ species in a porous media as follows:

∂C

∂t= ∇(D∇C) − V ∇C (2)

where C is the concentration of solute, D is the hy-drodynamic dispersion of porous media, V is the porevelocity and ∇ is the divergence or gradient operator.The analytical solution of the advective–dispersiveequation for the case of dispersive–advective leach-ing with retardation, evaluated at x = L (L: columnlength), is given as follows (Shackelford & Glade1997):

C

C0= 1 − 1

2erfc

{R − T ′

2(T ′R/P)1/2

}(3)

where C/C0 is the relative concentration, R is the re-tardation factor, T ′ is the dimensionless time (numberof flowing pore volumes), ‘erfc’ is the complemen-tary error function and P is the Péclet number ofthe column. This equation was used to compare theexperimental results for a solute of R = 1 (withoutretardation).

The RQF model can be considered to be the lim-iting case of advective–dispersive transport, and canbe evaluated by comparison with the analytical solu-tion to the RQF differential equation. The equationobtained by treating the column media as a perfectlymixed flow reactor (RQF) is (Brusseau 1996):

RV0dC

dt= −QC (4)

where V0 is the pore volume of the laboratory column(the water associated with the void volume column),and Q is the flow-rate through the column. The analy-tical solution of Equation (4), if one treats the columnas a perfectly mixed flow reactor is:

C

C0= e−KT ′

(5)

where K is a kinetic constant that is determined byfitting it to the experimental data and T ′ is the porevolumes or dimensional time.

Thus, the experimental curves obtained from theleaching experiments show a mobilization of Ba, Cu,Pb and Zn, so that there is a negligible relationshipbetween the predicted curves from the advective–dispersive leaching model (Eq. (3)) and the experi-mental data. However, for many heavy metals, as issuggested by the curves for Pb and Zn (Figures 5 and6), a significant fitting of experimental data is givenby the RQF model for R = 1. This, is evidence ofphenomenon such as the probable rapid dissolution ofsecondary mineral phases present in the contaminatedsoil.

Controls on dissolved contaminants

The concentrations and mobility of dissolved contam-inants within the SA mining area seem to be controlledby a series of precipitation–dissolution reactions, inwhich the secondary phases are of great importance.

Thus, the geochemistry data of distilled water ex-tracts, interpreted using the equilibrium geochemicalmodel PHREEQCI (Parkhurst 1995), suggest that theextractions are saturated to supersaturated with manyminerals: calcite, dolomite, celestite, gypsum, jaro-site, natrojarosite and iron (oxy)hydroxides (Table 5).In the same way, the leachings from column exper-iments are saturated to supersaturated with jarositeKFe3(SO4)2(OH)6, amorphous ferric hydroxide andgoethite. These phenomena have been detected inother tailings areas, showing the pore water satura-tion with respect to jarosite, natrojarosite and gypsum(McGregor et al. 1998; Al et al. 2000).

Most of these secondary phases were detected byX-ray powder diffraction (Table 5) suggesting the re-mobilization in the soil of Ba, Cu, Pb and Zn fromsecondary phases that had precipitated when the soiland pore water conditions favored their precipitation.

Along these lines, Bigham and Nordstrom (2000)indicate that ‘the rapid dissolution’ of accumulatedsoluble salts during subsequent rainfall events mayrelease acidity and produce ‘pulses’ of contaminantsinto the environment.

The column experiments correspond to the geo-chemical and X-ray data, which show a ‘kinetic’dissolution of most secondary phases and the mobi-lization of Cu, Pb and Zn, while the low/negligiblemobility of Ag, As and Sb can be associated with sec-ondary phases of low solubility or sulfides of a low

IMPACT OF MINING ACTIVITIES 391

Fig. 5. Elution curve of Pb and analytical models used.

Fig. 6. Elution curve of Zn and analytical models used.

oxidation rate. The low mobilization of Ag, As andSb in the column experiments is consistent with thosephase minerals that were detected by X-ray, since wehave not detected soluble secondary phases like Fe-sulfoarsenates, or valentinite, whose dissolution canproduce the mobilization of these elements. Further-more, the geochemistry of groundwater in this area(Navarro et al. 1998) reflects either the absence or verylow concentrations of these particular contaminants inthe proximity of dumping areas.

These results suggest that the mobilization of Pbcan be caused by the dissolution of secondary angle-site and plumbojarosite, which was also detected asa secondary phase and/or desorbed from reactive sur-faces (Al et al. 2000), or by the dissolution of jarosite,in which the Pb content may reach high amounts(Gieré et al. 2003).

The mobilization of Cu can be caused by langite,melanterite or brochantite/cal- cantite (also detected inminor quantities) dissolution and desorption from iron

392 ANDRES NAVARRO ET AL.

Table 5. Saturated phases of water extractions from soils, calculatedby PHREEQCI (1), saturated phases calculated by PHREEQCIfrom lixiviates (2) and phase minerals detected by X-ray powderdiffraction (3).

Mineral Component Formula 1 2 3

Ag metal Ag X X

Alunite K[Al3 (OH)6 (SO4)2] X

Anglesite Pb SO4 X

Anhydrite CaSO4 X X

Aragonite CaCO3 X

Artinite Mg2(CO3)(OH)2 · 3H2O X

Barite BaSO4 X X X

Birnessite Na4Mn14O27 · 9H2O X

Boehmite AlOOH X

Brucite Mg(OH)2 X

Calcantite CuSO4 · 5H2O X

Calcite CaCO3 X X X

Celestite SrSO4 X X

Cupricferrite CuFe2O4 X X

Cuprite Cu2O X

Cupruosferrite CuFeO2 X X

Diaspore AlOOH X

Dolomite CaMg(CO3)2 X X

Strontianite SrCO3 X

Ferrihydrite 5Fe23+O3 · 9H2O X X X

Gibbsite Al(OH)3 X X

Goethite α–Fe3+OOH X X X

Goslarite ZnSO4 · 7H2O X

Hausmannite Mn2+Mn23+O4 X X

Hematite ∀-Fe2O3 X X

Hercynite Fe2+CO3 X X

Hydromagnesite Mg5(CO3)4(OH)2 · 4H2O X

Hydronium jarosite (H3O+)2Fe3+6 (SO4)4(OH)12 X X

Huntite CaMg3(CO3)4 X

Jarosite K2Fe63+(SO4)4(OH)12 X X X

Langite [Cu4 (OH)6 (H2O)(SO4)]H2O X

Lepidocrocite γ–Fe3+OOH X X X

Maghemite Fe2O3 X X

Magnesite MgCO3 X

Magnetite Fe2+Fe23+O4 X X

Manganite MnOOH X

Melanterite [(Fe, Zn, Cu)SO4 · 7H2O X

Mg-Ferrite MgFe2O4 X

Natrojarosite Na2Fe3+6 (SO4)4(OH)12 X X

Other oxy-hydroxide X X X

of Fe

Plumbojarosite Pb[Fe3 (OH)6 (SO4)2]2 X

Pyrocroite Mn2+(OH)2 X

Pyrolusite Mn4+O2 X

Siderite Fe2+CO3 X X

Tenorite CuO X X

Gypsum CaSO4 · 2H2O X X X

Uklonskovite Na[Mg F (H2O)2 (SO4)] X

oxyhydroxides and organic matter. Whether mobiliza-tion of Ba occurs is still unclear. The mobilization ofZn can be originated by the dissolution of melanteriteand goslarite and/or desorption processes.

Summary and conclusions

The analysis of tailings taken from uncontrolleddumping sites in El Arteal (SA) shows high amountsof Ag (29.8 ppm), As (285.4 ppm), Ba (5.4%), Cu(57.7 ppm), Pb (2687.5 ppm), Sb (179 ppm), and Zn(2269 ppm). The geochemical data of contaminatedsoils and stream sediments (Tables II and III), in-dicate elevated levels of these metals. The concen-trations of As, Pb and Zn exceed the interventionvalues established in the Andalusian Regulations forContaminated Soil.

The oxidation of sulfides and sulfosalts causesthe precipitation of secondary phases like jaro-site (KFe3(SO4)2(OH)6), natrojarosite (NaFe3(SO4)2(OH)6), crystalline oxyhydroxide of Fe (goethite),amorphous ferric hydroxide (Fe (OH)3), clay miner-als, anglesite (PbSO4), alunite (KAl3(SO4)2(OH)6),gypsum (CaSO4 · 2H2O), and other less abundant min-eral phases.

The experimental curves obtained from leach-ing column experiments show a negligible relationbetween the advective–dispersive model and the ex-perimental data and a good fitting with the RQFmodel, and suggest the mobilization of Ba, Cu, Pb andZn, because of the probable rapid dissolution and/ordesorption of these elements from secondary mineralphases present in the contaminated soil.

The geochemistry data of distilled water extractsfrom soils and column experiments indicate that theextractions and leachings are saturated to supersatu-rated with many minerals, namely calcite, dolomite,gypsum, jarosite, and iron (oxy)hydroxides (Table 5).

These results suggest that the secondary phaseshave the potential to serve as a sink for dissolvedcontaminants in dry periods and also to serve as astock of mobilizable elements when, in humid periods,rainwater dissolves the secondary minerals.

Acknowledgements

This work was funded by the UPC Fluid MechanicsDepartment and the UAL Agricultural Chemistry andSoil Science Department, with partial financing from

IMPACT OF MINING ACTIVITIES 393

DSM-Deretil, through University-Enterprise contractC-3358, undertaken between 1998 and 2002. The au-thors would also like to thank the anonymous reviewerwhose comments and suggestions greatly enhancedthe manuscript.

References

Al TA, Martin CJ, Blowes DW. 2000 Carbonate-mineral/waterinteractions in sulphide-rich mine tailings. Geochimica etCosmochimica Acta 23, 3933–3948.

APHA. 1989 Standard Methods for the Examination of Water andWastewater, 17th edn. Washington: APHA.

Appelo CAJ, Postma D. 1996 Geochemistry Groundwater andPollution. Rotterdam: Balkema, 536 pp.

Azcue JM. 1999 Environmental Impacts of Mining Activities.Berlin: Springer, 300 pp.

Bear J. 1988 Dynamics of Fluids in Porous Medi. New York: DoverPublications Inc., 764 pp.

Bigham JM, Nordstrom DK. 2000 Iron and aluminium hy-droxysulfates from acid sulfate waters. In Alpers CN, Jambor JL,Nordstrom DK, eds. Sulfate Minerals – Crystallography, Geo-chemistry, and Environment Significance 40. Washington: Min-eralogical Society of America, Geochemical Society, Reviews inMineralogy and Geochemistry, pp. 351–403.

Blowes DW, Jambor JL. 1990 The pore-water geochemistry and themineralogy of the vadose zone of sulfide tailings, Waite Amulet,Québec, Canada. Applied Geochemistry 5, 327–346.

Brusseau ML. 1996 Evaluation of simple methods for estimatingcontaminant removal by flushing. Ground Water 34, 19–22.

Collado D. 2002 Movilización de contaminantes en el terreno apartir de Suelos Contaminados, MSc Thesis, Universidad deGranada.

EPA. 1991 Description and Sampling of Contaminated Soils. AField Pocket Guide. EPA/625/12-91/002.

EPA. 1996 Managing Environmental Problems at Inactiveand Abandoned Metal Mine Sites, E.P.A./625/R-95/007,Washington.

Förstner U, Wittman TW. 1979 Metal Pollution in the AquaticEnvironment. Berlin: Springer-Verlag, 486 pp.

Gieré R, Sidenko NV, Lazareva EV. 2003 The role of second-ary minerals in controlling the migration of arsenic and metalsfrom high-sulfide wastes (Berikul gold mine, Siberia). AppliedGeochemistry 18, 1347–1359.

Herbert RB. 1996 Metal retention by iron oxide precipitation fromacidic ground water in Dalarna, Sweden. Applied Geochemistry11, 229–235.

López Ruiz J, Rodríguez E. 1980 La región volcánica neógena delsureste de España. Estudios Geológicos 36, 5–63.

Martínez J, García J, López Ruiz J, Reynolds GA. 1992 Discoveryof fossil fumaroles in Spain. Economic Geology 87, 444–447.

Martínez J, Navarro A, Lunar R. 1997 First reference of pyriteframboids in a Hg-Sb mineralization: the Valle del Azogue min-eral deposit (SE Spain). Neues Jahrbuch Mineral Monatsh 4,175–184.

McGregor RG, Blowes DW, Jambor JL, Robertson WD. 1998 Thesolid-phase controls on the mobility of heavy metals at the Cop-per Cliff tailings area, Sudbury, Ontario, Canada. Journal ofContaminant Hydrology 33, 247–271.

Navarro A, Viladevall M, Font X, Rodríguez P. 1994 Las min-eralizaciones auríferas de Sierra Almagrera (Almería). Estu-dio geoquímico y modelos de yacimientos. Boletín Gológico yMinero 105–1, 85–101.

Navarro A, Collado D, Sánchez JA. 1998 Análisis de la contam-inación por actividades mineras de los suelos de la cuenca bajay delta del río Almanzora. Boletín Geológico y Minero 109,69–87.

Navarro A, Martínez J, Font X, Viladevall M. 2000 Modellingof modern mercury vapor transport in an ancient hydrothermalsystem: environmental and geochemical implications. AppliedGeochemistry 15, 281–294.

Parkhurst DL. 1995 Users Guide to PHREEQC – a ComputerProgram for Speciation, Reaction-Path, Advective-Transport,and Inverse Geochemical Calculations, U.S. Geological Survey,Water-Resources Investigations Report 95–4227.

Ribet I, Ptacek CJ, Blowes DW, Jambor JL. 1995 The potential formetal release by reductive dissolution of weathered mine tailings.Journal of Contaminant Hydrology 17, 239–273.

Shackelford CD, Glade MJ. 1997 Analytical mass leaching modelfor contaminated soil and stabilized waste. Ground Water 35,233–241.

Stollenwerk KG. 1994 Geochemical interactions between constitu-ents in acidic groundwater and alluvium in an aquifer near Globe,Arizona. Applied Geochemistry 9, 353–369.

Viladevall M, Font X, Navarro A. 1999 Geochemical mercury sur-vey in the Azogue Valley (Betic area, SE Spain). Journal ofGeochemical Exploration 66, 27–35.

Wray DS. 1998 The impact of unconfined mine tailings and anthro-pogenic pollution on a semi-arid environment – an initial study ofthe Rodalquilar mining district, south east Spain. EnvironmentalGeochemistry and Health 20, 29–38.