This Article Appeared in a Journal Published by Elsevier. the Attached Copy Is Furnished to the Author for Internal Non-Commerci

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright Author's personal copy

Journal of Geochemical Exploration 116 (2012) 1–7

Contents lists available at ScienceDirect

Journal of Geochemical Exploration

journal homepage: www.elsevier.com/locate/jgeoexp

Mercury vapor emissions from the Ingenios in Potosí (Bolivia)

P. Higueras a,⁎, W. Llanos a, M.E. García b, R. Millán c, C. Serrano d a Instituto de Geología Aplicada-Laboratorio de Biogeoquímica de Metales Pesados, Universidad de Castilla-La Mancha, Pl. Manuel Meca, 1, 13400 Almadén (Ciudad Real), Spain b Universidad Mayor de San Andrés, Facultad de Ciencias Químicas, La Paz, Bolivia c Centro de de Investigaciones Energéticas, Medio Ambientales y Tecnológicas (CIEMAT), Madrid, Spain d Universidad Tomás Frías, Potosí, Bolivia article info abstract

Article history: Potosí (South West Bolivia) is a well known historical mining site in the world, with mining activity centered Received 27 October 2010 at the so-called Cerro Rico. It is an impressive mount formed by rhyolitic rocks affected by intensive Accepted 5 May 2011 hydrothermal alteration, and hosting a complex vein deposit including mainly Ag and Sn minerals. From the Available online 13 May 2011 start of the mining activity, in the late 16th century, to 1850, the main ore was silver minerals, and from 1850 the silver ores exhausted, and mining activity centered on tin minerals. During the ﬁrst stage, the silver Keywords: minerals were treated by amalgamation, using the so-called “método de patio”, which implied the usage of Mercury emission Contaminated soils mercury and other compounds as metallurgical agents, and supposed the release of important quantities of Potosí, Bolivia mercury to the local environment. This work was carried out at the “ingenios”, milling and mercury processing facilities located next to streams, in order to have the water and mechanical energy needed for the process, and nowadays in ruins. Our results put forward very low mercury vapor concentrations in the region, reaching only occasionally values over 4 ng m−3, as well as in the town area, were maximum values reach 31 ng m−3 with an average of 5.5 ng m−3; detailed surveys at the “Ingenios” demonstrated that in these facilities mercury vapor concentrations were also low, but the excavation of the topsoil causes an important release of the elemental vapor, reaching concentrations over 3000 ng m−3. Causes of this low emission of unmodiﬁed soil are here interpreted as caused by biological and physicochemical transformation of the metallic mercury accumulated in the soil, to mineral phases such as cinnabar/metacinnabar and/or schuetteite, in reactions mediated by the formation of methylmercury. © 2011 Elsevier B.V. All rights reserved.

1. Introduction (1981), Cunningham et al. (1991), Steele et al. (2003) and Rice et al. (2005), among others. Potosí, located in South West Bolivia, is a world-class Ag–Sn deposit, There is evidence for pre-colonial mining activity in the area having yielded more than 56,000 tons of silver and 100,000 tons of tin (Abbott and Wolfe, 2003; Cooke et al., 2010), but the most intensive (Rice et al., 2005). The deposit is hosted at an impressive volcano-like mining activities of the deposit started in 1574, during the Spanish structure, the so-called Cerro Rico (rich mount, after the Spanish name), Colonial period, and exploitation was centered in the rich silver veins. constituted by a rhyolitic dome of Miocene age (13.77±0.03 Ma, In 1850, and coinciding with the independence of Bolivia, silver 40Ar–39Ar age), Rice et al. (2005), in the Cordillera Oriental area (Fig. 1). production declined sharply, and was replaced by tin production due The origin of the deposit is hydrothermal, and it consists in a complex of the exhaustion of the higher Ag-rich vein system. During this vein system, including two main groups: one comprising the silver rich “Silver period”, between 1574 and 1850, silver metallurgy was based mineralogy, including the ore minerals: stannite (Cu2FeSnS4)chalco- in the so called “método de patio”, a procedure implemented by pyrite (CuFeS2), tetrahedrite ((Cu,Fe)12Sb4S13), sphalerite ((Zn,Fe)S), Bartolomé de Medina (1497–1585; an early Spanish metallurgist) in andorite (PbAgSb3S6), matildite (AgBiS2), ruby silvers (chieﬂypyrar- San Luis Potosí (México) in 1550 based on amalgamation with gyrite, Ag3SbS3), jamesonite (Pb4FeSb6S14), and boulangerite metallic mercury. A large number of processing installations were set (Pb5Sb4S11); and a deeper vein system including pyrite (FeS2), up at Potosí, all along the so-called “Ribera de la Vera Cruz”, and also cassiterite (SnO2), and arsenopyrite (FeAsS) with some wolframite “Ribera de los Ingenios”, a channel crossing the town in order to have ((Fe,Mn)WO4) and bismuthinite (Bi2S3)(Sillitoe et al., 1975). More the water and mechanical energy needed for the process. Cooke et al. detailed descriptions of the deposit geology can be found in Francis et al. (2010) reported silver production, probably still based on amalgamation until 1930, but at a minor scale than that of the Colonial period. Up to thirty two “Ingenios” (Strosnider et al., 2008) were in activity “ ” ⁎ Corresponding author. during the main Silver Period of Potosí. All of them are presently in E-mail address: [email protected] (P. Higueras). ruins.

2 P. Higueras et al. / Journal of Geochemical Exploration 116 (2012) 1–7

Fig. 1. Geographic location of the Potosí area, and of the soil sampling sites. Sampling sites (Ingenios): A: Agua de Castilla; B: Laguacayo; C: Cantumarca. Historic sites: 1: Universidad Tomás Frías; 2: Casa de la Moneda; 3: Nuestra Señora de la Merced Church; 4: Cathedral; 5: San Francisco Church.

From an environmental point of view, the area has been subject of (Chumacero,2007, with a yearly average temperature of 9.3 °C, a a number of studies, including those of Cooke et al. (2010), analyzing maximum around 14 °C in summer (December–January) and a mercury contents in historic series of sediments in a lake near Potosí, minimum around 4.5 °C in winter (July). Average annual rainfall in which allowed detection of mercury pollution as old as the twelfth the studied area is 336 mm, whereas evapotranspiration calculated by century; Archer et al. (2005) study arsenic pollution in the Pilcomayo means of the Turc expression, is 302 mm year− 1; thus, effective River basin, draining the Potosí and Cerro Rico areas; Miller et al. rainfall is only 34 mm year− 1, which is mainly recorded in February (2004) study multi-heavy metals pollution of the Pilcomayo River and March. Dominant winds, especially during winter, blow from SW basin, draining the Potosí and Cerro Rico areas, including mercury and are usually cold winds from the huge Uyuni Salar. On the other analyses of irrigation waters, soils from agricultural fields adjacent to hand, during spring and summer more template winds blow from the the river and tributaries, and local vegetable samples, finding high to East. Population is in the order of 133,000 inhabitants, being among very high concentrations of Pb, Zn, Cu, Sb, Cd and Ag, but with the most important Bolivian cities. The town is crossed from East to mercury values always very low. Finally, Hagan et al. (2011) have West by the “Estero de los Ingenios”, a river canalized in the urban area, made an estimation of historical mercury concentrations in the and used traditionally to provide mechanical power and water supply atmosphere of Potosí, on the basic of historic production data, and to the “Ingenios”. analyzed mercury concentrations in the local soils, finding values between 0.105 and 155 mg kg− 1; an important conclusion expressed 3. Materials and methods in the Hagan et al. (2011) paper is the need for studies correlating the present day soil Hg concentrations with atmospheric Hg concentra- The work consisted in a car-driven mercury vapor survey tions, which is precisely the main goal of our study. including regional, local and site scale minor surveys. All of them were performed by using a LUMEX RA-915+ analytical device, that 2. Climatic and physiographic setting consist in an Atomic Absorption Spectrometer using Zeeman effect to perform background correction, and with high frequency modulation The Potosí area is located in the Eastern Cordillera range of the of light polarization (ZAAS-HFM) (Sholupov and Ganeyev, 1995). The Central Andes (South West Bolivia). The town is one of the world's device has a measuring range between 2 and 20,000 ng Hg m− 3, and highest cities by elevation, with an average height of 4067 m above it is able to analyze and register total gaseous mercury concentration sea level, while the Cerro Rico peak reaches an altitude of 4824 m. every second, data which is stored in a connected laptop computer or Local climatic conditions are typical of a semiarid mountainous area in an internal datalogger. Author's personal copy

P. Higueras et al. / Journal of Geochemical Exploration 116 (2012) 1–7 3

In parallel to mercury determinations, a registry of geographic center of Potosí (20K 211023, 7832276), with an average altitude of locations was performed by means of a GPS, registering also the 3870 m, including a maximum of 4285 m and minimum of 3384 m. position every second, so that a detailed georeferenciation of every The results (Figs. 2A and 3A and Table 1) show very low Hg analytical data was carried out merging the analytical and GPS files. concentrations, with an average below detection limit (2 ng Hg m− 3), Regional and local surveys were performed from a moving car, at very low standard deviation, and just some scattered values above speeds always below 50 km h− 1, and site surveys were performed by 4ngm− 3 (Fig. 3A). The latter correspond to the crossing of the “Ribera feet. Regional survey included a 200 km long track following the de los Ingenios” by the Oruro-Potosí road and to the arrival to the town Oruro-Potosí road; two local surveys were carried out, in the Potosí of Potosí. town and the Cerro Rico mount. Site surveys included the “Agua de Castilla” Ingenio, located some 900 m far from the town limits 4.2. Local scale surveys approximately following the SE direction, the “Laguacayo” Ingenio, located into the town, and the “Cantumarca” Ingenio (also called The local scale surveys were carried out at the Potosí urban area “Ingenio del Rey”, King's Ingenio), located some 1500 m NW of the and Cerro Rico. The Potosí urban area corresponds to a rather town limits. All of them are located in the proximity of the “Ribera de extensive sector (11 km2) located in a soft slope (0–5° in average) los Ingenios” before mentioned, “Agua de Castilla” upstream from the with maximum and minimum altitudes of 3384 and 4285 m, town, and “Cantumarca” downstream. respectively. As shown in Fig. 1, the town is crossed from SE to NW Statistical treatment of the data included the calculation of by the “Ribera de los Ingenios” water channel, where all the 32 facilities averages (common and geometric) and standard deviation, and the were located (the ruins can still be visited). The city track crossed the plotting of the logarithmic frequency histograms of mercury data. For Ribera three times, at different sectors of the town, always close to the data below detection limit (at 2 ng Hg m− 3) we followed two “Ingenios”. The results, shown in Table 1 and plotted in Figs. 2B and 3B, procedures. Those at 1 ng Hg m− 3 were regarded as statistically valid show higher Hg concentrations than those measured in the regional for calculations (e.g. Singh and Nocerino, 2002), and those below that survey, with geometric mean above 5.5 ng Hg m− 3, and a maximum figure were discarded but indicated as % in Table 1. concentration of 30.9 ng Hg m− 3. The histogram (Fig. 3B) represents a Soils samples were taken using hoe and spade from the three study single mode population, which cannot be properly defined as log- sites, each one composed of three subsamples taken at different normal, although it is close to it, and do not display any indication of depths: 0–5, 5–20 and 20–25 cm. Samples (~0.5 kg) were kept in differentiated anomalous population(s). Plot of mercury concentra- plastic bags until preparation. Samples were dried at room temper- tion vs. time (Fig. 2B) shows this variability, but also a minor ature, sieved to b2 mm, and grinded in an agate mortar to less than relationship with the proximity or crossing the above mentioned 50 μm. Aliquots of the samples were obtained for Hg analysis, and also “Ribera de los Ingenios”. for determinations of organic matter and soil texture. The chemical The Cerro Rico survey covers a track carried out on this mountain, analysis was carried out by Atomic Absorption Spectrometry (AAS) at 3907–4500 m of altitude. Mercury mean concentration during this with thermal decomposition, using AMA254 equipment (Leco In- survey was 3.08 ng Hg m−3 (2.45 ng Hg m− 3 as geometric mean), struments). In this technique, soil samples are dried before thermally with a standard deviation of 2.68, and with a maximum peak of decomposed on a catalytic tube. The decomposition products are then 24.4 ng Hg m− 3. These concentrations are not far from the regional carried to an amalgamator for a selective trap of mercury at 120 °C to averages, confirming that mercury is not being used nowadays in prevent water condensation. After that, mercury present in the mineral processing. The mercury profile is rather flat (Fig. 2C), and a amalgamator is released by a short heating-up, and transferred by decrease in background mercury levels is noted when leaving the carrier gas (oxygen) through the longer measuring cuvette. Mercury town of Potosí. The peaks reaching 25 ng Hg m− 3 are coincident with is quantified on a dynamic range of 0.05–600 ng Hg, with a detection active mine galleries, indicating either the possible usage of limited limit of 2 ng g− 1. Quality control was assessed by analysis of amounts of the metal by present day miners or the presence of an duplicates and reference material: (SRM) NIST 2710 and (SRM) unreported Hg mineral phase. The histogram (Fig. 3C) suggests the NIST 2711. Equivalent surveys and procedures have been used in presence of two well differentiated populations: one background Almadén (Spain) by Higueras et al. (2006) and Martínez-Coronado et population and an anomalous one including the Potosí town values al. (2010); as well as in Chile, IV Region by Higueras et al. (2005). and the peaks registered in the Cerro Rico area.

4. Results and discussion 4.3. Site scale surveys

4.1. Regional scale survey We include here three surveys carried out by feet in the above mentioned Ingenios of “Agua de Castilla”, “Laguacayo” and “Cantu- The regional scale survey served as baseline for the general survey, marca”. The main difference with previous ones is the more detailed and was performed along 200 km of the road from Oruro to Potosí, in a inspection of the points where the presence of mercury in the soil can track from the locality of Ventilla (19K 793863, 7888290) to the urban be more evident. The proper visit to the sites also allows soil excavation and in-situ measurements. Statistical data from these sites are to be taken by caution because the sampled air is not as Table 1 homogenous as that taken from a car in motion. The data are also Summary of mercury concentrations during the six surveys carried out in the Potosí shown in Table 1 and Figs. 2D to F and 3D to F, in order to analyze area. N: Number of determinations. StDev: Standard Deviation. G.Aver.: Geometric average. Max: Maximum value during survey. b1 indicates the percentage of values them in descriptive terms. below this measured concentration, considered as below the instrumental detection The “Agua de Castilla” Ingenio is located in the outskirts of Potosí, − limit. All values (except N and b1) in ng m 3. some 900 m to the SE (Fig. 1), at an altitude of 3671–3736 m. This area ﬁ Survey N Average StDev G. Aver. Max. b1 (%) was the rst studied in detail, and it was surprising to record extremely low mercury contents. However, everything changed when the soil Regional 7512 1.83 0.95 1.71 25.8 32.85 Potosí 11,740 7.53 6.01 5.53 30.9 21.13 was excavated in a point where ancient pottery fragments were found. Cerro Rico 7228 3.08 2.68 2.45 24.3 27.79 The excavation of the topsoil resulted in a remarkable increase of Agua de Castilla 8237 36.72 126.49 8.97 3379.0 18.42 mercury concentrations in the air (see point SR1 in Fig. 2D). Close to Laguacayo 2576 166.12 1075.53 10.19 16108.0 0.99 this point, there were discovered some other points where the removal Cantumarca 4018 59.71 227.24 14.16 2527.9 17.34 of the topsoil also resulted in important increases of mercury in air Author's personal copy

4 P. Higueras et al. / Journal of Geochemical Exploration 116 (2012) 1–7 A B

RI RI1 RI2

RI3

C D SR1 MG2 MG6

Pot

MG1 MG3 MG5 MG7 SR2 MG4

SR3

E F SR2 SR4 SR3

SR6 SR7 SR5 SR2 SR4 SR1 SR3 SR1 SR5

Fig. 2. Mercury concentration in the air versus time, along the six surveys carried out: A: Regional scale (last 200 km of the road from Oruro to Potosí, by car); B: Potosí city, by car; C: Cerro Rico area, by car; D: Agua de Castilla Ingenio, by foot; E: Laguacayo Ingenio, by foot; F: Cantumarca Ingenio, by foot. Abbreviations: RI: Ribera de los Ingenios crossing before arriving to Potosí; AP: Arrival to Potosí; Pot: Potosí urban area before going to Cerro Rico area; MG: Mine galleries passed by during the Cerro Rico survey; SR: Points of topsoil removal during the local by foot surveys at the Ingenios. All Hg concentrations in ng m− 3. contents. This is shown in the mercury profile (points SR2 and SR3 in The “Laguacayo” Ingenio is located inside the town, at an altitude Fig. 2D). General results (Table 1) show values that, as indicated in the 3879–4042 m, close to the “Ribera de los Ingenios”, and at present methodological section, cannot be compared to those from the car serves as a car-mechanics workshop. The facility is inside a closed surveys. First, a longer time was spent at these particular sites and precinct, with 2 m high walls enclosing the whole area. Following the second but not least, it is the excavated soil that generates the highest criteria put forward in the “Agua de Castilla” Ingenio, we measured the concentrations of mercury. Whatever the case, we measured Hg Hg concentrations near the undisturbed soil and after removal of the concentrations above 3000 ng Hg m− 3, which is three times the topsoil layer, which resulted in extremely high gaseous mercury maximum concentration level recommended by World Health concentrations (Table 1), as shown also in the mercury profile Organization (WHO, 2000) for chronic exposure to Hg vapors. On the (Fig. 2E). Analysis of the histogram (Fig. 3E) puts forward what could other hand, this value is of low significance in terms of local risk be a local background population and an anomalous one, with a assessment concerning the inhalation route of exposure, since it threshold separating both populations at 60 ng Hg m− 3. corresponds to punctual concentration, measured at approximately 10 “Cantumarca” Ingenio is located some 1150 m to the west from to 50 cm high with respect to the soil surface and immediately after the Potosí, also in the “Ribera de los Ingenios”, but downstream from the excavation. The histogram (Fig. 3D) is of complex analysis, and could town, and it was built on top of the alluvial sediments of the Ribera. be an indication of the sum of a number of different populations, Thus, the site corresponds to a plain area, with average altitude of separated by thresholds of also different entity. 3692 m. Mean results are in the same order of magnitude as in “Agua Author's personal copy

P. Higueras et al. / Journal of Geochemical Exploration 116 (2012) 1–7 5

A 350 B 300

300 250

250 200

200 150 150 Frequency Frequency 100 100

50 50

0 0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.00 1.58 2.51 3.98 6.31 10.00 15.85 25.12 1.00 1.58 2.51 3.98 6.31 10.00 15.85 25.12

C 250 D 350

300 200 250

150 200

150 100 Frequency Frequency 100 50 50

0 0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 1.00 1.58 2.51 3.98 6.31 10.00 15.85 25.12 1.00 3.16 10.00 31.6 100.0 316.2 1000 3162

E 400 F 400

300 300

200 200 Frequency Frequency

100 100

0 0 0.0 0.6 1.2 1.8 2.4 3.0 3.6 4.2 0.0 0.5 1.0 1.5 2.0 2.5 3.0 1.00 3.98 15.85 63.10 251.2 1000 3981 15849 1.00 3.16 10.00 31.6 100.0 316.2 1000

Fig. 3. Logarithmic histograms for measured mercury concentrations in the air for the six surveys carried out, in same order as in Fig. 2. In abscises it is indicated both the logarithmic and the corresponding natural values (below) for the intervals considered. All Hg concentrations in ng m− 3. de Castilla” Ingenio. Following similar procedures we measured the Hg Although the mineral speciation was beyond the scope of this concentrations near the undisturbed and excavated soil. The histo- work, the results from a preliminary speciation study carried out by gram (Fig. 3F) suggests the presence two populations, one with Hagan et al. (2011) by means of sequential selective extractions on background concentrations, and a minor anomalous one, separated by one soil sample from one of the local mills indicated the following a threshold located between 100 and 200 ng Hg m− 3,and results: 1) approximately 80% of the mercury was extracted in the corresponding to the peaks obtained above the excavated soil. ﬁnal step (aqua regia extraction), and 2) the smaller amount extracted

4.4. Mercury contents in soils Table 2 Mercury and organic matter (OM) concentrations in soil samples. Table 2 shows the total mercury content measured in the three soil − 1 − 1 samples (9 subsamples). Contents from “Laguacayo” are the highest, Sample Depth (cm) Hg (mg kg )OM(gkg) although all values can be considered low as compared to contents AC-1 (Agua de Castilla) 0–5 31.2 52 from heavily Hg contaminated areas, such as the Almadén (Spain) 5–20 29.5 26 – mercury mining district, but high as compared to baseline levels 20 30 27.5 24 – fi MW-1 (Laguacayo) 0 5 54.0 24 (Table 3). The distribution of Hg in the soil pro le appears to be 5–20 101.2 14 regular, with local variation, low significance, and maxima concen- 20–30 32.0 8 trations at the 5–20 cm depth level in those areas deeply influenced C-1 (Cantumarca) 0–5 13.6 23 by man activities. The latter can thus be more properly regarded as 5–20 18.4 22 20–30 21.1 17 Anthrosols than undisturbed soils where Hg concentrations are low. Author's personal copy

6 P. Higueras et al. / Journal of Geochemical Exploration 116 (2012) 1–7

Table 3 sublimation capacity, so preventing the mercury emissions from the Reference values of mercury concentrations in soils. topsoil. Mercury contents in soil profiles (Table 2) support the overall Area Hg (mg kg− 1) Reference process: mercury in these areas should have arrived to the soils from atmosphere, in the form of gas produced during the retorting Potosí 0.105–155 Hagan et al. (2011) Almadén (Spain) 6–9000 Higueras et al. (2003) processes, and possibly also in liquid form, because of accidental Almadenejos (Spain) 4–15 900 Martínez-Coronado et al. (2010) leakages. Later microbial activity in the soil could have induced Baseline levels 0.01–0.03 Senesi et al. (1999) formation of mobile forms, possibly including methylmercury, and favoring the infiltration of the element into deeper horizons. Sunlight and also bacterial activity may have favored the formation of in the fourth step (nitric acid) suggests that most mercury in the soil schuetteite and cinnabar/metacinnabar, respectively, in the topsoil. was in the form of cinnabar and/or metacinnabar. The key question here is where this cinnabar and/or metacinnabar came from. It is 5. Conclusions unlikely that the Spaniards have brought with them as raw minerals from Almadén (Spain), as no records were found in the so-called Our study has shown very low mercury vapor emissions from soils Archivo de Indias, at Seville (Cortés Alonso, 1988). In-situ formation of contaminated with metallic mercury in an area characterized by the either cinnabar or metacinnabar via reactions of the type S+Hg→HgS long-lasting use of mercury, but also for the cessation of its use is thermodynamically favored, however, sufficient heat has to be 250 year ago. This anomaly could be explained in terms of the provided to reach the activation energy required by the reaction formation at the topsoil of new compounds and mineral phases. These (López et al., 2010), which make the Ingenios old metallurgical would comprise methylmercury as an intermediate and infiltration- facilities an unlikely scenario. Alternatively, a far more probable and favoring phase, and schuetteite and/or metacinnabar, concentrating simpler scenario would be provided by the extraction, at some point in the topsoil and with no mercury sublimation capacity, so limiting in the mining history of Cerro Rico, of small amounts of cinnabar that the possibilities of vapor emissions. Further detailed studies on the passed mostly unnoticed by a group of colonial workers more mineralogy and chemical speciation of the topsoil compared to deeper interested in silver than in the subtleties of Hg mineralogy. This call soil levels are needed in order to confirm this idea. Nevertheless, the for future and enlightened Hg mineral speciation studies at Cerro Rico, atmospheric mercury concentrations found during this work shows because if cinnabar was indeed extracted from the mountain, then its that atmospheric mercury pollution is far from being a real local existence could much better explain, or at least, shed some more light, problem in this area, unless topsoil is removed at decommissioned on the present distribution of Hg in soils and atmosphere. mills areas.

4.5. Causes of the low emission values Acknowledgements Mercury emission from the Potosí local soils are much lower than expected for an area with so intensive mercury pollution. Values This study was funded by Projects PII1I09-0142-4389 (Regional above the reference concentration (300 ng Hg m− 3) of the US Agency Viceconsejería de Universidades e Investigación, Consejería de Medio for Environmental Protection (U.S. EPA, 2009) would not be detected Ambiente, JCCM) and CGL2009-13171-C03 (Spanish Ministerio de in the area, if topsoil was not previously removed. It may be easy to Ciencia e Innovación). The contribution to the improvement of the think that after 250 years of cessation of the intensive mercury usage, original manuscript by two anonymous reviewers and in particular the topsoil should become depleted in mercury via emission of the guest editor of this issue, Manuela Abreu (Technical University of mercury vapors to the atmosphere. However, as shown by the facts, Lisbon, Portugal), is also acknowledged. An anonymous colleague considerable emissions from the excavated soil still persist. An (and friend) contributed to the improvement of English redaction. explanation for this anomalous behavior could be provided by the chemical transformations of the mercury contained in the topsoil into References species immobilizing the element. Concerning these possible transformations, the best documented one is the formation of methylmer- Abbott, M.B., Wolfe, A.P., 2003. Intensive pre-Incan metallurgy recorded by lake – cury ([CH Hg]+) by means of physic-chemical and microbial sediments from the Bolivian Andes. Science 301 (5641), 1893 1895. 3 Archer, J., Hudson-Edwards, K.A., Preston, D.A., Howarth, R.J., Linge, K., 2005. Aqueous reactions (Gray et al., 2004; Jacobs and Keeney, 1974; King et al., exposure and uptake of arsenic by riverside communities affected by mining 2002; Siciliano and Lean, 2002); however, the low organic contents of contamination in the Río Pilcomayo basin, Bolivia. Mineralogical Magazine 69, these soils (Table 2) should be a limitation for this reaction, and in any 719–736. Baldi, F., 1997. Microbial transformation of mercury species and their importance in the case it will probably not limit the soil capacity for mercury vapor biogeochemical cycle of mercury. Metal Ions in Biological Systems 34, 213–257. emission to the extend recognized in our study. Besides, formation of Barnett, M.O., Harris, L.A., Turner, R.R., Stevenson, R.J., Henson, T.J., Melton, R.C., this Hg highly mobile compound should have resulted in the eventual Hoffman, D.P., 1997. Formation of mercuric sulfide in soil. Environmental Science – fi and Technology 31 (11), 3037 3043. in ltration of the element into deeper levels in the soil, even in an area Bernaus, A., Gaona, X., Esbrí, J.M., Higueras, P., Falkenberg, G., Valiente, M., 2006. with so low effective rain. Furthermore, Higueras et al. (2003), Kim et Microprobe techniques for speciation analysis and geochemical characterization of al. (2004), Bernaus et al. (2006) and Esbrí et al. (2010) describe the mine environments: the mercury district of Almadén in Spain. Environmental Science and Technology 40, 4090–4095. presence of schuetteite (Hg3O2(SO4)) in the supergene environment, Chumacero, J., 2007. Diagnóstico Previo—Aporte al Plan de Desarrollo Municipal de which forms from mercury vapor condensed on the sun-exposed Potosí 2007–2011. Honorable Alcaldía Municipal Potosí, Programa BOL/AIDCO/ surface of broken rocks in waste dumps. Baldi (1997) explained the 2002/0467 APEMIN II. http://www.apemin.eu/PDM/PDMPotosi.pdf. formation of metacinnabar (HgS) from methylmercury by means of Cooke, C.A., Balcom, P.H., Kerfoot, C., Abbott, M.B., Wolfe, A.P., 2010. Pre-Colombian mercury pollution associated with the smelting of argentiferous ores in the Bolivian microbial activity, and Barnett et al. (1997) describe the formation of Andes. Ambio 40 (1), 18–25. metacinnabar in mercury contaminated soils in the presence of other Cortés Alonso, V., 1988. Archival education in Spain. American Archivist 51, 330–335. fi Cunningham, C.G., McNamee, J., Pinto Vasquez, J., Ericksen, G.E., 1991. A model of metal sul des or sulfhydryl groups in the soil. Both possibilities are – fi dome-hosted precious metal deposits in Bolivia. Economic Geology 86, 415 421. realistic, since the rst (schuetteite formation in mining waste) is Esbrí, J.M., Bernaus, A., Ávila, M., Kocman, D., García-Noguero, E.M., Guerrero, B., Gaona, supported by observations from many other old mining areas, and the X., Álvarez, R., Perez-Gonzalez, R., Valiente, M., Higueras, P., Horvat, M., Loredo, J., latter (cinnabar/metacinnabar formation in the soil) is locally 2010. XANES speciation on mercury on three mining districts—Almadén, Asturias (Spain), Idria (Slovenia). Journal of Synchrotron Radiation 17, 179–186. supported by Hagan et al. (2011) data on mercury sequential selective Francis, P.W., Baker, M.C.W., Halls, C., 1981. The Kari Kari caldera, Bolivia, and the Cerro extractions. Both processes imply the formation of minerals with no Rico stock. Journal of Volcanology and Geothermal Research 10, 113–124. Author's personal copy

P. Higueras et al. / Journal of Geochemical Exploration 116 (2012) 1–7 7

Gray, J.E., Hines, M.E., Higueras, P.L., Adatto, I., Lasorsa, B.K., 2004. Mercury speciation Miller, J.R., Hudson-Edwards, K.A., Lechler, P.J., Preston, D., Macklin, M.G., 2004. Heavy and microbial transformations in mine wastes, stream sediments, and surface metal contamination of water, soil and produce within riverine communities of the waters at the Almadén Mining District, Spain. Environmental Science and Río Pilcomayo basin, Bolivia. Science of the Total Environment 320, 189–209. Technology 38, 4285–4292. Rice, C.M., Steele, G.B., Barfod, D.N., Boyce, A.J., Pringle, M.S., 2005. Duration of Hagan,N.,Robins,N.,Hsu-Kim,H.,Halabi,S.,Morris,M.,Woodall,G.,Zhang,T.,Bacon,A., magmatic, hydrothermal, and supergene activity at Cerro Rico de Potosi, Bolivia. Richter, D de B., Vandenberg, J., 2011. Estimating historical atmospheric mercury Economic Geology 100, 1647–1656. concentrations from silver mining and their legacies in present-day surface soil in Potosí, Senesi, G.S., Baldassare, G., Senesi, N., Radina, B., 1999. Trace element inputs into soils by Bolivia. Atmospheric Environment 45 (40), 7619–7626. anthropogenic activities and implications for human health. Chemosphere 39, Higueras, P., Oyarzun, R., Biester, H., Lillo, J., Lorenzo, S., 2003. A first insight into 343–377. mercury distribution and speciation in soils from the Almadén mining district, Singh, A., Nocerino, J., 2002. Robust estimation of mean and variance using Spain. Journal of Geochemical Exploration 8, 95–104. environmental data sets with below detection limit observations. Chemometrics Higueras, P., Oyarzun, R., Lillo, J., Oyarzún, J., Maturana, H., 2005. Atmospheric mercury and Intelligent Laboratory Systems 60, 69–86. data for the Coquimbo region, Chile: influence of mineral deposits and metal Sholupov, S.E., Ganeyev, A.A., 1995. Zeeman atomic-absorption spectrometry using recovery practices. Atmospheric Environment 39, 7587–7596. high-frequency modulated light polarization. Spectrochim Acta Part B—Atomic Higueras, P., Oyarzun, R., Lillo, J., Sánchez Hernández, J.C., Molina, J.A., Esbrí, J.M., Spectroscopy 50, 1227–1236. Lorenzo, S., 2006. The Almadén district (Spain): anatomy of one of the world's Siciliano, S.D., Lean, D.R.S., 2002. Methyltransferase: an enzyme assay for microbial largest Hg-contaminated sites. Science of the Total Environment 356, 112–124. methylmercury formation in acidic soils and sediments. Environmental Toxicology Jacobs, L.W., Keeney, D.R., 1974. Methylmercury formation in mercury treated river sediments and Chemistry 21, 1184–1190. during in situ equilibration. Journal of Environmental Quality 3, 121–126. Sillitoe, R.H., Halls, C., Grant, J.N., 1975. Porphyry tin deposits in Bolivia. Economic Kim, C.S., Rytuba, J.J., Brown Jr., G.E., 2004. Geological and anthropogenic factors Geology 70, 913–927. influencing mercury speciation in mine wastes: an EXAFS spectroscopy study. Steele, G.B., Boyce, A.J., Rice, C.M., Falliek, A.E., 2003. Metallogenesis of the world's Applied Geochemistry 19, 379–393. largest silver deposit—Cerro Rico, Bolivia. Transactions of the Institution of Mining King, J.K., Harmon, S.M., Fu, T.T., Gladden, J.B., 2002. Mercury removal, methylmercury and Metallurgy, Section B: Applied Earth Science 112, B217–B218. formation, and sulfate-reducing bacteria profiles in wetland mesocosms. Chemo- Strosnider, W.H., Llanos, F., Nairn, R.W., 2008. A Legacy of Nearly 500 Years of Mining in sphere 46, 859–870. Potosí, Bolivia: Stream Water Quality. In: Barnhisel, R.I. (Ed.), Published by ASMR, López, F.A., López-Delgado, A., Padilla, I., Tayibi, H., Alguacil, F.J., 2010. Formation of 3134 Montavesta Rd., Lexington, KY 40502. metacinnabar by milling of liquid mercury and elemental sulfur for long term U.S. EPA, 2009. Graphical Arrays of Chemical-Specific Health Effect Reference Values for mercury storage. Science of the Total Environment 408, 4341–4345. Inhalation Exposures (Final Report). EPA/600/R-09/061 http://cfpub.epa.gov/ncea/ Martínez-Coronado, A., Oyarzun, R., Esbrí, J.M., Llanos, W., Higueras, P., 2010. Sampling cfm/recordisplay.cfm?deid=211003. high to extremely high Hg concentrations at the Cerco de Almadenejos, Almadén WHO, 2000. Air quality guidelines for Europe. WHO Regional Publications European Series mining district (Spain): the old metallurgical precinct (1794 to 1861 AD) and 91, World Health Organization Regional Office for Europe, Copenhagen. 288pp. surrounding areas. Journal of Geochemical Exploration 109 (1–3), 70–77.