Rio Tinto (Iberian Pyrite Belt): A world-class mineral field reopens

Abstract

The deposits of Rio Tinto are located in the Spanish segment of the Iberian Pyrite Belt and are hosted within felsic porphyritic volcanic rocks and tuffs. The orebodies comprise a spectrum from sedimentary exhalative (San Antonio) to sub-sea floor replacement of the volcanic host rocks (Filon Norte); the two largest masses (San Dionisio and Filon Sur) are closely associated with black shale and probably formed by partial replacement of these units in an anoxic setting. Alteration associated with ore deposition is typified by marginal sericitic (white mica) alteration, and central chloritisation and silicification in a multi-phase alteration history. Structures formed during initial plate convergence may have acted as controls on ore deposition, and evidence of such controls is retained in the distribution of various elements in the sulphide deposits. Tectonism followed the mineralisation, and overprinted previous extensional events and resulted in the development of slaty cleavage in the pelitic rocks with partial remobilisation of sulphides, and tight folds with associated shearing of the southern limb.

Keywords

Iberian Pyrite Belt Rio Tinto Volcanogenic massive sulphide deposits Copper

Introduction and previous work

The Iberian Pyrite Belt (IPB) forms an arcuate belt 250 km long and up to 60 km wide extending north-westwards from Seville in Spain to southern Portugal (Fig. 1). The Belt contains one of the greatest concentrations of sulphide mineralisation on Earth, with resources in excess of 1700 mt of sulphide totalling 14·6 mt Cu, 13·0 mt Pb, 34·9 mt Zn, 46 100 t Ag and 880 t Au (Almodovar et al., 1998; Leistel et al., 1998; Saez et al., 1999; Carvalho et al., 1999; Tornos, 2006). Palaeontological data (Nehlig et al., 1998) constrain the age of the volcano-sedimentary formation between 360 and 342 Ma and this is supported by Re–Os dating of the sulphides (Mathur et al., 1999) and U–Pb dating of associated porphyries (Barrie et al., 2002).

Figure 1

Geological map of the Iberian Pyrite Belt showing location of major sulphide deposits

The deposits of Rio Tinto hold a prominent position in the IPB, with an estimated 500 Mt of massive sulphide prior to erosion and mining (Barriga, 1990; Boulter, 1993; Ribeiro, 1996), and have been worked since ancient times, with the most intense mining activity initiated in 1873 following the acquisition of the mines by a British syndicate headed by Hugh Matheson (Avery, 1974).

Numerous workers have examined the geology of Rio Tinto and associated deposits and contributed to the understanding of their setting and genesis. The classic paper of Williams (1934) provides detailed information on the setting of all the deposits. Lithological aspects were examined by Kinkel (1962) and Schermerhorn (1970) with the latter interpreting the felsic volcanic rocks as submarine ash flow tuffs. Early works on the petrography and mineralogy of the Rio Tinto deposits are those of Finlayson (1910)</citref> and Bateman (1927). Correlation of the deposits of the IPB with sill–sediment complexes such as the Guaymas Basin in the Gulf of California was suggested, because of the presence of peperitic associations with doleritic sills incorporating fragments of slate (Einsele, 1985; Boulter 1993). The environment of formation of the sulphide deposits was interpreted by Halsall (1989) to be close to the centre of rhyolitic activity with coherent rhyolite spatially associated with tuffaceous units. Morphological and genetic aspects of the deposits were discussed by Garcia Palomero (1976, 1990). A more recent contribution to the understanding of the tectonics of the Rio Tinto region is that of González–Clavijo and Díez–Montes (2010), and the impact of historical mining at Rio Tinto on the coastal environment was discussed by Koski (2012)</citref>.

Present investigation

This publication aims to provide some information on various aspects of the geology of Rio Tinto based on the author's work as part of the continuing operations for the reopening of the mines by EMED Mining Public Ltd through its local subsidiary EMED Tartessus S. L. U. The work involved, inter alia, the wireframing of all underground workings, compilation and wireframing of the lithology, alteration, and mineralisation for all deposits, and geological mapping of the area immediately surrounding the mineralised ground. The mapping covered part of the area previously described by Williams (1934) and provides an updated view of local geology. Additional work included detailed logging of selected historical diamond drill holes for the further understanding of the lithological and alteration characteristics of the host rocks and the nature of the associated mineralisation.

Petrographic investigations were carried out by the author on both surface samples and drill core collected as part of an MSc thesis and information on the location of samples is contained therein (Cosme, 2011). Treatment of geochemical data from historical blast-hole sampling aimed at the clarification of the distribution and content of selected elements in the mineralised ground. The entire work focused on a better understanding of the nature and controls of the mineralisation as an aid in future mining and exploration for additional resources.

Regional setting

The IPB (Fig. 1) is an epicontinental volcano-sedimentary belt developed during transpression following the oblique collision between the Ossa Morena Zone to the north and the South Portuguese Zone (Solomon et al., 2002; Soriano and Casas, 2002). The belt comprises rocks that range in age from Devonian to Middle Carboniferous, with a thick bimodal volcano-sedimentary formation (Volcano-sedimentary Complex) intercalated between Late Devonian phyllitic quartzite (PQ Group; Schermerhorn, 1970) and the detrital Culm Group of Dinantian age, a turbidite sequence derived from a northern land mass (Halsall, 1989).

The disposition of rock units on either side of Rio Tinto is in the form of a synclinal structure, with rocks of the PQ Group exposed in the area of Campofrio to the north and El Campillo to the south, and the Volcano-sedimentary Complex occupying the core of the syncline. Near Rio Tinto the felsic volcanic rocks are in the form of a doubly-plunging anticline with a central core of mafic volcanic rocks surrounded by the Culm Group.

Local geology

A geological map of the area around the Rio Tinto deposits shows the main lithological units comprising a sequence of porphyritic felsic volcanic rocks underlain by mafic volcanic rocks and associated meta-sedimentary rocks (Fig. 2). These are overlain by the pelitic and psammitic turbidite sequence of the Culm Group. Locally, a ferruginous conglomerate comprising clasts of varying lithology forms a prominent feature and rests with strong angular unconformity on the older rocks.

Figure 2

Generalised geological map of the Rio Tinto area of Spain

Felsic volcanic rocks

There are relatively unaltered examples of these rocks in the northern and eastern parts of Cerro Colorado and the southern wall of Corta Atalaya. They are represented by massive porphyritic rock of cream or grey colour with scattered phenocrysts of quartz and subordinate feldspar, in a fine-grained felsic matrix. The lithologies are mostly rhyolite, however locally densely feldspar-phyric dacite has been identified in core. Domal structures characterised by variations in morphology and alteration intensity have been interpreted as cryptodomes (Diez-Montes, pers., com., 2012). Flow layering highlighted by colour variations has been observed north of the Salomón open cut (Fig. 3A) and there are local vesicles in porphyry on the southern side of Corta Atalaya; however most of the porphyry presently exposed is massive and structureless.

Figure 3

Photographs of selected outcrops from the environs of Rio Tinto: A: Flow layering in felsic porphyry, northeast wall of Cerro Colorado; B: Peperitic associations between felsic porphyry and mafic metasediment, Rio Tinto–Aracena road section; C: Interbedded mafic metapelite and bedded tuff, southwest corner of Cerro Colorado; D: Thin bedded mafic tuffs, north wall of Corta Atalaya; E: Interbedded pelitic and sandy units highlighting style of folding in Culm units, south wall of Corta Atalaya; F: Well-developed pyritic stockwork in felsic porphyry, north wall of Corta Atalaya

The felsic volcanic rocks are locally sheared, with the development of strong slaty cleavage and only the quartz phenocrysts testifying to the nature of the original protolith. In the western extensions of Cerro Colorado, close to the Rio Tinto–Aracena road, peperitic interbedding between weakly mineralised porphyry and black pelitic meta-sedimentary rocks suggests a locally intrusive relationship of the porphyry (Fig. 3B).

Hydrothermal alteration and the earlier low-grade metamorphism mask the original lithological character of the rocks, and may give rise to the formation of false textures (Allen, 1988; Halsall, 1989; Gifkins et al., 2005). Nevertheless, it is possible to distinguish primary volcanic textures through the alteration and, based on their comparison with the lithologies in equivalent settings (Cas and Wright, 1993; McPhie et al., 2010), to determine the depositional environment.

Although massive structureless porphyritic felsic rock is the most common rock type, a variety of other lithologies are distinguished in core. One of the most common comprises monomictic breccia superficially resembling epiclastic material (Fig. 4A), but differing from the latter by the interstitial matrix which consists of a microcrystalline mixture of quartz and white mica and the contacts between the clasts and the matrix are diffuse, suggesting an alteration origin rather than primary volcanic one. The breccias are commonly transected by sulphide veins confirming an early genesis in the hydrothermal history, and are interpreted as probably the result of in situ brecciation and concomitant silica flooding as a precursor to the main mineralisation-related alteration event.

Figure 4

Lithological associations observed in drill core: A: Monomictic breccia composed of felsic porphyry in fine-grained siliceous matrix; B: Tuffaceous layering in chloritic quartz-phyric rock; C: Lapilli tuff texture in felsic porphyry; D: Hyaloclastite, with rounded lava clasts displaying chilled margins and coarser interior; E: Conglomerate at contact between mafic and felsic units; F: Ferruginous vesicular weakly mineralised mafic volcanic

A very common texture comprises dark shard-like bodies up to a few mm in dimensions enclosed in the uniformly fine-grained variably quartz-phyric matrix. These probably represent altered volcanic glass and make the rock texture superficially similar to ignimbrite; however the absence of associated vesicularity, the envisaged subaerial setting of ignimbrites (Holmes, 1964) and the generally structureless nature of the Rio Tinto lithologies, with the exception of local layering (Fig. 4B), suggest a genesis as mass flow tuffs (Schermerhorn, 1970).

Lapilli tuffs, spatially associated with laminated tuffs (Fig. 4C), and hyaloclastites comprising irregular to rounded lava clasts in a matrix of altered glass (Fig. 4D) are relatively rare.

Differentiation during transport results in some cases in the winnowing of the finer matrix and concentration of phenocrysts resulting in the creation of coarse quartz-rich volcanogenic sandstone (crystal tuff) in association with more typical matrix-rich tuffs and hyaloclastic units. The assemblages may indicate interaction of a variety of environments with proximal volcanism adjacent to sites of sedimentation and mass-flow deposition.

Distinct from the hydrothermal breccia mentioned above, there are also angular breccia enclosed in a matrix of sulphide, with alteration halos present around some of the fragments suggesting an epiclastic genesis.

The contact between the felsic sequence and the underlying mafic volcanic rocks is marked by a thin conglomeratic sequence comprising angular clasts of felsic lithology set in a finer lithic matrix (Fig. 4E). The sequence beneath the conglomerate comprises an association of pelite, tuffaceous units and amygdaloidal mafic lava (Fig. 3C). Early ferruginisation in the lava is overprinted by the mineralisation-related alteration. The vesicularity of the lava (Fig. 4F) and the spatially associated pillows exposed in the open cut suggest a subaqueous environment with the parallel laminations in the pelite indicating deposition below storm-wave base.

Carbonaceous slate

On the southwestern corner of Corta Atalaya an outcrop of highly tectonised and quartz-veined black carbonaceous slate is in fault contact with felsic porphyry. Similar slate closely associated with massive sulphide is present in the southeastern corner of Corta Atalaya and it is possible that a closer spatial association initially existed between the two occurrences, now separated by faulting, with the carbonaceous slate being part of the Transition Series (Garcia Palomero, 1990) which hosted the massive sulphide of San Dionisio and Filon Sur.

Mafic volcanic rocks

Outcrops of these units in the Cerro Colorado opencut are confined to its southwestern corner where they are represented by the aforementioned interbedded vesicular mafic lavas, laminated pelite and associated tuffs and local pillow lavas. At Corta Atalaya, the mafic units occupy the northern wall of the open cut and exhibit a more thinly layered appearance with local tuffaceous affinities and associated pelite (Fig. 3D).

Culm Group

This unit comprises a monotonous sequence of highly cleaved blue-grey slate with local intercalations of fine sandy and silty layers. The unit is laterally extensive with an estimated thickness of 3000 m. In the area of Corta Atalaya, sandstone layers are intercalated with the slate highlighting the style of folding (Fig. 3E). The sequence is interpreted as a turbidite infilling a subsiding basin, with sediment derivation from both the Ossa Morena Zone and the IPB (Schermerhorn, 1970; Saez et al., 1999).

Bog iron ore

South of Filon Sur an extensive outcrop of indurated highly ferruginous conglomerate rests unconformably on acid-leached slate. The rock type contains clasts of varying lithology cemented with ferruginous material. It is interpreted as bog iron ore formed from the cementation of the products of weathering of the surrounding rocks by percolating iron-bearing waters on a peneplain surrounding the more elevated mineralised outcrops of Cerro Colorado and Salomón (Bateman, 1927; Williams, 1934). Similar outcrops are present west of the Rio Tinto creek as well as close to the Nerva cemetery. At the latter locality the rock type rests on highly ferruginous slate which preserves its original fabric. Present day formation of similar rocks is observed in the river channel.

Structure

Although three roughly homo-axial phases of deformation have been postulated by some workers for the IPB characterised by south-verging folds, axial planar cleavages, and thrusts (Soriano and Casas, 2002), at Rio Tinto the field evidence supports a single phase associated with the closure of the volcano-sedimentary basin during Variscan times, probably overprinting the earlier transpressional tectonics.

The general structure of Rio Tinto may be described as a gentle doubly plunging anticline defined by the annular arrangement of the felsic units which are in turn surrounded by the slate of the Culm Group. This antiformal structure is bisected by the northwest-trending Eduardo Fault (Fig. 2) which is a regional-scale structure probably inherited from the original tectonic regime associated with the evolution of the volcano-sedimentary complex.

The various structural elements recorded from field mapping are plotted on stereograms and are consistent with one period of compression resulting in the closure of the volcano-sedimentary basin following the period of volcanism and mineralisation (Fig. 5). Slaty cleavage is particularly developed in the pelitic lithologies and along more localised domains in the volcanic rocks (Fig. 5A). This cleavage is axial-planar to folds highlighted by the sandy layers of the Culm Group, particularly evident in the environs of Corta Atalaya (Fig. 3E). The tectonic style comprises almost isoclinal folding with vergence to the south-southwest and shearing of the southern limb with quartz veining commonly highlighting the shearing.

Figure 5

Stereographic projections (Fisher, equal area, lower hemisphere) of various structural elements from the Rio Tinto mining area: A: Poles to cleavage; B: Poles to bedding or layering; C: Poles to joints; D: Poles to quartz veins; E: Poles to L1 lineations and minor anticlines; F: Poles to shear zones

The dominant lineation is L1, defined by the intersection of the slaty cleavage with the bedding, which highlights the overall structure of the antiform, with a gentle easterly dip in the eastern parts of the structure and the closure of the antiform in the west suggesting a similarly gentle plunge in that direction (Fig. 5E). The main direction of shearing is parallel to the orientation of S1 suggesting contemporaneous formation with the slaty cleavage (Fig. 5F). A dominant northerly orientation of jointing is mirrored by similar orientation of quartz veins suggesting a genesis of both elements as a result of stress relaxation during north–south compression (Fig. 5C and D). Overprinting of the hydrothermal event by the tectonism is suggested in the field by the alignment of sulphide veining parallel to the cleavage, and under the microscope by the brittle brecciation of pyrite and infilling of interspaces by the more malleable chalcopyrite (Fig. 13A).

Mineralisation

The Rio Tinto deposits comprise a number of groups:

San Dionisio in the west, separated from Filon Sur by the Eduardo Fault which is spatially associated with a small deposit (the Eduardo Mass–Williams, 1934)

Filon Sur (the South Lode) located on the south limb of the Rio Tinto Anticline

iii

Filon Norte (North Lode) which includes the deposits of Salomón, Quebrantahuesos, Lago, Dehesa, Mal Año, and Algamasilla

The Planes–San Antonio couple which occupies the eastern nose of the Rio Tinto Anticline

The small Valle deposit located within slate south of San Dionisio (outside of Fig. 2).

At present, as a result of mining, the geology is well exposed in a major open cut (Corta Atalaya) which exploited part of San Dionisio, and in the Cerro Colorado open cut which exploited part of Filon Sur and Filon Norte groups. A view of the distribution of workings from the exploitation of the deposits is shown in Fig. 6.

Figure 6

Plan view of all underground workings at Rio Tinto showing the extent of mining activity and distribution of ore concentrations. Data compiled from historical archives

Previous studies have described the varying morphology and lithological relationships between the sulphide deposits and their host rocks (Williams, 1934; Williams et al., 1975; Garcia Palomero, 1990). Both Filon Sur and San Dionisio are closely associated with black slate and tuffs – the Transition Zone (Garcia Palomero, 1990), with local incorporation of slate in the massive sulphide and dense impregnations of sulphide in the slate suggesting a genesis by sub-seafloor replacement close to the rock/seawater interface.

The mineralisation exposed in Cerro Colorado at the present level of mining displays a number of distinct facies. In the west, the most common occurrence is in the form of sulphide disseminations in chlorite and silica rich country rocks. Local typical stockwork is only observed near Mal Año and at Filon Sur, and this veining diminishes rapidly with stratigraphic depth (Fig. 2). However the best development of stockwork is in the Corta Atalaya open cut which directly underlies the massive sulphide zone of San Dionisio and is characterised by intense pyritic veining in sericitic, chloritic, and silicified felsic country rock, locally known as the Cloritas (Fig. 3F).

A transitional passage from altered generally unmineralised country rocks to semi-massive or massive mineralisation characterises the Salomón–Lago area (Fig. 7). Replacement textures are well documented in mine outcrop and in core, from initial veining of the country rock by sulphides to complete replacement by massive pyrite. These relationships, together with the location of the massive sulphide at some depth below the felsic volcanic/sediment contact clearly suggest a genesis by sub-sea floor replacement for the Salomón–Quebrantahuesos deposits.

Figure 7

North–south section through Cerro Colorado, highlighting rock relationships and the disposition of mineralisation in the Filon Norte area

Morphology of the deposits

Wireframing of the mineralisation from historical data has highlighted the varying morphology and lithological associations of the sulphide deposits. The general form of San Dionisio suggests a synclinal disposition with massive sulphide exposed both on the northern wall of the Corta Atalaya open cut where it is directly underlain by the aforementioned Cloritas stockwork, and in the southern side where it is overlain by altered but generally unmineralised felsic volcanic rocks (Fig. 8A). Slate of the Culm Group occupies the core of the syncline, and the geometry of the deposit may suggest structural duplication of the massive sulphide zone resultant from the thin-skinned tectonics characteristic of the IPB (Tornos et al., 1998; Castroviejo et al., 2011).

Figure 8

Three-dimensional views of selected deposits of Rio Tinto, derived from the wireframing of data from underground mining: A: Massive sulphide of San Dionisio and Filon Sur, and relationship to the Eduardo Fault; B: View of the massive sulphide of Planes and San Antonio deposits

A sheet-like morphology with a large length to width ratio characterises Filon Sur, which is linked with San Dionisio by the Eduardo deposit which, in turn, occupies the plane of the Eduardo Fault (Fig. 8A). The location of the deposit at the contact between the felsic volcanic rocks and slate and the local incorporation of slate within the massive sulphide (Williams, 1934) suggests possible deposition just beneath the sea floor by replacement of the pelitic lithologies.

In contrast to the close spatial relationship between the felsic volcanic rocks and the sulphide mineralisation exhibited by the Filon Sur and San Dionisio deposits, the Planes–San Antonio pair mark the transition into a more typically sedimentary association with interbedding of massive sulphide and tuffaceous lithologies suggesting an interplay between hydrothermal and sedimentary processes (Fig. 8B). In this pair of deposits, Planes represents proximal facies of sulphide deposition underlain by porphyry (Williams, 1934), whereas San Antonio represents the more distal facies with interfingering of oligomictic breccia, volcanogenic shale and massive sulphide (Williams et al., 1975; Garcia Palomero, 1990).

Information on the detailed morphology of the deposits of the Filon Norte group is not available because of the early mining before accurate records were kept. Remnants of stockwork-type mineralisation within chloritic felsic lithologies near Dehesa and the presence of sericitic volcanic rocks in the hanging wall side suggest a replacement genesis for the deposit. Similar support is provided by the transition from peripheral altered felsic volcanic rocks into semi-massive mineralisation in the Lago–Salomón area and the disposition of stopes of historical mining (Fig. 7).

The above relationships between ore and country rocks suggest that the Rio Tinto deposits display a spectrum of ore types from massive sulphide deposited by subsurface replacement of the volcanic host rocks, through deposition at or close to the rock/seawater interface in close association with black shales, into typical sedimentary exhalative deposition at some distance from the source of the hydrothermal fluids.

Geochemical aspects

Immobile elements

The use of elements which are considered immobile under conditions of mild hydrothermal alteration and low-grade metamorphism has proved useful in the characterisation of the volcanic rocks, particularly where hydrothermal alteration may mask their original nature. A small number of samples collected from various parts of the open cut as part of a MSc thesis (Cosme, 2011) were plotted on the discriminant diagram of Winchester and Floyd (1977). A distinct grouping of the analyses in two populations confirms the bimodal nature of the rocks, with most of the samples plotting in the rhyolite field, and mafic lithologies, representing pillow lavas and associated mafic tuff, plotting in the andesite field (Fig. 9). The results are consistent with the conclusions of workers in the general Rio Tinto area (Pascual et al., 2005).

Figure 9

Plot of analyses of samples from Cerro Colorado on the discriminant diagram of Winchester and Floyd (1977) highlighting the bimodal nature of the volcanic rocks

Blast hole data

A suite of elements routinely assayed in blast holes during mining for grade control purposes was evaluated by using their spatial distribution and other statistical characteristics to constrain the genesis of the deposits. The analyses represent composite assays over a 10- or 12-metre interval for 91 694 individual blast holes and define a slice of the upper part of Cerro Colorado for which detailed records were kept. This, in addition to the effects of erosion and previous mining, results in only partial conclusions being derived. The frequency distribution of the various elements (Fig. 10) suggests:

Figure 10

Frequency distribution patterns from blast holes in Cerro Colorado

S: Maximum concentration of values in the region 2–3% sulphur, with a Gaussian distribution around this value up to 5% S, representing the concentration of most samples in the western poorly mineralised part of Cerro Colorado, with logarithmic drop of values towards the high-end range. Weakly bimodal distribution, with a second peak centred around 18% S reflects the concentration of high-sulphur ore in the Quebrantahuesos–Argamasilla area.

Cu: The frequency distribution is positively skewed, with a clear concentration of most of the values at the interval below 0·5%, and a rapid decrease in values at the higher end of the scale with very limited values exceeding 1%Cu.

Zn: The frequency distribution displays highest concentration of values at the interval 0–0·1%Zn, thereafter decreasing steeply to very low values beyond 0·5%Zn.

Pb: The element displays very low values in the Cerro Colorado open cut with the majority of the samples within the range 0–0·05%Pb and only very few samples exceeding 0·25%Pb. Data are not available for the Lago–Quebrantahuesos area.

As: The frequency distribution displays a maximum at the 100–200 ppm level with a positive skewness and a generally regular decrease in values towards the higher end of the scale with few samples exceeding 2, 000 ppm and reaching a maximum of 10 000 ppm.

Bi: The frequency distribution is weakly bimodal with positive skewness. A minor peak at 5–10 ppm Bi (2300 samples) is followed by a major peak at 15–20 ppm (5500 samples), thereafter the values steadily falling to almost 0 at 200 ppm.

Sb: Highest concentration of values is at the interval 5–10 ppm thereafter displaying a sharp decrease for the interval 10–15 ppm and a more gentle positively skewed drop in values up to a level of 400 ppm beyond which values taper off to negligible levels.

Compilation of data on the spatial distribution of selected elements within Cerro Colorado highlights a number of interesting features: Cu defines a number of linear zones of enhanced concentration oriented northeast and northwest, and this pattern is substantiated by the distribution of Zn and to a lesser extent S (Fig. 11A). A pattern of greatly enhanced As in the Salomón–Quebrantahuesos area with a northwest-oriented abrupt break between Salomón and Dehesa suggests a possible fault (Fig. 11B). A further narrow northwest-trending zone of elevated As values is observed in the Mal Año area.

Figure 11

Compilation of the spatial distribution of Cu (A) and As (B) in blast holes of Cerro Colorado

The above patterns, which are further supported by more detailed plots of element distribution, suggest the influence of a structural control in the channeling of hydrothermal fluids, with the northwest and northeast directions probably representing conjugate fracture systems inherited from the original configuration of stress during creation of the volcano-sedimentary basin by the oblique transpressive forces.

Petrography and mineralogy

Host rocks

Unaltered felsic rocks commonly comprise two generations of euhedral to subhedral quartz set in a fine matrix of quartz and white mica. Large feldspars are replaced by clay minerals with only the outlines of the phenocrysts preserved. Many of the quartz phenocrysts have fractures healed by mica. Orthogonal fracturing of the quartz phenocrysts is a common feature of the felsic units (Fig. 12A). A thin rim of microcrystalline silica, a result of pressure dissolution, surrounds many of the phenocrysts.

Figure 12

Selected photomicrographs (transmitted light) of assemblages from Rio Tinto: A: Orthogonal fracturing of quartz phenocryst in felsic porphyry. Crossed polars; B: Mafic pillow lava displaying two generations of plagioclase in a fine-grained matrix. Parallel polars; C: Replacement of quartz by hydrothermal well-crystallised chlorite. Parallel polars; D: Pressure fringes of lamellar quartz around core of pyrite crystal. Crossed polars; E: Replacement of quartz phenocryst by carbonate in the matrix of felsic porphyry. Crossed polars; F: Association of well-crystallised muscovite and microcrystalline carbonate in the matrix of hydrothermally altered felsic porphyry. Crossed polars

The mafic pillow lava, spatially associated with laminated mudstone and volcanogenic tuff, is extensively altered, with prismatic domains replaced by chlorite defining the location of original clinopyroxene phenocrysts, and leucocratic domains defining original feldspars (Fig. 12B). Cleavages are defined by the preferred orientation of domains of chlorite. The chlorite is very weakly pleochroic, with low first order grey birefringence characteristic of the iron-poor variety (Deer et al., 1966).

Several forms of chlorite have been observed in the hydrothermally altered rocks, based on their optical characteristics: A matrix chlorite of fine grain size and low birefringence, and coarser crystalline hydrothermal types with anomalous blue and purple birefringence, or green strongly pleochroic iron-rich types. Well-crystallised chlorite commonly replaces phenocryst quartz or occurs interstitial to the sulphides (Fig. 12C).

Carbonate is widespread in veins and appears to be paragenetically late, traversing all other mineral associations. The carbonate, probably ankerite, is commonly associated with ferruginous veining. Together with ubiquitous barite, they are probably the result of the expulsion of alkalis from the ore zone during hydrothermal activity (Almodovar et al., 1998).

Sulphide mineralogy

The sulphide mineralogy of the deposits is dominantly pyrite with minor chalcopyrite, sphalerite, and galena in proportions which vary with location in individual deposits as well as between deposits. Cerro Colorado is dominated by chalcopyrite, in contrast to San Antonio and San Dionisio, where Pb and Zn become important components of the assemblage (Garcia Palomero, 1990). In the present work, very limited petrographic work was carried out on a small number of samples from the eastern part of Cerro Colorado (Quebrantahuesos area), and the following descriptions are based on that work.

The main minerals identified in the core and under the microscope comprise pyrite and chalcopyrite with subordinate sphalerite and galena. Previous detailed investigations, however, have identified an array of minerals including Bi–Sb–Pb–As sulphosalts and other minor phases considered to be hosts for a variety of minor elements (Bateman, 1927).

Pyrite is the dominant sulphide, and is generally euhedral, albeit commonly brecciated (Fig. 13A), with development of lamellar quartz fringes (Fig. 12D) in pressure shadows (terminology from Spry, 1969) commonly in association with muscovite (Fig. 12E) and colourless iron-poor chlorite.

Figure 13

Photomicrographs of sulphide associations from Cerro Colorado. All images in parallel polars. A: Cataclastic brecciation of pyrite (PY) with chalcopyrite (CP) filling spaces interstitial to the pyrite clasts; B: Euhedral pyrite with interstitial spaces occupied by chalcopyrite, sphalerite (SP;with chalcopyrite disease) or galena (GN); C: Association of chalcopyrite, galena, and sphalerite in a vein, with the three sulphides displaying mutual boundaries suggesting simultaneous deposition; D: Supergene replacement of chalcopyrite by bornite (BN) and subsequent replacement of bornite by covellite (CV)

Chalcopyrite commonly fills spaces interstitial to euhedral pyrite and partly replaces it (Fig. 13B). Association of all the main sulphides in veins is typically in the form of euhedral pyrite with the other sulphides interstitial to the pyrite displaying anhedral mutual boundaries suggesting contemporaneous deposition (Fig. 13C). Chalcopyrite disease in sphalerite (Barton and Bethke, 1987) is common and suggestive of the high-temperature deposition of this assemblage (Fig. 13C). Supergene oxidation of chalcopyrite is characterised by fracture-guided replacement by bornite which is in turn replaced by covellite occupying the central parts of the fractures (Fig. 13D).

Wall rock alteration

The general distribution of the alteration, compiled from the results of historical logging by Edmund Sides in the early 1980s, highlights a zonal arrangement from outer sericitisation to inner chloritisation associated with the sulphide mineralisation (Fig. 14). This pattern has been confirmed by other workers (Halsall, 1989; Garcia Palomero, 1990; Ribeiro, 1996; Saez et al., 1999; Sánchez–España et al., 2000), as well as by the results of drill hole logging during the current work which has also disclosed a complex multi-phase alteration history.

Figure 14

Northeast view of the Rio Tinto deposits and associated alteration, as derived from historical logging. The simpler pattern west of the Eduardo Fault and east of Filon Norte is due to lack of detailed data

The initial product of alteration is likely to be represented by a form of angular hydrothermal breccia which is a widespread lithology, and interpreted as due to contemporaneous in situ brecciation and silica flooding at low temperature (Fig. 4A). These breccias are commonly transected by pyrite veins without evidence of reaction, suggesting a similarly low-temperature emplacement at the marginal, cooler, parts of the system.

Sericitic alteration is particularly evident in the felsic lithologies in the form of pervasive light yellow-green to light grey colouration of the host rock. In places the white mica alteration may be intense, however in most cases it is accompanied by other types of alteration, particularly chlorite and silica. The effects of silicification are partly masked by the originally siliceous nature of the host rocks, however the destruction of the original textures, and the attainment in some cases of a completely glassy aspect, confirms the introduction of silica from the hydrothermal fluids (Fig. 15A). Vein associations of sulphide occupying the central parts surrounded by siliceous rim are common in the mineralised rocks (Fig. 15B).

Figure 15

Selected views of alteration associations in drill core: A: Entirely silicified felsic tuff with abundant glassy clasts, traversed by pyritic vein; B: Sulphide vein traversing siliceous felsic volcanic rock, with associated inner silica and marginal chlorite alteration; C: Spotted limonitic alteration in siliceous felsic rock; D: Early ferruginisation (?weathering) in mafic volcanic, reduced by later sulphide vein with associated silicification

Chlorite is visually distinguished from white mica during logging by its darker colour (Fig. 15B), and massive black chlorite is locally intimately associated with the massive pyritic mineralisation, with the pyrite occurring as equant or more elongate clusters of crystals in the black chloritic matrix. Rare talc alteration suggests locally enhanced magnesium metasomatism and is evident in the form of brown wax-textured replacement of the rock.

Assemblages characteristic of advanced argillic alteration, with vuggy silica, kaolinite, and secondary copper minerals, are noted along narrow zones and may equally be the result of descending fluids from the oxidation of overlying pyritic bodies, or of ascending late-stage acid fluids during the final stages of hydrothermal activity.

Ferruginisation, commonly associated with carbonate alteration, is present in a variety of forms, either as pervasive alteration of the basalt or fine networks of hematite or limonite associated with shearing (Fig. 15C). In the mafic sequence, early ferruginisation evident as a uniform brown goethitic oxidation of the rocks is overprinted by the reducing alteration associated with the mineralisation (Fig. 15D).

Discussion and conclusions

The data presented in the preceding paragraphs suggest formation of the Rio Tinto deposits in a setting characterised by a mafic sequence of interbedded volcanic rocks, tuffs, and pelite overlain by a felsic volcaniclastic sequence. The felsic rocks are dominated by massive porphyritic units of predominantly rhyolitic composition, however proximal tuffs comprise a major part of the sequence at Rio Tinto and their generally massive unsorted and mostly unstratified internal structure suggests a mass-flow genesis. There is no support either from the field mapping or drill hole logging for a sill–sediment complex setting as suggested by some workers (Einsele, 1985; Boulter, 1993), however the data support a depositional environment dominated by mass-flow tuffs as envisaged by Schermerhorn (1970).

The close spatial relationships between black shale and massive sulphide (Williams, 1934), and confirmed in the field at San Dionisio, suggest deposition of the Filon Sur and San Dionisio deposits partly by sub-sea floor replacement of pelitic units. This is considered the most efficient mechanism of ore deposition, in contrast to the exhalative deposits where most of the metal load may be lost into the ambient seawater (Franklin et al., 2005). In this respect, the deposits of Rio Tinto are closer to the shale-hosted type than the Cyprus-type, which were formed in a sediment-starved spreading environment (Galley and Koski, 1999). Ore deposition by replacement of the felsic host rocks is the mode of occurrence of the deposits of the Filon Norte group.

The limited depth extent of the stockworks is probably a result of the shale-hosted affinities of the Rio Tinto deposits. Historical records suggest that the stockwork zones of both Filon Sur and San Dionisio merge into the background of chloritic rocks with disseminated sulphide within a few tens of metres of the base of the massive sulphide. This is in contrast with ophiolite-hosted deposits where footwall alteration pipes may extend hundreds of metres beneath the massive sulphide (Franklin et al., 2005). This difference is attributed here to the strong control exercised in ore localisation at oceanic spreading centres by deep extensional axis-parallel structures. In the epicontinental setting of the Rio Tinto deposits (Solomon et al., 2002), the equivalent control was provided by transcurrent structures resulting from the oblique collision between the south Portuguese terrane and the Iberian continental block (Leistel et al., 1998; Solomon and Quesada, 2003), and these would be expected to close rapidly at depth. Support for this inference is provided by the patterns observed in the distribution of elements within Cerro Colorado, which suggest that an inherited set of structures guided the hydrothermal fluids, and the location of San Dionisio and Filon Sur with respect to the Eduardo Fault which may imply that this structure played an active role in ore deposition.

The occurrence of pillow lavas as well as laminated pelitic units and tuffs associated with the mafic sequence underlying the felsic volcanic rocks suggests subaqueous deposition below wave base. Furthermore, the close association between black shale and massive sulphide deposits at Filon Sur and San Dionisio suggest ore deposition in an anoxic and probably deep-water setting. Although a shallow marine environment and even emergence has been suggested by some workers (Halsall, 1989; Garcia Palomero, 1990), the general absence of vesicularity in the felsic volcanic rocks may suggest volatile retention due to high water pressure (Valenzuela et al., 2011), and the association with black shale favours a deep anoxic setting.

The associations between the massive sulphides and the host rocks lead to some interesting conclusions with regard to the potential for extensions of the known mineralisation. The relationship between Planes as the proximal massive sulphide zone, replacing felsic tuffs (Williams, 1934), and San Antonio as its distal equivalent, with massive sulphide interlayered with tuffs (Williams et al., 1975; Garcia Palomero, 1990), suggests the presence of equivalent extensions to the known mineralisation in the other deposits. Filon Sur as well as San Dionisio lack the distal facies which may have been disrupted by faulting. The occurrence of the small and little-known Valle deposit within slate south of San Dionisio (Williams, 1934) may imply the existence of extensions to the known mineralisation beneath the Culm, in a setting probably similar to other shale-hosted deposits such as the Filon Norte (Tharsis) orebody (Tornos et al., 1998). At Filon Sur, historical geological mapping displays steeply dipping quartz veins within the Culm south of the deposit, and this may suggest shearing of the southern limb in agreement with the tectonic style of the deformation, with the more distal facies of the deposit as yet undetected.

Footnotes

Acknowledgements

The author wishes to thank the management of EMED Mining Public Ltd and in particular Harry Adams (Managing Director) and Ron Cunneen (EMED Chief Geologist) for permission to publish this work. Thanks are also due to EMED colleagues John Ingram and Angelo Farci for discussions on various aspects of the project, and to Daniel de Oliveira and Gul Seddel for constructive reviews. Editorial comments by Neil Phillips improved the final presentation and are gratefully acknowledged.

References

Allen

. 1988. False pyroclastic textures in altered silicic lavas, with implications for volcanic-associated mineralization Econ. Geol., 83, 1424–1446.

Almodovar

, Saez

, Pons

, Maestre

, Toscano

, Pascual

. 1998. Geology and genesis of the Aznalcollar massive sulphide deposits, Iberian Pyrite Belt, Spain Miner. Depos., 33, 111–136.

Avery

. 1974. Not on Queen Victoria's birthday, Collins, 464.

Barrie

, Amelin

, Pasqual

. 2002. U-Pb geochronology of VMS mineralization in the Iberian Pyrite Belt Miner. Depos., 37, 684–703.

Barriga

FJAS

. 1990, Metallogenesis in the Iberian Pyrite Belt, in Pre-mesozoic geology of Iberia, (ed. , Dallmeyer

R D

, Garcia

E Martinez

), 369–379.

Barton

, Bethke

. 1987. Chalcopyrite disease in sphalerite: pathology and epidemiology Am. Mineral., 72, 451–467.

Bateman

. 1927. Ore deposits of the Rio Tinto (Huelva) District, Spain Econ. Geol., 22, 569–614.

Boulter

. 1993. Comparison of Rio Tinto, Spain, and Guayamas Basin, Gulf of California: an explanation of a supergiant massive sulfide deposit in an ancient sill–sediment complex, Geology, 21, 801–804.

Carvalho

, Barriga

FJAS

, Munha

. 1999. Bimodal siliciclastic systems—the case of the Iberian Pyrite Belt, (ed. , Barrie

, Hannington

), Rev. Econ. Geol., 8, 375–408.

10.

Cas

RAF

, Wright

. 1993. Volcanic successions modern and ancient, Chapman and Hall, 528.

11.

Cosme

. 2011. Analisis de distintos aspectos de la alteración, mineralogía y geoquímica de un segmento de la secuencia vulcanoclástica, Proyecto Rio Tinto, unpublished Master's thesis, Universidad Internacional de Andalucia, 65 (Draft version).

12.

Deer

, Howie

, Zussman

. 1966. An introduction to the rock-forming minerals, 528, London, Longman.

13.

Einsele

. 1985. Basaltic sill–sediment complexes in young spreading centres: Genesis and significance, Geology, 13, 249–252.

14.

Finlayson

. 1910. The pyritic deposits of Huelva, Spain, Econ. Geol., 5, 357–372.

15.

Franklin

, Gibson

, Jonasson

, Galley

. 2005. Volcanogenic massive sulfide deposits, Econ. Geol. 100th Anniv. Vol., 523–560.

16.

Galley

, Koski

. 1999. Setting and characteristics of ophiolite-hosted volcanogenic massive sulphide deposits, (ed. , Barrie

, Hannington

), Rev. Econ. Geol., 8, 221–246

17.

Garcia Palomero

., 1976. Caracteres geologicos y relaciones morfologicas y geneticas de los yacimientos del ‘Anticlinal de Rio Tinto’, PhD thesis, University of Salamanca (in Spanish).

18.

Garcia Palomero

., 1990. Rio Tinto deposits–geology and geological models for their exploration and ore-reserve evaluation, in Sulphide deposits – their origin and processing, (ed. , Gray

P M J

), 17–35, Inst. Min. and Metal.

19.

Castroviejo

, Quesada

, Soler

. 2011. Post-depositional tectonic modification of VMS deposits in Iberia and its economic significance Miner. Depos., 46, 615–637

20.

Gifkins

, Herrmann

, Large

. 2005. Altered volcanic rocks, a guide to description and interpretation, Centre of Ore Deposits Research, University of Tasmania, 274.

21.

González-Clavijo

, Díez-Montes

. 2010. Proposal of a structural setting for the giant volcanogenic massive sulphide deposit of Rio Tinto, Iberian Pyrite Belt, Geogaceta, 48, 203–206 (In Spanish)

22.

Halsall

. 1989. The relationship between intrusive magmatism, volcanism, and massive sulphide mineralisation at Rio Tinto, Spain, Unpublished PhD thesis, University of London Imperial College, 291.

23.

Holmes

. 1964. Physical Geology, Thomas Nelson and Sons, 1288.

24.

Kinkel

. 1962. Observations on the pyrite deposits of the Huelva District, Spain, and their relation to volcanism Econ. Geol., 57, 1071–1080.

25.

Koski

. 2012. Metal dispersion resulting from mining activities in coastal environments: a pathways approach, Oceanography, 25, (2), 170–183.

26.

Leistel

, Marcoux

, Deschamps

. 1998. Chert in the Iberian Pyrite Belt Miner. Depos., 33, 59–81.

27.

Mathur

, Ruiz

, Tornos

. 1999. Age and sources of the ore at Tharsis and Rio Tinto, Iberian Pyrite Belt, from Re–Os isotopes Miner. Depos., 34, 790–793.

28.

McPhie

, Doyle

, Allen

. 2010. Volcanic textures: a guide to the interpretation of textures in volcanic rocks, ARC Centre of Excellence in Ore Deposits, University of Tasmania, 196.

29.

Nehlig

, Cassard

, Marcoux

. 1998. Geometry and genesis of feeder zones of massive sulphide deposits: constraints from the Rio Tinto ore deposit (Spain) Miner. Depos., 33, 137–149.

30.

Pascual

, Donaire

, Valenzuela

. 2005. Local and regional geochemical variations in VHMS-related felsic volcanic series in the Iberian Pyrite Belt, Abs., SGA meeting, Beijing, 671–673.

31.

Ribeiro

IMS

. 1996. Efeitos mineralógicos e geoquímicos da alteração mineralizante em rochas vulcânicas félsicas de Rio Tinto (Faixa Piritosa Ibérica, Espanha), unpubl. Master's thesis, Univ. Lisbon, Portugal, 200. (In Portuguese, English abstract)

32.

Saez

, Pasqual

, Toscano

, Almodovar

. 1999. The Iberian type of volcano-sedimentary massive sulphide deposits Mineral. Depos., 34, 549–570.

33.

Sánchez-España

, Velasco

, Iñaki

. 2000. Hydrothermal alteration of felsic volcanic rocks associated with massive sulphide deposition in the northern Iberian Pyrite Belt (SW Spain) Appl. Geochem., 15, 1265–1290.

34.

Schermerhorn

LJG

. 1970. The deposition of volcanics and pyritite in the Iberian Pyrite Belt Mineral. Depos. (Berl.), 5, 273–279.

35.

Solomon

, Quesada

. 2003. Zn–Pb–Cu massive sulfide deposits: Brine pool types occur in collisional orogens, black smoker types occur in backarc and/or arc basins, Geology, 31, 1029–1032.

36.

Solomon

, Tornos

, Caspar

. 2002. Explanation for many of the unusual features of the massive sulfide deposits of the Iberian Pyrite Belt, Geology, 30, 87–90.

37.

Soriano

, Casas

. 2002. Variscan tectonics in the Iberian Pyrite Belt, South Portuguese Zone Int. J. Earth Sci., 91, 882–896.

38.

Spry

. 1969. Metamorphic textures, Pergamon Press, 350.

39.

Tornos

. 2006. Environment of formation and styles of volcanogenic massive sulfides: The Iberian Pyrite Belt Ore Geol. Rev., 28, 259–307.

40.

Tornos

, Clavijo

, Spiro

. 1998, The Filon Norte orebody (Tharsis, Iberian Pyrite Belt): a proximal low-temperature shale-hosted massive sulphide in a thin-skinned tectonic belt Miner. Depos., 33, 150–169.

41.

Valenzuela

, Donaire

, Pin

, Toscano

, Hamilton

, Pascual

. 2011. Geochemistry and U–Pb dating of felsic volcanic rocks in the Riotinto–Nerva units, Iberian Pyrite Belt, Spain: crustal thinning, progressive crustal melting and massive sulphide genesis J. Geol. Soc. Lond., 168, 1–15.

42.

Williams

. 1934. The geology of the Rio Tinto Mines, Spain Trans. Inst. Min. Metal., 43, 593–640.

43.

Williams

, Stanton

, Rambaud

. 1975. The Planes–San Antonio pyritic deposit of Rio Tinto, Spain: its nature, environment and genesis, Appl. Earth Sci. (Trans. Inst. Min. Metal. London B), 84, B73–B82.

44.

Winchester

, Floyd

. 1977. Geochemical discrimination diagrams of different magna series and their differentiation products using immobile elements Chem. Geol., 20, 325–343.