Abstract

Fibrous NaFe3+-amphiboles (winchite, richterite, and magnesioriebeckite) form primarily by alkali metasomatism from magmatic fluids expelled from carbonatite or peralkaline silicate magmas, and have been implicated in high rates of death and disease at Libby, Montana (USA). Fibrous NaFe3+-amphiboles, principally winchite and magnesioriebeckite, are found as fracture-fill veins and as replacement of magmatic hornblende in faulted margins of the dominantly subalkaline, metaluminous Miocene Wilson Ridge pluton, Mohave County, Arizona (USA). Here, the fibrous NaFe3+-amphiboles formed from hypersodic, high-forumla hydrothermal fluids, which circulated through active faults as the pluton cooled through subsolidus temperatures. Halite deposits in adjacent Miocene sedimentary basins are the likely source of Na in the hydrothermal fluid. Amphibole fibers are <1 µm in diameter (typically <0.5 µm), vary from tens to hundreds of microns in length with length-to-width aspect ratios of 20:1 to over 100:1, are capable of dust transport and human inhalation, and should be considered hazardous. Transport and deposition of sediment eroded from primary pluton sources significantly increase the areal distribution of the fibrous amphiboles. Mitigation strategies require an understanding of the geologic settings where hazardous geologic materials are found. Our results suggest that fibrous NaFe3+-amphibole may be present in areas not previously considered at risk for naturally occurring asbestos.

INTRODUCTION

Fibrous NaFe3+-amphiboles are known human carcinogens (Aust et al., 2011; Antao et al., 2012). Although stable over a wide range of temperatures and pressures, NaFe3+-amphibole occurrences are restricted to rare hypersodic compositions combined with high oxygen fugacity (forumla) near the hematite-magnetite buffer (Evans, 2007). In magmatic systems, this hypersodic oxidized environment is thought be related to differentiation of carbonatite magmas or alkaline to peralkaline silicate magmas that release Na-rich metasomatizing fluids (Van Gosen, 2007). Here we document an unusual occurrence of fibrous NaFe3+-amphiboles that formed within and adjacent to the Miocene metaluminous Wilson Ridge pluton in Arizona, USA. The Na in the hydrothermal fluids that formed these amphiboles by metasomatic processes was not magmatic in origin, but was dissolved from nearby Miocene halite (evaporate) deposits during cooling, faulting, and uplift of the pluton. We integrate this new finding with the recently reported finding of fibrous actinolite associated with Miocene plutons in an adjacent area of southern Nevada (Buck et al., 2013). Understanding the origin of these amphiboles is critical to predicting fibrous amphibole occurrences and protecting public health, particularly with respect to environmental exposures to naturally occurring asbestos.

FIBROUS MINERALS AND DISEASE

Occupational exposures to commercial asbestos led to the recognition of its carcinogenic effects. In most developed countries, six fibrous silicate minerals are regulated as asbestos (Van Gosen, 2007); five of these are amphiboles, including actinolite and the blue NaFe3+-amphibole riebeckite. Regulatory classification of amphibole does not keep pace with mineralogical classifications (Meeker, 2008). Many non-regulated fibrous minerals have been linked to disease, including the fibrous NaFe3+-amphiboles winchite, richterite, and magnesioriebeckite at the Libby Asbestos Superfund site, Montana, USA (Meeker et al., 2003). At Libby, both environmental and occupational exposure to fibrous NaFe3+-amphiboles is causing significantly increased rates of mortality through diseases such as asbestosis, malignant mesothelioma, and respiratory cancers (Antao et al., 2012).

Although occupational exposure to asbestos has been steadily declining, awareness of possible health risks from environmental exposure of naturally occurring asbestos (NOA) is increasing (Harper, 2008). NOA is a natural component of rocks and soils but may include fibrous minerals that do not meet the regulatory definitions of asbestos (Harper, 2008). The primary pathway for human exposure from NOA is through dust emissions, which are particularly problematic in arid regions. Dust emissions from soils containing these minerals can be generated by both natural wind erosion and anthropogenic activities (Buck et al., 2013). Because NOA is often disseminated through geologic materials and may not be readily apparent in outcrop, an understanding of the geologic settings where NOA may occur is critical to mitigating human health risk.

GEOLOGIC SETTING

Primary outcrops of fibrous NaFe3+-amphiboles are hosted in the Miocene Wilson Ridge pluton, Mohave County, Arizona, in veins and as replacement of magmatic hornblende particularly in the proximity of faults that cut the west and east margins of the pluton (Potts, 2000) (Fig. 1). The pluton crops out near the northern termination of the northern Colorado River extensional corridor (NCREC) within the Basin and Range province (Larsen and Smith, 1990) (Fig. 1). In the NCREC, large-scale extensional faulting from ca. 20 to 12 Ma resulted in a number of fault-bounded north-northwest–trending blocks (Felger and Beard, 2010). Proterozoic igneous and metamorphic rocks form the basement to these blocks. Their superstructure consists of a sequence of pre- and syn-extensional mafic to silicic volcanic rocks intercalated with continental sediments. Coeval with the volcanic sequence is a suite of plutons (17–12 Ma) that are predominantly quartz monzonite, but include some more mafic rocks, and are commonly cut by aphanitic dikes (Larsen and Smith, 1990; Metcalf, 2004). All these rocks have been severely tilted by the extensional faulting.

Isotopic ages presented below bracket the onset of fibrous NaFe3+-amphibole mineralization to between 13.73 and 11.7 Ma, with active mineralization ending by 5.9 Ma. Ion probe 238U/206Pb zircon data collected at the Secondary Ionization Mass Spectroscopy Laboratory at the University of California–Los Angeles (USA) for two samples of Wilson Ridge pluton yielded ages of 13.78 ± 0.43 Ma and 13.73 ± 0.43 Ma (2σ) (Fig. DR1 and Table DR1 in the GSA Data Repository1). These data establish the pluton solidification age, which represents the maximum age of NaFe3+-amphibole mineralization. Erosion of the pluton and overlying volcanic rocks during syn-extensional uplift deposited sediment in an adjacent growth fault, forming the late Miocene to Pliocene Black Mountain conglomerate (Fig. 1). Clasts in the conglomerate include gneiss, Miocene volcanic rocks, and lithologies of the Wilson Ridge pluton including clasts containing NaFe3+-amphibole (Fig. 1). Volcanic clasts dominate the lower conglomerate section, but plutonic clasts increase up section, where 7%–28% of the total clast count includes NaFe3+-amphibole (Williams, 2003). Underlying the conglomerate are volcanic rocks as young as 12.7 Ma (Mills, 1994), and intercalated within the conglomerate near its base is a volcanic ash layer with a reported 40Ar/39Ar sanidine age of 11.72 ± 0.06 Ma (Williams, 2003). Capping the conglomerate are post-extensional 5.9 Ma alkali basalts (Mills, 1994) (Fig. 1). These data show that the Wilson Ridge pluton was uplifted and eroded soon after crystallization, and constrain the formation age of fibrous amphibole to between 13.73 and 5.9 Ma.

Buck et al. (2013) recently reported the occurrence of fibrous actinolite, a regulated asbestos mineral, in the Las Vegas, Nevada, metropolitan area (Fig. 1). Bedrock outcrops of actinolite NOA are found in hydrothermally altered Miocene plutons as veins and replacement of primary magmatic hornblende. Actinolite fibers were also found in alluvial fans eroded from the Miocene plutons and in dust generated from their surfaces.

CHARACTERIZATION OF FIBROUS NaFe3+-AMPHIBOLES

In the field, NaFe3+-amphiboles, easily recognized by their distinctive blue color, are found as veins in fractured and brecciated hornblende-bearing plutonic rock (Fig. DR2). Individual veins are millimeters to centimeters wide; planar networks of veins and breccia up to several meters wide can be traced for hundreds of meters parallel to mapped faults. NaFe3+-amphiboles in fractures are observed as both slip fibers (parallel to fracture margins) and cross fibers indicative of growth during active faulting (Van Gosen, 2007). NaFe3+-amphiboles also are found disseminated within phaneritic granitoid rocks and aphanitic dikes in proximity to veins. Disseminated NaFe3+-amphiboles have fibrous habit either as epitaxial overgrowths on, or as replacement of, magmatic hornblende. Hematite-after-magnetite replacement textures were observed in thin section adjacent to NaFe3+-amphibole veins. Veins and breccia zones filled with hematite are found in the field near NaFe3+-amphibole vein networks; in thin section, minor amounts of blue amphibole were found in the margins of hematite veins. Slickenlines in hematite and blue NaFe3+-amphibole can be observed on fault surfaces.

Amphibole mineralogical classification (Fig. 2A; Fig. DR3; Table DR2) is based on quantitative chemical analyses measured on a JEOL 8900 electron microprobe at the University of Nevada–Las Vegas (UNLV) using polished thin sections. Data were collected from five phaneritic samples where fibrous NaFe3+-amphiboles were present as veins and two aphanitic dike samples where fibrous NaFe3+-amphiboles replaced magmatic amphibole. Structural formula calculations, including Fe3+ estimates, and amphibole classifications followed recommendations of the International Mineralogical Association (Leake et al., 1997) (Fig. DR3; Table DR2). The Leake et al. (1997) classification scheme was used to facilitate comparison with published data for Libby amphibole. The fibrous amphiboles are classified as winchite (52%), magnesioriebeckite (24%), actinolite (16%), and richterite (8%). Significant compositional variation exists in single samples. Substitution of Na+ for Ca2+ on the B site is largely charge balanced by substitution of Fe3+ for divalent cations on the C site (Fig. 2B; Table DR2). Compositions of Wilson Ridge fibrous amphiboles are similar to that of fibrous amphibole linked to asbestos-related diseases at Libby (Meeker et al., 2003) (Fig. 2). Primary magmatic amphiboles, which appear unaltered in thin section, classify as magnesiohornblende (Figs. DR2d and DR3; Table DR2).

Fibers scraped from three vein samples were examined and had their dimensions measured using a field emission scanning electron microscope at UNLV. Imaged samples exhibit bundles of parallel fibers with splayed ends and masses of matted fibers (Fig. 3). Individual fibers typically have widths of <0.5 µm and vary from tens to hundreds of microns in length with length-to-width aspect ratios of 20:1 to over 100:1 (Fig. 3A). Curved fibers are common in bundles and in mats, with greater curvature observed in fiber mats (Fig. 3B). Based on their morphology, such fibers should be considered pathogenic and hazardous (Aust et al., 2011).

ORIGIN OF FIBROUS NaFe3+-AMPHIBOLE

We envision a metasomatic origin for NaFe3+-amphibole in a hydrothermal system where saline, high- forumla aqueous fluids circulated through faults and fracture networks. In order to evaluate a metasomatic origin, whole-rock chemistry for unmineralized samples of the Wilson Ridge pluton (Larsen and Smith, 1990) was compared to compositions for mineralized pluton samples with visible blue NaFe3+-amphibole observed in hand sample or thin section (Table DR3; Fig. DR4). The plutonic rocks containing NaFe3+-amphibole show elevated Na2O (mean = 8.2 wt%, standard deviation [s.d.] = 0.9 wt%, n = 9) and depleted K2O (mean = 0.41 wt%, s.d. = 0.45 wt%, n = 9) compared to unmineralized plutonic rock (Na2O mean = 4.2 wt%, s.d. = 0.6, n = 28; K2O mean = 2.7 wt%, s.d. = 1.1 wt%, n = 28). Using TiO2 as a differentiation index, we found elevated Na2O regardless of the pre-mineralization host-rock composition (Fig. DR4). Electron microprobe analysis shows complete albitization of feldspars. These results are consistent with Na metasomatism. The narrow time interval between solidification of pluton host rocks (13.75 Ma) and exposure of NaFe3+-amphibole to surface erosion (ca. 11.7 Ma) suggests mineralization driven by residual heat as the pluton cooled through subsolidus temperatures. Slickenlines in hematite and NaFe3+-amphibole, and NaFe3+-amphibole slip fibers, demonstrate mineralization contemporaneous with pluton faulting. Hydrothermal fibrous NaFe3+-amphibole mineralization involved both direct precipitation in veins and alteration of pre-existing magmatic magnesiohornblende. Magnesium, a crucial component in the formation of these NaFe3+-amphiboles (Van Gosen, 2007), likely was scavenged from magnesiohornblende along fluid pathways in the pluton host.

Reported NaFe3+-amphibole plutonic occurrences include metasomatic rocks called fenites found in association with carbonatite–alkaline silicate complexes (Rainy Creek alkali-ultramafic plutonic complex, Libby, Montana [Van Gosen, 2007]; Mountain Pass rare-earth-element carbonatite-ultrapotassic intrusive complex, southern California [Castor, 2008]; Fen carbonatite, Norway [Kresten, 1988]), and peralkaline A-type granitic complexes (Bryansky complex, Russia [Litvinovsky et al., 2002]; Serra do Meio complex, Brazil [Plá Cid et al., 2001]). Fenites typically contain alkali pyroxene and/or alkali amphiboles, including K, K-Na, and, more rarely, Na types, as fracture-fill veins or replacement of Ca-pyroxene or Ca-amphibole (Kresten, 1988). In A-type granitic intrusions, NaFe3+-amphiboles are present as primary magmatic crystals or as deuteric alteration products of Na-pyroxene (Plá Cid et al., 2001). In both settings, high-forumla, Na-metasomatizing fluids are inferred to be magmatic in origin, emanating from fractionated mantle-derived carbonatite magma or silicate magmas fractionated to peralkaline compositions (Fig. 4).

Carbonatite and peralkaline silicate intrusions are absent in the NCREC magmatic record. In contrast, Miocene NCREC plutonic complexes, including Wilson Ridge, evolved by fractional crystallization of lithospheric mantle–derived alkali basalt combined with open-system mixing of large volumes of crustal-derived subalkaline felsic magma (Larsen and Smith, 1990; Metcalf, 2004). Miocene NCREC plutonic suites are metaluminous in composition (Fig. 4), including portions of Wilson Ridge affected by Na metasomatism; late-stage fractional crystallization in NCREC plutons produced metaluminous, high-silica leucogranites (Walker et al., 2007). Lacking an obvious magmatic source for Na-metasomatizing fluids, an alternate petrogenetic model is needed.

Halite deposits intercalated within adjacent Miocene conglomerate are the most likely source for Na in metasomatizing fluids (Potts, 2000). During Miocene time (16–9 Ma), thick evaporite (halite, gypsum) and lacustrine (limestone) deposits formed in the NCREC in closed growth-fault basins fed by groundwater recharge systems (Faulds et al., 1997). Thick non-marine halite deposits are recognized in cores and seismic profiles from several syn-extensional Miocene basins, including >500 m of halite beneath Detrital Valley (Faulds et al., 1997) east of Wilson Ridge (Fig. 1). While no cores penetrate the Black Mountain conglomerate, gypsum is present in surface exposures, indicating that evaporate deposits formed in the basin west of Wilson Ridge. The juxtaposition of an active evaporate growth-fault basin with active faulting of the Wilson Ridge pluton as it cooled through subsolidus temperatures appears to have generated the hypersodic, high-forumla hydrothermal system; NaFe3+-amphibole formed where fluids encountered Mg-rich (hornblende-bearing) plutonic rocks. The scavenging of Na from sedimentary deposits has been suggested as an origin of saline fluids in the Salton Sea geothermal field, California (McKibben et al., 1988), and for saline fluids associated in the Yerington porphyry copper district, Nevada (Dilles and Einaudi, 1992).

IMPLICATIONS FOR HEALTH RISK ASSESSMENT

Mitigation strategies for reducing human exposure to hazardous geologic materials require an understanding of the geologic settings where such materials are found. Thus, the broader significance of our results is the identification of a previously unrecognized mechanism and geologic setting for the formation fibrous NaFe3+-amphiboles. Settings where hydrothermal fluids have accessed both halite deposits and Mg-rich (hornblende-bearing) lithologies need to be considered as potential sites for hazardous fibrous NaFe3+-amphiboles. The presence of fibrous actinolite in the Boulder City pluton to the west of Wilson Ridge (Arizona) carries a similar implication, as fibrous actinolite is typically formed by dynamothermal metamorphism of mafic to ultramafic rock (Van Gosen, 2007). An obvious question is: Why were fibrous NaFe3+-amphiboles formed in one pluton and fibrous actinolite in the other? Saline hydrothermal fluids produced fibrous actinolite at both the Salton Sea geothermal field (Yau et al., 1986) and the Yerington district (Dilles and Einaudi, 1992). We hypothesize that differences in fibrous mineralogy between the Boulder City and Wilson Ridge plutons resulted from higher Na+ activities in the hydrothermal system affecting the Wilson Ridge pluton.

As with all naturally occurring asbestos, natural erosion processes have released the fibrous amphiboles from bedrock outcrops and redistributed them. Here, they are found in Miocene–Pliocene and Quaternary deposits (Fig. 1). Human exposures are primarily from both natural wind erosion and anthropogenic disturbances that release fibers from soil into the air. These processes are enhanced by the arid climate of the region. Fibrous NaFe3+-amphibole sources include public lands with popular hiking and off-road-vehicle trails, and fibrous actinolite sources overlap a portion of the greater Las Vegas metropolitan area (Fig. 1; see Table DR4 for sample coordinates and descriptions).

Mingua Ren provided technical support for scanning electron microscope and electron probe microanalyzer work. This research developed from an earlier University of Nevada–Las Vegas M.S. thesis by D.A. Potts, conducted under the direction of R.V. Metcalf. We gratefully acknowledge Jayson Medema’s work in producing Figure 1. The paper was improved by thoughtful comments from three anonymous reviewers.

1GSA Data Repository item 2015035, methods, Figure DR1 (U-Pb geochronology plots), Figure DR2 (field and microscope images), Figure DR3 (amphibole classification plots), Figure DR4 (whole-rock chemistry plots), Table DR1 (U-Pb geochronology data), Table DR2 (mineral chemistry), Table DR3 (whole-rock chemistry), and Table DR4 (sample locations and descriptions), is available online at www.geosociety.org/pubs/ft2015.htm, or on request from editing@geosociety.org or Documents Secretary, GSA, P.O. Box 9140, Boulder, CO 80301, USA.