Exploration, Discovery, and Geology of the Ntaka Hill Nickel Sulfide Deposits: A Potential New Proterozoic Nickel District in the Nachingwea Area, Mozambique Belt, Tanzania
Published:January 01, 2010
- PDF LinkChapter PDF
P. A. Tirschmann, C. S. MacDougall, G. R. Katchen, 2010. "Exploration, Discovery, and Geology of the Ntaka Hill Nickel Sulfide Deposits: A Potential New Proterozoic Nickel District in the Nachingwea Area, Mozambique Belt, Tanzania", The Challenge of Finding New Mineral Resources: Global Metallogeny, Innovative Exploration, and New Discoveries, Richard J. Goldfarb, Erin E. Marsh, Thomas Monecke
Download citation file:
© 2010 Society of Economic Geologists, Inc. Special Publication 15, pp. 487–512
The Ntaka Hill nickel sulfide deposits are hosted in the peridotitic to pyroxenitic Ntaka ultramafic intrusion located in the Nachingwea area and are the first significant occurrence of nickel sulfides in the Tanzania portion of the Late Proterozoic Mozambique belt. High-grade nickel sulfide mineralization was first discovered at Ntaka Hill in 2006. Six near-surface sulfide deposits have since been delineated containing a measured and indicated mineral resource of 1.8 million tonnes (Mt) @ 1.82 percent Ni and 0.31 percent Cu. The recent discovery history can be traced back to the presence of a historic surface copper oxide malachite showing. Subsequent soil sampling defined a large coincident Ni-Cu anomaly that provided the impetus for airborne and ground geophysical surveys and ultimately diamond drilling leading to the initial discovery. Further ground electromagnetic surveys were successful in defining additional moderate to high conductance anomalies resulting, upon drill testing, in the discovery of five additional nickel sulfide zones.
The Ntaka intrusion is postulated to have formed from a relatively primitive, high MgO magma, dominated by the crystallization and accumulation of olivine and pyroxene. The intrusion is characterized by high MgO contents, low CaO, Al2O3, Cu, PGE, and incompatible element contents, and relatively flat chondrite-normalized REE profiles lacking Eu anomalies. The geologic setting is similar to that of the Early Proterozoic Thompson Nickel belt in Canada. There supracrustal rocks formed on a continental margin platform and were intruded by ultramafic sills that interacted with the sulfidic metasedimentary rocks to produce the resulting nickel deposits.
The Ntaka Hill sulfide zones occur in separate south-plunging lenses but may represent remnants of a dismembered original basal sulfide zone. The zones consist of magmatic, remobilized, and graphite-bearing mineralization with variable nickel grades of as much as 17 percent. Mineralization consists of disseminated, net-textured, and massive magmatic sulfides, as well as remobilized semimassive and massive sulfide veins and stringers composed of pyrrhotite, pentlandite, pyrite, chalcopyrite, and violarite. Pentlandite is the main nickel-bearing sulfide mineral occurring as coarse grains and eyes that are as large as 5 cm in diameter. Pyrrhotite-rich, nickel-poor, graphite-bearing, disseminated to massive sulfide mineralization occurs at several locations within the Ntaka intrusion and is thought to have formed by assimilation of graphitic metasedimentary rocks.
The Ntaka intrusion possesses a number of elements critical to the formation of nickel sulfide deposits and good potential exists to discover additional nickel sulfide deposits, both in the Ntaka Hill area and regionally. Exploration challenges in this underexplored belt include a complex deformation history, an abundance of graphitic metasedimentary rocks, and a paucity of outcrop.
Ultramafic rock-hosted, high-grade nickel sulfides were discovered in the Late Proterozoic Mozambique belt of southeastern Tanzania during an exploration drilling program in 2006 to follow up historic occurrences of low-grade nickel sulfide and surface copper oxide mineralization within the Ntaka intrusion. To date, six mineralized zones totaling 1,835,000 tonnes (t) grading 1.82 percent Ni, 0.31 percent Cu, and 0.05 percent Co (Continental Nickel Limited, 2009) have been discovered and delineated. They represent the first significant occurrence of nickel sulfides within the Tanzanian portion of the Mozambique belt. This paper describes the exploration history and geology of the Nachingwea area Ntaka Hill nickel sulfide deposits and comments on the potential for additional nickel deposits within this mainly unexplored and little documented geologic terrane.
Regional Geologic Setting
The Ntaka Hill nickel sulfide deposits are located in the Nachingwea district of southeastern Tanzania and approximately 400 km south of Dar es Salaam, 180 km west of the coastal port of Mtwara, and 100 km north of the border with Mozambique. Ntaka Hill is located 45 km north of the town of Nachingwea. The area lies within the Late Proterozoic Mozambique belt, which is one of several Proterozoic mobile belts bordering and surrounding the granite and greenstone assemblages of the central part of the Archean Tanzania craton. The Tanzania craton is bounded to the northwest by the Middle Proterozoic Kibaran belt, to the southwest by the Early Proterozoic Ubendian-Usagaran belt, and to the east and southeast by the Late Proterozoic Mozambique belt (Fig. 1).
Mining and exploration activity in Tanzania has historically focused on the Archean greenstone belts of the Lake Victoria gold district. However, the surrounding mobile belts are becoming increasingly recognized for their potential to host Ni-Cu-PGE deposits. The Kibaran belt hosts the large Kabanga nickel deposits located in northwestern Tanzania, near the border with Burundi, with mineral resources totalling 51.7 million tonnes (Mt) @ 2.67 percent Ni and 0.38 percent Cu (XStrata Nickel, 2009). The Kapalagulu layered intrusion located in western Tanzania and situated at a contact between underlying Ubendian gneisses and overlying Kibaran metasedimentary rocks hosts both PGE and massive nickel sulfide mineralization. Both the Kabanga and Kapalagulu intrusions have been dated at ˜1400 Ma and are postulated to be part of the same magmatic belt (Maier et al., 2007). With the recent discoveries of ultramafic rock-hosted, high-grade nickel sulfide mineralization in the Nachingwea area, the Mozambique belt represents a new Proterozoic terrane in Tanzania with the potential to host significant nickel sulfide deposits. The definitive age of the Nachingwea ultramafic bodies and related mineralization is, as yet, undetermined.
The Mozambique belt is a dominantly north-south trending orogenic domain of highly deformed and metamorphosed rocks that formed during oblique collision of East and West Gondwana and are part of the Pan African orogenic system. Peak metamorphic conditions to granulite facies are dated at 640 Ma (Muhongo et al., 2001; Sommer et al., 2003). The following geologic summary includes data summarized from Tenczer et al. (2005) and Bauernhofer et al. (2003). The Mozambique belt is bounded to the west by the 2.7 Ga Tanzanian craton with a narrow intervening band of 2.0 to 1.8 Ga Usagaran belt rocks. The first stage of Mozambique belt formation occurred between 1,000 and 700 Ma (Tenczer et al., 2005) and was marked by large-scale magmatic intrusive activity relating to a long period of island arc accretion. Evidence of this activity can be found in the eastern part of the Mozambique belt, where meta-anorthosites and meta-igneous granulites record magmatic ages ranging from 950 to 820 Ma. The ultramafic-mafic intrusions observed on the Nachingwea property may have formed during this period, but no geochronology has been done to confirm this. The initial phase of Mozambique belt formation was followed by two collisional phases recorded at 640 to 620 and 580 to 530 Ma, which resulted in west-directed thrust propogation and regional deformation and metamorphism. The 640 to 620 Ma phase involved the onset of deep-seated thrusting and lateral shearing, whereas the 580 to 530 Ma phase of final collision involved thrust propagation and exhumation. Regional metamorphic gradients range from greenschist facies in the west to granulite facies in the east. In the western part of the Mozambique belt, magmatic rocks have Archean (2.7–2.5 Ga) crystallization ages, similar to those of the Tanzanian craton, and Early Proterozoic (2.0–1.8 Ga) ages reflecting contributions from the Usagaran belt.
In the Nachingwea area, Mozambique belt lithologic units consist of a mixed assemblage of mafic to felsic granulites, gneisses, and migmatites that are interlayered with amphibolites and metasedimentary rocks, including quartzites, banded magnetic quartzites, pelites, graphitic schists, and marbles (Fozzard and Quinnel, 1957). These lithologic units are crosscut by poorly documented mafic to ultramafic intrusions of unknown age and include the Ntaka ultramafic intrusion. All units are complexly deformed and metamorphosed to amphibolite and granulite grades of metamorphism and occur in blocks bounded by major northwest-, east-, and northeast-striking fault zones. The high degree of deformation in this area is best illustrated by aeromagnetic data that reveal a complex pattern of folding, which is further disrupted by faulting. The regional geologic and structural setting of the Ntaka Hill nickel sulfide deposits bears a marked resemblance to that of the nickel deposits in the Thompson Nickel belt, Canada, where boudinaged and dismembered ultramafic bodies are hosted within high-grade gneissic and schistose metasedimentary rocks of the Ospwagan Group in an Early Proterozoic continental margin setting (Bleeker, 1990a, b; Layton-Matthews et al., 2007).
Exploration and Discovery History
The discovery of the Ntaka Hill sulfide deposits can be traced back to the presence of a historic surface copper oxide showing within the Ntaka ultramafic intrusion. Subsequent soil sampling surrounding the original showing defined a much larger coincident Ni-Cu anomaly, which provided the impetus for more focused and rigorous nickel sulfide exploration programs comprised of airborne and ground geophysical surveys and diamond drilling.
Three ultramafic intrusions were documented in the Nachingwea area on a 1:100,000 scale map published by the Geological Survey of Tanzania (Fozzard and Quinnel, 1957). These included an ultramafic body near the town of Nachingwea, a serpentinized body located ˜5 km south-southwest of Mnero Mission, and the Ntaka intrusion cropping out 3 km west of Nditi village and forming a topographic high known locally as Ntaka Hill. Between 1950 and 1953, INCO and Selection Trust carried out mapping, trenching, and drilling at Ntaka Hill in the immediate area of a historic surface malachite showing (Fig. 2). Six diamond drill holes (1,307 m) were completed and intersected a best value of 1.60 percent Ni and 0.56 percent Cu over 3.7 m.
Between 1996 and 1998, BHP carried out a regional exploration program in the Nachingwea area searching for base metals that included Broken Hill-type Pb-Zn-Ag deposits. They completed a regional magnetic and radiometric airborne survey, as well as a large stream sediment and soil sampling program but do not appear to have completed any work in the vicinity of Ntaka Hill.
Goldstream Mining NL
In 2000, Goldstream Mining NL of Australia, now known as IMX Resources, acquired their first prospecting license in the Nachingwea area and rediscovered the historic malachite showing and pits at Ntaka Hill. In 2004, Goldstream carried out soil and stream sediment surveys that included a detailed soil survey over Ntaka Hill. Samples were collected at 100-m centers on east-west lines spaced 400 m apart with detailed sampling completed at 25-m centers on 200-m-spaced east-west lines. The soil survey returned highly anomalous values of Ni and Cu defining a large, 2- × 4-km coincident Ni-Cu anomaly (Fig. 3A). Many of the soil samples contained >1,000 ppm Ni, with a maximum of 8,360 ppm Ni, and >500 ppm Cu, with a maximum of 6,940 ppm Cu. Based on these encouraging results, Goldstream flew a small, 450-line-km helicopter-borne versatile time domain electromagnetic (VTEM) survey in 2005 and obtained a number of electromagnetic (EM) anomalies, including an area of strong conductivity situated over the north-central part of the Ntaka intrusion (Fig. 3B). In 2006, selected VTEM anomalies were followed up with small, fixed and moving loop ground EM surveys and selected anomalies were drill tested.
Goldstream completed 17 diamond drill holes (2,153 m), including 14 holes at Ntaka Hill and three holes located ˜8 km southwest of Ntaka Hill, in an area called Lionja, where soil samples also returned elevated nickel and copper values. Twelve of the 14 widely spaced drill holes at Ntaka Hill intersected sulfide mineralization returning values >0.5 percent Ni, with a number of samples assaying between 0.5 and 2.0 percent Ni. Drill hole NAD013 intersected high-grade nickel sulfides returning 11.23 percent Ni, 1.74 percent Cu, and 0.15 percent Co over 3.0 m, including 15.87 percent Ni, 2.61 percent Cu, 0.21 percent Co, 0.49 g/t Pt, and 0.32 g/t Pd over 1.68 m and is considered the discovery hole at Ntaka Hill. Hole NAD013 was drilled 1.3 km to the southwest of the historic malachite showing on the western side of the Ntaka intrusion in an area of minimal outcrop.
Continental Nickel Limited
Continental Nickel Limited of Canada formed a 70/30 venture with Goldstream in 2007 and between 2007 and 2009 carried out extensive exploration of the Ntaka intrusion and surrounding regional land holdings. In 2007, Continental Nickel completed a large surface time domain EM survey over the entire Ntaka intrusion, utilizing the Crone 4.8 kWatt Pulse EM time domain system. The survey was designed to look for high conductance, long time constant, slow decaying anomalies within the intrusion. Surveying was conducted using nine large (1,000- to 1,500- × 800- to 1,000-m) and two small (400- × 400-m) fixed rectangular to square transmitter loops. The X and Z component data were collected in both in-and out-of-loop configurations on 100- to 200-m spaced lines at a station interval of 25 m. The EM survey resulted in the identification of multiple anomalies that were subsequently modelled as moderate to high conductance (1,500–7,100 siemens) targets ranging from 45 to 375 m in strike length and 25 to 225 m in dip extent.
Drill testing of the new ground EM anomalies was carried out in late 2007 and resulted in the discovery of five new sulfide zones, including the J, M, H, L, and G zones, as well as additional mineralization at the NAD013 discovery zone. The relative locations of the sulfide zones and selected 2007 drill intersections are shown in Figure 4. Delineation drilling was carried out at all six sulfide zones at Ntaka Hill in 2008 and resulted in the estimation of measured and indicated mineral resources totaling 3,085,000 t grading 1.31 percent Ni, 0.24 percent Cu, and 0.04 percent Co at a Net Smelter Return (NSR) cut-off of US$23/t (Table 1). At a higher NSR cut-off of US$50/t, measured and indicated mineral resources total 1,835,000 t grading 1.82 percent Ni, 0.31 percent Cu, and 0.05 percent Co. To date, Continental Nickel has completed 169 drill holes totaling 25,555 m at Ntaka Hill. Drilling in 2009 continued to intersect new zones of sulfide mineralization, which have yet to be evaluated for their resource potential. Regional exploration carried out by Continental Nickel in 2008 and 2009 has led to the identification of new ultramafic occurrences. Follow-up drilling in the Lionja area has intersected nickel sulfide mineralization assaying as high as 2.03 percent Ni and 0.41 percent Cu over 2.2 m.
The proceeding geologic and geochemical descriptions and interpretations are synthesized from data collected during the 2006 to 2009 exploration programs carried out at Ntaka Hill and include a database of 189 drill holes totaling 28,598 m. Surface mapping of the Ntaka intrusion was carried out by Goldstream Mining. Two samples suites were assembled by Continental Nickel for detailed analysis, including a suite of 27 unmineralized to weakly mineralized samples for petrological and geochemical characterization (Table 2) and a suite of 28 mineralized samples for sulfide mineral characterization.
Unmineralized to weakly mineralized samples
The suite of unmineralized to weakly mineralized samples was comprised principally of ultramafic rocks from the Ntaka intrusion and mafic to ultramafic country rocks but also included one sample of a siliceous graphitic metasedimentary inclusion from the J zone. Mafic and ultramafic country rocks included gabbros, amphibole gneisses, and amphibolites and were sampled for comparison to lithologic units of the Ntaka intrusion and to assess possible genetic relationships. Country-rock amphibolites are often visually similar to the amphibolite-altered pyroxenites of the Ntaka intrusion and contact relationships can be ambiguous due to apparent tectonic interfingering of units.
Whole-rock major and trace element analyses, as well as Pt, Pd, Au, and S analyses, were completed for all samples. Whole-rock analyses were conducted by lithium borate fusion and ICP-MS techniques. The Pt, Pd, and Au contents were determined by lead bead fire assay and ICP-MS finish and S by Leco furnace and infrared spectroscopy. Samples typically consisted of a 15- to 25-cm-long piece of one-half or one-quarter NQ size (47.6 mm diam) drill core and were selected to exclude alteration, particularly in the form of veining or fracture filling.
Preparation and petrographic analyses of polished thin sections were carried out by Vancouver Petrographics Ltd. for a selected subset of 12 of the whole-rock samples. Mineralogical descriptions of the various mafic and ultramafic rocks are summarized mainly from Leitch (writ. commun., 2009). Additional petrographic information was obtained from previous work carried out by Goldstream Mining. Amphibole and pyroxene mineral species are interpreted from petrography but are also supported by electron microprobe analyses of silicate minerals observed in the second sample suite of mineralized samples from the various Ntaka sulfide zones.
Twenty-eight drill core samples were collected from five of the Ntaka sulfide zones (J, M, H, NAD013, and L) and were examined using a scanning electron microscope and electron probe microanalysis. Analytical work was carried out by Xstrata Process Support in Sudbury, Ontario, Canada. Samples typically consisted of a 3- to 7-cm-long piece of one-half NQ-size drill core and were selected to be representative of the various styles of mineralization observed in the sulfide zones. Samples from each zone represented sulfide mineralization, but examples of transition and oxide mineralization were also included for the J zone.
Geology of the Ntaka Hill Area
The Ntaka Hill nickel sulfide deposits are underlain by highly deformed and metamorphosed rocks of the Mozambique belt. The nickel sulfide deposits are hosted within the dismembered remnants of a package of deformed ultramafic rocks referred to as the Ntaka intrusion, one of three ultramafic bodies originally documented in the area (Fig. 5). The country rocks of the Ntaka intrusion consist of felsic to mafic gneisses and amphibolites, which are interpreted to represent, at least in part, metasedimentary and metavolcanic supracrustal rocks.
All lithologic units have undergone polyphase deformation and high-grade metamorphism and are extensively recrystallized. The peak metamorphic event lies within lower (hydrous) granulite facies as determined from mineralogical observations and petrographic studies (M. Prendergast, writ. commun., 2006). The Ntaka Hill area consists of a complex basin with refolded isoclinal folds. The north- to northwest-trending fold axes contain the ultramafic rocks that make up a shallow, complexly infolded, synformal body about 5 km long and as much as 3 km wide, which plunges gently to the south (M. Prendergast, writ. commun., 2006). This structural interpretation was based on an analysis of patterns from a regional aeromagnetic survey, as well as from geologic mapping. Subsequent drilling has shown that the mineralized zones uniformly plunge to the south.
The age of the Ntaka intrusion is not known and age relationships with the surrounding gneisses and amphibolites are obscured by the high degree of the deformation and metamorphism. The intrusion could be as old as 1.8 Ga (i.e., Usagaran) or, alternatively, could be related to the much younger 1000 to 700 Ma magmatic activity that occurred early in the development of the Mozambique belt.
The Ntaka intrusion is best exposed at Ntaka Hill, which is a 50-m topographic high feature representing the northern surface expression of the intrusion. The intrusion is ˜ 3 × 5 km in size and can be traced to the south and west of Ntaka Hill in sporadic outcrop and drill intersections and as interpreted from a subtle moderate magnetic high signature. The ultramafic rocks comprising the Ntaka intrusion include peridotite, olivine pyroxenite, pyroxenite, and amphibole pyroxenite. These lithologic units are typically massive to weakly foliated and exhibit granoblastic and poikiloblastic textures. Altered ultramafic is a term applied during logging to the pyroxenitic units that have undergone substantial recrystallization and are now comprised mainly of amphiboles ± biotite and/or phlogopite. Units of ultramafic amphibolite are locally present and can be difficult to distinguish from country-rock amphibolites.
The ultramafic lithologic units are heterogeneously distributed, in some cases on the scale of only a few meters, and often exhibit indistinct or gradational boundaries attributed to secondary metamorphic and structural modification. The intrusion does not exhibit any well-defined layering and likely represents a compositionally zoned ultramafic body. The pyroxenitic units are typically nonmagnetic, whereas the more olivine-rich peridotitic units have a weak to moderate magnetic signature due to alteration of the olivine to a mixture of serpentine and magnetite. The distribution of these latter magnetic units is thought to be correlated with a subtle magnetic high that forms an oval-shaped ring in the center of the intrusion, as well as several satellite magnetic highs.
Where observed in drilling, the external contacts of the ultramafic body often appear to be gradational in nature, consisting of alternating decimetre- to meter-scale intervals of ultramafic and gneissic country rocks occurring within a mixed zone as much as several tens of meters in width. Decimeterto meter-scale intervals of gneissic rock with sharp external contacts also occur within the intrusion and the majority of these are interpreted to represent either inclusions or tectonically interfingered units. One notable exception occurs at the H zone where a 5- to 10-m-thick unit of garnet-amphibole gneiss bounded on both sides by ultramafic rocks was intersected along a 100-m-dip extent in five holes drilled on one section. This unit appears to be an in situ and traceable stratigraphic horizon.
Although the Ntaka ultramafic rocks have historically been interpreted to represent an intrusion, they could, alternatively, be a sequence of ultramafic extrusive flows. Definitive textural and stratigraphic evidence for such an interpretation is, however, lacking due to the extensive effects of deformation and metamorphic recrystallization.
The Ntaka peridotites are medium to coarse grained and have a mottled or spotted black and gray appearance. They can be classified as harzburgites and are comprised of approximately subequal amounts of olivine and orthopyroxene, with accessory phlogopite, chlorite, spinel, ilmenite, and sulfide minerals. Olivine forms rounded anhedral to euhedral crystals as much as several millimeters in diameter and partially altered to very fine grained serpentine and magnetite. Ortho -pyroxene (enstatite) forms colorless to pale green or brown, rounded to tabular subhedral crystals as much as 2 mm in size, which are locally mantled by clinoamphibole (hornblende) and/or crosscut by serpentine-filled fractures. Ilmenite, sulfide minerals, and phlogopite, the latter variably altered to chlorite, all occur interstitial to the olivine and orthopyroxene. Pyrrhotite is the dominant sulfide phase, but small amounts of chalcopyrite and pentlandite are also present. Spinel forms ragged, irregular, to skeletal crystals as much as 1.5 mm across, which are poikilitically enclosed within amphibole.
The Ntaka pyroxenites are medium to coarse grained, possess poikiloblastic and granoblastic textures, and have a medium brown color on weathered outcrop surfaces. They range from orthopyroxenites to clinopyroxenites to websterite and are locally olivine bearing. The main pyroxene mineral species are interpreted, from petrography, to be enstatite and augite, which is in agreement with mineral species identified in electron microprobe analysis of mineralized pyroxenite samples from several of the nickel sulfide zones (L. Kormos et al., writ. commun., 2009).
Partial alteration of pyroxene to amphibole is typical and, in some cases, the pyroxenites are completely altered to amphibolite, resulting in a similar appearance to the country-rock amphibolites. The least altered pyroxenites are comprised of interlocking, rounded, subhedral orthoproxene and/or clino -pyroxene ± finer grained, irregular, anhedral to euhedral crystals of olivine with interstitial amphibole and accessory sulfide minerals and iron oxides. Sulfides include pyrrhotite and lesser amounts of pentlandite and chalcopyrite. With increasing amphibolitization, the pyroxenes are observed to contain small inclusions of amphiboles increasing in amount until the pyroxenes themselves become poikitically enclosed in or are completely replaced by amphiboles.
Amphibolite-altered pyroxenites are typically comprised of remnant ragged to irregularly shaped, subhedral orthopyroxene and/or clinopyroxene crystals as much as 2.5 mm in size, which are poikilitically enclosed in clinoamphibole and lesser orthoamphibole. Clinoamphibole occurs as subhedral crystals as much as 4 mm long with ragged terminations. The mineral species is interpreted to be hornblende based on pale olive green pleochroism and extinction angles of as much as 25°. Clinoamphibole is intimately intergrown with slender, lathlike to tabular crystals of orthoamphibole, as much as 2.5 mm long and interpreted to be anthophyllite. Hornblende and anthophyllite amphiboles were routinely noted in previous petrographic analyses (M. Prendergast, writ. commun., 2006). However, hornblende and actinolite were identified by electron microprobe analysis of mineralized pyroxenite samples (L. Kormos et al., writ. commun., 2009). Accessory minerals include sulfides, ilmenite, and biotite and/or phlogopite, the latter partially altered to chlorite. Trace amounts of carbonate minerals occur along thin fractures and as alteration products after pyroxene.
The country rocks to the Ntaka intrusion consist of felsic and intermediate gneisses, gabbros, amphibole gneisses, amphibolites, and pegmatite. The felsic and intermediate gneisses are comprised of variable amounts of quartz, feldspar, biotite, garnet, amphibole, and graphite. They are interpreted to represent highly metamorphosed sedimentary rocks that include quartzites, semipelites, pelites, and graphitic metasedimentary rocks. Narrow intervals of carbonate-rich rock, typically altered to skarn, as well as siliceous, iron-rich, garnet-bearing rocks have been intersected locally in drill holes and appear to be calcareous metasedimentary rocks and silicate facies banded iron formation, respectively. The amphibole gneisses and amphibolites are most likely of mafic intrusive or extrusive igneous origin. Medium- to coarse-grained felsic pegmatites occur as veins and irregular bodies within both the intrusion and in the country rocks and are remobilized melt products derived from the gneisses during high-grade metamorphism.
Graphite occurs in both graphitic gneisses and locally within the ultramafic lithologic units, contributing to the conductive EM signature over the Ntaka intrusion. Within the graphitic gneisses, graphite occurs as fine to coarse, disseminated and foliated grains, as well as semimassive to massive laminations and bands. In the ultramafic rocks at the J and G zones, graphite occurs locally as disseminated grains, as well as millimeter- to centimeter-scale massive clots or patches, usually in close proximity to relict, but recognizable inclusions of graphitic metasedimentary rock.
Mafic country rocks bordering the Ntaka intrusion include gabbro, amphibole gneiss, and amphibolite. A group of massive to foliated mafic rocks occurring both in the country-rock package and as narrow lenses within the Ntaka intrusion were loosely termed gabbros (Fig. 5) by Prendergast (wit. commun., 2006). He described them as being comprised of granoblastic plagioclase and hornblende, with variable amounts of biotite and lesser clinopyroxene, garnet, quartz, cumming-tonite, and magnetite. These gabbroic rocks are relatively well developed along the western flank of the Ntaka intrusion and appear to form a semicontinuous, broad marginal zone to the ultramafic rocks. Lenticular outcrops of gabbroic rock were also mapped by Prendergast (writ. commun., 2006) within the central part of the Ntaka intrusion and were interpreted to be erosional windows exposing marginal lithologic units within shallow fold hinges. Massive to foliated metagabbros, with a unique mottled to striped appearance, have been intersected in drilling along the western side of the intrusion at the M, H, and L zones. These gabbros are mineralogically and geochemically similar to the country-rock amphibole gneisses and amphibolites and are described, along with the amphibolitic gneisses, in more detail below.
Medium- to coarse-grained, massive to weakly foliated gray and green metagabbros have been intersected in the vicinity of all sulfide zones along the western side of the Ntaka intrusion and their distinctive appearance makes them a potential stratigraphic marker. The metagabbros are comprised primarily of plagioclase (55%) and amphibole (40%), with accessory biotite, ilmenite, sulfide minerals, and apatite. Plagioclase forms rounded, subhedral to euhedral crystals, as much as 2 mm in diameter, which do not show any traces of zoning due to metamorphic recrystallization. Compositions are interpreted to range from calcic to sodic andesine (AN = 37−45) based on Y/Z extinction angles. Locally, crystals show weak alteration to very fine grained carbonate and sericite. Amphibole (hornblende) also occurs as rounded, subhedral to euhedral crystals, as much as 3 mm in diameter, but is commonly concentrated into irregularly shaped aggregates as much as 2 cm long. Locally, small <0.2-mm to 20-μm inclusions of ilmenite display a Schiller-like structure, suggesting possible replacement of former pyroxene. Biotite forms subhedral flakes as much as 1 mm in length, which are intergrown with amphibole and locally altered to chlorite. Sulfide minerals include pyrrhotite, pyrite, and chalcopyrite.
Amphibole gneiss and/or amphibolite
Amphibole gneisses and amphibolites have been intersected in many of the drill holes and are an integral part of the country-rock stratigraphy, typically occurring in contact with, and in close proximity to, the ultramafic rocks of the Ntaka intrusion. These amphibole-rich country rocks are fine to medium grained, weakly to moderately foliated, and green to black on weathered surfaces. Lithologic units form a continuum from banded to foliated amphibole-plagioclase gneiss through to foliated amphibolite and may represent metamorphosed mafic to ultramafic metavolcanic rocks or, in some cases, ultramafic intrusive rocks.
The amphibole gneisses and amphibolites are mineralogically very similar to the gabbroic rocks, differing mainly in texture and higher concentrations of amphibole. They are comprised of variable amounts of amphibole (50−90%) and plagioclase (10−45%), with accessory biotite, ilmenite, rutile, sphene, sulfide minerals, and apatite. Amphibole (hornblende) forms rounded to irregular subhedral crystals or aggregates of crystals, as much as 5 mm in diameter and exhibiting pale green to, locally, brown pleochroism. Inclusions of ilmenite and rutile in amphibole occur as subhedral to euhedral grains, as well as minute (<20-μm) grains with relict Schiller structure as in the gabbros. Weak carbonate alteration of amphibole occurs along fractures and cleavage traces. Plagioclase forms rounded, subhedral to euhedral interlocking crystals as much as 1 mm in diameter. Zoning is typically absent and compositions are interpreted to be that of sodic andesine (˜An35). Biotite occurs as subhedral to euhedral foliated flakes as much as 1.5 mm in diameter and locally intergrown with amphibole. Accessory opaque grains include tabular ilmentite and acicular rutile crystals less than 0.5 mm in diameter and intergrown with amphibole. Pyrrhotite is present as fine (<0.5-mm) subhedral grains locally containing chalcopyrite rims or inclusions.
Geochemistry of the Ntaka Intrusion and Mafic-Ultramafic Country Rocks
Major and trace element compositions of the Ntaka ultramafic rocks and spatially related mafic to ultramafic country rocks are listed in Table 3 and depicted in the series of geochemical plots shown in Figures 6 through 8. Major element oxides have been recalculated to 100 percent on a volatile-free basis for plotting purposes.
The ultramafic rocks of the Ntaka intrusion bear geochemical similarities to other high MgO, including komatiitic, nickel-bearing rocks occurring in Proterozoic continental margin terranes and Archean greenstone belts elsewhere in the world, including the Thompson (Peredery, 1979), Raglan (Barnes et al., 1982; Lesher, 2007) and Namew Lake (Menard et al., 1996) deposits in Canada, the Kabanga deposit in Tanzania (Evans et al., 2000), and various komatiite-hosted deposits in Western Australia (Lesher, 1989). The Ntaka peridotites and pyroxenites exhibit geochemical signatures characterized by high MgO contents (typically 20−39 wt %), low SiO2 and CaO contents, low incompatible element contents (both LILE and REE), very low Pt-Pd contents, moderate Ni/Cu ratios, and relatively flat chondrite-normalized REE patterns.
Major element oxides
The major element compositions are illustrated by the series of variation diagrams shown in Figure 6 with major oxides plotted against MgO. The peridotite and pyroxenite samples each plot in distinct, nonoverlapping fields with respect to MgO content, which ranges from 30 to 39 wt percent in the peridotites and from 15 to 28 wt percent in the pyroxenites. The Al2O3, SiO2, CaO, Na2O, and TiO2 are all negatively correlated with MgO and increase in abundance with decreasing MgO from the peridotitic samples through to the pyroxenitic samples, which reflects the higher percentage of olivine relative to pyroxene and amphibole in the peridotites. In contrast, plots of K2O, MnO, and Fe2O3 show more scattered distributions. One pyroxenite sample (14008) has an anomalously high K2O content (˜2 wt %) and SiO2 content (57.1 wt %), as well as elevated Rb, Ba, Nb, Pb, Th, U, and Zr, which likely reflects alteration.
The major element oxide compositions of the gabbroic country rocks are similar to and overlap with those of the amphibolitic country rocks, suggesting that these rock types could be genetically related, representing intrusive and extrusive equivalents. When compared to the Ntaka ultramafic rock samples, the amphibolitic and gabbroic country rocks have higher Al2O3, CaO, Na2O, and TiO2 contents, reflecting the higher percentage of plagioclase and amphibole relative to pyroxene in these latter units. The Al2O3, CaO, Na2O, and TiO2 plots also show continuous, nonoverlapping trends, with the exception of anomalous pyroxenite sample 14008, progressing from the Ntaka ultramafic rocks through to the mafic country rocks.
Chalcophile elements—Ni, Cu, Co, Pt, and Pd
Nickel and cobalt contents decrease with decreasing MgO from the peridotites through the pyroxenites to the amphibolites and/or gabbros, and generally plot in three distinct fields collectively defining continuous trends for both Ni and Co (Fig. 7A, B). Several of the pyroxenite samples have elevated Ni and Co contents, which can be correlated with the slightly higher sulfide content of these samples. Pyroxenite sample 14006 was excluded from the plots in Figure 7 because it contains 1 to 2 percent disseminated to blebby visible sulfides and measurements indicated anomalously high Ni, Cu, and Co contents.
Copper shows a scattered distribution when plotted against MgO and Ni (Fig. 7C, D). However, each of the mafic and ultramafic rock types exhibit distinct Ni/Cu ratios that increase systematically from the mafic rocks through to the most MgO rich ultramafic rocks. Peridotitic samples contain the highest Ni/Cu ratios (avg = 13.6, n = 4) followed by the pyroxenites (avg = 5.4, n = 13) and the amphibolites/gabbros (avg = 2.1, n = 9). Copper, Pt, and Pd contents are low for all of the mafic to ultramafic rock types, with Cu typically <500 ppm and Pt + Pd <50 ppb.
Rare earth elements (REEs)
The Ntaka ultramafic rocks have rare earth element (REE) contents comparable to rocks of komatiitic affinity (Table 4), with total REEs averaging <30 ppm and La/LuCN ratios ranging from 1.3 to 3.6. In comparison, the mafic to ultramafic country rocks contain higher abundances of REEs (54−138 ppm) and are more LREE enriched with La/LuCN ratios ranging from 2.8 to 9.0.
A series of chondrite-normalized REE plots for the various Ntaka rock types are shown in Figure 8. Figure 8A shows the average range for each rock type, excluding anomalous samples possibly affected by either alteration or contamination. All samples, including anomalous ones, are plotted by rock type in Figure 8B through E. The Ntaka peridotite and the pyroxenite samples display similar and overlapping flat profiles, exhibiting only slight LREE enrichment and lacking any pronounced Eu anomalies. In comparison, the mafic to ultramafic country rocks contain higher abundances of REEs and display steeper, more LREE enriched patterns but also lack Eu anomalies. One peridotite sample, 14004 (Fig. 8B) is anomalous, possessing very low total REE abundances (ΣREE = 7.94 ppm) but displaying a higher degree of LREE enrichment compared to the other peridotite samples. The pyroxenite field shown in Figure 8A was defined by the unaltered samples shown in Figure 8C. All anomalous pyroxenite samples are plotted separately in Figure 8D and include three samples (14002, 14008, and 14010) with steep LREE enriched patterns and two samples (10684 and 14001) possessing unusual concave and extremely low total REE abundances (ΣREE = 6−7) that are similar to peridotite sample 14004. The three LREE-enriched pyroxenite samples are interpreted to be altered, in particular sample 14008 that contains elevated SiO2 and K2O contents and sample 14002 that contains 5 to 10 percent phlogopite as estimated in hand sample. The two pyroxenite samples with concave REE patterns are interpreted to be contaminated and are discussed in more detail in the next section. The amphibolite and gabbro country-rock samples shown in Figure 8E have very similar REE compositions and plot in a tight grouping, further supporting a genetic relationship between these rock types.
Evidence of contamination
Inclusions of siliceous, graphitic metasedimentary rocks, as well as occurrences of disseminated to clotty graphite, are observed in several locations within the Ntaka intrusion (e.g., J and G zones), suggesting at least localized assimilation and contamination. This interpretation is supported by the REE data shown in Figure 9. Samples 10683, 10684, and 10685 are from drill hole NAD08-073 in the J zone. Sample 10683 is a normal pyroxenite located 1.7 m uphole from anomalous pyroxenite sample 10684 that displays the concave depleted REE profile. Sample 10685 is a siliceous, graphitic metasedimentary rock inclusion located 6.3 m downhole of sample 10684. Sample 14001 is a pyroxenite sample from drill hole NAD08-077 in the J zone and displays the same unusual concave REE pattern as sample 10684. The REE pattern exhibited by sample 10684 is transitional between the patterns exhibited by the uphole pyroxenite and the downhole metasedimentary rock inclusion, thus indicating that its REE contents have been modified by contamination resulting from partial assimilation of metasedimentary material into the intrusion. The REE pattern of sample 14001 is virtually identical to that of 10684 and can be similarly attributed to contamination.
The metasedimentary rock inclusion sample also appears to be significantly altered. It exhibits extremely low total REE contents (ΣREE = 2.9), an atypical REE pattern, very high silica (83.1 wt %) and LOI values (12.5 wt %), and anomalously high Ni (555 ppm), Cu (203 ppm), Cr (120 ppm), U (7.3 ppm), and Mo (69 ppm) values. This geochemistry is compatible with a siliceous restite resulting from partial melting and interaction with the ultramafic intrusion. A similar origin is postulated by Evans et al (2000) for quartz xenoliths occurring in gabbro-norite from the Kabanga intrusion.
Nickel Sulfide Mineralization
The Ntaka intrusion appears to have been sulfur saturated upon emplacement based on the widespread sulfide mineralization intersected in drilling at Ntaka Hill. Most exploration drill holes contain occurrences of disseminated to blebby sulfide minerals, including those holes that did not specifically target EM anomalies. Six near-surface nickel sulfide deposits have so far been discovered and delineated including the J, M, H, NAD013, L, and G zones. A number of additional nickel sulfide zones have been intersected and continue to be intersected but have either not been fully delineated or are less continuous in nature. Nickel tenors, defined here as the percentage nickel in 100 percent sulfides, are highly variable from one sulfide zone to another, ranging from 2 to 17 percent. Nickel tenors quoted in proceeding sections are based either on direct assays of massive sulfides or on assays of net-textured to semimassive sulfides extrapolated to 100 percent sulfides.
Three main types of nickel sulfide mineralization have been identified in the Ntaka intrusion, including magmatic sulfides, remobilized sulfide veins, and magmatic sulfide-graphite contamination zones (Fig. 10). The main sulfide mineral phases are pyrrhotite, pentlandite, pyrite, chalcopyrite, and violarite. Table 5 lists the average geochemical compositions of the sulfide minerals for each sulfide zone with the exception of the G zone.
Pentlandite typically occurs as free grains and is the main nickel-bearing sulfide mineral. Nickel contents in pentlandite are, on average, between 35.5 and 37 percent Ni but are higher in the remobilized veins than in the magmatic mineralization. Nickel contents in pyrrhotite are variable from zone to zone, ranging from a low of 0.47 percent in the J zone to 1.26 percent in the NAD013 zone. Copper contents in chalcopyrite lie within a restricted range averaging 34.3 to 34.7 percent. Examples of the various sulfide textures are shown by the series of scanning electron microscope images in Figure 11.
Magmatic sulfide mineralization
Ultramafic rock-hosted magmatic mineralization consists of disseminated, blebby, net-textured, and locally semimassive to massive sulfides hosted within pyroxenites and more rarely peridotites. Zones J and M, located on the eastern and northwestern edges of the intrusion, respectively, are typical of this type of mineralization. Both of these magmatic zones are characterized by high nickel tenors ranging from 6 to 13 percent in massive sulfides, Ni/Cu ratios ranging from 2 to 10, and low PGE contents averaging <1 g/t combined Pt + Pd.
Rocks of the J zone crop out at the historic malachite pits, which represent the oxidized updip extent of the zone. The zone strikes north-south, dips 15° to the west, and plunges shallowly to the south. Drilling has delineated the zone over a strike length of 325 m (Fig. 12). Mineralization consists of a primary sulfide zone occurring below the base of weathering, a transition zone containing partially weathered and oxidized sulfide minerals, and an overlying Cu-Ni−enriched, weathered oxide zone. This oxide zone includes the historic malachite showing, and large blocks of gossan can be seen in the area of the pits. The primary sulfide minerals occur at a depth of 20 to 30 m and extend downdip for 50 to 75 m. The J zone is one of the largest known sulfide lenses (˜1.2 Mt @ 1.245 Ni and 0.20% Cu) within the Ntaka intrusion, although a significant part of the mineralization has been oxidized and likely lost to erosion at the northern upplunge and eastern updip extents.
Primary and transition sulfide mineralization consists of disseminated to net-textured and massive sulfide minerals hosted within a medium-grained, olivine-bearing orthopyroxenite. Primary sulfide minerals are pyrrhotite and pentlandite, with minor chalcopyrite and trace pyrite. Within the transition zone, pyrrhotite and pentlandite are partially to extensively altered to pyrite and violarite, respectively. In the oxide zone, sulfide minerals are rare and nickeliferous mineral phases include malachite, chalcocite, and nickel-bearing chlorite.
A small zone of massive sulfide mineralization, as much as 4 m in thickness, is preserved as a remnant in several holes immediately below or near the base of weathering. The massive sulfides are transitional downhole to net-textured sulfides, followed by disseminated sulfides, which suggests the J zone is overturned. Pentlandite in massive sulfides forms coarse-grained eyes or patches that are as large as 5 cm in diameter and the mineralization typically assays 6 to 10 percent Ni and 0.2 to 1.0 percent Cu.
In the central core of the J zone, disseminated to net-textured sulfide mineralization attains thicknesses of as much as 25 m and typically assays 1.0 to 3.0 percent Ni and 0.2 to 0.5 percent Cu, with Ni/Cu ratios ranging from 4 to 10. Nickel tenors of the net-textured mineralization typically range from 6 to 9 percent Ni and are similar to the massive sulfides in the J zone. A halo of weaker disseminated pyrrhotite-rich mineralization grades <0.2 to 0.5 percent Ni. Pentlandite occurs as blebs and along the grain boundaries of pyrrhotite. Together, these two sulfide minerals form a net texture surrounding orthopyroxene grains. Chalcopyrite occurs as small grains associated with the pyrrhotite and pentlandite.
A significant part of the sulfide zone appears to have been lost to weathering and now exists as a copper-rich oxide cap. The current mineral resource estimate includes only the sulfide and transition mineralization and excludes a large volume of copper oxide mineralization presumably derived from the weathering of former sulfide mineralization. In addition, a part of the massive sulfide zone was also likely lost to both weathering and erosion because it has only been intersected in a few drill holes, and blocks of gossanous material are observed in the area of the malachite pits.
The M zone is located approximately 1.5 km west of the J zone on the northwestern edge of the Ntaka intrusion. The zone trends 015°, dips 70° to the southeast, and plunges moderately to the south. Mineralization can be traced along a strike length of 225 m and subcrops near the northern upplunge extent of the zone. The zone ranges in thickness from 1.5 to 10 m, averaging 7.5 m (Fig. 13).
Mineralization in the M zone consists of medium- to coarse-grained, disseminated, blebby, net-textured, and semimassive sulfides hosted in altered orthopyroxenites, clinopyroxenites, and peridotite. The sulfide minerals have undergone partial remobilization forming localized stringer mineralization. Assays range from <1 to 8 percent Ni but typically average 2 percent Ni through the central part of the zone. Nickel tenors of the net-textured to semimassive mineralization are higher than at the J zone and typically range from 10 to 13 percent Ni.
The main sulfide minerals are pyrrhotite, pentlandite, and chalcopyrite, with minor pyrite and violarite. Pentlandite occurs along grain boundaries of pyrrhotite and also as coarser masses interstitial to silicate minerals. Chalcopyrite is slightly more abundant in the M zone compared to the J zone and this is reflected in lower Ni/Cu ratios ranging from 2 to 9, but averaging ˜5.
Remobilized massive sulfide veins
The term remobilized is used here as a local term to describe a form of mineralization consisting of massive sulfide veins and stringers that are typically displaced from, or are in sharp contact with, ultramafic rocks and magmatic mineralization in the intrusion. The veins and stringers are commonly in contact with intervals of country rock occurring within or near the margins of the intrusion.
Three zones located on the western edge of the intrusion, NAD013, H, and L, are characterized as remobilized mineralization comprised of vein systems that consist of single to multiple veins and stringers of coarse-grained massive sulfides. Individual veins pinch and swell from 10 cm to 4 m and are hosted within pyroxenite or along the contacts between ultramafic rocks and intervals of amphibolite or felsic gneiss. Where several veins occur, the vein system can attain thicknesses of as much as 8 m. Local zones of disseminated to net-textured magmatic sulfide mineralization carrying lower, but significant Ni-Cu grades, locally occur in hanging wall and footwall positions for tens of meters away from the vein systems. Magmatic mineralization has also been observed both up- and downplunge from veins, specifically in the H and L zones.
Massive sulfide vein mineralization consists of coarse-grained pyrrhotite, pentlandite, pyrite, and chalcopyrite, with pentlandite occurring as coarse eyes and aggregates ranging from 1 to 3 cm in diameter. The remobilized sulfide zones typically contain 5 to 15 percent pyrite that occurs (1) as intergrowths with pentlandite; ( 2) as coarse grains commonly containing a very fine grained, complex intergrowth of pyrite-chalcopyrite; and (3) in complex vein structures with pentlandite. Nickel grades of the remobilized sulfide mineralization are exceptionally high, with massive sulfide samples returning values of 10 to 17 percent Ni and 1 to 5 percent Cu. Despite the high Cu grades, PGE contents are low (<1 ppm Pt + Pd).
The veins are interpreted to have been initially derived from magmatic mineralization in the intrusion based on the proximity to up- and downplunge magmatic mineralization observed in the H and L zones. However, the high percentage of pyrite indicates fluids likely played a role in the formation and alteration of these zones (A. Naldrett, writ. commun., 2010) and is further supported by some of the observed pyrite textures.
The NAD013 zone is located between the H and L zones and is named for the discovery drill hole completed by Goldstream Mining in 2006. The zone strikes ˜155° and dips ˜44° to the southwest. It has been traced along strike for 225 m and exhibits a minor kink fold about midway along its defined strike. The vein system extends downdip for 50 to 100 m and varies from <1 to 7 m in thickness, averaging 2.75 m. Mineralization has been intersected at shallow depths of less than 50 m in the north down to a depth of 140 m at the downplunge extent to the south.
Mineralization consists primarily of narrow, remobilized, massive sulfide veins with associated sulfide stringers and disseminations. Veins consist of coarse-grained pentlandite, pyrrhotite, pyrite, and chalcopyrite, which return exceptionally high grades of as much as 17 percent Ni and 5 percent Cu. Magnetite is observed to fill the partings within pentlandite grains and also occurs along pyrrhotite grain boundaries. Chalcopyrite occurs as a complex eutectic intergrowth with pyrite, as well as irregular and discontinuous veins. Narrow, centimetre-scale massive chalcopyrite veins represent the copper-rich terminations of the vein system. In a footwall position, approximately 10 m below the remobilized vein system, a variably developed zone of lower grade disseminated magmatic mineralization occurs and averages close to 1 percent Ni + Cu.
The H zone is the southernmost of the three remobilized sulfide zones and is located 300 m southeast of the NAD013 zone. The zone strikes 100°, dips 35° to the southwest, and plunges to the southeast. Rocks in the zone crop out along it northern updip edge, in a small creek bed, as oxidized gossan and malachite mineralization. The vein system has been traced along a strike length of 150 m and a dip extent of 100 m.
Mineralization consists of remobilized, massive sulfide veins that pinch and swell from <1 m to several meters in thickness (Fig. 14). The vein system varies from single to multiple closely spaced veins and stringers, the latter occurring over widths of as much as 8 m, averaging 4.8 m. The sulfide mineralogy of the veins is similar to that of the NAD013 zone and consists of coarse-grained, intergrown pentlandite, pyrite, chalcopyrite-pyrite intergrowths, and pyrrhotite. Massive sulfide veins in the H zone return exceptionally high nickel grades of 15 to 17 percent. Up- and downhole of the vein systems, in both the hanging wall and footwall, there are several zones of disseminated to net-textured magmatic sulfide minerals that carry lower grades ranging from 0.5 to 1 percent Ni + Cu.
The L zone is located 500 m northwest of the NAD013. The zone strikes approximately north-south, dips 68° to the west, and plunges moderately to the south, subcropping at its northern extent. It has been delineated along a strike length of ˜150 m and dip extent of 75 m.
Unlike the H and NAD013 zones, the L zone consists of both intrusion-hosted magmatic sulfide minerals and massive sulfide veins that are remobilized as two parallel zones along an apparent lithologic contact downdip from the magmatic mineralization. The magmatic mineralization consists of disseminated to net-textured sulfides dominated by pyrrhotite, with lesser pentlandite and chalcopyrite, and is hosted in altered pyroxenite. Where the ultramafic rocks are in contact with a narrow mixed unit of amphibolitic to felsic gneiss, high-grade massive sulfide veins composed of coarse-grained pyrrhotite, pentlandite, and pyrite, with lesser chalcopyrite and trace violarite, have been remobilized along both gneissic contacts as two subparallel zones. Remobilization appears to be stronger along the lower contact. The whole zone varies from 3 to 12 m in true thickness and individual massive sulfide veins within the zone are as much as 3 m thick.
Massive sulfide veins are similar to those of the NAD013 and H zones. Pentlandite grains form a cauliflower-like texture occurring as fine to coarse chains surrounding the edges of large pyrrhotite grains. Pentlandite also occurs as flames oriented along the crystallographic planes in pyrrhotite and extending out from thin magnetite veins that cut through pyrrhotite grains. Chalcopyrite occurs as coarse masses interstitial to other sulfide minerals, as thin veinlets within pentlandite or pyrrhotite, and within the complex chalcopyritepyrite intergrowth texture. Magnetite occurs as euhedral grains, as much as 1 mm in diameter, as well as fracture fillings within pentlandite and pyrrhotite. Grades in the massive sulfide veins at the L zone are lower than in the other remobilized zones, usually ranging from 10 to 14 percent Ni but locally reaching 16 percent Ni.
Magmatic sulfide-graphite “contamination zones”
Within the Ntaka intrusion, low-grade nickel sulfide zones intermixed with graphite and rafts of metasedimentary rocks have been intersected in drilling. These zones are interpreted to have resulted from the contamination of the intrusion through assimilation of the graphitic metasedimentary rocks. This process has resulted in the segregation and accumulation of extensive sulfide zones of which the G zone represents the most significant delineated to date. Inclusions of graphitic metasediments are also observed locally at the J zone.
The G zone is located 200 m west of the J zone and is deeper than the other to sulfide zones, occurring 75 to 130 m below surface. The zone strikes 150°, dips 15° to 20° to the southwest, plunges shallowly to the southeast, and has been traced along a strike length of 400 m. Mineralization varies from 2 to 22 m, but averages 9 m in thickness. The zone consists of disseminated, net-textured, and, locally, massive sulfide minerals intermixed with graphite lenses and disseminations hosted in pyroxenite and altered pyroxenite. The sulfide mineralization is dominated by pyrrhotite, with lesser amounts of finer grained pentlandite and chalcopyrite that occur as inclusions and interstitial grains, as well as narrow 0.5- to 2.0-mm-wide fracture fillings cutting pyrrhotite. Assays commonly range from 0.4 to 1 percent Ni and 0.15 to 0.3 percent Cu. Semimassive and near-massive sulfides typically grade between 2 and 3 percent Ni. This equates to nickel tenors that range from ˜2.8 to 4 percent Ni, which is significantly lower than those observed in either the magmatic or remobilized zones described above.
Discussion and Deposit Model
The Ntaka Hill nickel sulfide deposits are hosted in a peridotitic to pyroxenite ultramafic body of interpreted, but not certain Proterozoic age, located in high-grade metamorphic supracrustal rocks of the Mozambique belt in southeastern Tanzania. The geochemistry of the Ntaka intrusion is characterized by high MgO contents (≤38.6 wt %), low CaO, Al2O3, Cu, PGE, and incompatible element contents, and relatively flat chondrite-normalized REE profiles lacking Eu anomalies. The observed geochemistry is compatible with derivation from a fairly primitive, hot, MgO-rich magma, where formation of the rocks was controlled by the crystallization and accumulation of olivine and pyroxene. The geologic setting at Ntaka Hill has similarities to a number of other significant nickel sulfide deposits located at Proterozoic plate margins elsewhere in the world.
Key factors in the formation of magmatic nickel sulfide deposits are the metal contents of the magma, the ability of the magma to reach sulfur saturation, and the mechanisms of sulfide concentration. The physical setting of such deposits can be either intrusive or extrusive. At Ntaka, the high nickel grades of the magmatic and remobilized mineralization indicate that nickel was abundant in the magmatic system and the widespread presence of sulfides within the intrusion further indicates that the magma was readily able to reach sulfur saturation. However, the exact processes that led to the formation of the deposits and the physical setting of the deposits are, at present, poorly understood.
The two most likely geologic models for the Ntaka Hill area are either a high MgO ultramafic intrusion with one or more basal stratiform sulfide zones that were subsequently deformed, boudinaged, folded, and partially eroded, or a package of komatiitic ultramafic flows containing separate sulfide lenses that were subsequently deformed, folded, and partially eroded. Unfortunately, textural evidence to support one setting over the other is lacking due to the extensive deformation, recrystallization, and paucity of primary lithologic contacts. The sulfide zones delineated thus far range in size from ˜200,000 t to just over 1,000,000 t. This is very similar to the size of massive to net-textured sulfide ore lenses in komatiitehosted deposits, such as those of the Raglan area (Lesher, 2007) and is smaller than many of the larger intrusive deposits, such as Voisey’s Bay (Inco, Annual Report, 2003) and the Thompson Nickel belt (Layton-Matthews et al., 2007). However, larger, as yet undiscovered, sulfide zones may be present at Ntaka based on observations at the J zone, where a significant part of the mineralization is thought to have been lost to weathering and erosion. It was hoped that the geochemistry of the gabbros, amphibole gneisses, and amphibolites, which are intimately associated with the Ntaka intrusion and are intersected in many of the drill holes, might help resolve the intrusive versus extrusive debate. The gabbroic and amphibolitic rocks plot in separate but continuous fields relative to the Ntaka ultramafic rocks on several major element variation diagrams. However, in REE plots, these rocks are distinctly more LREE enriched and, therefore, could not have been derived from the same magma source.
Determining the precise geologic setting is difficult given the high degree of secondary modification that has taken place. Isotopic dating would be beneficial in defining the age of the Ntaka ultramafic rocks and the relationship to Proterozoic magmatic events. Detailed geochemical sampling to produce chemical profiles may help determine the difference between komatiitic flows and igneous cyclic units (A. Naldrett, writ. commun., 2010).
Based on the existing data, the Ntaka Hill nickel sulfide deposits are interpreted to have formed in an intrusive environment, mainly as basal stratiform zones. The intrusion also contains a number of internal disseminated sulfide zones that have not yet been fully delineated by drilling. The setting closely resembles that of the nickel sulfide deposits of the Thompson Nickel belt in Canada, as described by Bleeker (1990a, b), Burnham et al. (2009), Layton-Matthews et al. (2007), Peredery (1979), and other workers. The Thompson Nickel belt is located at a collisional plate margin and forms the boundary between the Archean Superior province to the east and the Proterozoic Churchill province to the west. The lithologic units within the belt represent an Early Proterozoic continental margin of the Superior craton that was reworked during the 1.9 to 1.7 Ga Hudsonian orogeny. The belt is comprised of reworked Archean gneisses that are unconformably overlain by the polyphase deformed Proterozoic Ospwagan Group supracrustal rocks. The ultramafic intrusions hosting the nickel deposits of the belt formed from komatiitic magmas (Burnham et al., 2009) and intruded the Archean basement gneisses and overlying Ospwagan Group metasedimentary and metavolcanic rocks. Many of the intrusions are interpreted to have undergone only limited amounts of differentiation and, therefore, represent conduits that were part of a dynamic, open magmatic system (Burnham et al., 2009). The Thompson Nickel belt deposits are thought to have formed where the ultramafic sills intruded the sulfide-facies iron formations of the Pipe Formation and assimilated sedimentary sulfur, resulting in the segregation and accumulation of magmatic nickel sulfides. Subsequent high-grade metamorphism and polyphase deformation has, in some cases, remobilized the massive nickel sulfide mineralization away from the boudinaged ultramafic source rocks into the host metasedimentary rocks. Table 6 compares various attributes of Thompson Nickel belt geology to that of the Ntaka Hill area.
It seems likely that the Ntaka parent magma experienced some degree of contamination. The pyrrhotite-rich, nickelpoor, graphite-bearing, disseminated to massive sulfide mineralization that occurs at several locations within the intrusion (e.g., G zone) is thought to have formed by assimilation of graphitic metasedimentary rocks. Evidence of this interaction exists in the form of siliceous graphitic metasedimentary inclusions, as well as concentrations of disseminated to clotty graphite within the ultramafic rocks, and is further supported by REE geochemical data from the J zone. Assimilation of graphitic and sulfidic metasedimentary rocks would have aided sulfur saturation of the magma by contributing silica and/or sulfur to the system. More detailed geochemical studies of the sulfide mineralization and host rocks would help determine the role contamination played in the formation of the sulfide minerals. In addition, sulfur isotope data and S/Se ratios would also be useful because high δ34S contents and S/Se ratios in nickel ores have been used to support contamination by sulfur-bearing sedimentary rocks for nickel deposits such as Raglan (Lesher, 2007), Thompson (Burnham et al., 2003), and Noril’sk (Ripley et al., 2003).
Given the high degree of deformation and metamorphism, it is also likely that the Ntaka sulfide zones underwent both structural and geochemical modification. Along the western side of the intrusion, high-grade massive sulfide stringers and veins have been concentrated into south-plunging shoots. Remobilization of sulfide minerals may have occurred by plastic deformation, by fluid transport, or by a combination of both processes. The involvement of fluids in the formation of the remobilized zones (e.g., NAD013, H, and L) is supported by the pyrite contents and textures and could have resulted in nickel enrichment of these zones. The remobilized sulfide zones have higher nickel grades than the magmatic zones (e.g., J and M) and the nickel contents in pentlandite and pyrrhotite increase from the magmatic to the remobilized zones. Secondary modification involving metal transport by metamorphic fluids has been postulated by Layton-Matthews et al. (2007) for the Thompson Nickel belt deposits.
The Ntaka Hill nickel sulfide deposits represent the first significant occurrence of nickel in the Tanzanian portion of the Mozambique belt. The geologic setting is similar to that of other Proterozoic nickel deposits located elsewhere in the world at rifted continental margins, in particular those of the Thompson Nickel belt deposits of the Circum-Superior belt in Canada. The Ntaka ultramafic rocks and related sulfide zones are interpreted to represent a high MgO ultramafic intrusion with one or more basal stratiform sulfide zones that were subsequently deformed and metamorphosed. The intrusion possesses a number of the elements thought to be critical in the formation of nickel sulfide deposits including nickel-rich, high MgO, sulfur- saturated ultramafic host rocks and evidence of assimilation.
The mineralized zones discovered thus far in the Ntaka intrusion are of limited size, but potential exists for larger sulfide lenses to be present. The currently defined mineral resources at Ntaka Hill all occur at shallow depths of <150 m and a number of other sulfide occurrences have yet to be fully tested and delineated. Prospects are good to increase the overall nickel resource and additional drilling will be required to fully evaluate the potential of the Ntaka intrusion.
Regional exploration in the Nachingwea area is complicated by the complex deformation history, the abundance of graphitic metasedimentary rocks, and the paucity of outcrop. Exploration success will depend on the ability to identify prospective ultramafic intrusions and apply geophysical tools to define high-quality conductive drill targets within these bodies. The intersection of nickel sulfides 8 km to the southwest of Ntaka Hill in the Lionja intrusion and the discovery of other new regional ultramafic rock-hosted occurrences bode well for the potential to discover additional deposits in this part of the Mozambique belt. Application of the Thompson Nickel belt model on a much larger scale involves identifying supracrustal sequences within the Mozambique belt that may have formed in a continental margin setting and that contain evidence of mafic to ultramafic magmatism potentially relating to a major rifting event. The Mozambique belt of southeastern Tanzania is virtually unexplored for nickel sulfide deposits but has the potential to host a new nickel district based on the recent discoveries at Ntaka Hill. Defining pre-Pan African orogeny stratigraphy and understanding the structural implications of Pan African deformation will be important challenges to future explorers.
The information in this paper is based on several years of exploration carried out by IMX Resources (formerly Goldstream Mining NL) and Continental Nickel Limited (CNI). The authors wish to acknowledge the entire exploration teams of these two companies whose efforts are directly responsible for the exploration successes. CNI and joint venture partner, IMX Resources, are also thanked for supporting this initiative, allowing staff the time to compile and interpret project data and consenting to the publication of information contained in this paper. QEMSCAN and EPMA analyses of mineralized samples were completed by the staff of Xstrata Process Support in Sudbury, who provided skilled data analysis and professional and confidential communication and reporting. Tony Naldrett and Wolf Maier are gratefully acknowledged and thanked for their constructive editorial input which served to improve the final version of this paper.
Figures & Tables
The Challenge of Finding New Mineral Resources: Global Metallogeny, Innovative Exploration, and New Discoveries
VOLCANIC-ASSOCIATED and sedimentary-exhalative massive sulfide deposits on land account for more than one-half of the world's total past production and current reserves of zinc and lead, 7 percent of the copper, 18 percent of the silver, and a significant amount of gold and other by-product metals (Singer, 1995). A new source of these metals is now being considered for exploitation from deep-sea massive sulfide deposits. Because the oceans cover more than 70 percent of the Earth's surface, many expect the ocean floor to host a proportionately large number of these deposits. However, there have been few attempts to estimate the global mineral potential. Significant accumulations of metals from hydrothermal vents have been documented at some locations (e.g., 91.7 Mt of 2.06% Zn, 0.46% Cu, 58.5 g/t Co, 40.95 g/t Ag, and 0.51 g/t Au in the Atlantis II Deep of the Red Sea: Mustafa et al., 1984; Nawab, 1984; Guney et al., 1988). Even more metal is contained in deep-sea manganese nodules. Current estimates in the U.S. Geological Survey (USGS) mineral commodities summaries indicate a global resource of copper in deep-sea nodules of about 700 Mt. In the Pacific "high-grade" area, an estimated 34,000 Mt of nodules contain 7,500 Mt of Mn, 340 Mt of Ni, 265 Mt of Cu, and 78 Mt of Co (Morgan, 2000; Rona, 2003). A number of countries, including China, Japan, Korea, Russia, France, and Germany, are actively exploring this area.