Search

Recent decades have seen an exponential rise in the application of machine learning in geoscience. Fundamental differences distinguish geoscience data from most other data types. Geoscience datasets are typically multi-dimensional, and contain 1-D (drillholes), 2-D (maps or cross-sections), and 3-D volumetric and point data (models/voxels). Geoscience data quality is a product of its resolution and the precision of the methods used to acquire it. The dimensionality, resolution, and precision of each layer within a geoscience dataset translates to limitations in spatiality, scale and uncertainty of resulting interpretations. Historically, geoscience datasets were overlaid cartographically, to incorporate subjective, experience-driven knowledge, and variances in scale, and resolution. The nuances and limitations that underpin the reliability of automated interpretation are well understood by geoscientists, but are rarely appropriately transferred to data science. However, for true integration of geoscience data, such issues cannot be overlooked without consequence. To apply data analytics to complex geoscience data (e.g., hydrothermal mineral systems) effectively, methodologies must be used that characterise the system quantitatively, using collocated analyses, at a common scale. This paper provides research and exploration insights from an innovative district-wide, scale-integrated, geoscience data project, which analysed 1,590 samples from 23 mineral deposits and prospects across the Cloncurry District, Queensland, Australia. Ten different analytical techniques, including density, magnetic susceptibility, remanent magnetisation, anisotropy of magnetic susceptibility, radiometrics, conductivity, scanning electron microscopy (SEM)-based automated mineralogy, geochemistry, and short-wave infrared (SWIR) hyperspectral data with 561 columns of scale-integrated data (+2151 columns of SWIR). All data were collected on 2 cm x 2.5 cm sample cylinders; a scale at which the confidence in coupling of data from techniques can be high. These data are integrated by design, to eliminate the need to downscale coarser measurements via assumptions, inferences, inversions, and interpolations. This scale-consistent approach is critical to the quantitative characterisation of mineral systems and has numerous applications to mineral exploration, such as linking alteration paragenesis with structural controls and petrophysical zonation. [ABSTRACT FROM AUTHOR]

The analysis of drill powders (drill fines) extracted from fluid that is returned during drilling can provide a sampling media that records cm-scale changes in the geochemistry of the rock being drilled through. In addition to being an ideal sample media (~ 78% of particles are < 38 μm; cf. conventionally pulverised samples where ~ 42% of particles are < 38 μm) that is ready for analysis once dried, diamond drill fines may produce a larger sample per meter drilled than recovered by the core itself, and thus be a better representation of the rock that has been drilled through. For an HQ hole size (for rocks with specific gravity = 3100 kg/m 3 ), the mass of drill fines produced in 1 m of drilling is 12.5 kg, whereas, the mass of the 1 m in length of core for the same interval is 9.7 kg. In this contribution we compare high-spatial resolution geochemistry collected by portable X-ray fluorescence (pXRF) and mineralogy collected by portable X-ray diffraction (pXRD) from 12 m of diamond drill core to the corresponding interval of drill fines to highlight the depth fidelity (at the cm-scale) that this sample source may record. We also demonstrate that the drill fines that were previously discarded show high potential to act as a representative sample media of the lithology intersected by the drill hole and can be successfully used for analysis in real time. The integrated pXRF-pXRD data can be used to constrain lithologies and contacts between various units, hydrothermal alteration and ore types. [ABSTRACT FROM AUTHOR]

Extensive, variable, and generally thick regolith is a major challenge to mineral exploration in regolith-dominated terranes. The landscape of the Mount Isa Inlier, NW Queensland, Australia, is the product of several episodes of complex weathering, erosion and sedimentation. Weathering profiles were studied from more than 100 drill holes along three traverses crossing the Lady Annie, Mount Kelly and Anthill sulfide deposits to assess the nature, lateral extent and thickness of the supergene Cu oxide mineralization. A typical regolith profile consists, from the base up, of saprock, ferruginous saprolite (Cu-oxide zone), kaolinitic saprolite and silcrete bands (leached zone), and a pedolith of mottled, lateritic duricrust and soil zones. Oxidation of sulfides (primarily pyrite) in host rocks has created an acidic pH gradient, mobilizing Fe, Mn, Cu, Mo, Zn, Co, Ni, As, Bi, In, Sb, Be, P, U and Cd. The acidic pH is buffered by calcareous lithologies of the host rocks, particularly in the vicinity of the primary sulfides at Lady Annie and Mount Kelly resulting in formation of thick, mushroom-shaped Cu oxide halo, in which Cu may reach > 15 wt.%. The Cu values decrease to < 500 ppm in the bleached kaolinitic saprolite and silcrete. Conversely, siliceous bedrock lithologies at Anthill act as a weak buffer to the acidic fluids; thus, Fe, Cu and other mobile elements are laterally dispersed from the sulfide ore body and precipitated at different lithological boundaries where they are buffered by carbonates. Therefore, thick kaolinitic saprolite and silcrete are formed over sulfide mineralization, particularly along faults. Ferruginous duricrusts marginal to fault-controlled sulfide mineralization are anomalous in Cu, Zn, Cd, Sb and Be, and thus can be used as useful medium in geochemical exploration. Gold, Ag, Pb and As are residually concentrated in pedolith and kaolinitic saprolite over the primary sulfide mineralization. Thus, these elements can be used effectively as pathfinder elements, particularly in truncated landscape, where pedolith is eroded and the exposed kaolinitic saprolite is depleted in Cu. Ferruginous soil (850–300, 300–80 and < 80 μm size fractions), lag (> 1.6 and > 7.6 mm) and termite mounds have been applied successfully as sampling media for Cu oxide exploration in Anthill. Soil and lag over sulfide mineralization are rich in Fe, As, Pb, Ag, Bi, In, V, U and Mo. The best response over sulfide mineralization is recorded in the > 7.6 mm ferruginous lag. However, broadest signature of well-defined Au, Bi and In anomalies are better detected using termite mounds than soil and lag. Successful use of soil in exploration for Cu oxide deposits in Mount Isa requires a genetic discrimination between residual and transported origin. [ABSTRACT FROM AUTHOR]

Abstract: Geological logging by mine and exploration geologists involves the application of a classification code that requires a geologist to make a subjective decision to classify a particular rock unit (e.g. rock type, alteration intensity, mineralisation content). Where multiple geologists carry out logging this approach will most likely result in inconsistencies that are difficult to monitor or assess over time. A potential solution to this problem is to have a method to determine the nature of a given sample based on a set of fixed and quantifiable parameters. Plutonic Gold Mine (Plutonic), in Western Australia, has a growing dataset of more than 150 000 portable X-ray fluorescence (pXRF) multi-element analyses from core and underground face samples, reconciled to Au fire-assay concentration, and logging codes. This study analyses and validates the consistency with which Plutonic's logging codes have been applied from 2009 to 2013. In doing so, we have developed a schema that allows objective logging by utilising elemental data from pXRF analyses to allocate a mineralisation code. A K/V ratio is used to determine the degree of alteration, while As+12Cu concentrations are used to determine how sulphide-rich a sample is. Thresholds were defined to maintain consistency with pre-existing logged dataset. While far from replacing the geologist, this method is complementary that adds significant value to a dataset. [Copyright &y& Elsevier]

Quantification of halloysite and kaolinite in clay deposits from X-ray diffraction (XRD) commonly requires extensive sample preparation to differentiate the two phyllosilicates. When assessing hundreds of samples for mineral resource estimations, XRD analyses may become unfeasible due to time and expense. Fourier transform infrared (FTIR) analysis is a fast and cost-effective method to discriminate between kaolinite and halloysite; however, few efforts have been made to use this technique for quantified analysis of these minerals. In this study, we trained machine- and deep-learning models on XRD data to predict the abundance of kaolinite and halloysite from FTIR, chemical composition, and brightness data. The case study is from the Cloud Nine kaolinite–halloysite deposit, Noombenberry Project, Western Australia. The residual clay deposit is hosted in the saprolitic and transition zone of the weathering profile above the basement granite on the southwestern portion of the Archean Yilgarn Craton. Compared with XRD quantification, the predicted models have an R2 of 0.97 for kaolinite and 0.96 for halloysite, demonstrating an excellent fit. Based on these results, we demonstrate that our methodology provides a cost-effective alternative to XRD to quantify kaolinite and halloysite abundances. [ABSTRACT FROM AUTHOR]

[Display omitted] • Geological map layers derived using machine learning reveal geochemical anomalies. • ASTER and airborne radiometric data and a SOM generate a map layer of regolith type. • Airborne geophysic data and Random Forests construct a map layer of bedrock geology. • Regolith and bedrock map layers inform normalisation of soil geochemical data. The primary aim of mineral exploration is to focus the search space from regional scale (1000 s km2) to prospect scale (1–10 km2), identify areas that warrant further investigation, and ultimately drill a target. Mineral deposits typically display geochemical anomalism related to the conditions of their formation, or the weathering and mass transport of sediment away from a mineral deposit. However, this anomalism can be obscured in soil-sample surveys due to background variations in host bedrock chemistry and the transformation and redistribution of original minerals during surface weathering, regolith development, and supergene processes. This contribution uses machine-learnt map products to aid in the interpretation of a regional-scale soil-sample dataset. ASTER (satellite spectral) data were combined with airborne radiometric data using Self-Organising Maps (SOM) to generate a map of regolith type. A map of bedrock geology was previously derived by the authors from airborne geophysical data using Random Forests. These two products are used to classify soil samples with an objective bedrock-geology and regolith-type classification. A z-score normalisation was implemented on the soil-sample geochemical data, using two different matrices (regolith type and bedrock geology) to remove the bias of bedrock geology and regolith type from the soil sample chemistry. The signal that remains likely reflects metasomatic processes, and associated elemental anomalism, potentially linked to mineralisation. These data can then be plotted to provide maps of (pathfinder) element anomalies, thus identifying potential exploration targets. We explore this approach to normalise geochemical data in a dataset of 9,924 soil samples collected over an area of 1,028 km2, from Kerkasha, Eritrea (Nubian Shield). The Kerkasha project is an early-stage exploration project, with a limited understanding of the local geology and mineral deposit types; however, it contains many Au and Cu prospects that have received minimal exploration work. [ABSTRACT FROM AUTHOR]

Abstract: Stratigraphy, structure and host-rock chemistry are dominant controls on the location of Au in Archaean greenstone-hosted Au deposits, but the stratigraphy in such deposits is seldom obvious due to the monotonous nature of the host rocks or pervasive alteration associated with Au mineralisation. Portable, hand-held, X-ray fluorescence (pXRF) spectrometry provides a method to rapidly collect large amounts of whole-rock geochemical data that can yield new insights into both stratigraphy and Au localisation. Here we present results of pXRF analyses of samples from a representative section through Au-mineralised amphibolite-facies metabasaltic rocks at Plutonic Gold Mine, Western Australia. These data illustrate a geochemical stratigraphy in which individual lava flows can be identified on the basis of element concentrations. The most evolved basalts are at the structural base of the succession, and the least evolved at the top of the sequence, confirming previous geochemical interpretations and textural evidence that the sequence is overturned, and demonstrating for the first time that the presented section does not involve significant structural repetition. In conjunction with Au assay data, the pXRF data reveal that Au commonly occurs along basalt flow boundaries. The elemental concentration data clearly demonstrates for the first time the stratigraphic control on Au mineralisation that is not readily apparent at the macroscopic level. The methods described in this paper are readily applied, and have the potential to enhance the understanding of otherwise unclear stratigraphy and its control on mineralisation in many different types of deposits worldwide. [Copyright &y& Elsevier]

Soil samples collected at 50–70 cm depth at 8 km-spaced sample sites over 40,000 km2 of southern New Zealand were analysed for their chemical composition. Nine pathfinder elements (Ag, As, Au, Bi, Mo, Sb, Sn, Te, W), which are useful in exploration for orogenic-style mineralisation, were z-score normalised using two different matrices to account for both lithology and metamorphic grade, which vary significantly across the survey area. Interrogation of interpolated plots of the normalised data, relative to known orogenic-style deposits (Au ± Sb ± W), highlights chemical associations that varied between deposit types, and between similar deposit types mineralised under varying pressure-temperature-time conditions. Active mining for orogenic-style Au along the Hyde-Macraes Shear Zone is coincident with a regionally-significant, multi-element soil anomaly (Ag, As, Au, Mo, Sn, Te, Sb, W). The chemical associations vary along strike of the mined-portion of the shear, such that the northwest has high Ag, Au, Mo, Sn, and Te relative to the southeast which has high Sb, W in soil anomalies, hinting at chemical complexity along the shear. The chemical footprint of this deposit is readily identified in the 8 km-spaced grid, considerably expanding the chemical footprint previously inferred from hard-rock studies. Historic orogenic W mines are associated with high As, Au, Bi, W in soil anomalies, and historic Sb mining with high Bi in soil anomalies. Chemical soil anomalies are likely related to the progressive metamorphic breakdown of minerals, especially rutile and pyrite. Several multi-element soil anomalies that are coincident with known orogenic mineralisation, also occur in areas that have not been mined, and explored only at a reconnaissance level. This study demonstrates the importance of normalisation of chemical data in complex geological settings and highlights the value that can be extracted beyond only plotting raw-concentration, single-element maps, and the significant opportunity in multielement studies of soil in mineral exploration for orogenic-style mineralisation. • Soil samples from 8 km-spaced sample sites over 40,000 km2 of New Zealand • Nine orogenic Au pathfinder elements standardised for lithology/metamorphic grade • The footprint of the Macraes deposit is readily identified in the 8 km-spaced grid. • Important to normalise chemical data, especially in complex geological settings • Significant value that can be extracted beyond raw-concentration, single-element maps [ABSTRACT FROM AUTHOR]

The sequence of coastal and river sands of Pleistocene to Holocene age at the north head of the Waikato River is ~80-m thick; the sands are located in a fault-angle depression at the southern end of the Awhitu Peninsula, a 40-km-long coastal sand barrier. The sequence at Waikato North Head (WNH) consists of three main formations: Awhitu Sands, Hood Sands and Mitiwai Sands. The Waiuku Black Sand member of the Hood Sands and the Entrican Sand member of the Mitiwai Sands contain titanomagnetite-rich dune sands, which are currently mined by New Zealand Steel and constitute a giant placer deposit that has a total resource of ~90 Mt Fe. A tephra at the top of the Awhitu Sands is correlated with the ~1000 ka Waiuku (Potaka) tephra. The Mitiwai Sands overlie the 1.85 ka Taupo Pumice. Comparison of electron probe microanalyses (EPMA) of titanomagnetites in heavy mineral separates of surface and drill-hole samples indicated that the bulk of the titanomagnetite in the Waiuku Black sand and the Entrican Sand is derived from andesitic rocks of the Taranaki Volcanoes 220 km to the south. The Waiuku Black Sand may correlate with an influx of andesitic clasts and mafic minerals in the cover beds of the ~400 ka Ararata Terrace in south Taranaki. There is also a contribution of titanomagnetite and minor ilmenite from the ignimbrites of the Taupo Volcanic Zone (TVZ), particularly in the Lower Hood and Entrican sands. Automated mineralogy data collected on selected sand samples contain silicate and resistate minerals that indicate other potential sources in addition to the Taranaki andesites and TVZ ignimbrites. Minor olivine was likely derived from the late Pleistocene basalts of the South Auckland volcanic field, whereas almandine and epidote were likely sourced from metasedimentary rocks of the Murihiku and Waipapa terranes. The formation of the giant placer at WNH was a multistage process involving (1) a supply of titanomagnetite-bearing sand eroded from the late Pleistocene–Holocene andesites of the Taranaki volcanoes that was transported northwards along the coast by the prevailing longshore drift; (2) concentration of titanomagnetite by wave action on beach faces and collection in a coastal embayment on the northern side of a headland of basement rocks at Port Waikato; (3) further concentration by wind action into dune sands during interglacial low sea levels; and (4) preservation by coeval subsidence in a fault-angle depression on the down-thrown north side of the Waikato Fault. [ABSTRACT FROM AUTHOR]

Mass balance equations can be used to understand metasomatism. Here we use 2726 multi-element chemical analyses to quantify the effect of metasomatism associated with Au mineralisation at the Hamlet Gold Mine (Hamlet), Yilgarn Craton, Western Australia. Principal components analysis is first used to identify covarying elements and six groups of elements are identified: (1) relatively immobile elements, with Zr and Hf providing an appropriate reference frame for mass balance calculations; (2) elements associated with mafic rocks, such as Ca, Co, Fe, Ge, Mg, Mn, and Zn; (3) elements related to potassic alteration, Ba, Cs, K, Rb, and Tl; (4) elements related to sodic alteration, Be, Na, and Sr; (5) chalcophile elements, including Bi, weakly associated with Au; and (6) Au, Te, and S representing the main Au-mineralising event. A statistically-rigorous approach is then used to quantify element addition and depletion and provides 95% confidence limits for changes in mass. To achieve this, the basaltic host rocks are divided into four subunits based on Cr content. Sample subsets are then created to select altered rocks and least-altered protolith rocks, using discrimination diagrams. Pairings of least altered-altered samples from each basalt subunit are used to construct matrices via bootstrapping. Mass balance results from these matrices indicate similar element mobility associated with Au mineralisation across the four basalt subunits, with up to 24% mass loss. Key potassic and sodic group elements are enriched in the Au mineralised zones at the expense of elements with a mafic-association. Spatial patterns of mass change are linear and repetitive within the steeply-dipping Hamlet Shear Zone, interpreted to represent a mesh-like geometry of metasomatism within the plane of the shear zone. Localised enrichment of K, Na, and Bi are related to Au distribution while Ca depletion zones reflect the movement of reactive fluid along the shear plane. • Element covariance and mass change are quantified within an Au mineralised shear zone. • Mass change estimations are statistically robust and include 95% confidence limits. • Up to 24% mass loss is recorded during metasomatism of basalt host rocks. • Element mobility patterns highlight fluid flow pathways in three-dimensions. [ABSTRACT FROM AUTHOR]

Here we report an early Pleistocene occurrence of the New Zealand freshwater crayfish (kōura) Paranephrops White 1842 in the marginal-marine sediments of the Hautotara Formation, southeastern Wairarapa. The disarticulated parts are tentatively assigned to P. ?planifrons. This, to our knowledge, is the first fossil occurrence attributable to an extant species of Paranephrops, and only the second fossil occurrence of Paranephrops reported in the literature. This find provides a maximum constraint on the age of divergence of the two extant species of Paranephrops, and shows that post-glacial recovery of freshwater ecosystems in the North Island may have been rapid if the Northland refugium hypothesis for survival and dispersal of Paranephrops planifrons is accurate. [ABSTRACT FROM AUTHOR]

Abstract Elemental mass changes were calculated for 154 samples from four rock types around the Ann Mason porphyry-Cu-(Mo Au) deposit, Yerington district, Nevada. The purpose of these calculations was to investigate addition and depletion patterns in two-dimensional (2-D) map space. Mass changes in Ca, Mg, Na, P, Cu, Mo, As, Sb, Sr, and Pb from samples in the pre- to syn -mineralisation Yerington batholith vary systematically with lateral and vertical distance from the Ann Mason deposit. At the core of the Ann Mason deposit, Cu and Mo were strongly added (>10,000% relative to average protolith values), with restricted lateral mass addition halos around the deposit. Copper is depleted in wall rocks surrounding the deposit. Calcium and Sr were removed (up to 98% depletion) from parts of the deposit centre due to the destruction of calcic-plagioclase and other calcium-bearing minerals during potassic alteration. Calcium (up to 1570%) and Sr (up to 315%) were added outside the deposit and were concentrated around 4 km from the deposit centre where epidote alteration is most intense. Arsenic and Sb were removed from the deposit centre (up to 99% depletion) where silicate-mineral alteration assemblages dominate. Antimony shows an addition halo (ranging from 25% up to 1900%) that extends from the deposit centre 500 m laterally away, and correlates spatially with the increased abundance of pyrite that typically surrounds porphyry deposits. Epidote-altered samples of Shamrock monzonite from outside the Casting Copper skarn show addition of Ca (up to 196%), Sb (up to 1100%), and depletion of Na (up to −98%). Samples from the central part of the post-mineralisation Shamrock monzonite record lower mass changes in Mg, Ca, P, and Mo, compared to other parts of the batholith. Some samples from the northern contact of the Shamrock batholith exhibit addition of Ca (up to 554%), Sr (up to 504%), and depletion in Na (up to −98%). These samples are coincident with strong epidote alteration. Our results provide a rock-normalised geochemical map that reflects quantified element addition and depletion patterns, which is useful for interrogating geological processes at a regional scale. Although interpreting relative element addition and depletion in map space is possible using conventional immobile element ratio approaches, mass balance calculations provide the benefit of quantified results. Highlights • Addition and depletion of elements can be calculated using Grant's Isocon approach. • Results show significant chemical mobility associated with epidote alteration. • Calcium and Sr mobility are strongly correlated with epidote alteration. • Albite in sodic-calcic alteration domains does not always indicate Na addition. [ABSTRACT FROM AUTHOR]

In Latin America many mine sites are located more than 2000 m above sea level, and some over 4000 m above sea level. Portable X-ray ray fluorescence (pXRF) is becoming a routine method for collecting chemical data at different altitudes during mineral exploration campaigns. As altitude increases, air density decreases, and the physics of X-rays being transmitted through air mean that the transmission of low-energy X-rays increases and accordingly the transmission effectiveness of low-atomic weight elements (e.g. Mg, Al, and Si) also increases. Here we assess the performance of pXRF units across a range of pressures that equate to 0–5000 m above sea level by conducting well-documented tests with changing pressure to assess the use of a pXRF unit in high-altitude environments. Utilising both field test work, and test work using a hypobaric chamber in a laboratory where external conditions could be better controlled we examine how changing altitude can affect the performance of pXRF units. Units that have in-built pressure corrections perform reasonably consistently as altitude increases, whereas those that do not perform exactly as X-ray transmission modelling suggests. That is to say, the increased count rates of low-atomic weight elements (e.g. Mg, Al, and Si) means that these elements are over-reported and as a result the unit may under-report heavy elements. [ABSTRACT FROM AUTHOR]

Metasomatism occurs when fluid interacts with rock to add, or remove, its chemical constituents; these processes form mineral deposits where economic element(s) are concentrated into small volumes of rock. It can be difficult, or impossible, to visually determine original rock types for samples that are significantly altered, e.g., when rocks have experienced texturally destructive metasomatism or deformation. A typical solution using chemical data involves the separation and labelling of chemically distinct rocks using discrimination diagrams. However, such approaches can be subjective, and manual sample-by-sample consideration of large mining or exploration databases is untenable. Here we present an example workflow to facilitate relating rocks with similar origins but differing geological histories. We employ a combination of unsupervised and supervised machine learning algorithms to automate classification tasks typically undertaken manually by a geologist with domain expertise. In this study, data are first normalised and then clustered into natural groupings that represent protolith lithologies or rock-type subunits. These clusters are then used to inform a classification algorithm that assigns protolith equivalent labels to samples of altered rocks. Applied to problems involving large chemical datasets, machine learning provides objectivity, reproducibility and rapidity; useful advantages as compared to geostatistical domaining methods that involve manual determination and selection of geochemically similar regions. We utilise k-means ++ unsupervised clustering to create objective and reproducible groupings of data points, with many geochemical variables considered simultaneously. Subsequently, Random Forests supervised classification is used to label samples while accommodating interactions and/or correlations between data points. We present a case study from the Minto Cu-Au-Ag mine, Whitehorse, Yukon, Canada. Interpretation of multi-element geochemical data using the approach that we have outlined here allows reconstruction of protolith geometry and an understanding of how rock type may have influenced later partitioning of hydrothermal fluids and ductile deformation. [ABSTRACT FROM AUTHOR]

Phosphate deposits can form in a range of environments, and despite similar P contents, their mineralogy, and hence major and trace element chemistry, can vary substantially. Continental margins with high surface-water productivity are recognised as a major environment for the formation of extensive phosphate deposits. The geochemical characterisation of continental-margin-derived phosphorites from Chatham Rise and south-eastern offshore New Zealand, as well as newly discovered seamount phosphorites from Bollons Seamount (first documented here), offer insights into the different mechanisms of phosphorite formation from these two different environments. Though less studied, seamounts are increasingly recognised for their phosphorite deposits as well as Fe-Mn crusts and Mn nodules. We propose the first model for the formation of nodules at Bollons Seamount, and an updated model for Chatham Rise nodules (and south-eastern offshore NZ phosphorites) is presented. Suboxic-to-anoxic enrichments of U and V are observed in Chatham Rise and south-eastern offshore phosphorites that contrast with oxic enrichments of Mn, Ce, Co, and Cr in Bollons Seamount phosphorites. Differences in uptake of Light Rare Earth Elements (LREE) vs. Heavy Rare Earth Elements (HREE) and variations in mineralogy (e.g., presence of glauconite vs. Fe-Mn crusts), allow phosphate deposits that formed in organic-rich, upwelling continental-margin environments to be distinguished from those that formed in oxic-suboxic organic-poor seamount environments. • The first documented presence of phosphorite nodules from Bollons Seamount. • Multivariate analysis of formation environment and redox proxies. • Chatham Rise phosphorites formed in continental shelf, high productivity environment. [ABSTRACT FROM AUTHOR]