How SWIR mineralogy and detailed volcano-stratigraphic log guides exploration drilling

01

Why this matters

A lithocap is a large alteration footprint that is typically associated to epithermal high-sulfidation systems. The exploration problem at Bakan is not recognising a lithocap. More than 12 km² of advanced-argillic and argillic alteration is already known. The exploration decision is to identify which parts record a hot fluid pathway, which are shallow or barren cover, and where permeability and structure concentrated metals.

The paper compares eight mineralized centres rather than treating the lithocap as one uniform system. It links SWIR (“Terraspec”) mineral chemistry to the textures shown in the photographs and sections, then tests those relationships against grade shells, oxidation, drilling and production data. The result is a local exploration model: an assemblage indicates system position, but structure, permeability and brecciation determine where grade is concentrated.

02

What the study found

The authors combine detailed mapping, 1,628 drill holes, more than 132,000 rock assays, blast-hole sampling, SWIR, XRD, petrography, ore petrology, geophysics and mine reconciliation. SWIR was applied to approximately 103,753 core and rock samples to identify alunite, chlorite, white mica, clay minerals and iron oxides. The alteration sections therefore show an integrated geological interpretation, not a direct conversion of spectra into polygons.

The host-rock observations are central to the result. Mineralization is mainly associated with the younger dacitic volcanic sequence: tuff, lapilli tuff, tuff breccia, domes and lavas. Diatreme breccias provide additional permeable pathways.

The mineralised centres are emplaced within different units of the volcanic sequence, and these seems to correlate with the geometry, extension and grade of the mineralised bodies (Table 1)

Table 1. Body configuration, diagnostic geometry and host textures

Table 1 separates the names of the Bakan centres from the configurations used by the authors. These are not four unrelated deposit types: geometry reflects the interaction of volcanostratigraphy, host-rock permeability, diatreme emplacement, structure, hydrothermal overprinting and erosion level.

The TerraSpec results then provide mineralogical information within these geometries. K-rich alunite is common around the exposed epithermal quartz centres, whereas Na-rich and mixed Na–K alunite, Fe-rich chlorite and high-crystallinity white mica occur in deeper parts of the Main Ridge–Villa–Jalina system. Pyrophyllite, diaspore, topaz and tourmaline reinforce the hotter advanced-argillic association. In contrast, illite–smectite, kaolinite and granular quartz commonly mark cooler, shallower or covered portions. These assemblages are used as indicators of position within the hydrothermal system, not as direct thermometers or grade thresholds.

The important link is therefore between spectral mineralogy and body geometry. The same broad advanced-argillic footprint can overlie a long, structurally controlled silicic body, surround a breccia-hosted conical body, accompany a sheeted quartz-vein system, or form barren cover above concealed ledges. Table 2 records the SWIR–grade relationships that the paper documents for each configuration.

Table 2. SWIR–grade relationships and exploration use

Across the centres, the paper defines a broad 0.1 g/t Au shell, a moderate approximately 0.5 g/t Au domain and central or structurally focused zones above 1 g/t Au. Vuggy quartz therefore identifies acid leaching and residual silicification, not necessarily high grade. Grade increases where silicification overlaps permeable beds, faults, diatreme bodies, crackle breccia or later quartz veins. Oxide grades also include supergene effects and should not be treated as direct measures of hypogene grade.

03

How the alteration sections were constrained

The sections combine geological interpretation and spectral mineralogy; they are not SWIR-only products. Detailed mapping, drill-core logging, blast-hole logging and pit mapping established lithology, breccia facies, quartz ledges, structures, oxidation and alteration contacts. SWIR from approximately 103,753 core/rock samples refined mineralogical variation. XRD on 13 samples checked clay mineralogy; petrography on 50 samples and ore petrology on 22 samples checked textures and sulfides. Assays, grade shells, resource models and production reconciliation then tested whether the interpreted domains were meaningful for mining.

In practical terms, 'alteration' is the integrated geological domain; 'SWIR' is the mineralogical measurement used to refine and vector that domain. This distinction matters: a SWIR spectrum can identify Na-rich alunite or paragonitic white mica inside a mapped advanced-argillic zone, but the section still needs host rock, structure, breccia, depth and grade to explain what that mineral means.

04

Our interpretation

The four configurations are best read as four ways of organising permeability. The elongated Main Ridge–Durian bodies follow NNE-trending structures and favourable dacitic tuff or lapilli-tuff horizons; their broad silicic domains can be low to moderate grade, with grade concentrated where beds, faults and crackle breccia overlap. The circular to conical Campsite–Gunung Botak–Tapagale bodies are different: matrix-supported diatreme breccia provides a vent-related, highly permeable host, so the breccia core is more important than the narrow alteration halo. Osela is a third case, where steep sheeted bodies and later crustiform–colloform quartz veins record structural focusing; high grade can therefore occur without a large classic vuggy-quartz core. Villa and comparable covered portions of North Main Ridge and Jalina represent a fourth case: shallow granular quartz and argillic cover may conceal deeper quartz ledges.

The clay assemblages help place these textures in the hydrothermal architecture. Quartz with pyrophyllite and dickite, locally with diaspore, topaz or tourmaline, is consistent with hotter advanced-argillic conditions; Na-rich alunite, Fe-rich chlorite and paragonitic or high-temperature white mica strengthen a deeper, hotter vector. K-rich alunite is common nearer exposed silicic centres, whereas illite–smectite, kaolinite and granular quartz can mark cooler or shallow cover. These minerals do not predict grade alone: the grade signal is the overlap between the appropriate assemblage and a permeable or structurally focused body.

The porphyry implication should remain qualified. The authors interpret parts of the system as proximal to a magmatic-hydrothermal source, and drilling below Villa and North Main Ridge encountered subeconomic porphyry-style mineralization. That supports a vector toward a concealed source, not a claim that an economic porphyry deposit has been demonstrated.