The research focuses on an area between longitudes 12°00’E and 12°30’E and latitudes 3°30’N and 4°30’N, drained by the Nyong River and underlain predominantly by gneisses, quartzites, schists, and micaschists cut by pegmatite and quartz veins. This sector sits along the northern margin of the Congo Craton, where it meets the Central African Fold Belt, and lies near two of the region’s most important structures: the Central Cameroon Shear Zone and the Sanaga Shear Zone. Previous work in southern Cameroon had highlighted faults sub-parallel to these mega-shears, but the relationships between the structures and the regional rock units needed updating — a gap the new study set out to close using modern filtering and modeling techniques applied to legacy airborne data.

That data itself has a remarkable pedigree. The aeromagnetic survey was flown in 1970 under a cooperative agreement between the governments of Canada and Cameroon, with north-south flight lines spaced a nominal 750 meters apart and a terrain clearance of 235 meters monitored by radar altimeter. Half a century later, the dataset remains a validated baseline for the region. The team processed it using Geosoft Oasis montaj software, first transforming the raw magnetic field into a Reduced-to-the-Equator (RTE) anomaly grid, which repositions anomalies so that their sources lie directly beneath the magnetic peaks regardless of the local inclination of the Earth’s field — a critical correction at low magnetic latitudes like Cameroon’s, where the field dips at roughly minus 17 degrees.

The RTE map revealed anomaly values ranging from minus 188.09 to plus 102.24 nanotesla and subdivided the study area into three southwest-northeast trending domains. The southern domain shows heterogeneous, high-frequency anomalies; the central domain is dominated by high-intensity anomalies that terminate abruptly along their northern edge; and the northern domain contains a sub-elliptical low along the Bana-Nguinda axis, bounded further north by a very strong positive anomaly. The sharpest lateral gradients — the telltale signatures of fault zones — cluster along the Akonolinga-Obis, Koundou-Loum, and Bana-Nguinda axes. A companion residual map reinforced the regional picture: high-amplitude blocks in the south reflect the shallowing of the rigid Archean Congo Craton basement, while alternating elongated highs and lows to the north trace the highly deformed metamorphic belts of the Pan-African Mobile Belt.

To sharpen this picture, the researchers deployed a battery of derivative filters, each with a distinct mathematical strength. The horizontal gradient magnitude (HGM), with amplitudes between 0.002 and 0.072 nanotesla per meter, enhanced geological contacts and zones of intense magnetic contrast, delineating both deep-seated and shallow faults. The first vertical derivative suppressed broad regional anomalies in favor of wave-like, high-frequency responses, confirming that the deeper regional signal is composed of many discrete structural units. The analytic signal, whose amplitude peaks symmetrically directly over the edges of magnetized bodies independent of magnetization direction, refined broad anomalies into linear and quasi-circular features with peaks reaching 0.101 nanotesla per meter — the circular ones pointing to plutonic intrusions of highly magnetized gneisses into the basement around Obis, Zoulou, Akonolinga, Bana, and Nguinda.

Perhaps the most elegant technique in the workflow was multi-scale edge detection. By mathematically elevating the observation plane upward in 500-meter steps from 1,000 to 5,000 meters, the team tracked how the local maxima of the horizontal gradient migrated with altitude. Structures whose maxima stack vertically across all levels are near-vertical; those whose maxima drift laterally in a consistent down-dip direction are oblique. The analysis revealed tight, vertically superposed maxima along the Akonolinga, Loum, Bibinda, and Zoulou axes, and systematically shifted configurations between Akonolinga, Koundou, Bana, and Alata, indicating obliquely dipping fault planes. Crucially, at shallow continuation levels the map captures the jagged geometry of the obducted Pan-African metamorphic nappes, while at 4,000 to 5,000 meters the peaks realign to trace the deep boundary of the cratonic basement underneath — separating massive transcurrent faults from shallow pull-apart structures and brittle fractures associated with late orogenic collapse.

Superimposing the maxima from the HGM, analytic signal, and tilt derivative onto a single map allowed the team to compile a definitive structural map containing 51 principal lineaments. Statistical analysis shows 20 of them trend northeast-southwest, 17 east-northeast-west-southwest, and 14 east-west, with subordinate north-south and northwest-southeast orientations — confirming the northeast-southwest grain as the dominant regional signature. Most lineaments dip steeply to vertically. The authors interpret the dominant ENE-WSW and NE-SW trends as the primary expression of the continental collision and subduction of the Congo Craton beneath the Pan-African Belt, subsequently overprinted by dextral strike-slip shearing, while the subordinate directions record later brittle reactivation. The mapped fault tracks align closely with the Sanaga and Central Cameroon Shear Zones and match published sheet geology of the Akonolinga and Ayos quadrangles, lending independent confidence to the geophysical interpretation.

The quantitative heart of the study lies in two 2.75-dimensional models along northwest-southeast profiles crossing the dominant anomalies. Profile P1, 66 kilometers long, resolves four units: a schist layer from the surface to about 1.2 kilometers depth with a susceptibility of 0.026 SI, micaschists between 0.1 and 1.8 kilometers thick, gneiss at 1.2 to 2.7 kilometers, and a basal granulite basement with a higher susceptibility of 0.035 SI, plus two steep quartzite intrusions. Profile P2, spanning 82 kilometers from Alata to south of Obis, adds an unsorted metasedimentary cover and resolves a pegmatite body at 1.1 to 2.0 kilometers depth and a quartzite body at 2.0 to 3.0 kilometers. The fits between observed and calculated magnetic curves are exceptionally tight, with normalized errors between 0.081 and 0.26 percent — far below the one percent threshold generally considered a sound fit in potential-field geophysics. The modeled depths and susceptibilities also agree closely with earlier gravity-based estimates of about 4,148 meters and 0.041 SI for buried high-susceptibility bodies in the region.

Together, the results portray a crust that has endured intense, polyphase tectonic deformation: subduction, terrane accretion, continental collision during the Pan-African orogeny around 620 million years ago, and later strike-slip shearing and brittle fracturing. The metamorphic stratigraphy — schists, micaschists, gneisses, and granulites intruded by pegmatite and quartz veins emplaced along active fault zones — matches the known geology of the Yaoundé Group and Bafia Group, whose metasediments were thrust onto the Congo Craton between roughly 616 and 590 million years ago. Beyond its geodynamic significance, the work has practical weight: structurally controlled shear networks are known to focus mineralizing fluids, so high-resolution lineament maps like this one are indispensable for identifying exploration targets in tropical terrains where thick soils hide the bedrock from direct observation.

The authors are candid about the limits of the method. All potential-field interpretations suffer from inherent mathematical non-uniqueness — multiple subsurface configurations can produce the same magnetic signal — and the structural models here would benefit from validation through core petrophysical sampling, high-resolution gravity surveys, or exploratory boreholes. Still, by combining derivative filtering, which provides two-dimensional delineation of structural tracks, with 2.75D modeling, which yields quantitative depths and geometries, the study delivers an updated geophysical framework for a poorly constrained sector of southern Cameroon. It demonstrates how fifty-year-old airborne data, reprocessed with modern tools, can still rewrite the map of a continent’s deep interior — and guide the next generation of mineral hunts across Central Africa.

Subject of Research: Aeromagnetic data enhancement and 2.75D modeling to map the structural and lithological architecture of the Akonolinga-Nguinda area in Centre-East Cameroon

Article Title: Enhancement and modeling of aeromagnetic data of Akonolinga-Nguinda area, Centre-East Cameroon

Article References: Enhancement and modeling of aeromagnetic data of Akonolinga-Nguinda area, Centre-East Cameroon. (n.d.). https://doi.org/10.1007/s44288-026-00648-6

Image Credits: AI Generated

DOI: 10.1007/s44288-026-00648-6

Keywords: aeromagnetic data, Congo Craton, Pan-African Belt, Cameroon, lineament mapping, 2.75D modeling, reduced-to-the-equator, analytic signal, horizontal gradient, shear zones, structural geology, mineral exploration

The research area sits within the Igarra Schist Belt, a segment of the West African Precambrian terrain dominated by migmatites, gneisses, quartzite mica schists, a psammite-pelite sequence and hornblende granite gneiss. These rocks were forged and deformed during two major orogenic episodes: the Eburnean event roughly two billion years ago and the Pan-African orogeny around 600 million years ago. That long tectonic history left the region riddled with faults, folds and shear zones, the classic plumbing through which hot, metal-bearing fluids migrate and deposit gold in quartz veins. The terrain also hosts limestone, granite, kaolin, feldspar, quartz, lateritic iron ore and pegmatites carrying gemstones such as tourmaline, garnet and zircon, yet the structural controls on its mineral wealth have remained only partially understood.

The dataset underpinning the study came from a nationwide airborne geophysical survey flown between 2005 and 2009 by Fugro Airborne Surveys, funded by the Federal Government of Nigeria and the World Bank and archived by the Nigerian Geological Survey Agency. Flight lines ran north-south at 500 metre spacing with east-west tie lines every 2,000 metres, at an average terrain clearance of just 80 metres. The researchers processed the total magnetic intensity data in Oasis Montaj, removing the International Geomagnetic Reference Field to isolate residual anomalies, applying tie-line and micro-leveling to suppress survey noise, and gridding the measurements at 125 metre resolution using the minimum curvature method.

Because the study area lies at low magnetic latitude, with an inclination of minus 13.91 degrees and a declination of minus 2.15 degrees, raw magnetic anomalies appear asymmetric and shifted away from the rocks that cause them. To correct this distortion, the team applied a Reduction to the Equator transformation, which mathematically repositions anomalies as if measured at the magnetic equator, centering them directly over their sources. The resulting map, with residual intensity values ranging from minus 79.1 to 60.3 nanotesla, showed noticeably realigned anomaly peaks and troughs, particularly in the southeast and along mapped lithological contacts, sharpening the definition of geological boundaries that the uncorrected data blurred.

The raw total magnetic intensity map itself, spanning 32,931.4 to 33,079.9 nanotesla, already told a geological story. Magnetic highs cluster in the northwest, southwest and northeast, while prominent lows occupy the central and southeastern portions around Aiyetoro, Iretutu, Igarra, Somorika, Ojirami Oke and Akuku. When overlaid on the geological map, the lows correspond to the psammite-pelite sequence and weathered hornblende granite gneiss, both depleted in magnetite through composition and leaching. Pockets of high magnetization within the quartzite mica schist terrain suggest either unaltered basement blocks or intrusive magmatic bodies emplaced during later tectonic events, precisely the kind of magnetization contrasts that flag lithological boundaries and potential mineralization zones.

To separate deep crustal architecture from shallow detail, the team decomposed the field into regional and residual components and extended the reduced grid upward to 10 kilometres above the survey plane. The regional map revealed broad northeast-southwest trending anomalies tied to crustal thickness and deep tectonic fabric, while the residual field, spanning minus 72.7 to 68.3 nanotesla, isolated shallow lithological changes and structural discontinuities superimposed on those regional trends. First and second vertical derivative filters then acted as edge detectors: the first derivative exposed north-south trending shallow faults and fractures in the western and eastern parts of the area, and the second derivative refined those boundaries with high precision, delineating shallow fault corridors and fracture networks that define prospective mineralization zones.

The most diagnostic tool proved to be the Analytic Signal Amplitude, which is independent of both the Earth’s field direction and the magnetization of the source, making it especially robust at low latitudes. Its peaks fall directly over the edges of magnetized bodies, and in this survey its values ranged from 0.01 to 0.294 nanotesla per metre, sorting the terrain into three magnetic domains. Low amplitudes of 0.01 to 0.045 nanotesla per metre mark quartz-rich metasediments and weathered granites; moderate values of 0.045 to 0.101 reflect the granite-gneiss complex; and high amplitudes above 0.294 concentrate along the western margin and parts of the southeast, coinciding with migmatite and the psammite-pelite sequence. Critically, the high-amplitude zones align with lithological boundaries and structural intersections, the very settings where hydrothermal fluids are channeled and gold is deposited.

Lineaments digitized from the derivative and analytic signal maps and plotted in rose diagrams revealed a dominant east-northeast to west-southwest structural grain, followed by west-northwest to east-southeast trends, with minor northeast-southwest and northwest-southeast orientations. This fabric matches the Pan-African tectonic trends documented across the Nigerian Basement Complex and the Benue Trough in earlier studies, indicating that regional deformation, not local accident, dictated where mineralizing fluids could travel. The synthesis yields a hierarchical framework: crustal-scale northeast-southwest basement grain overlain by east-northeast and west-northwest fault corridors where edge-detection anomalies, derivative lineaments and mapped geology converge.

The decisive test came from the ground. Nine known artisanal gold mining sites were overlaid on the analytic signal map, and most sit directly on or adjacent to high-amplitude anomalies and major structural intersections. The northeastern workings in particular follow the contact between the schist and porphyritic granite, close to a major mapped structure. This spatial correlation validates the geophysical interpretation and confirms that gold emplacement here reflects the combined influence of favourable lithologies, including biotite granite, granite gneiss, quartzite-mica schist and migmatites, and tectonic deformation rather than lithology alone. Areas around Ojirami, Akuku, Somorika, Ogbe and the northeastern sector of the study area emerge as priority exploration targets where structural complexity, magnetic signatures and known occurrences converge.

Beyond its immediate findings, the study makes a case for aeromagnetic interpretation as a cost-efficient, non-invasive first step in exploring crystalline basement terrains where surface exposure is poor and conventional mapping struggles. The authors position their structural framework as the essential foundation for future three-dimensional magnetic inversion and magnetization vector inversion modelling, and for integration with aeroradiometric, gravity, remote sensing and geochemical datasets within a GIS-based prospectivity model. For a nation seeking to grow its solid mineral reserves, the message is that the gold-bearing architecture of the Igarra Schist Belt was already recorded in the airborne data; it simply needed the right filters to be seen.

Subject of Research: Aeromagnetic delineation of structurally controlled gold mineralization zones in the Igarra Schist Belt, southwestern Nigerian Basement Complex

Article Title: Preliminary assessment for delineation of mineralized structures and potential gold alteration zones in Igarra Schist Belt southwestern Nigerian Basement Complex

Article References: Ogbamikhumi, A., Salami, S. A., Ehinlaiye, A. O., Worenwu, W. E., Aigbadon, G. O., Ojeaga, K., Igbinigie, N. S., & Tawey, M. D. (2026). Preliminary assessment for delineation of mineralized structures and potential gold alteration zones in Igarra Schist Belt southwestern Nigerian Basement Complex. Discover Geoscience, 4(1), Article 321. https://doi.org/10.1007/s44288-026-00695-z

Image Credits: AI Generated

DOI: 10.1007/s44288-026-00695-z

Keywords: aeromagnetic data, gold mineralization, Igarra Schist Belt, Nigerian Basement Complex, analytic signal, vertical derivatives, Reduction to the Equator, structural geology, Pan-African orogeny, mineral exploration, shear zones, hydrothermal fluids

The Bou Azzer–El Graara inlier occupies a special place in the geology of North Africa. It lies within the Pan-African suture zone, the scar left behind when an ancient ocean closed during the assembly of the West African Craton more than half a billion years ago. The inlier preserves one of the most representative Neoproterozoic ophiolite successions known, a fragment of oceanic crust and mantle rocks thrust onto continental margins during arc–continent collision. These same serpentinite bodies later became the host of the famous cobalt–nickel arsenide ores that have been mined at Bou Azzer for nearly a century, making the district a globally significant source of cobalt and a natural laboratory for studying how ancient plate boundaries control the distribution of metals.

Despite decades of geological mapping, structural analysis and geophysical surveying in the region, no previous study had combined quantitative machine learning classification of magnetic edges with three-dimensional magnetic inversion inside a single multiscale structural framework. That gap is what motivated Ayoub Soulaimani of Morocco’s Directorate of Geology, Mines and Hydrocarbons, Saâd Soulaimani of the Rabat School of Mines, and Abdelhalim Miftah of Hassan First University of Settat to design a four-stage workflow built around high-resolution aeromagnetic data acquired under the Moroccan national airborne survey program. The result is a methodology that turns raw magnetic measurements into a ranked, probability-based map of where hidden structures, and potentially hidden ores, are most likely to lie.

The first stage of the workflow involved computing a suite of potential-field derivatives from the total magnetic intensity data. These included the horizontal gradient, the analytic signal, three orthogonal first-order directional gradients, the tilt derivative, the horizontal gradient of the tilt angle, and a residual map reduced to the pole. Each of these filters emphasizes different aspects of the magnetic field. The horizontal gradient and analytic signal highlight the edges of magnetized bodies, the directional gradients accentuate lineaments oriented in specific azimuths, and the tilt derivative balances the responses of shallow and deep sources so that weak signals from deeply buried contacts remain visible. Together they form a multidimensional description of where the magnetic fabric of the region changes abruptly, which in turn corresponds to lithological boundaries and faults.

Those derivative maps then served as the input features for training a feed-forward artificial neural network, optimized through the Levenberg–Marquardt backpropagation algorithm, a classical and computationally efficient training method for small to medium networks. The researchers tested two weighting schemes for the input data: one in which all derivatives contributed equally, and another in which derivatives known to be especially sensitive to lithological contacts received selective weighting. Both configurations reached an accuracy of roughly 86 percent, and the very similar area-under-the-curve values obtained in the two cases suggested to the authors that the solution is robust rather than an artifact of any particular parameterization. In other words, the network was genuinely learning the structural signature of the region, not merely fitting noise.

The output of the trained network is a set of weighted probability surfaces that quantify, pixel by pixel, the likelihood that a structural discontinuity exists at that location. When compared against existing geological maps, these surfaces reproduced all of the major structural corridors already documented in the Bou Azzer region, providing an independent validation of the approach. More intriguingly, they also revealed zones of structural complexity that had never been mapped before, places where the magnetic fabric implies intersecting fault sets and rotated blocks that surface geology alone had not resolved. It is precisely such corridors of intersecting structures that tend to channel hydrothermal fluids and localize mineralization, which makes the newly identified zones prime candidates for future exploration.

In parallel with the machine learning analysis, the team performed a three-dimensional inversion of the magnetic data over the Foum Zguid dyke zone, one of the most striking magmatic features of the Anti-Atlas. The inversion reconstructed the volume distribution of magnetic susceptibility and allowed the researchers to recover the geometry of the dyke structure down to a depth of about two kilometers. This kind of quantitative depth constraint is rare in regional structural studies and demonstrates how modern inversion algorithms, originally developed for mineral deposit scale problems, can be scaled up to interrogate the deep architecture of entire tectonic corridors.

The final synthesis combined gradient analysis, Euler deconvolution for estimating source depths, the neural network prospectivity results, and the three-dimensional susceptibility model into a single high-resolution structural interpretation. The dominant fault orientations that emerged are north–south and WNW–ESE, a pattern that carries direct geodynamic significance. The north–south set is consistent with structures related to the Pan-African collision and its later reactivation, while the WNW–ESE set aligns with the major shear zones that dissect the Anti-Atlas belt. The study thus links the fine-scale magnetic fabric of the inlier to the broad tectonic evolution of the northern margin of the West African Craton, from Neoproterozoic ocean closure through Paleozoic deformation.

The economic implications are immediate. Bou Azzer remains one of the few operating cobalt arsenide districts in the world, and cobalt has become a strategically critical metal for battery technologies. The cobalt–nickel–arsenic ores at Bou Azzer are hosted in serpentinites along fault-controlled hydrothermal veins, so the identification of previously unrecognized structural corridors directly translates into new underexplored target zones for cobalt, copper, silver and gold. Rather than drilling blindly, exploration geologists can now prioritize the areas where the neural network’s probability surfaces intersect favorable host rocks and where Euler deconvolution indicates structures extending to mineable depths.

Beyond Morocco, the study offers a template that other exploration geologists and structural mappers can adapt. The workflow requires nothing more exotic than a standard aeromagnetic dataset, a set of well-established potential-field filters, and a modest feed-forward neural network, yet it delivers validated structural probability maps and quantitative three-dimensional models in a single coherent framework. As airborne magnetic surveys become cheaper and machine learning tools become more accessible, the approach demonstrated in the Central Anti-Atlas suggests that many of the world’s poorly mapped Precambrian terranes could be re-examined at high resolution, revealing both the deep tectonic history recorded in their magnetic fabrics and the mineral deposits that history has left behind.

Subject of Research: Machine learning analysis of aeromagnetic data for structural mapping and mineral prospectivity in the Moroccan Anti-Atlas

Article Title: Structural and tectonic analysis of the Moroccan Central anti-atlas using predictive targeting with machine learning and inversion of aeromagnetic data: geodynamic and mining implications

Article References: Soulaimani, A., Soulaimani, S., & Miftah, A. (2026). Structural and tectonic analysis of the Moroccan Central anti-atlas using predictive targeting with machine learning and inversion of aeromagnetic data: geodynamic and mining implications. Earth Science Informatics, 19(11), Article 203. https://doi.org/10.1007/s12145-026-02257-w

Image Credits: AI Generated

DOI: 10.1007/s12145-026-02257-w

Keywords: aeromagnetic data, machine learning, artificial neural network, Bou Azzer, Anti-Atlas, Morocco, cobalt mineralization, magnetic inversion, structural geology, Pan-African suture, mineral prospectivity, geophysics

The Bornu Basin is no ordinary piece of real estate. It is the Nigerian portion of the vast intracontinental Chad Basin, which touches five countries: Cameroon, the Central African Republic, Chad, Niger and Nigeria. Bordered to the south by the Benue Trough and to the west by the Central Nigerian Basement Complex, the basin belongs to the broader West and Central African Rift System, a network of rift-related sedimentary basins that share a common tectonic history. Renewed interest in Nigeria’s inland basins has been fuelled by recent petroleum discoveries in neighbouring basins, including the Upper Benue Trough, the Muglad Basin of Sudan and South Sudan, Chad’s Doba, Doseo and Salamat basins, and Niger’s Termit Basin, all of which evolved under similar tectonic regimes.

Earlier geochemical and heat-flow studies had already suggested that the Bornu Basin holds promise, particularly for gas generation, but a viable petroleum system requires more than mature source rocks. Explorers need to know where sediments are thick enough to have buried organic matter to the right temperatures, and where faults and fractures can act as traps and migration pathways. That is precisely where the new study makes its mark: by combining basin-wide coverage with modern interpretation techniques, the researchers produced a far more complete picture of the basin’s hidden architecture than previous, more localised efforts.

The data came from the Nigerian Geological Survey Agency, acquired by Fugro Geophysical Company, and the team applied a battery of processing steps designed to sharpen the magnetic signal. They first separated regional and residual magnetic components using Gaussian filtering, then applied a reduction-to-pole transformation to shift anomalies directly above their source bodies. Because the study area sits at low magnetic latitude, where that transformation is notoriously unstable, the researchers also computed the analytic signal, an enhancement that works regardless of the direction of magnetisation. Total horizontal derivative maps, upward-continued to 500 metres and 2 kilometres, highlighted lithologic contacts and basin boundaries, while tilt-derivative analysis and CET Grid Analysis extracted linear structures representing faults and fractures.

For the crucial task of estimating how deep the buried basement lies, the team turned to three-dimensional Located Euler Deconvolution, a technique prized for its accurate source positioning and its independence from assumptions about magnetisation direction. The resulting depth-to-basement map spans values from about 5,800 metres below the reference datum to 220 metres above it, with the deepest sedimentary accumulations concentrated in the central and eastern portions of the basin, around Baga, Monguno, Maiduguri, Gashua, Geidam, Potiskum, Azare and east of Damaturu. The western part of the study area, by contrast, is dominated by shallow or exposed basement with less than two kilometres of sedimentary cover.

The basement topography reconstructed from the Euler solutions turned out to be strikingly undulating, especially in the eastern half of the basin. Six sub-basins emerged from the analysis, separated by igneous and volcanic highs. Two of them dominate the picture. The largest, located in the northeastern part of the basin, trends northeast to southwest, stretches roughly 250 kilometres in length and 95 kilometres in width, and reaches thicknesses of about six kilometres in places. The second major sub-basin sits in the central region, extends about 140 kilometres and exceeds three kilometres in thickness. The remaining four sub-basins are shallower, with sedimentary fills of less than 2.5 kilometres.

The extracted lineaments, ranging from about 10 to more than 150 kilometres in length, weave an interconnected structural network across both the exposed basement and the sediment-covered basin. Their dominant orientation is northeast to southwest, particularly east-northeast to west-southwest, with subordinate northwest to southeast and north-northeast to south-southwest trends. This structural grain matches the regional tectonic fabric documented in earlier studies of the Bornu Basin and the nearby Upper Benue Trough, and is interpreted as reflecting reactivated Pan-African basement fabrics and Cretaceous rifting. In petroleum terms, such fault and fracture systems matter enormously: they can enhance reservoir permeability, provide migration pathways for hydrocarbons, and create structural traps where faults are sealed laterally or capped by impermeable strata.

One of the most intriguing findings concerns volcanic intrusions. The analytic signal and Euler deconvolution results suggest that volcanic bodies within the Gongola arm of the Northern Benue Trough extend northward into the Bornu Basin, corroborating earlier reports and even the penetration of such intrusions by the Kanadi exploratory well. Igneous intrusions are a double-edged sword in petroleum geology. On one hand, they can locally heat adjacent sediments and accelerate the maturation of organic matter into oil and gas; on the other, they can alter reservoir quality and disrupt fluid pathways. The authors stress that their study did not specifically assess thermal maturity, so the precise role of these intrusions in the basin’s petroleum system remains an open question that demands geochemical, thermal and basin-modelling follow-up work.

The depth estimates also invite comparison with earlier studies, which reported maximum basement depths beneath the giant northeastern sub-basin ranging from about 3 to 6.2 kilometres depending on the method used. The 5.8-kilometre maximum identified in the new work sits near the upper end of that range and is consistent with estimates from spectral and source-parameter imaging techniques, while some prior estimates in the central part of the basin appear to have overestimated depths relative to the new results. The discrepancies underscore how sensitive depth estimates are to data resolution, processing choices and spatial coverage, which is exactly why a consistent, basin-wide dataset and a single, robust interpretation method can add so much value.

Taken together, the findings point to two highly prospective sub-basins characterised by substantial sedimentary thickness, intersections of regional fault systems and proximity to volcanic intrusions, a combination that the authors argue makes them the most favourable segments of the basin for hydrocarbon occurrence. The exploration wells Kanadi and Albarka, which encountered total sedimentary thicknesses of 3,048 and 3,470 metres respectively within the Bima, Gongila, Fika and Chad Formations, provide independent support for the substantial accumulations inferred from the magnetic data. The sedimentary succession itself, spanning continental Bima sandstones, shallow-marine Gongila Formation, the organic-rich Fika Shale, deltaic Gombe deposits and the younger Kerri-Kerri and Chad Formations, includes the source-rock intervals that geochemists have long flagged as promising. What the new study adds is a roadmap: by pinpointing where thick sediments coincide with dense structural fabric, it gives explorers a rational basis for prioritising seismic acquisition and, eventually, exploratory drilling. For a nation eager to grow its petroleum reserves beyond the mature Niger Delta, the buried depocentres of the Bornu Basin may represent one of the most compelling frontier plays on the African continent, and the next chapter of the story will be written with seismic surveys, geochemistry and, ultimately, the drill bit.

Subject of Research: Aeromagnetic re-evaluation of petroleum prospectivity in the Nigerian sector of the Chad (Bornu) Basin

Article Title: Petroleum prospect re-evaluation of the Nigerian sector of the Chad (Bornu) Basin using high-resolution aeromagnetic data

Article References: Petroleum prospect re-evaluation of the Nigerian sector of the Chad (Bornu) Basin using high-resolution aeromagnetic data. (n.d.). https://doi.org/10.1007/s44288-026-00730-z

Image Credits: AI Generated

DOI: 10.1007/s44288-026-00730-z

Keywords: Bornu Basin, Chad Basin, aeromagnetic data, depth to basement, hydrocarbon prospectivity, sub-basins, Euler deconvolution, analytic signal, lineaments, volcanic intrusions, West and Central African Rift System, Nigeria

The research, led by Sharon Christa and Tushar Mane of MIT Art Design and Technology University in Pune together with Ketut Tomy Suhari of Universiti Geomatika Malaysia, focuses on the Dharwar Craton, an ancient block of continental crust spanning roughly 42,291 square kilometers in Karnataka and Andhra Pradesh. The Dharwar Craton is one of India’s most storied mineral provinces, hosting gold deposits and metallogenic belts that have been worked and debated for well over a century. Rather than relying on scarce deposit inventories to teach the model what mineralized ground looks like, the team built a framework that first teaches itself the fundamental patterns of the landscape, then applies that knowledge to the prospectivity task with only limited labels.

The technical heart of the approach is masked image modelling, a self-supervised pre-training strategy popularized in computer vision. The idea is elegantly simple: take an image, hide most of it, and ask the network to reconstruct the missing pieces. To succeed, the model must internalize the spatial grammar of the data, the way rock units adjoin one another, how magnetic anomalies trace buried structures, and where geological boundaries align with geophysical gradients. The authors pre-trained a ViT-Base encoder separately for each fold of a spatial cross-validation scheme, meaning the network developed its own understanding of the terrain before ever seeing a single prospectivity label. Crucially, the pre-training data for each fold excluded the spatial region reserved for testing, a deliberate design choice that eliminates what the authors call transductive data leakage, a subtle but serious flaw in many geospatial machine learning studies.

That leakage concern deserves emphasis, because it is where many published geospatial AI results quietly inflate their performance. When training and test data come from geographically adjacent or overlapping areas, the model can effectively memorize the answer key rather than learn transferable geological relationships. The team addressed this with fivefold spatial cross-validation and a 20-kilometer exclusion buffer around each test region, a rigorous protocol borrowed from ecological modelling literature that ensures the model is genuinely predicting unseen territory. The result is a performance estimate that reflects real-world exploration conditions, where a geologist asks the algorithm about ground it has never been shown.

Once pre-training was complete, the framework faced the second great challenge of geoscience machine learning: fusing data that come in different forms and at different scales. Geological maps exist at 1:25,000 and 1:50,000 scales, capturing lithology, structure, and alteration at complementary resolutions, while aeromagnetic survey data record the magnetic fingerprint of subsurface rocks from the air. The architecture the researchers devised is dual-branch, with one pathway processing the multi-scale geological information and another handling the aeromagnetic features. These branches communicate through cross-modal attention, a mechanism that lets each data type learn which features of the other are relevant, and through deep canonical correlation analysis, a technique that finds shared structure between the two representations. In effect, the model learns how surface geology and magnetic signatures conspire to reveal hidden mineral systems.

The performance gains are striking. Across the five spatial cross-validation folds, the full framework achieved a mean area under the precision-recall curve, or AUC-PR, of 0.7172, with a standard deviation of 0.2020. That figure towers over the baselines: a Random Forest classifier managed only 0.3672, the classical Weights of Evidence method reached 0.3619, and a Vision Transformer without self-supervised pre-training scored 0.3994. The comparison isolates the value of pre-training itself, confirming a mean improvement of roughly 0.32 AUC-PR attributable to masked image modelling. Because mineral exploration is a classic imbalanced-data problem, where prospective ground is a tiny fraction of the total landscape, the precision-recall metric is the appropriate yardstick, and the margin over the baselines is not incremental but transformative.

The authors were equally careful about uncertainty and honesty in their predictions. The framework employs a Positive-Unlabelled learning setup, which acknowledges that the map contains confirmed mineralized zones, the positives, and vast areas whose mineral status is simply unknown, rather than known to be barren. This is a more truthful description of exploration reality than binary labelled classification. On top of that, Monte Carlo dropout provides uncertainty-aware outputs, running the network repeatedly with different dropout configurations to quantify how confident each prediction should be. The model also produces three prospectivity prediction heads at different depths, but the researchers are explicit that only the 0 to 500 meter surface head is the primary validated output; the deeper heads are exploratory and should not be treated as validated predictions. That kind of methodological candor is rare and welcome.

Perhaps the most reassuring result is geological rather than statistical: the high-prospectivity zones identified by the model are spatially consistent with known metallogenic belts in the Dharwar Craton. The algorithm, trained largely on unlabeled data, independently rediscovered the regions that generations of field geologists have flagged as fertile ground. That convergence suggests the network is learning genuine geological structure rather than exploiting artifacts of the data pipeline. It also hints at the framework’s real promise, which is not confirming what is already known but highlighting analogous, previously overlooked terrain where the same structural and geophysical signatures appear without any recorded deposit.

The practical implications extend well beyond one craton in southern India. Global demand for critical minerals is accelerating, and greenfield exploration increasingly depends on extracting maximum insight from existing geophysical and geological datasets before committing to expensive drilling campaigns. Airborne magnetic surveys cover enormous territories, and geological maps exist at multiple scales for most of the planet’s exposed land, yet the labelled deposit inventories needed for supervised learning remain sparse. A framework that learns rich representations from the abundant unlabeled data and fuses heterogeneous sources through attention mechanisms offers a template for prospectivity mapping in exactly those data-poor settings. The authors have also made their work reproducible, releasing the code publicly on GitHub, and the underlying aerogeophysical magnetic dataset from the Geological Survey of India is available through the AI Kosh platform, lowering the barrier for other teams to build on the approach.

There are, of course, caveats. The high variance across cross-validation folds, reflected in that 0.2020 standard deviation, reminds us that prospectivity prediction in some regions remains much harder than in others, and the exploratory status of the deeper prediction heads means three-dimensional targeting is not yet validated. The framework is also surface-focused, so deposits concealed deep beneath cover sequences may evade detection. But the core demonstration stands: self-supervised representation learning, applied with rigorous spatial validation and honest uncertainty quantification, can nearly double the detection performance of established methods under realistic data scarcity. As exploration geologists confront ever subtler targets and shrinking discovery rates, the marriage of Vision Transformers, masked image modelling, and multi-modal geophysical fusion may prove to be one of the most consequential tools to enter the mineral exploration toolkit in decades. The Earth’s next great ore deposits, the study suggests, may be found first by machines that taught themselves to read the rocks.

Subject of Research: Self-supervised multi-modal deep learning for mineral prospectivity mapping in the Dharwar Craton, India

Article Title: Self-Supervised Multi-modal Fusion for Mineral Prospectivity Mapping Using Vision Transformers Integrating Multi-scale Geological Maps with Aeromagnetic Data

Article References: Christa, S., Suhari, K. T., & Mane, T. (2026). Self-Supervised Multi-modal Fusion for Mineral Prospectivity Mapping Using Vision Transformers Integrating Multi-scale Geological Maps with Aeromagnetic Data. Natural Resources Research. https://doi.org/10.1007/s11053-026-10771-3

Image Credits: AI Generated

DOI: 10.1007/s11053-026-10771-3

Keywords: mineral prospectivity mapping, Vision Transformer, self-supervised learning, masked image modelling, multi-modal fusion, aeromagnetic data, geological maps, Dharwar Craton, Positive-Unlabelled learning, spatial cross-validation, uncertainty quantification, deep learning