A sinkhole opened in Bangkok yesterday. It’s a reminder that in karstic ground or areas prone to subsidence, we don’t always spot change until it’s too late.
A simple idea: mount a single-channel GPR antenna under municipal vehicles (refuse lorries, street sweepers, buses). They’re already driving the streets—why not have them stream low-bandwidth data to the cloud for continuous change detection? Systems like those built by GPR.com for navigation show the hardware exists; a lightweight algorithm could flag anomalies for a closer look.
It’s not a silver bullet, but routine, rolling surveys could help catch early warning signs and prioritise inspections—quietly, affordably, and without disrupting traffic.
#gpr#georadar#groundpenetratingradar
✅ New Results from My Geostatistical Analysis for Gold (Au) Exploration 🚨
I’m excited to share the outcomes of my recent work using geostatistical techniques to analyze Au (gold) concentration in soil samples. The project focused on identifying geochemical anomalies and mapping their spatial distribution.
🔍 Key Steps:
1️⃣ Threshold Determination
Application statistical methods to establish a robust background level and define a threshold value for Au (in ppb).
2️⃣ Variogram Analysis
Construction and modeled the semivariogram using an exponential model.
3️⃣ Kriging Interpolation
Using Ordinary Kriging, generation a prediction map that reveals significant gold anomalies.
📌 Software Used: ArcMap (Geostatistical Analyst)
📈 The results confirm the importance of combining spatial statistics with geochemical data to extract valuable insights and reduce exploration risk.
#Geostatistics#GoldExploration#ArcGIS#Kriging#Variogram#Geology#MineralExploration#EconomicGeology#DataDrivenExploration
Award-winning Journalist, Content Creator and Industry Commentator. Host of On-Site Videos, Webinars & Live Events. Owner of Contentforindustry.com and correspondent for Earthmovers Magazine
Scanning above and below ground whilst creating advanced 3D models at Highways UK.
I caught up with James Tindall from Castle Surveys Ltd to talk about its real showstopper. Their fully wrapped mobile mapping unit, equipped with Leica Geosystems part of Hexagon mobile mapping and ground penetrating radar solutions.
James: “Our TRK mobile mapping system captures data at highway speeds, up to 70 miles an hour, making it perfect for topographical surveys, asset management, vegetation encroachment, and pavement analysis.
"In conjunction with the Stream UP ground penetrating radar, we’re now able to capture above and below ground utility information simultaneously.”
What’s equally important is what happens next, the processing. For that, Castle Surveys has chosen TopoDOT, as James explained:
“We wanted a solution that could give us everything we needed, with no compromise. TopoDOT lets us extract, assess and verify our data in one place. It’s the reassurance that what we hand over to clients is completely accurate.”
To find out more, I spoke with Filipe Pinto from TopoDOT, who explained how their software turns raw data into actionable insights.
“TopoDOT empowers any LiDAR user from mobile mapping to UAV and static scanning to transform complex point clouds into vector data for decision-making.
"Users can extract features like kerbs, signage, and pavement condition, calculate volumes, assess bridge clearances, and even identify potholes automatically.”
And it doesn’t stop there. Filipe added:
“Our collaboration platform means clients don’t need CAD or GIS software. They can view and query LiDAR derived data through a simple web link, making it accessible to designers, engineers and maintenance teams alike.”
It’s great to see how Leica Geosystems cutting-edge capture technology, Castle Surveys’ surveying expertise, and TopoDOT’s powerful processing tools have come together at Highways UK.
And we look forward to visiting the Castle Surveys team to learn how they put everything together.
#surveying#highwaysUK#highways#mobilemapping#infrastructure#pointclouds
“𝑯𝒐𝒘 𝒎𝒖𝒄𝒉 𝒊𝒔 𝒓𝒆𝒂𝒍𝒍𝒚 𝒊𝒏 𝒕𝒉𝒆 𝒈𝒓𝒐𝒖𝒏𝒅?”
This is the single most important question in mining. To answer this, mining engineers and geologists use different resource estimation methods. Each method has its own accuracy, data requirements, and ideal use case.
1. Polygonal / Triangular Methods (Classical)
Draw polygons (or triangles) around sample points (e.g., drill holes).
Assign the grade of the sample to the whole polygon or use the average of three samples for a triangle.
Used in : Early exploration, very sparse data, quick first-look estimates.
2. Inverse Distance Weighting (IDW)
Estimate the grade at an unsampled point by averaging nearby samples.
Closer samples have more weight (weight decreases with distance, often by distance²).
Used in : Moderate drill density, mid-stage projects needing a straightforward interpolator.
3. Ordinary Kriging (OK)
Use a semivariogram to model how grades correlate with distance and direction.
Calculate optimized weights from that model to produce an unbiased estimate and error measure.
Used in: Advanced exploration, feasibility studies, and formal resource reporting (JORC/NI 43-101).
4. Indicator Kriging (IK)
Convert grades into indicators (e.g., above/below a cutoff).
Krige those indicators to estimate probabilities that blocks exceed cutoffs; combine probabilities to infer grade classes.
Used in : Highly variable deposits, modelling cutoffs for ore/waste, probabilistic resource classification.
5. Sequential Gaussian Simulation (SGS) / Multiple Simulations
Generate multiple equally-probable realizations of the grade distribution that honour data and spatial continuity.
Use the ensemble of realizations to assess uncertainty and preserve local variability.
Used in : Uncertainty / risk analysis, complex or highly heterogeneous ore bodies, mine planning with scenario testing.
6. Machine Learning (ML)–Based Estimation
Use supervised learning algorithms (e.g., random forests, gradient boosting, neural networks) to predict grades or classes from many inputs: drill data, geology logs, geophysics, remote sensing, structural interpretations, and derived features.
ML models learn non-linear relationships and can incorporate large multi-source datasets. Often used together with spatial methods (e.g., ML predictions as inputs to kriging or as features in simulations).
Used in : Complex datasets with many predictors, integrating geophysics/chemistry/structural data, rapid scenario testing, and when non-linear patterns are suspected. Increasingly used for feature engineering, anomaly detection, and to augment traditional geostatistics.
#mining#geology#resources#resourceestimation#geostatistics#Kriging#IDW
The Leapfrog Geo Mastery Series – Session 2 focused on Advanced Control with Structural Trends, offering valuable insights into creating geologically realistic and reliable 3D models.
A key takeaway was the application of structural trends and anisotropy to guide interpolation, ensuring geological surfaces reflect the natural orientation of the subsurface. By integrating strike and dip data into the modeling workflow, we can achieve a more accurate representation of folds, faults, and mineralized zones.
Key learnings include:
● Structural trends are essential for maintaining geological continuity and minimizing unrealistic model behavior
● Anisotropy aligns models with the actual geological fabric
● Structural frames offer a robust method for modeling complex, folded terrains
● Continuous validation with drillhole and field data is crucial for accuracy
Practical impact: These techniques are particularly beneficial in structurally controlled deposits, such as gold systems, where mineralization is heavily influenced by faults and shear zones. Implementing these methods will enhance targeting precision and boost confidence in exploration outcomes.
This session has reinforced my approach to geological modeling, highlighting the importance of integrating technical tools with a solid geological understanding.
#LeapfrogGeo#GeologicalModeling#StructuralGeology#Mining#GoldExploration#GeoAI#ProfessionalDevelopment
Revolutionizing Mineral Exploration with Remote Sensing (RS) &GIS 🌍⛏️
RS and GIS are essential tools in mineral exploration. RS utilizes satellite capture spectral and spatial data, identifying mineralogical features and geological structures. GIS integrates this data with spatial analysis, enabling the creation of detailed geological models, enhancing target identification, and optimizing exploration workflows.
RS Workflow
1. Image Acquisition: Utilizing advanced satellite platforms (e.g., Landsat, Sentinel-2, ASTER) and hyperspectral imagery to meet specific spatial, spectral, and radiometric resolution requirements.
2. Pre-processing: Applying radiometric calibration, geometric corrections, and atmospheric removal techniques to eliminate noise and enhance data fidelity.
3. Image Enhancement: Leveraging advanced methods like band rationing, decorrelation stretching, and principal component analysis (PCA) to emphasize key geological features.
4. Spectral Analysis: Extracting diagnostic absorption features from reflectance spectra to characterize mineralogical compositions and identify alteration zones.
5. Feature Extraction: Employing algorithms such as edge detection, thresholding, and object-based image analysis to delineate structural controls, lithologies, and mineralization patterns
6. Spatial Analysis: Quantifying spatial relationships between features and integrating remote sensing-derived anomalies with ground-truth data to validate exploration models
7. GIS Integration: Combining multisource datasets (RS , geophysical, geochemical, and field data) within GIS for spatial modeling, 3D visualization, and predictive analysis
Applications in Mineral Exploration
Regional Exploration
Lithological Mapping: Discriminating and classifying rock units based on spectral properties and reflectance curves.
Structural Mapping: Accurately identifying faults, lineaments, folds, and fracture systems critical for ore deposition using high-resolution imagery and LiDAR data
Alteration Mapping: Detecting hydrothermal alteration minerals (e.g., clays, iron oxides, carbonates) indicative of mineralization processes
Target Generation
Anomaly Detection: Recognizing spectral and geophysical anomalies associated with buried ore bodies or surface mineralization signatures
Data Integration: Fusing remote sensing data with magnetometry, radiometry, and gravity surveys to refine geological interpretations
Target Prioritization: Applying multi-criteria decision analysis and machine learning algorithms to rank prospective targets based on integrated datasets
Benefits of RS and GIS
Efficiency: Rapid acquisition of regional-scale data, enabling large-area exploration in diverse terrains
Precision: High-resolution and multispectral imaging improve the accuracy of mineralogical mapping and resource estimation
Cost-effectiveness: Reduction in fieldwork expenditures through preliminary remote investigations
#RemoteSensing#GIS#GeospatialAnalysis
A layman with wisdom from above. Man who does things as the spirit directs. An apple of God's eye, protected and taken care of by God himself. Oasis of the barren land.
The essence of geological field mapping is that it serves as the foundation of geology—it is the direct, hands-on way geologists understand the Earth. Everything else (models, lab analyses, geophysics, geochemistry) builds on or ties back to it.
Here are the core points:
1. Observation at the Outcrop
Field mapping means reading the rocks where they are exposed.
Rock type, texture, structure, mineralogy, orientation, and relationships are recorded. Outcrops are the “pages” of the Earth’s book, and mapping is the act of reading and compiling them.
2. Spatial Relationships
Rocks are not isolated; they form patterns.
Mapping traces contacts between units, tracks faults, folds, dykes, and reconstructs the 3D geometry beneath the surface.
3. Synthesis into a Map
The field notes are distilled into a geological map: a two-dimensional picture of three-dimensional reality.
The map reveals the distribution of rock units, structures, and history.
4. Interpreting Geological History
Mapping is not just “drawing rocks.”
It reconstructs geological evolution: what came first, what was deformed, intruded, uplifted, or eroded.
It explains processes like magmatism, sedimentation, tectonics, metamorphism.
5. Applied Value
Resource exploration (minerals, water, hydrocarbons).
Engineering geology (dams, tunnels, foundations).
Environmental management and hazards (landslides, earthquakes, groundwater).
6. Ground-Truthing
In an era of satellite imagery, AI, and remote sensing, field mapping remains the ground truth—the reality check against assumptions.
👉 In short: The essence of geological field mapping is to decode Earth’s story by observing and recording rocks in their natural setting, then synthesizing those observations into a coherent map that explains both the present distribution and past evolution of geology.
80,000+ drilling activities. 200 years of exploration. One map that shows where gold exploration may have stopped short.
Last week's map on the successes and failures of precious metals drilling sparked an insightful comment: "Back in the day, exploration focused on open-pitable oxide material. There are areas with pin cushions of drill holes that are still fertile."
That inspired me to map these "pin cushions" - districts where shallow oxide-focused drilling dominated, but deeper tests for primary mineralization were never done.
The 'oxide gold rush' is a classic exploration filter. Many areas were effectively sterilised because drilling stopped once fresh rock was encountered. By highlighting where this shallow drilling was never followed up at depth, we get a new kind of target map: not where gold is, but where exploration stopped short.
How the scoring works:
- Pale Yellow (Low Score): Few shallow holes compared to deeper ones. Either not a focus of the oxide rush, or they were tested at depth later.
- Orange (Medium Score): Moderate imbalance. Plenty of shallow drilling, limited deeper follow-up. Worth a closer look.
- Deep Red (High Score): Bullseye targets. Intense shallow drilling in the oxide era with little or no subsequent diamond drilling - the most likely "fertile but overlooked" districts.
What the map reveals:
- The Oxide Rush Heartland: Red and orange hexagons cluster in the Yilgarn Craton, neatly outlining the 1980s-90s boom.
- Structural Corridors: The red zones trace NW-SE belts like the Boulder-Lefroy Fault and Zuleika Shear, showing the historical exploration mindset.
- Frontier Regions: Northern areas (Pilbara, Paterson) remain pale, confirming they weren't the focus during the oxide rush. A lone red patch in the far North (Tanami?) suggests a specific historical campaign worth revisiting.
The real power comes when this is combined with the earlier "sentiment map" of historic drilling outcomes. Together, they highlight districts that are both untested at depth and historically written off as "failures" - places less likely to have been re-examined by competitors.
What do you think of this workflow? Does it capture the key exploration blind spots, or are there trends and pitfalls I might be missing?
I enjoyed Scott Halley’s excellent recent post where he emphasizes that “a well-constrained geology map remains the most important layer in any exploration project”. After nearly 40 years of conducting applied geological mapping on diverse projects globally I fully concur with that opinion.
Simply, the project geology map provides essential context for all other information and the meaningful interpretation of that information. That is the basic premise behind my applied mapping training courses for industry and why I devote a large part of my time to educating and motivating exploration geologists in the field.
Attached is one of my more recent outcrop/interpretation-style maps highlighting the spatial relationship between mapped mineralized features and an earlier phase of soil/rock geochemical sampling with Cu values shown in this case. Many of the mineralized features were not depicted on previous generations of mapping, or simply shown as generalized stratabound "gossan" zones with no structural information or indications of actual mineralization style.
More significantly, though the main proposed target here was VMS-related Cu associated with ostensibly “concordant “ gossan zones, the mapping actually indicates many of the key Cu geochemical anomalies to be associated with a later, strongly structurally controlled Cu-bearing vein mineralization overprint developed well after build-up of the volcanic pile, with veins swarms oblique to stratigraphy caught up in and extensively transposed along late shear zones developed at major lithological contacts where pronounced rheological contrasts are present. The mapping observations in this case impact massively on the target model and exploration strategy going forward.
This is precisely why Fred Graybeal, a highly seasoned and well-respected explorationist, commented (see SEG Discovery, no 128, 2022): “ The importance of a geological map can’t be over-emphasized. A geological map is a decision-making document, and a reliable map is as important as a resource estimate or financial analysis….”
I would also emphasize the cost-benefit aspect and value-adding nature of geological mapping - which is typically a relatively low-cost component in most exploration programmes - yet one that has major ramifications for programme design and target generation, and not least of all ensuring every hard-won dollar expended really counts.
So once again if anyone out there wants to up their game in applied geological mapping to really optimize their exploration and targeting process, I'm happy to discuss on-site training/mentoring! Good luck to all in their 2025 endeavours.