Advancing Radiometric Mapping in Australia: Bridging Gaps and Enhancing Mineral Exploratio

A study advancing radiometric mapping in Australia by applying machine learning and deep learning to enhance potassium concentration predictions from airborne gamma-ray spectrometric surveys. The research compares traditional interpolation methods with Random Forest, XGBoost, and deep learning models, using a dataset integrated with 25 geospatial layers. It was presented at the IGARSS 2025 symposium and published by Geoscience Australia Data in 2026.

Use Cases

• Improving mineral exploration targeting based on enhanced potassium concentration predictions.
• Evaluating machine learning model performance for geospatial interpolation across various data gap scenarios.
• Comparing traditional geostatistical methods (like Kriging) with advanced algorithms for environmental mapping.

Strengths

• Integrates 25 geospatial layers to evaluate model predictions.
• Compares a range of methods including Inverse Distance Weighting, Kriging, Random Forest, XGBoost, and deep learning models.
• Addresses a specific national data gap in Australia's radiometric coverage.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and specific data formats (beyond PDF/HTML) are unknown, which may limit suitability assessment.

Provenance

Source

Geoscience Australia Data

Collection Method

Airborne gamma-ray spectrometric surveys integrated with 25 geospatial layers.

Freshness

Last updated 2026-05-19 22:27:43.166417; freshness should be verified.

Geography

Australia