Mineral mapping of the Australian continent using the new generation of hyperspectral satellite imagery

For over four decades, the Commonwealth Scientific and Industrial Research Organisation (CSIRO) has been at the forefront of developing and evaluating Earth Observation (EO) data tailored to the needs of the mineral resources sector. A significant milestone was achieved in 2012 with the release of the world’s first ASTER-derived continental scale mineral maps [1,2]. The multispectral ASTER satellite (METI-NASA, JPL) was the first ever geoscience-tuned EO system that contained in total 14 bands distributed in visible to near (3 bands in the VNIR: 500-1000 nm), shortwave (6 bands in the SWIR: 1000-2500 nm), and in thermal (5 bands in the TIR: 8000-12,000 nm) infrared wavelength range [3]. These maps facilitated critical greenfield discoveries [4], marking a breakthrough in geological research and resource exploration.

Since 2019, advanced hyperspectral spaceborne sensors such as PRISMA (ASI), EnMAP (DLR), DESIS (DLR), and EMIT (JPL) has revolutionised EO. These sensors provide unprecedented spectral resolution (between 235 and 285 bands within the VNIR-SWIR wavelength range, dependent on the sensor configuration) and spatial coverage (30 or 60 m) that are comparable to those derived from rock specimens by field and laboratory sensors, enabling the extraction of detailed mineralogical and other geologically significant parameters. These datasets are freely accessible for research purposes, and as part of ongoing efforts, CSIRO has been pioneering new methodologies to harness these hyperspectral data for enhanced mineral exploration and mining applications [5 – 8]. The multi-sensor comprehensive coverage is maintained through regular ~28-day reacquisition cycles, ensuring up-to-date, cloud-free and high-resolution data. The hyperspectral satellite imagery can be processed into a suite of geoscience layers, showing the relative abundance of key mineral species and their variations in terms of mineral chemistry and crystallinity. A preliminary example of the PRISMA-EnMap-derived continental-scale Al-sheetsilicate abundance index is shown in the Figure below. New PRISMA and EnMap tiles will be added to the CSIRO-developed image processing workflow as soon as they become available. Furthermore, green and dry vegetation abundance indices are developed as part of the processing workflow, representing valuable information for environmental, agriculture and fire fuel load applications [9]. The resulting maps are powerful tools for geologists, providing insights into the continent’s mineralogical landscape.

References

[1] T. Cudahy et al. (2008). Next Generation Mineral Mapping: Queensland airborne HyMap and satellite ASTER surveys 2006-2008, CSIRO report P2007/364, 161pp.

[2] Caccetta, M., Collings, S., Cudahy, T. (2013). A calibration methodology for continental scale mapping using ASTER imagery, in Remote Sensing of Environment, 139, pp. 206-317.

[3] Abrams M., Hook, S., Ramachandran, B. (2002). ASTER User Handbook, Version 2, Jet Propulsion Laboratory, 4800, 135.

[4] Laukamp C., Cudahy, T., Caccetta, M., Thomas, M., Close, D., Lennartz, R. (2013). Successful mineral exploration using multispectral remote sensing data – ASTER Geoscience Map of Australia, in SGA Uppsala 2013 “Mineral deposits research for a high-tech world – volume 1”, pp. 32-36.

[5] Laukamp, C., Huang, F., Kohan Pour, F., Lampinen, H.M. (2024): Machine learning-supported regolith landform mapping using satellite imagery and geophysical data. IGARSS 2024 conference, 07.07.2024-11.07.2024, Athens, Greece.

[6] Laukamp, C., Thomas, M., Lau, I.C. (2023): Review of currently available multispectral and hyperspectral imaging spectroscopy satellite sensors for critical minerals exploration. Whispers 2023 conference, 31.10.23-02.11.23, Athens, Greece.

[7] Laukamp, C. (2022): Geological Mapping using mineral absorption feature-guided band-ratios applied to PRISMA satellite hyperspectral level 2D imagery. IGARSS 2022 conference, 17.07.22-22.07.22, Kuala Lumpur, Malaysia.

[8] Chirico, R. et al. (2023): Mapping hydrothermal and supergene alteration zones associated with carbonate-hosted Zn-Pb deposits by using PRISMA satellite imagery supported by field-based hyperspectral data, mineralogical and geochemical analysis. Ore Geology Reviews, 152, 1052444.

[9] Lau, I.C., Laukamp, C., Miles, J. (2024): Hyperspectral satellite imaging of electronic and vibrational modes of minerals and vegetation. The 2024 Advancing Earth Observation Forum, 10.09.2024-12.09.2024, Adelaide, Australia.

[10] Laukamp, C., Miles, A.J., Williams, M., Lau, I.C., Mule, S., Lampinen, H.M., Hunt, A., Bradley, S., Devaraju, A., Ong, C., Caccetta, M. (2025): Towards mapping the Australian continent with the next generation of hyperspectral satellite imagery.- IGARSS’25, Brisbane, 3.-8.08.25.