Hyperspectral Surveying for Mineral Resources in Alaska

Figure 1. Regional mineral classification map draped over a digital elevation model of the Orange Hill area, Wrangell–St. Elias National Park and Preserve, Alaska. Colours show the spectrally dominant minerals; data at 6-metre resolution. USGS.

Alaska is a major producer of base and precious metals and has high potential for more undiscovered mineral resources. Its vast size, remoteness and rugged terrain make discovery hard, and new methods are needed to assess its geology and mineral potential fully. Hyperspectral surveying is one: a fast way to map surface materials, including different kinds of bedrock and ground cover.

In 2014 the USGS began the Alaska Hyperspectral Project to test the method in Alaska, working with the Alaska Division of Geological and Geophysical Surveys, the University of Alaska Fairbanks and the National Park Service. The main study area is a remote part of the eastern Alaska Range where porphyry deposits are exposed. The aims are better geologic mapping, and methods for finding and characterising mineral deposits elsewhere in Alaska.

In short:

• Hyperspectral surveys give information about bedrock and the materials covering the Earth's surface.
• Data collection is fast, at scales from whole regions down to a single hand specimen.
• The rock composition they reveal can add to existing geologic maps.
• Maps of mineral classes help economic geologists find areas with mineral deposits.

What is hyperspectral analysis?

When light hits a material, some of it is reflected. A spectrometer measures how strongly, and measuring across a range of wavelengths gives a spectrum. Its shape works like a fingerprint, showing which minerals are present, roughly how abundant they are, and their chemistry. Some common minerals, such as quartz and feldspars, have no distinctive sharp absorption features at the wavelengths of sunlight and must be detected with a thermal spectrometer.

Example spectra from 1.6 to 2.5 micrometres, showing the diagnostic absorption features (dips) of calcite, antigorite, muscovite and gypsum. USGS.

From kilometres to micrometres

The project combines spectral measurements with geology in the field and the laboratory, collecting hyperspectral (imaging spectrometer) data at three scales:

Electron microprobe and X-ray diffraction analyses of rock and soil samples check the mineral and spectral readings of the hand specimens, and the hand-specimen and outcrop results sharpen the interpretation of the regional survey.

Figure 2. Outcrop mineral map of the western side of the Orange Hill porphyry copper deposit, from HySpex data at 30-centimetre resolution. The colours show a predominance of clays, muscovite and gypsum (likely related to porphyry mineralisation) across the hillside. USGS.

Figure 3. Hand-specimen photographs (top) and mineral classification maps (bottom) from Corescan data at the Orange Hill deposit, at 500-micrometre resolution. USGS.

Dominant minerals, and more

The mineral classification maps match each spectrum to standards in the USGS spectral library and show the predominant mineral or minerals. The USGS is developing ways to estimate how abundant those minerals are across the survey area, and studying subtle changes in the shapes of spectral features to map differences in the chemistry of certain minerals. Such differences generally reflect changes in rock type. In this area, as studies elsewhere have also shown, the spectral features of muscovite change with nearness to porphyry copper deposits.

Figure 4. The Orange Hill and Bond Creek areas, showing systematic changes in the wavelength position of the muscovite feature, related to geology and to nearness to porphyry deposits and occurrences. Base from the USGS National Elevation Dataset 5-metre digital elevation model. USGS.

Early observations

• Regional mineral maps generally show changes in rock type that agree with earlier geologic mapping.
• Mineral groupings agree from the regional (6 m) to the outcrop (30 cm) scale, with more detail at the hand-specimen scale (500 µm).
• Differences in the wavelength position of the muscovite feature (different chemistry) can correspond to particular geologic units and to general nearness to known porphyry occurrences.
• The results support earlier studies showing that hyperspectral data can reveal surface signs of undiscovered mineral resources.

Sources

• Raymond F. Kokaly, Garth E. Graham, Todd M. Hoefen, Karen D. Kelley, Michaela R. Johnson and Bernard E. Hubbard, Hyperspectral Surveying for Mineral Resources in Alaska, USGS Fact Sheet 2016–3029, July 2016, prepared with the Alaska Division of Geological and Geophysical Surveys, the University of Alaska Fairbanks and the National Park Service. https://doi.org/10.3133/fs20163029
• R. N. Clark and others, USGS Digital Spectral Library splib06a, USGS Digital Data Series 231, 2007.
• The figures are taken from the fact sheet's PDF.