HYPERSPECTRAL REMOTE SENSING

Spaceborne optical imagers are currently either panchromatic or multispectral, providing just a few spectral bands and limited resolving power. Hyperspectral imagers typically collect data in numerous (sometimes several hundred) contiguous narrow bands spanning a vast region of the electromagnetic spectrum ranging from .001µm to 14.0 µm. Hyperspectral imagers produce vast quantities of data because of the number of bands simultaneously imaged.

One of the first operational applications of remotely sensed earth observation data in the 1980s was in the field of mineral exploration.  Investigations made in the last 20 years indicate that hyperspectral remote sensing can significantly contribute to geological investigations, especially in the identification and mapping of minerals and lithology in arid (non-vegetated) environments.  This is possible because hyperspectral sensors, in contrast to other existing broad-band multi-spectral scanners, are able to better resolve absorption features unique to specific mineral species.

Analysis of  hyperspectral data involves using the reflectance of each pixel in each of the hundreds of bands, and and representing this data as ‘spectral reflectance curves’. The identity of material constituting the target is determined by comparison of its spectral reflectance curve with 'library' spectra of known materials measured in the field or in the laboratory.  It is envisaged that hyperspectral data will enable the identification of terrestrial features with greater accuracy.

The NASA EO-1 Hyperion sensor was the first satellite to collect hyperspectral data from space (November 2000). The spaceborne hyperspectral imagers to date have been technology demonstrators. There are currently no commercial spaceborne hyperspectral sensors in orbit, although some are planned.

In addition to these benefits, hyperspectral imaging has the potential to augment expensive surveying methods.  The cost of hyperspectral surveys is less than 10% as compared to that of traditional surveying techniques.

Applications of Hyperspectral Imaging:

Spaceborne multi-spectral imaging provides gross lithologic information or identification of stressed vegetation.  It has severe limitations in so far as detailed mineralogy or geobotany of importance for mineral exploration is concerned. Specific mineralogical and geobotanical information is currently provided by field crews.  This is expensive and time-consuming. Hyperspectral remote sensing has the potential of highlighting significant mineralogical and geobotanical anomalies or trends.In mineral exploration, hyperspectral data finds two major applications:

1.      Lithologic mapping

2.      Geobotanical mapping

Lithologic Mapping The key elements in planning a mineral exploration program prior to undertaking intensive field exploration activities are: 1) to obtain a preliminary understanding of a geographic area through lithological mapping, and 2) to help in the identification of potential exploration targets. Bedrock mapping and identification of the presence and abundance of particular minerals are facilitated by hyperspectral data.  Lithologic maps help geologists decipher the lithologic and structural history of a region.  This is particularly valuable for areas for which no maps or very generalized maps exist.

Minerals that can be successfully identified with hyperspectral imaging are: OH-bearing minerals, carbonates, sulfates, olivines, pyroxenes, iron oxides and hydroxides. The identification of these minerals and mapping their distribution provides a framework for exploration of precious and base metals, diamonds, etc.  Whereas hyperspectral products can be used for characterization of lithology in arid (non-vegetated) environments, basic research is required to find ways to utilize these in vegetated terrains.

Geobotanical Mapping:  Lithologic mapping through hyperspectral data is effective only in arid regions.  Surface geology is obscured to varying degrees by vegetation in most areas of the world. Specific element associated spectral changes in vegetation, which in turn are related to lithology and soil chemistry, help in the identification, distribution and spatial relationships of anomalous zones. Geobotanical mapping thus holds promise in mineral exploration activity.  This approach makes use of the fact that the spectral reflectance of vegetation is affected in the presence of heavy metals or alteration zones. For example, accumulation of heavy metals induces stress on the vegetation causing a shift of the red-edge (680 nm - 800 nm). Such a shift is only detectable with a hyperspectral imager.  Geobotanical anomalies associated with ore bodies may sometimes be evident as abrupt changes in plant species, being thereby indicative of lithological changes rather than stress induced physiological changes.  The ability of acieving element specific geobotanical products is not yet well developed, and considerable research is required before geobotanical mapping can be effectively utilized in mineral exploration.

Although the use of geobotanical mapping is very promising, the ability of achieving element specific geobotanical products is not yet well developed and requires further basic research and development before it achieves operational status.

Availibility of Hyperspectral Data:

Remote Sensing has advanced the ability to map the location of altered areas, specify many mineral alteration species, and designate regions of hydrothermal upwelling and outflow zones related to mineralization. The increased spatial and spectral resolution of satellite and airborne sensors provide significant tools to the exploration, evaluation and understanding of the genesis of mineral deposits.

Exploitation of Landsat Thematic Mapper (TM) data continues to enable explorationists to make new discoveries in well studied and mapped mineralized regions. Landsat TM enables maps of clay and iron oxide alteration to be produced. Analysis of ASTER satellite data has provided information on mineral groups and a few specific mineral species. Importantly ASTER data can effectively map propylitic, argillic, and advanced argillic assemblages. Hyperspectral data from airborne scanners can map many specific alteration mineral species as well as cation substitution in certain clay minerals important to the evaluation of mineralization.

A major challenge for exploration and mineral genesis studies is being able to distinguish and map hydrothermal upwelling and outflow zones. Currently available ASTER and hyperspectral data have the ability to provide coherent maps that can distinguish paleo-fluid paths that can lead to a greater understanding of ore system genesis and aid exploration efforts. Images that provide synoptic and spatially coherent data provide an increasingly important tool when combined with field and laboratory spectral analysis, petrography and critical field observations. ASTER (Advanced Spaceborne Thermal Emission and Reflection Radiometer) is an imaging instrument  flying on Terra, a satellite launched in December 1999 as part of NASA's Earth Observing System (EOS). ASTER is a cooperative effort between NASA,  Japan's Ministry of Economy, Trade and Industry (METI) and Japan's Earth Remote Sensing Data Analysis Center (ERSDAC). ASTER is being used to obtain detailed maps of land surface temperature, reflectance and elevation.

About ASTER:

The Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) is one of the five state-of-the-art instrument sensor systems on board the Terra satellite, launched in December 1999.  ASTER is a cooperative effort between NASA and Japan's Ministry of Economy, and has been designed to acquire land surface temperature, emissivity, reflectance, and elevation data.  An ASTER scene covers an area of approximately 60 km by 60 km and data is acquired simultaneously at three resolutions. The repeat cycle is 16 days or less.  The images are georeferenced to the WGS84 datum and Universal Transverse Mercator projection.  ASTER monitors cloud cover, glaciers, land temperature, land use, natural disasters, sea ice, snow cover and vegetation patterns at a spatial resolution of 15 to 90 meters.

A complete ASTER scene consists of 15 bands of data; including one band which points backwards to create parallax. The three bands in the visible and near infrared (VNIR) part of the spectrum have a 15m resolution and an 8-bit unsigned integer data type. This dataset also features a second near infrared backward-scanning band labeled Band 3B. This is used to create a stereo view and to develop elevation information and is unsuitable for analysis or classifications.

The six bands in the short wave infrared region (SWIR) have a 30m resolution and also have an 8-bit unsigned integer data type. Finally there are five thermal infrared bands (TIR) with a 90m resolution and have a 16-bit unsigned integer data type.

Since its launch, ASTER has acquired over 1,100,000 images of the land surface. The data are first processed in Japan, then they are archived and distributed in both Japan, and in the US through the EROS Data Center Distributed Active Archive Center. Data are available as Level 1 images, geometrically and radiometrically corrected, and on-demand higher level, geophysical data products.

ASTER was designed largely with improved geologic mapping capabilities in mind, compared to previous Landsat Thematic Mapper sensors. ASTER is the first multispectral satellite with multiple bands in the VNIR, SWIR and TIR wavelength regions. VNIR bands help distinguish various iron oxide exposures. The SWIR bands were placed to help separate phyllosilicate, carbonate, and sulfate minerals. The TIR bands were selected to help distinguish various silicates by detecting changes in the position of SiO2 restrahlen bands.

Geoscience Australia has developed a new remote sensing tool that will will use ASTER data and will assist exploration geologists to map alteration mineralogy.  While previous studies required many months of detailed mapping and sample collection, ASTER maps enable alteration to be identified before field work is undertaken, maximising the value of time spent in the field.

The table below lists the specific wavelength information of ASTER data.  

ASTER Band Specifications

Sensor

Band #

Wavelength

Resolution

View

VNIR

B1 VNIR_Band1

0.52 - 0.60 μm

15 m

 Nadir view

B2 VNIR_Band2

0.63 - 0.69 μm

15 m

 Nadir view

B3 VNIR_Band3N

0.76 - 0.86 μm

15 m

 Nadir view

B4 VNIR_Band3B

0.76 - 0.86 μm

15 m

 Backward scan

SWIR

B5 SWIR_Band4

1.60 - 1.70 μm

30 m

 Nadir view

B6 SWIR_Band5

2.145 - 2.185 μm

30 m

 Nadir view

B7 SWIR_Band6

2.185 - 2.225 μm

30 m

 Nadir view

B8 SWIR_Band7

2.235 - 2.285 μm

30 m

 Nadir view

B9 SWIR_Band8

2.295 - 2.365 μm

30 m

 Nadir view

B10 SWIR_Band9

2.36 - 2.43 μm

30 m

 Nadir view

TIR

B11 TIR_Band10

8.125 - 8.475 μm

90 m

 Nadir view

B12 TIR_Band11

8.475 - 8.825 μm

90 m

 Nadir view

B13 TIR_Band12

8.925 - 9.275 μm

90 m

 Nadir view

B14 TIR_Band13

10.25 - 10.95 μm

90 m

 Nadir view

B15 TIR_Band14

10.95 - 11.65 μm

90 m

 Nadir view

 

Notes & Handouts

The Himalayas

Kumaon Himalayas

Askot Basemetals

University

   


This website is hosted by

S. Farooq

Department of Geology

Aligarh Muslim University, Aligarh - 202 002 (India)

Phone: 91-571-2721150

email: farooq.amu@gmail.com