RADARSAT GEOLOGY HANDBOOK - pudn.comread.pudn.com/downloads149/ebook/645429/rghb.pdf · 2009-01-30 · RADARSAT GEOLOGY HANDBOOK Reproduction of the contents of this Handbook is permissible
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
R A D A R S A T G E O L O G Y H A N D B O O K
Reproduction of the contents of this Handbook is permissible and encouraged.
1 - 2 R A D A R S A T G E O L O G Y H A N D B O O K
C O M P A R I S O N O F S A T E L L I T E I M A G I N G S Y S T E M S 1 - 3
T H E R A D A R S A T S A T E L L I T E
RADARSAT, launched November 4, 1995 was the result of a joint venture
between the Canadian Government, private industry and NASA. As Canada’s
first Earth observation satellite, and the world’s first operationally-oriented
radar sensor, RADARSAT is providing valuable information for use in monitoring
the world’s environmental and natural resources.
Addressing Canada’s first major need for a radar sensor, RADARSAT provides
effective surveillance of the Canadian Arctic, which is characterized by long
periods of darkness in the winter. Major shipping routes cross this vast region
which recently gained importance due to its significant mineral and petroleum
reserves. Another need met by RADARSAT is to monitor Canada’s coastline,
which is one of the longest in the world, and is perennially cloud-covered
and foggy.
RADARSAT was launched in a sun-synchronous, dawn-dusk orbit with a
24-day repeat cycle. It provides regular imaging opportunities as frequently as
daily above the Arctic, and every five days over equatorial latitudes.
For geologists who have been trained to work with optical images, a transition
must be made to understand the unique features of radar imagery and to
successfully utilize the radar data. An introductory discussion of the
RADARSAT satellite is provided in Chapter Two, followed by an overview of
radar image interpretation in Chapter Three.
2
3
4
5
6
COMPARISON OF SATELLITE IMAGING SYSTEMS
THE RADARSAT SATELLITE
VISUAL INTERPRETATION OF RADARSAT IMAGERY
IMAGE ENHANCEMENT OF RADARSAT DATA
VALUE-ADDED RADARSAT PRODUCTS
SUMMARY
REFERENCE MATERIALS
1
T H E R A D A R S A T S A T E L L I T E 2 - 1
RADARSAT differs from research-oriented radar sensors such as ERS and
JERS-1, given that RADARSAT is the first radar sensor totally dedicated to
operational applications and it offers a variety of beam modes to meet
requirements of the particular application at hand. Using a single frequency
C-Band (5.3 Ghz frequency or 5.6 cm wavelength), the RADARSAT SAR has
the ability to shape and steer its radar beam over a 500 kilometre range.
RADARSAT provides complete global coverage with the satellite’s orbit repeated
every 24 days. The Arctic is imaged daily, while equatorial areas achieve complete
coverage approximately every 5 days. The following sections discuss
RADARSAT’s unique features and the benefits these pose for geologists.
U N D E R S T A N D I N G R A D A R I M A G E R Y
What is a radar image?
Radar images are single frequency representations of the Earth, which highlight
changes in the terrain’s roughness, relief, and moisture levels. They are similar
to other types of Earth observation imagery in that they represent the reflectivity
portion of the electromagnetic spectrum (Figure 2.1). However, radar imagery
is derived from a portion of the light spectrum that human vision is unable to
detect. This special wavelength is capable of penetrating rain, cloud and haze, to
provide a continually clear view of the Earth.
FIGURE 2.1: Electromagnetic spectrum
OPTICAL RADAR
Gamma Rays X-Rays U.V. Rays Visible
Near and MidInfrared
Thermal Infrared Microwave TV/Radio
(cm) 10-10
10-9
10-8
10-7
10-6
10-5
10-4
10-3
10-2
10-1
1.0 10 100
FIGURE 2.2: Energy interaction in a SAR system
Why are radar images black and white?
Radar images are black and white not because they are fundamentally different
from other data sources (i.e LANDSAT or SPOT optical sensors), but because
they do not have a multispectral component necessary for false-colour formation.
LANDSAT TM is sensitive to the Earth’s reflectivity at seven different
wavelengths, hence the seven bands. Colour is achieved by combining any three
of the optical bands. RADARSAT contains only one spectral band, and thus
offers a unique dataset for the exploration geologist.
What do radar images consist of?
A radar image is the ratio of microwave energy transmitted to the Earth to the
energy which is reflected directly back to the sensor. The energy returning to
the sensor is called backscatter (see Figure 2.2). The backscatter of an imaged
area is dependent upon local topography, centimetre-scale roughness, and
dielectric properties, which are directly affected by moisture levels. Low
2 - 2 R A D A R S A T G E O L O G Y H A N D B O O K
T H E R A D A R S A T S A T E L L I T E 2 - 3
backscatter values are portrayed as dark image tones or grey levels which
approximate black, while high backscatter values are shown as light image tones
or grey levels approximating white.
FIGURE 2.3: Surface roughness from varying terrains
What sort of information is provided by radar images?
Radar imagery provides valuable information to a broad user community.
Geology, agriculture and landcover mapping are just a few of the applications
which benefit from the way in which a radar image portrays landcover types.
Although no two land units are the same, there are general rules which apply to
certain landcover classes. Water is usually dark- a product of specular reflection
and resulting weak return, while urban areas are always very bright due to the
presence of corner reflectors (see Figure 2.3). Everything else falling between
these two extremes is represented in various shades of grey. By interpreting the
various tones, textures and patterns on the image, the user can unlock information
pertaining to geologic structure and lithology.
R A D A R S A T ’ S S T A T E - O F - T H E - A R T F E A T U R E S
Range of Beam Modes
RADARSAT is equipped with seven beam modes, which offer image resolutions
ranging from 8 to 100 metres. RADARSAT is designed in such a way that its
beam can be steered at incidence angles ranging from 10-60 degrees. It offers
spatial coverage ranging from 50-500 kilometre swaths, and can be used for
mapping at scales of 1:1,000,000 to 1:50,000. These features are described in
Figures 2.4 and 2.5.
FIGURE 2.4: SAR operating beam modes
2 - 4 R A D A R S A T G E O L O G Y H A N D B O O K
T H E R A D A R S A T S A T E L L I T E 2 - 5
FIGURE 2.5: RADARSAT beam positions
BEAM MODE BEAM POSITION INCIDENCE APPROXIMATE NOMINALANGLE RANGE (º) RESOLUTION (m) AREA (km)
Fine F1 near 36.4 - 39.6 8 * 50 x 50F1 36.8 - 39.9F1 far 37.2 - 40.3F2 near 38.8 - 41.8F2 39.2 - 42.1F2 far 39.6 - 42.5F3 near 41.1 - 43.7F3 41.5 - 44.0F3 far 41.8 - 44.3F4 near 43.1 - 45.5F4 43.5 - 45.8F4 far 43.8 - 46.1F5 near 45.0 - 47.2F5 45.3 - 47.5F5 far 45.6 - 47.8
Space Agency/Agence spatiale canadienne 1996. Received by the Canada Centre for Remote
Sensing. Processed and distributed by RADARSAT International. Data acquired during the
commissioning phase and may not conform to system specifications.
V I S U A L I N T E R P R E T A T I O N O F R A D A R S A T I M A G E R Y 3 - 1 3
G E O L O G I C A L A P P L I C A T I O N S G U I D E L I N E S
TABLE 3.1: Geological Activities and RADARSAT Recommendations
GEOLOGICAL ACTIVITY RESPONSE AND RECOMMENDATIONS
Geological RADAR RESPONSE Structure Mapping Geological structures often have characteristic forms, which, if located
near the Earth’s surface, may be manifested topographically as the side-looking configuration of radar highlights relief.
RECOMMENDATIONSRADARSAT Beam Mode: All beam modes are suitable for geological structure mapping. The final beam mode selection is dependent on the areal coverage and the level of detail required. Generally, Fine and Standard beam modes are best suited for detailed geological structure mapping, while Wide and ScanSAR are better for basin-wide geological structure mapping.
RADARSAT Incidence Angle: Shallow incidence angles are ideal for enhancing subtle terrain features through shadowing. In areas of high relief, too much shadowing may occur with shallow incidence angles and therefore, intermediate incidence angles may be more suitable.
Look Direction: Orientation of geological structures relative to look direction should be considered.
When to Acquire RADARSAT Data: Acquire data when geological structure information is required, regardless of seasons. Avoid periods of heavy snowcover.
Lineament RADAR RESPONSEIdentification Lineaments, such as folds and faults may be manifested as topographic
relief. The ability to image lineaments is a result of the side-looking configuration of radar which highlights relief.
RECOMMENDATIONSRADARSAT Beam Mode: The beam mode chosen is dependent on the width of the lineaments, and on areal coverage and the level of detail required. Generally, Fine and Standard beam modes are best suited for detailed lineament identification, while Wide and ScanSAR are better for regional identification.
3 - 1 4 R A D A R S A T G E O L O G Y H A N D B O O K
GEOLOGICAL ACTIVITY RESPONSE AND RECOMMENDATIONS
RADARSAT Incidence Angle: Shallow incidence angles are ideal for enhancing the subtle topographic relief of lineaments.
Look Direction: A look direction perpendicular to the direction oflineaments enhances their detectability. The acquisition of both ascending and descending passes maximizes the number of lineaments that can be identified.
When to Acquire RADARSAT Data: Acquire data when lineament information is required.
Seismic Zones RADAR RESPONSEIdentification Seismic zones are often characterized by the presence of faults which
may be manifested topographically. The ability to image seismic zones is a result of the side-looking configuration of radar, which highlights this topography.
RECOMMENDATIONSRADARSAT Beam Mode: The beam mode chosen is dependent on the width of the lineaments associated with seismic zones, and on areal coverage and the level of detail required. Generally, Fine and Standard beam modes are best suited for detailed seismic zones identification, while Wide and ScanSAR are better for regional seismic zone identification.
RADARSAT Incidence Angle: Shallow incidence angles are ideal forenhancing the subtle topographic relief of lineaments.
Look Direction: A look direction perpendicular to the direction of lineaments will enhance their detectability. The acquisition of both ascending and descending passes maximizes the number of lineaments that can be identified.
When to Acquire RADARSAT Data: Acquire data when lineamentinformation associated with seismic zones is required.
Landform Delineation RADAR RESPONSELandforms often have characteristic shapes which may be manifested as topographic relief. The ability to image landforms is a result of the side-looking configuration of radar, which highlights relief.
RECOMMENDATIONSRADARSAT Beam Mode: All beam modes are suitable for landform
V I S U A L I N T E R P R E T A T I O N O F R A D A R S A T I M A G E R Y 3 - 1 5
GEOLOGICAL ACTIVITY RESPONSE AND RECOMMENDATIONS
delineation. The final beam mode selection is dependent on the areal coverage and the level of detail required. Generally, Fine and Standard beam modes are best suited for detailed landform delineation, while Wide and ScanSAR are better for regional landform delineation.
RADARSAT Incidence Angle: Shallow incidence angles are ideal for enhancing subtle terrain features through shadowing.
Look Direction: Orientation of land forms relative to look direction should be considered.
When to Acquire RADARSAT Data: Acquire data when landform information is required, regardless of season. Avoid periods of heavy snowcover.
Surficial Bedrock RADAR RESPONSEGeological Mapping Depending on the type of physical weathering, surficial bedrock may
characteristically fracture to produce fragment sizes, which are a function of elements such as rock fabric, texture and mineral composition. Individual rock units may break down differentially, resulting in unique backscatter.
RECOMMENDATIONSRADARSAT Beam Mode: All beam modes are suitable for surficialbedrock mapping. The final beam mode selection is dependent on the areal coverage and the level of detail required. Generally, Fine and Standard beam modes are best suited for detailed surficial bedrock mapping, while Wide and ScanSAR are better for regional surficial bedrock geological mapping.
RADARSAT Incidence Angle: The main parameter that may differentiate rock fragment size associated with surficial bedrock units is surface roughness. Shallow incidence angles maximize the contrast in backscatter resulting from variances in roughness.
Look Direction: Orientation of geological structures relative to look direction should be considered.
When to Acquire RADARSAT Data: Acquire data when moisture levels are low in order that the backscatter be more closely correlated to surface roughness than to moisture content.
GEOLOGICAL ACTIVITY RESPONSE AND RECOMMENDATIONS
Surficial Material RADAR RESPONSEAssessment Non-vegetated, unconsolidated surficial material contains different
fragment sizes which may produce a characteristic soil roughness and soil moisture holding capability. Radar is sensitive to changes in roughness and moisture, and the result is contrasting backscatter between different surficial units.
RECOMMENDATIONSRADARSAT Beam Mode: All beam modes are suitable for the assessmentof surficial materials. The final beam mode selection is dependent on the areal coverage and the level of detail required. Generally, Fine and Standard beam modes are best suited for detailed surficial assessment, while Wide and ScanSAR are better for regional surficial material assessment.
RADARSAT Incidence Angle: If surficial material is assessed based on soil moisture, steep incidence angles are preferred to minimize thebackscatter associated with soil roughness. If surficial materials are assessed based on soil surface roughness, shallow incidence angles are better suited.
Look Direction: Orientation of geological structures relative to look direction should be considered.
When to Acquire RADARSAT Data: If surficial material assessment is based on soil surface roughness, then acquire data when the moisture levels are low to ensure that the backscatter is more closely correlated to surface roughness than it is to moisture content.
Sedimentology RADAR RESPONSEMapping Unconsolidated sediments, such as those deposited by glaciers, are often
manifested as topographic relief. The ability to image sedimentological units is a result of the side-looking configuration of radar which highlights topographic relief. Sediments also have characteristic grain sizes with different moisture holding capacities, and they may produce a characteristic surface roughness. Radar is sensitive to changes in moisture and roughness, which results in contrasting backscatter between different sediments. Each consolidated type shows unique erosional patterns, including karsting in carbonates and bedding in clasticenvironments.
3 - 1 6 R A D A R S A T G E O L O G Y H A N D B O O K
V I S U A L I N T E R P R E T A T I O N O F R A D A R S A T I M A G E R Y 3 - 1 7
GEOLOGICAL ACTIVITY RESPONSE AND RECOMMENDATIONS
RECOMMENDATIONSRADARSAT Beam Mode: All beam modes are suitable for sedimentology mapping. The final beam mode selection is dependent on the areal coverage and the level of detail required. Generally, Fine and Standard beam modes are best suited for detailed sedimentology mapping, while Wide and ScanSAR are better for basin-wide sedimentology mapping.
RADARSAT Incidence Angle: If sedimentology mapping is carried out based on the delineation of topographic relief, shallow incidence angles are ideal for enhancing subtle terrain features. If sedimentology mapping is carried out based on soil moisture differences, steep incidence angles are preferred to minimize the backscatter associated with soil roughness. If sedimentology mapping is carried out based on the determination of surface roughness, shallow incidence angles are better as they maximize the contrast in surface roughness.
Look Direction: In areas of moderate to high relief, acquisition of both ascending and descending passes allows the true form of topographic features to be represented.
When to Acquire RADARSAT Data: If sedimentology mapping is carried out based on soil moisture or surface roughness, then the data should be acquired when the vegetation is at a minimum to avoid having the vegetation response to RADARSAT’s energy dominate the backscatter.
Landslide Hazard RADAR RESPONSEAssessment RADARSAT Beam Mode: Landslide hazard areas are defined when the
locations of past landslides are identified. Landslides change the landscape through the transportation of vegetation and soil, thus affected areas have different canopy and soil roughness than surrounding unaffected areas. Radar is sensitive to these variances in roughness, and produces contrasting backscatter between affected and unaffected areas.
RECOMMENDATIONSRADARSAT Beam Mode: Both Standard and Fine beam mode can be usedto obtain detailed information on individual landslides.
RADARSAT Incidence Angle: Shallow incidence angles will minimize geometric distortions associated with areas of moderate to high relief.
GEOLOGICAL ACTIVITY RESPONSE AND RECOMMENDATIONS
Look Direction: For moderate to high relief terrain, acquisitions of both ascending and descending passes maximizes the amount of landslide information.
When to Acquire RADARSAT Data: Acquire data when landslide information is required.
Coastal Erosion RADAR RESPONSEAssessment A smooth water surface is a specular reflector, which results in low
backscatter. The rougher surface of the land however, is a diffuse scatterer and produces relatively high amounts of backscatter. Evaluationof the change in backscatter over time allows the assessment of coastal erosion.
RECOMMENDATIONSRADARSAT Beam Mode: Depending on the level of detail required, all of RADARSAT’s beam modes are suitable for coastal erosion assessment.
RADARSAT Incidence Angle: Shallow incidence angles create the greatest contrast between water and land. At these angles there is specular reflection from the water surface while the roughness of the land surface is enhanced.
Look Direction: N/A
When to Acquire RADARSAT Data: Coastal erosion assessment deals with the delineation of the land/water boundary over time. Acquire data when the coastline needs to be monitored.
3 - 1 8 R A D A R S A T G E O L O G Y H A N D B O O K
2
3
4
5
6
COMPARISON OF SATELLITE IMAGING SYSTEMS
THE RADARSAT SATELLITE
VISUAL INTERPRETATION OF RADARSAT IMAGERY
IMAGE ENHANCEMENT OF RADARSAT DATA
VALUE-ADDED RADARSAT PRODUCTS
SUMMARY
REFERENCE MATERIALS
1
I M A G E E N H A N C E M E N T O F R A D A R S A T D A T A 4 - 1
RADARSAT’s multiple beam modes and positions combined with the available
levels of processing provide geologists with an extensive selection of digital
products. To efficiently access the information contained within these products,
one must understand the digital characteristics and the implications of the
selected image processing procedures. Before the imagery is interpreted, the
RADARSAT image is calibrated to compensate for errors related to satellite
movement, the SAR antenna and ground reception. Calibration provides a
means of ensuring the consistency of data quality both within the image
(relative calibration) and from one image to another (absolute calibration).
Various digital image processing techniques used to analyze radar images are
discussed in the following sections.
H A R D C O P Y P R O D U C T S
Hardcopy products are often the most convenient to use. They are portable and
easy to handle in remote locations. These print and film products can also be
utilized as field maps and interpreted directly to extract application information.
Hardcopy RADARSAT products contain the actual image plus auxiliary
information.
D I G I T A L P R O D U C T S
Digital RADARSAT products can provide the geologist with greater flexibility
in how the data is manipulated and utilized. Digital interpretations are less
labour-intensive and less costly than visual interpretations. Digital 8-bit images
contain 256 grey levels, while the human eye can only distinguish approximately
16 to 32 distinct grey tones. Therefore, a visual interpretation alone may not be
sufficient to notice all the subtle radiometric differences in a radar image.
Speckle
SAR imagery is characterized by the presence of speckle, which is a “salt and
pepper” effect occurring throughout the radar image. The SAR sensor transmits
thousands of pulses for a given area on the ground defined as the resolution
cell. The pulses are reflected from many scattering points within the resolution
cell, but the motion of the satellite causes each received pulse to be phase-shifted.
When all the pulses are added vectorially, resolution cells within a homogeneous
ground region will have a different backscatter value. A seemingly
homogeneous surface area will have an irregular distribution of light and dark
pixels, producing a granular effect. The brightness variation is called speckle.
Speckle is a phenomenon of the technology used in the creation of the image
and is not related to the imaged target. There will always be high pixel
variability even if the landcover is uniform.
Speckle can be controlled at the initial image processing stage by data sampling.
Any attempts to rid the image of speckle will result in a loss of potentially
important information; therefore, one must carefully analyze the methodology
used, such that the information loss is minimized and the overall smoothing
effect increases image interpretability.
Speckle Removal
The most common form of speckle removal is filtering, which employs various
techniques of pixel averaging to smooth out the speckle. The extent of the
smoothing varies with the filtering technique and how it is implemented.
Filtering techniques are included in the radar modules of commercially available
image analysis software packages. Mean and median filters are examples of
non-adaptive filters which can be used to despeckle, and the Frost and Kuan are
examples of adaptive filters. Figure 4.1 is an example of a RADARSAT subscene
which has been filtered to remove the speckle. The sedimentary geological
feature in this remote area of Indonesia is more clearly emphasized when the
Kuan and/or Gamma filters are applied to remove speckle. These adaptive
filters smooth speckled areas while preserving point target detail. Additional
information regarding the digital manipulation of RADARSAT imagery can be
obtained from the RADARSAT Data Processing and Integration Handbook.
4 - 2 R A D A R S A T G E O L O G Y H A N D B O O K
I M A G E E N H A N C E M E N T O F R A D A R S A T D A T A 4 - 3
FIGURE 4.1: Unfiltered vs. filtered RADARSAT images
Fine beam position 4: acquired March 21, 1996, area: 3.5 x 5 km, subscene. RADARSAT data
Brown, R.J., B. Brisco, et al., 1996. RADARSAT Applications: Review of
GlobeSAR Program. Canadian Journal of Remote Sensing, Vol 22, No. 4,
pp. 404-419.
CASI, 1993. Special Issue: RADARSAT. Canadian Journal of Remote Sensing,
Vol 19, No. 4, entire issue.
CASI, 1994. Special Issue on Radar Geology. Canadian Journal of RemoteSensing, Vol 20, No. 3, entire issue.
Chigne, N., F. Dekker, et al., 1996. Spaceborne Radar (ERS-1 and
RADARSAT-1) Tests for Hydrocarbon Exploration in Venezuela.
II AAPG/SVG International Congress and Exhibition, Caracas, Venezuela,
8-11 September, pp. A9-A10.
Dekker, F., and D. Nazarenko, 1994. Radar Offers Many Unique Benefits as an
Exploration Tool in Tropical Environments. Earth Observation Magazine, Vol 3, No. 5, May, pp. 26-29.
Dekker, F., 1994. A Comparison of Various Remote Sensing Tools, IncludingAirborne and Satellite SAR, for Hydrocarbon Exploration in Tropical RainForests: Projected Usefulness of RADARSAT For Geological Applications. Report
prepared for RADARSAT International, #93-110, 54 pp.
Evans, D.L., T.G Farr, et al., 1986. Multipolarization Radar Images for
Geologic Mapping and Vegetation Discrimination. IEEE Transactions onGeoscience and Remote Sensing, Vol GE-24, No. 2, pp. 246-257.
Harris, J.R., R. Murray, and T. Hirose, 1990. IHS Transform for the
Integration of Radar Imagery with Other Remotely Sensed Data.
Photogrammetric Engineering and Remote Sensing, Vol 56, No. 12, December,
pp. 1631-1641.
Lillesand, T.M, and R.W. Kiefer, 1994. Remote Sensing and ImageInterpretation. John Wiley and Sons, Inc., Toronto, ON, Canada, 750 pp.
Mahmood, A.S., Carboni et al., 1996. Potential Use of RADARSAT in
Geologic Remote Sensing. Eleventh Thematic Conference and Workshop onApplied Geologic Remote Sensing, Las Vegas, NV, 27-29 February, 1995,
pp. I475-I484.
Thompson, M.D., and B.J. Mercer, 1996. Digital Terrain Models. EarthObservation Magazine, March, pp. 22-26.
RADARSAT International, 1995. RADARSAT Illuminated: Your Guide toProducts and Services. Unpublished manual, 60 pp. (available from RADARSAT
International).
RADARSAT International, 1996. RADARSAT Data Processing andIntegration Handbook. Unpublished manual, 30 pp. (available from
RADARSAT International).
RADARSAT International, 1997. Hydrocarbon Exploration UsingRADARSAT: Looking for Oil and Gas in Venezuela. Unpublished
brochure (available from RADARSAT International).
Rossignol, S., and K.P. Corbley, 1996. Reconnaissance by Radar. CanadianMining Journal, Vol 117, No. 6, December, pp. 13-16.
Sabins, F.F. Jr., 1987. Remote Sensing: Principles and Interpretation.W.H. Freeman and Company, 3rd Ed., New York, 404 pp.