After you clearly understood the problem that needs to be solved, you need to determine or choose the right imagery to be obtained for extracting the information we need. There a large number of data types are available to suit various needs.
1. If we are studying thrust belts, we may wish to order 1:1,000,000 B/W single-band Landsat multispectral scanner (MSS) images with 80 m resolution and large area coverage (185 × 185 km) to make a mosaic of part of a continent.
2. If we are doing a basin analysis, then a color MSS image at 1:250,000 or 1:500,000 should be acceptable.
3. If we are mapping details of lithologic and facies changes, or vegetation patterns and wildlife habitats, we can use color SPOT multispectral (XS) imagery with 20 m resolution and 60 × 60 km (vertical view) coverage at a scale of 1:100,000 or 1:50,000, or Color Landsat Thematic Mapper (TM) imagery, with 30 m resolution, and covering 185 × 185 km at a scale of 1:100,000 would also work.
4. For mapping alteration associated with mineral deposits, the ideal choice is high resolution (1–10 m) airborne hyperspectral imagery. If that is not available or is too expensive and the area is large and remote, we may wish to use Landsat TM for a reconnaissance look.
5. If we want to know where a well was drilled several years ago in a poorly mapped part of the world, we chose a SPOTP high resolution (10 m) panchromatic image or 5 m resolution.
6. For very fine detail, GeoEye or WorldView images have approximately > 1 m resolution. These can be enlarged to > 1:25,000 and still appear clear and sharp.
The following factors should be taken into account when ordering data (Dekker, 1993).
One factor to keep in mind is that the smaller the area covered, the higher the cost per unit area. Airphotos or airborne imagery will almost always cost more per unit area than satellite images.
If the imagery is needed immediately, one can screen grab free but rather low-quality imagery from internet sources such as Google Earth or purchase off-the-shelf data from a government agency or vendor. Large image archives exist, and data can often be obtained quickly. Purchasing custom images from vendors and consultants, or processing digital data in your own shop can take up to several weeks.
A- Large area coverage can be obtained using weather satellites such as GOES (covers a full hemisphere), the Advanced Very High Resolution Radiometer, which covers a 2700 km swath on the Earth’s surface, or the SeaWIFS instrument, with a 1502–2801 km swath width.
B- Moderate size areas can be covered using some handheld Shuttle (and other mission) photos (variable area coverage), as well as Landsat images (MSS and TM), which cover 185 × 185 km. Systems that cover 50 × 50 to 500 × 500 km include MK-4 photos (120–270 km), KATE-200 photos (180 × 180 km), KFA-1000 photos (68–85 km), and the SPOT systems (XS and P) that cover 60 × 60 km.
Satellite images generally cover larger areas than airborne photos or images, and the synoptic view is one of their greatest advantages.
C- For field or local studies, airborne surveys or small area satellite images will save cost and/or provide more detail. Recent satellites, such as Worldview 2, 3 and GeoEye-1, have very high resolution (up to 31 cm for the panchromatic sensor and 1.65 m for the multispectral instrument) and cover correspondingly smaller areas (16.4 and 15.2 km swaths, respectively).
The scale of the final image will to some extent be a function of the sensor system resolution, in that one cannot enlarge, say, an image with 80 m resolution to a scale of 1:100,000 without the image becoming “pixelated,” that is, breaking up into the individual resolution elements that appear as an array of colored squares.
5. Nighttime Surveys
Thermal and radar surveys can be flown effectively at night because neither system relies on reflected sunlight: the radar instrument illuminates the surface by providing its own energy source, and thermal energy is radiated from the surface.
Predawn thermal imagery reveals, among other things, lithologic contrasts related to differing rock and soil densities or color tones (light versus dark). Nighttime thermal imagery can reveal shallow groundwater and moist soil (generally warmer than background), can detect oil spills on water, and can help map underground coal mine fires. Because radar illuminates the ground with microwaves, it can be flown at night to map oil spills, for example, or during polar night to map the movement of ice floes that could threaten an offshore oil rig or platform.
6. Seasonal/Repetitive Coverage
Certain seasons are better for specific surveys. For example, a geologic mapping project in an area covered by temperate forest would see more of the ground in spring before deciduous plant leaf-out or in the fall, after leaves have dropped.
High sun elevation angle (summer) provides images with the best color saturation, which can be useful when mapping lithologies in low contrast areas. On the other hand, low sun angle images (flown during the morning or in winter), especially with a light snow cover, enhance geologic features in low-relief terrain.
If repetitive coverage is needed to monitor natural (e.g., flooding, ice floes) or man-made (e.g., drilling, roads) changes, it is often most cost effective to use satellites because of their regular repeat cycles. Repeated aircraft surveys provide more detail but are much more costly.
Low-relief terrain may require low sun angle or grazing radar imagery to enhance subtle topographic and structural features.
On the other hand, high-relief terrain poses the potential problem of large shadowed areas that can obscure important areas or details. These areas should be flown during mid-day or using radar with a steep depression angle to minimize shadows.
8. Vegetation Cover
Color infrared images are very sensitive to changes in vegetation type or vigor, since the peak reflection for vegetation is in the near infrared region. Combinations of infrared and visible wavelengths have been used to map changes in vegetation related to underlying rock types and even hydrocarbon seepage (Abrams et al., 1984).
Lidar which will also provide an image of the top of the vegetation canopy that looks like topography. There are some image processing methods, called “vegetation suppression” techniques, which appear to remove vegetation and reveal subtle changes in the underlying soil or bedrock. These algorithms tend to remove the reflectance attributed to vegetation and enhance the remaining wavelengths.
9. Water-Covered Areas
Shorter wavelengths (blue and green light) penetrate water farther than longer wavelengths. The euphotic or light-penetrating zone is known to extend to 30 m in clear water (Purser, 1973).
Infrared light, which has longer wavelengths than visible light, is absorbed by water and does not provide information on bottom features.
Landsat TM, with its blue band, is excellent for mapping shallow water features such as shoals, reefs, or geologic structures. Likewise, true color and special water penetration films such as Kodak Aerocolor SO-224 have excellent water penetration capabilities (Reeves et al., 1975). Side-scan sonar is available for shallow and deep water mapping, and produces images of the sea bottom reflectance using acoustic energy, much like radar uses microwave energy to produce an image.
Abrams, M.J., J.E. Conel, H.R. Lang. 1984. The Joint NASA/Geosat Test Case Project Sections 11 and 12. Tulsa: AAPG Bookstore.
Dekker, F. 1993. What is the right remote sensing tool for oil exploration? Earth Obs. Mag. 2: 28–35.
Junge, C.E. 1963. Air Chemistry and Radioactivity. New York: Academic Press: 382 p.
Purser, B.H. 1973. The Persian Gulf. New York: Springer-Verlag: 1–9.
Reeves, R.G., A. Anson, D. Landen. 1975. Manual of Remote Sensing, 1st ed, Chap. 6. Falls Church: American Society of Photogrammetry.
U.S. Bureau of Land Management. 1983. Aerial Photography Specifications. Denver: U.S. Government Printing Office: 15 p.