Civil, Commercial, and International Remote Sensing Systems and Geoprocessing
David L. Glackin and Gerard R. Peltzer
Chapter 1: Civil, Commercial, and International Remote Sensing (cont.)
1.2 Commercialization and Small Satellites
The concept of hybrid government/commercial systems was developed during the 1980–1997 period. These systems typically involve series of satellites whose development is initially funded by the government. The data are sold by commercial entities, and the satellites are increasingly taken over by those commercial entities. The notable examples are Landsat in the United States, which has met with decidedly limited commercial success; SPOT in France; and RADARSAT in Canada.
The concept of fully commercial systems was also developed during the 1980–1997 period, but the systems that were launched were restricted to satellites of the microsat class (10–100 kg). Perhaps most notable in this class were the commercial microsats from SSTL in the United Kingdom. The first of these microsats, UoSAT-5, was launched in 1991. UoSAT-5 provided "snapshots" of the Earth acquired under the control of the United Kingdom but made available to ham radio operators worldwide. SSTL built and sold similar microsats to South Korea (KITSAT-1) and Portugal (PoSat) in a program of international technology transfer. The number of countries poised to own SSTL microsats or similarly sized microsats is large. The first commercial weather satellite, OrbView-1/Microlab-1, was developed in the United States. This satellite proved the value of microsats in performing new science, by observing global positioning system (GPS) satellites as they set behind the Earth's limb, using them as beacons to probe the atmosphere and ionosphere.
The concept of remote sensing satellites in the minisat class (100–500 kg) was developed during the 1980–1997 period. The first minisat was launched by the United States as part of Mission to Planet Earth (now called the Earth Science Enterprise). This minisat was the TOMS/Earth Probe, designed to measure the total ozone content of the atmosphere. Many countries developed plans during this period to field their own minisats (Fig. 1.2), to be built either in their country or by other countries. Small satellites can provide a significant cost savings because their smaller mass requires a less powerful launch vehicle. The launch vehicle can represent a large portion of the total mission cost, and large launch vehicles are very expensive. Small satellites provide a good way for the less experienced countries to gain technological expertise by starting small, then working up to more capable satellites, if desired. Much beneficial work can be performed with small satellites, which can be focused on specific areas of interest for a given user community.
Fig. 1.2. Proliferation of small satellites for Earth observation, 1980–2007. |
1.2.1 Spatial and Spectral Resolution
Through 1995, the number of spectral bands available in spaceborne visible/infrared (vis/IR) imaging instruments for Earth observation remained relatively small (under 10). This number increased substantially in 1996, most notably with the launch of the first spaceborne imaging spectrometers for Earth observation. These spectrometers were built in Germany. They were launched on the Indian IRS-P3 satellite and on the Priroda module that mated with the Russian Mir space station. These initial imaging spectrometers were limited to imaging in under 20 spectral bands. Several missions involving more ambitious hyperspectral imaging spectrometers were being planned (Fig. 1.3), and the number of spectral bands available from spaceborne instruments will increase dramatically in the near future.
Fig. 1.3. Diverging trends in spectral bands for civil spaceborne visible/IR imaging instruments, 1980–2007. |
The spatial resolution of imagery systems operating in the visible region of the spectrum increased slowly during the 1980–1997 period. If we examine the spatial resolution of spaceborne systems having a stereo imaging capability (Fig. 1.4), we see that the Resurs film-based systems of the former Soviet Union already provided up to a 4-m capability in 1980. These systems do not, however, provide imagery in real time, and there have been notorious difficulties associated with acquiring the archival imagery from the systems. If we restrict ourselves to digital imaging systems, we see that the first substantial increase in capability over SPOT was provided by the 6-m PAN instrument on the Indian IRS-1C satellite. The situation in spatial resolution is poised to change dramatically, as numerous commercial 1-m systems come on line in the future, and as systems with resolutions from 0.8 to 20 m having a stereo capability are about to be launched.
|
Fig. 1.4. Proliferation of stereo visible capability for topographic mapping, 1980–2007. |
Visible and IR instruments operate in the electro-optical region of the spectrum, where the location within the spectrum is denoted by the wavelength of radiation, measured in microns. The visible region extends from 0.4 to 0.7 µm; the near-IR, from 0.7 to 1.5 µm; the short-wave IR, from 1.5 to 3.0 µm; the mid-wave IR, from 3.0 to 5.0 µm; and the long-wave IR, from 5.0 to 15 µm. The visible through short-wave IR is dominated by sunlight, so imaging is normally restricted to the daytime. The long-wave IR is dominated by natural thermal radiation emanating from the Earth and its environment, which can be seen day or night. The mid-wave IR contains a combination of both solar and thermal radiation in the daytime. Because of basic physics, it is far easier to achieve high spatial resolution in the visible than in the IR. Clouds, with the exception of thin cirrus clouds, cannot be substantially penetrated with visible or IR imagery.
There are three basic classes of electro-optical imagery:
- Panchromatic imagery in a single broad spectral band ("black and white"), which is useful for maximum spatial resolution because more light is available over the wide spectral range
- Multispectral imagery in a few to tens of moderately wide spectral bands ("color"), which adds information that allows discrimination, classification, and measurement of objects based on their spectral properties
- Hyperspectral imagery in hundred(s) of narrow contiguous spectral bands
Hyperspectral imagery is collected with instruments called "imaging spectrometers." Normal spectrometers typically measure the shape of the spectrum of a single point in the scene, but they do not create images. Imaging spectrometers measure the shape of the spectrum for many points in the scene, and create images in multiple narrow spectral bands. They can be designed to output either multispectral data for, say, tens of narrow selected spectral bands, or to output hyperspectral data for, typically, 100–200 narrow contiguous spectral bands across an entire spectral range. Hyperspectral instruments offer improvements in commercial ventures such as oil and mineral exploration, precision farming, crop assessment, and forestry.
1.2.2 Stereo Imaging
The systems shown in the figure on high-resolution stereo visible systems (Fig. 1.4) are all capable of providing topographic data, through analysis of their stereo imagery. All of the digital imaging systems on free-flying satellites during this period obtained their stereo imagery through the cross-track method. In this method, the same region on the Earth is imaged at different times with a single optical system, from different vantage points on different orbits.
The film-based systems of the former Soviet Union were the only free-flying satellite systems that acquired their stereo imagery through the in-track method. In this method, separate optical systems obtain imagery by looking fore and aft along the satellite's ground path. However, film is more difficult to work with, as it must first be digitized and registered. Digital in-track stereo imaging instruments, in which the looks of the fore and aft (and nadir) at the same regions are almost inherently registered, were pioneered during this period by the Germans. The Germans built the MOMS-02/D2 instrument that flew on the Shuttle, as well as the MOMS-02P instrument that is now flying on the Russian Mir-Priroda. These instruments paved the way for the higher resolution in-track digital systems that are now poised to fly. These systems, when routinely available, will be useful in mapping, road route selection, communications equipment siting, disaster response, and news "fly-throughs." They will even be useful in the visualization of real estate, with its associated potential problems of mud slides or flooding on a given piece of property. In many cases, 3- to 5-m resolution is sufficient, while in other cases, 1-m resolution is needed.
1.2.3 Ocean Color
The fishing industry could benefit substantially from the availability of timely ocean color imagery because this imagery can indicate where fish congregate and feed. Appropriate selection of spectral bands in the visible spectrum allows ocean phytoplankton to be mapped in terms of a color code that is easily understood by personnel with a minimum of training. The nutrient-rich regions, which contain the phytoplankton upon which fish feed, are easy to discern, but they also tend to be the cold-water regions in which most fish do not like to live. The fish tend to congregate along the boundaries of these regions, swimming into the cold-water regions to feed.
During the 1980–1995 period, the only source of ocean color imagery was the coastal zone color scanner (CZCS) instrument on NASA's Nimbus-7 scientific satellite. When this instrument failed, there was a decade-long gap with no spaceborne ocean color imagery. Then in 1996, ocean color imagers were launched by India on IRS-P3, by Japan on ADEOS-1, and by Russia on the Mir-Priroda module (Fig. 1.5). Many other countries are poised to join this group.
|
Fig. 1.5. Proliferation of ocean color imagers, 1980–2007. |
1.2.4 Airborne Sensing
Much of the commercial remote sensing market remained the province of airborne remote sensing instruments and companies during this 1980–1997 period. In particular, for applications that required a spatial resolution greater than that which could be provided by SPOT (i.e., 1-m rather than 10-m imagery), the airborne companies had the market locked up. The companies that are poised to field 1-m spaceborne systems (e.g., Space Imaging, EarthWatch, and ORBIMAGE) will be attempting to penetrate that market.
The ultimate impact of these high-resolution airborne commercial systems will be determined by a combination of cost effectiveness, product quality, refresh rate, and timeliness of product delivery to the user. The user community will want end products that can be applied transparently to their problems, regardless of the source of the raw data. The companies that plan to field high-resolution commercial systems have identified niche markets and have specifically targeted those markets with their planned systems. The commercial markets include mapping, urban planning, disaster assessment for the insurance industry, pipeline and power line monitoring, communications equipment siting, news reporting, and real estate visualization. If it turns out to be more cost effective to acquire a strip of imagery along a pipeline with a spaceborne system rather than an airborne system, then the 1-m spaceborne system should significantly penetrate that particular market. For markets where very fast tasking and timeliness is of the essence, certain airborne systems may retain an advantage.
Ultimately, it can be expected that market forces will cause data from airborne systems or data from spaceborne systems to be used where they are most effective and available. The spaceborne data will surely displace some fraction of the existing airborne market, but in the end, the two sources of data may prove to be complementary.
1.2.5 Synthetic Aperture Radar
The other part of the electromagnetic spectrum in which remote sensing imaging is routinely carried out is the microwave region, and the techniques used for this imaging include the active technique of synthetic aperture radar (SAR) as well as passive microwave imaging. Active techniques emit beams of radiation that interact with the environment, while passive techniques use the radiation that is emitted naturally by the environment. One advantage of microwave systems is that they can operate day or night. In addition, they can penetrate nearly all kinds of weather phenomena, including most clouds. Thus, portions of the globe can be viewed that are usually cloud covered, such as the polar ice caps and many tropical countries. Microwave instruments are ideal for imaging sea ice during the polar winter when the polar region is not illuminated by the sun, and they can see through the persistent cloud cover that often plagues these regions. SAR offers higher spatial resolution than passive techniques. For these reasons, Canada's remote sensing program is focused on RADARSAT. The capabilities of SAR also explain the interest of many tropical countries.
The commercial applications of microwave remote sensing lie predominantly in SAR. In addition to the capabilities already mentioned, SAR systems can be used to monitor sea ice for ship routing and drilling rig safety, to monitor river ice buildup for shipping, to map oil spills, and to assess agricultural and forestry issues. Interferometric SAR, employing two views that are sufficiently close in time, from two different locations, can be used to monitor surface motion resulting from earthquakes and volcanoes, and to create digital elevation models (DEMs). The commercial exploitation of SAR is yet to be fully realized.
Table 1.1. SAR Summary, 1980–1997
| System | Country | Bandsa | Polarizationb | Best resolution (m) | Best swath width (km) |
|---|---|---|---|---|---|
| Almaz-1a | Russia | S | H/H | 15 | 45 |
| ERS-1 and -2 | Europe | C | V/V | 20 | 100 |
| JERS-1 | Japan | L | H/H | 18 | 75 |
| Kosmos-1870 | Russia | S | H/H | 25 | 35 |
| SIR-A | United States | L | H/H | 40 | 50 |
| SIR-B | United States | L | H/H | 25 | 60 |
| SIR-C | United States | L, C | H/H, H/V, V/H, V/V | 25 | 60 |
| RADARSAT-1 | Canada | C | H/H | 9 | 500 |
| Travers | Russia | L, S | H/H, V/V | 50 | 50 |
| X-SAR | Germany/ Italy | X | V/V | 25 | 60 |
|
aS = 3 GHz, L = 1 GHz, C = 5 GHz, X = 10 GHz. bH = horizontal V = vertical (send/receive). |
|||||
Throughout the 1980–1997 period, SAR remained in its infancy relative to vis/IR instrumentation, especially with regard to its commercialization. The first spaceborne SAR for Earth observation flew just before this period, on the U.S. Seasat in 1978. The United States then flew instruments on the Shuttle, while the former Soviet Union operated various free-fliers. Later in this period, ESA flew ERS-1 and ERS-2, Japan flew JERS-1, and Germany and Italy participated with the United States in the SIR-C/XSAR Shuttle mission. ERS-1 and ERS-2 are partially commercial, in the sense that users outside of the research community must pay for the data. Russia flew Almaz-1a in 1991 and attempted to commercialize it by selling the data, but did not do well in the marketplace. Commercialization of SAR came closer to reality when Canada launched RADARSAT-1 in 1995. The RADARSAT International company was formed to market the data. It is still too early to tell, because the program was mostly government funded, but RADARSAT-1 may be the harbinger of true commercialization of SAR imagery.
The SAR systems flown in this period (Table 1.1 and Fig. 1.6) variously had operating frequencies in the L, S, C, and X bands (
1–10 GHz). With the exception of SIR-C, the systems all transmitted with a single polarization (horizontal or vertical) and received with the same polarization. SIR-C was the first fully polarimetric system, in that it could send and receive with either horizontal or vertical polarization.
Fig. 1.6. International proliferation of spaceborne SAR systems, 1980–2007. |
1.2.6 Lidar
Remote sensing in the visible and IR may also be done either by passive or active techniques. The passive electro-optical imagery techniques were described previously. The active technique is called lidar (light detection and ranging), in which a laser beam is emitted by an instrument, and the returning radiation is collected by a telescope in the instrument. Lidars may range in complexity from the simple "backscatter" variety that measures clouds and atmospheric aerosols, to the complex Doppler variety that measures wind.
The application of lidar to spaceborne remote sensing is in its infancy. NASA's LITE (Lidar In-space Technology Experiment), which flew on the Shuttle in 1994, was the first lidar to fly in space that was designed to perform remote sensing of the Earth's atmosphere. The Russian Balkan-1 lidar was launched on the Mir-Spektr module in 1995, while the French ALISSA lidar was launched on the MIR-Priroda module in 1996, to study clouds, aerosols, and the Earth's surface. A potential commercial application for Doppler lidar might be to measure the wind profile throughout the troposphere and up into the stratosphere. Of all of the meteorological parameters that cannot currently be measured, wind would have the greatest impact on numerical weather prediction. Measurement of wind, in turn, could lead to improved routing of aircraft and substantial fuel savings, as well as other commercial benefits that would accrue from improved weather forecasts. However, Doppler lidar, as envisioned during much of this period, represented an extremely expensive instrument concept—one that must overcome various technical and physical hurdles to become a reality. Another potential commercial application may lie in the use of a simple backscatter lidar to map volcanic plumes, so that commercial airliners may be routed around these very dangerous phenomena.
The application of active microwave and electro-optical techniques to remote sensing during this period is summarized in Table 1.2. All of the space-based systems that employed these techniques are shown.
Table 1.2. Overview of Active Sensing Techniques, 1980–1997
| Technique | Flown In space | Systems | Applications |
| Microwave Spectrum | |||
|---|---|---|---|
| Altimetry | Yes | GEOSAT Mir-Priroda TOPEX/Poseidon |
Sea surface topography |
| Scatterometry | Yes | ADEOS 1 ERS 1,2 |
Sea surface wind |
| Real-Aperture Radar | Yes | Okean O1 Sich 1 TRMM |
Rain, cloud layers, and bases |
| Synthetic Aperture Radar (SAR) | Yes | Almaz 1 ERS 1,2 JERS 1 Mir-Priroda RADARSAT 1 Resurs 01-1 |
Sea ice, soil moisture, snow, vegetation, ocean waves and currents, geology, underground structures (dry soil) |
| Interferometric SAR (single pass) | No | — | Land surface topography, Earth movement |
| Visible/Infrared Spectrum | |||
| Backscatter lidar | Yes | LITE (Shuttle) Mir-Spektr Mir-Priroda |
Clouds, aerosols, visibility, ice topography |
| DIAL (Differential Absorption Lidar) | Noa | — | Atmospheric temperature, moisture, trace species |
| Doppler Lidar | Noa | — | Wind |
| Fluorescence Lidar | Noa | — | Oil spills, biological growth on buildings |
aLidar remains primarily airborne.
Next: 1.3 State of the Art: 1998
Previous: 1.1 Evolution of the Field: 1980–1997
