|
Satellite Usage in Natural Disaster Management
By Louis S. Walter, Ph.D. As this article is written, the media carry news of the death toll resulting from a Magnitude 6 earthquake in Colombia. Initial reports said that several hundred people died as a result of the disaster. However, as often occurs, the numbers increase greatly as the full impact of the calamity is assessed. Currently, the figure has extended to over 2,000 fatalities.
This is not an isolated event. Each year, on average, over 30,000 people lose their lives to natural disasters. Hundreds of thousands are left homeless or jobless. Economic losses are claimed to be near one billion dollars a week in the U.S. alone-though, in view of ripple effects, the full costs are very difficult to assess accurately. For example, the earthquake that struck Kobe, Japan, closed a major port, affecting the shipment of automobile parts and thus, production in the U.S. Similarly, Hurricane Mitch, which recently devastated Central America, has, for the time being, reversed economic growth and sustainability in that region.
 TRMM image of Hurricane Bonnie and her chimney cloud on 22 August 1998. Accurate and early predictions of such disasters can be accomplished using ESE data.
As we become more interconnected commercially, our vulnerability to disasters in other parts of the world expands. Likewise, as we become more dependent on technology, we become more susceptible to the destruction or interruption of sensitive communications systems. Urbanization, expanding population, and the movement of people to hazardous areas such as coastal zones increase our exposure to hazards. Our perception of the frequency and magnitude of disasters is marked by major events such as today's earthquake, which periodically make headlines. But disasters with more localized impact in remote areas, or in which only a few are killed, are not generally noted, yet their cumulative effect is as great, or greater, as those of the larger events. All in all, the cost of natural disasters is rising exponentially, so advanced technologies are increasingly being put to use to manage these calamities. Disaster Management
To understand how these technologies may be applied, we should first understand how disasters are managed and who manages them. Disaster management may be defined as the set of activities used to combat disasters, including elements of mitigation (actions taken so that hazards such as earthquakes will not result in damage), preparedness (actions to avoid a specific hazard such as warning and evacuation before a cyclone), and response after a disaster occurs.
The exponential growth in the cost of disasters is leading to a new paradigm in disaster management-one that focuses on avoiding disasters rather than responding to them and cleaning up the debris afterward. Avoiding disasters, disaster mitigation, is not a new concept. Natural hazards-hurricanes, earthquakes, volcanic eruptions-will occur and there is little we can do to stop them. Indeed, what sets natural hazards apart from other (man-made) hazards is that humans cannot control them. However, we can control our exposure to natural hazards, or mitigate the effects they can have, by adjusting to them. For example, new, though costly, construction practices can make our cities and towns much less vulnerable to earthquake shocks. But practitioners need hard information with which to convince land use planners or builders to incur such costs. Helping provide this information will be an intrinsic role for remote sensing.
Organizations involved in disaster management range from local units (police and fire departments) to state and national government emergency management offices to voluntary and international organizations. These organizations need different information, often in different scales. National emergency agencies require broad assessment of damage while local officials need detailed information in real-time. Thus, assessing information requirements must take into administrative level, as well as hazard type and disaster management element. Disaster Management Applications of Satellite Remote Sensing
Given the complexity of disaster management-the various activities, functional levels, and types of hazards-and in view of the increasing cost and importance of natural disasters, the field of remote sensing applications is particularly interesting and fertile. Hurricanes
The approaching landfall of a cyclonic storm usually is the cause of intense preparation and, sometimes, evacuation. These storms can cause intense destruction along 50 miles or more of coastline and occasionally reaching far inland. The cost of preparations and evacuation has been estimated at about one million dollars per coastline mile, making accurate and early prediction of landfall extremely important. The issue can become particularly critical in the many coastal areas connected by narrow causeways such as the Florida Keys.
 3D image of Cyclone Susan as taken by TRMM. ESE data can be used to warn and evacuate communities before a cyclone. Ideally, we would wish to be able to predict both the storm's diameter, maximum wind velocity, and landfall to within 100 miles or better up to three days before landfall. Unfortunately, this is not often attained. At present, ground-based radar and hurricane tracking aircraft provide important data. Observations from geosynchronous satellites are used to track a storm and to study the sharpness of its "eye," an indication of the storm's intensity. The ADEOS NSCAT demonstrated the utility of scatterometry for measuring sea surface wind velocities and using these data to improve weather forecasts.
NASA's Quickscat will soon replace NSCAT, which is no longer operational. In the future, satellite data, including measurement of ocean surface temperature, soundings of atmospheric humidity and temperature, and lidar measurements of tropospheric and stratospheric winds, will be analyzed through data assimilation. This will greatly improve hurricane prediction accuracies. Floods
The most frequent type of disaster, flooding, is readily detected in satellite imagery. Synthetic Aperture Radar (SAR) imagery is often advantageous because it can penetrate the clouds that often exist during flooding. However, more frequent observations are often needed to catch flooding at various stages so, in these cases, NOAA AVHRR visible/infra-red imagery can be used.
Mitigating flood disasters before they happen is becoming rec ognized as more important and effective than assessing the damage afterwards and here, remote sensing can make its biggest contribution. At the basis of flood disaster mitigation is the flood plain map which, in the U.S., provides the fundamental data for the Flood Insurance Program. The FPM is used to define the 100 year flood level and the area within which special insurance and re-construction provisions may apply. With changing land use patterns, construction, and other changes such as sedimentation and obstruction in river reaches, FPMs change and require costly and time-consuming revision, leaving many out of date. High-resolution imagery can provide the needed land-cover maps and interferometric SAR the topography for digital elevation maps (DEMs). NASA and FEMA are currently embarking on a joint test of these technologies to see if they can be used operationally. Earthquakes
In contrast with flooding, earthquake disaster mitigation is much more complex but certainly no less important. The difficulties lie not only in the high cost of earthquake-proof construction but also in our poor knowledge of the risk. At present, successful earthquake predictions are insufficiently reliable and there is a real question as to whether they will ever be adequately dependable except under special circumstances. Thus, attention focuses on risk assessment. Reliable assessments are needed so that builders and planners can accurately judge whether preventive measures (e.g., structural reinforcement) are needed or warranted. But, how do we assess risk?
Past and current seismicity provide the basic information that earthquakes happen where they have happened before and that major earthquakes are more likely where small earthquakes are occurring now. For this, seismic and geologic data are used. Thanks to space technology and the Global Positioning System (GPS), we are now able to make precise positioning measurements to determine the rate of movement along faults. Where there is no movement, these faults may be locked, and, it is reasoned, earthquakes are more likely when the lock ruptures. Another important clue comes from the study of earthquake source mechanisms. Here, the technique of SAR interferometry can reveal new information. This technique directly compares two SAR waveforms acquired before and after an earthquake. Using such an interferogram, the illustration shows the surface movement-and inferred tectonic changes-which took place during the California Landers earthquake in 1992. This information helps geophyicists better understand seismic vulnerability in such active areas. The Future
As national, international, and commercial interests combine to set new records in the number and quality of Earth observing satellites, we may hope to turn the corner on the growing frequency and cost of disasters. Using their data, by working closely with disaster management practitioners, using newly developed understanding and models, as well as increasingly rich geographic information data bases, we can look forward to providing information which will be used, not only to respond to disasters, but to prevent them. About the Author:
Dr. Louis S. Walter works in the Institute for Crisis, Disaster, and Risk Management at George Washington University.
Back
|
