PE&RS September 2002

Highlight Article

Lending A Helping Hand: Using Remote Sensing to Support the Response and Recovery Operations at the World Trade Center
Ray A. Williamson and John C. Baker

Please click any of the images if you would like to view a larger version

This one-meter resolution satellite image of Manhattan, New York was collected at 11:43 a.m. EDT on Sept. 12, 2001 by Space Imaging's IKONOS satellite. The image shows an area of white and gray-colored dust and smoke at the location where the 1,350-foot towers of the World Trade Center once stood. (courtesy of Space Imaging)

On the morning of last September 11th, the nation was left stunned and outraged by ruthless attacks, resulting in thousands of fatalities from passenger aircraft crashes into the World Trade Center towers and the Pentagon. In the hours, days, and weeks following these disastrous events, the nation rallied to support the emergency responders and other professionals and volunteers who participated in the round-the-clock rescue and recovery operations. This article offers an overview of the role that remote sensing played in providing timely data and information on the nature of the World Trade Center (WTC) site in the weeks following the 9/11 attacks. The imagery collected and distributed was largely intended to provide practical information for use by emergency response teams operating at the WTC site, as well as to give decision makers a better sense of the nature and magnitude of the disaster during a time of substantial uncertainty. The imagery and other geospatial data were used to assess the conditions in and around the collapsed and burning rubble and to support emergency response operations.

The nation’s response to the emergency illustrated the considerable value that remote sensing technologies have for gathering timely, critical information. It also exposed numerous problems in the sharing of geospatial and other information among the many groups that were trying to help. The response further illustrated the importance of being better prepared to use these technologies, not only to respond to such catastrophic events, but also to prevent them. It is especially important now, one year after the episode, to examine the historical events in detail, to bring together insights from on-site personnel, and to review the lessons learned from the WTC experience that could be applied to precluding and handling future disasters. This essay offers a modest contribution by reviewing the diverse types of remote sensing technologies and techniques that were brought to bear under short notice and very challenging circumstances to support the rescue and response operations.

A Catastrophe of Unexpected Magnitude
The terrorist attacks succeeded in inflicting unimaginable devastation on the World Trade Center. The crash of American Flight #11 into the North Tower (1 WTC) was followed moments later with the crash of United Airlines Flight #175 into the South Tower (2 WTC). The raging fires ignited within each structure by the impacts of these fuel-laden aircraft led to the collapse of each tower within 105 minutes of the first crash. Along with the heartbreaking loss of nearly 3,000 lives, including an unprecedented number of firefighters, police, and other public safety officials, the surrounding buildings in the WTC complex suffered extensive damage from debris and fire. Some of these buildings eventually collapsed as well.

From the first hours after the attack, therefore, the emergency responders from various New York City departments, New York state agencies, and the Federal Emergency Management Agency (FEMA) and other federal agencies were confronted with a scene of previously unimaginable devastation, very much like a war zone. Communications, power, and transportation facilities were severely disrupted and it was impossible to know with certainty where it was safe to travel. Rescue operations commenced with the search for survivors among the rubble or in the remaining buildings and structures in the WTC complex. Unfortunately, the search for the missing soon became a protracted recovery operation as firefighters, urban search and rescue (USAR) teams, construction workers, and others shifted their efforts to recovering and identifying those missing and presumed dead.¹

The response and recovery operations were complicated by the loss of the City’s Emergency Management Center, located at 7 World Trade Center, which was wrecked by the falling debris from the twin towers. Within a few days, the City’s geospatial capabilities were largely reconstituted at the Pier 92 Command Center in the form of the Emergency Mapping and Data Center (EMDC) thanks to prompt assistance from other city departments, volunteers, private firms, and local centers of geospatial expertise.

Along with a broad range of other types of geospatial data, up-to-date overhead imagery played a significant role in helping the emergency response teams to orient themselves to the confusing and dangerous “Ground Zero” landscape that was created in the aftermath of the collapsed towers and several of the surrounding buildings. These images were important for several reasons: (1) current images of the changing scene at the WTC location were useful for orienting the emergency worker in an environment nearly devoid of familiar landmarks; (2) smoldering fires within the rubble created potential hot spots that could flare up to threaten the safety of emergency workers operating at the site; (3) emergency responders needed an accurate method for estimating the changing volume of the rubble pile as material was carefully screened and trucked away; and (4) imagery was valuable in helping planners create open transportation routes for moving in and out of the affected zone.

Satellite Imagery

Figure 1. Space Imaging IKONOS image of World Trade Center site acquired on September 15, 2001. (Courtesy of Space Imaging)

Earth observation satellites were able to acquire images in the first days following the attacks at a time when civil and commercial aircraft were largely prohibited from flying. One of the first remote sensing satellite images of the disaster site was acquired by a French SPOT satellite on September 11th about three hours following the first aircraft crash. The multispectral image with 20-meter resolution revealed the plume of smoke and dust, as well as identified the fire hot spots using the infrared band.² Similarly, Space Imaging’s IKONOS commercial observation satellite acquired very timely images of the devastated location on September 12th. Several versions of the pan-sharpened panchromatic and multispectral images (Figure 1) were placed on Space Imaging’s Web site to provide broad public access. The company donated IKONOS imagery data of the affected location to various city, state, and federal agencies involved in the response and recovery operations.³ Other civilian imaging satellites, including Landsat 7 and NASA’s Moderate-resolution Imaging Spectroradiometer (MODIS) instrument on its Terra satellite also acquired lower-resolution, multispectral images in the first few cloud-free days following the September 11th attacks. These satellite images provided a useful perspective view of the disaster site. Nevertheless, emergency responders looked to aerial sources to provide even more detailed remote sensing data for supporting their specific information needs for rescue and recovery operations.

Aerial Data
Several types of aerial imagery were provided to the city, state, and federal organizations supporting the rescue and response operations during the first days and weeks after the attacks. Both government agencies and commercial firms played important roles in making available and operating airborne sensors that produced overhead data as part of the supply of time-urgent information needed by the emergency response organizations operating at Ground Zero. Early images of Ground Zero and surrounding areas were collected by the Photo Unit of the Fire Department of New York (FDNY), which took digital images from low flying police helicopters using handheld cameras.⁴ The unit provided CDs of the images of the city’s changed topography to the Department’s Phoenix Unit for supporting its emergency logistics planning and transportation routes.  

Figure 2. Lidar image visualization of Ground Zero at the World Trade Center (courtesy of DoD's Joint Precision Strike Demonstration Project Office).

In the first days and weeks following the 9/11 attacks, several aerial imaging operations were commissioned or approved by New York City and relevant New York State authorities, such as the New York Office for Technology (NYSOFT). One of the first imaging operations to begin operating used a combination of sensors flown on a Navajo Chieftain aircraft operated by EarthData, a private firm (see also the article and figures on page 877 in this issue of PE&RS). Over the course of the five weeks, EarthData flew more than 40 missions to collect timely information needed by the emergency responders and other personnel working at Ground Zero. The EarthData sensor suite included a LIght Detection and Ranging (lidar) system, a high-resolution digital camera, and a thermal camera. Lidar is an active laser remote sensing system that can rapidly collect highly accurate spot elevation data.⁵ These data can be used to generate a digital three-dimensional perspective of surface features through post-processing with very accurate platform position and orientation information. Such images provided emergency planners at Ground Zero with a highly accurate means of measuring the substantial topographic change resulting from the collapse of the twin towers and the subsequent removal of the rubble pile. In addition, the thermal camera played a role in assisting the firefighters and rescue crews to identify and measure potentially hazardous hot spots in the rubble. Although the EarthData aircraft operated out of Albany, the EarthData capability for rapidly processing the data collected during each mission ensured that the resulting image products could be delivered to the data analysts at the EMDC in New York City in less than one day.

Certain national capabilities were also brought to bear on supplying remote sensing data to support the operations at Ground Zero. On September 14th, the Department of Defense (DoD) directed that the Rapid Terrain Visualization (RTV) project provide support to recovery efforts both at the WTC and the Pentagon. RTV is a technology demonstration program that is developing an operational capability for high-resolution terrain mapping and visualization using both lidar and interferometric synthetic aperture radar (IFSAR) data for supporting military operations.⁶ The testbed airborne system, a de Havilland DHC-7, was already engaged abroad at the time. The program manager, the U.S. Army’s Joint Precision Strike Demonstration Program Office, took steps to provide an expedient capability by assembling a team from government, from industry, and from the University of Florida’s Geoscience and Remote Sensing Department. Lidar equipment provided by Optech International was integrated into a Cessna Citation jet provided by the National Oceanic and Atmospheric Administration (NOAA). The aircraft flew missions to collect lidar data (Figure 2) of the WTC site on September 23rd and 24th, as well as two follow-up missions on October 15th and 16th that also included collection of overhead data of Lower Manhattan.

Figure 3. High-resolution, oblique image of damaged buildings near the World Trade Center location. (courtesy of Pictometry)

In addition, the NOAA aircraft acquired high-resolution photographs of the same area using a Leica/LH systems RC30 camera. The University of Florida experts assisted with GPS collection and the processing and analysis of the lidar data. The resulting images achieved 15-centimeter accuracy based on a combination of lidar data, aerial photography, and accurate Global Positioning System (GPS) measurements tied to the National Spatial Reference Systems. The data were delivered in both point cloud and grided format, along with software for visualization and analysis. This highly accurate spatial reference data provided city and federal organizations (e.g., New York City Office of Emergency Management, Fire Department, FEMA) operating at Ground Zero with a reliable 3-dimensional perspective of the disaster scene that emergency responders could use to locate destroyed structures, such as elevator shafts, stairwells, and utilities, as well as support recovery efforts by estimating rubble volume and monitoring the structural movement of damaged buildings. This sensor platform also flew subsequent missions over the Pentagon on September 26th and 28th.

Another national program involved collections using NASA’s Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) instrument, which is flown aboard a de Havilland Twin Otter aircraft.⁷ The AVIRIS collects data in 224 channels of wavelengths (0.37 to 2.5 micrometers) in the visible to short-wavelength infrared portion of the spectrum. This instrument was flown over Ground Zero at the request of the Environmental Protection Agency (EPA) through the U.S. Geological Survey (USGS). Two flights were undertaken over lower Manhattan at mid-day, one on September 16th and the other on September 23 rd. These images provided a useful pair for identifying and comparing significant thermal hot spots at Ground Zero, which apparently had substantially diminished by the 23rd. The Jet Propulsion Laboratory (JPL) calibrated the imagery and made corrections for aircraft movements, while the USGS’s Imaging Spectroscopy Group made atmospheric and ground calibrations for the resulting image maps. The imagery data also helped in identifying and locating potential asbestos fallout from the WTC plume.

Finally, the nature of the widespread devastation surrounding Ground Zero created a need to leverage other remote sensing methods in helping emergency workers undertake damage assessments of the nearby buildings and neighborhoods. Pictometry, a private firm located in Rochester, N.Y., was commissioned to provide high-resolution color orthophotographs and controlled, geolocated oblique images (Figure 3) of Ground Zero and the surrounding area. A single 4-hour (1200 image) mission was flown on September 27th using a Cessna 172 carrying a digital aerial sensor capable of producing 6-inch ground sample distance (GSD) color georegistered oblique images and 1-foot GSD color orthophotos. The resulting imagery and associated measurement/management software were delivered to NYSOFT within 72 hours and eventually made available to local, state, and federal users at command centers in the city. The oblique images taken at angles were useful for assessing the damaged state of building facades, as well as provided image libraries that could be used for modeling, assessing insurance claims, and supporting the city’s remediation activities.

Broader Implications
The experiences in using remote sensing technologies and data over the World Trade Center and the Pentagon underscore the utility of these methods for responding to terrorist acts and, by extension, to most emergencies, including the aftermath of tornados, hurricanes, and flooding. These technologies also have salience in mitigating or even preventing some of the worst disasters. Nevertheless, the experiences of September 11th also remind us that there is much to do to wring the full utility out of such technologies.

As the previous paragraphs show, several groups were able to respond quickly to provide imagery and other remotely sensed data. Yet, data providers and officials at the scene experienced several problems in actually using data to the optimum. For example, imagery and analysis delivered to one set of emergency responders were not always available to others, creating occasional disconnects in the emergency operations. Further, relatively few of the response planners were familiar with the use of such data and had difficulty interpreting what they were being shown.⁸

In March of this year, the National Consortium on Safety, Hazards, and Disaster Assessment of Transportation Lifelines held a workshop devoted to exploring how remote sensing and other geospatial technologies can best assist in the task of improving transportation security.⁹ Although the focus of the workshop was transportation security, most of its findings apply broadly to the use of remotely sensed data and methods to the security of all critical infrastructure.

Workshop participants noted that remote sensing technologies are especially capable where spatial issues are of major concern; for example, relative orientation and placement of infrastructure elements, and placement and structure of ground cover. However, they cited a number of structural or institutional barriers that make it difficult to make effective use of remote sensing technologies, especially under emergency conditions.

In the case of the World Trade Center response, institutional barriers against sharing geospatial information sometimes prevented close cooperation. The many responders deserve congratulations for their accomplishments in extremely dangerous and difficult conditions. Nevertheless, the lack of established coordination plans and the impediments to sharing data quickly and efficiently among the many groups working the problem hindered their efforts. Fortunately after several days, because of the willingness of many individuals and organizations to help, the teams were eventually able to receive mapping, GPS, and imagery information in a rapid fashion.¹⁰ Nevertheless, this coordination was more a reflection of individuals committed to taking action in the face of a massive problem than the result of prior planning. These organizations could have put teams in place and worked effectively if the institutional policies of their various agencies had encouraged more extensive sharing of information and responsibility for response. A continuing problem in sharing remotely sensed data and other geospatial information is the lack of uniform standards in data formats.¹¹

Lack of familiarity with viewing and interpreting remotely sensed data was also an issue in the ability of emergency responders to use the information they were given. People who are not used to looking at locations from above or at color coded thermal images may have difficulty interpreting what they see. In the emergency of the moment, they may fail to make best use of the information collected. Hence, in addition to collecting and maintaining up-to-date databases of overhead imagery of the nation’s critical infrastructure, emergency response planners should receive imagery and other geospatial data in formats that are readily accessible to them.

Clearly, combating the threat of terrorism and responding to any future attacks will require more effective sharing and use of geospatial data and information in a coordinated effort across the agencies of the federal government; among federal, state and local governments; and among government and private sector geospatial data providers and analysts.

This one-meter resolution satellite image of Manhattan, New York was collected June 30, 2000 by Space Imaging's IKONOS satellite. The image, taken from the south, prominently features the 110-story World Trade Center twin towers. IKONOS travels 423 miles above the Earth's surface at a speed of 17,500 miles per hour. (Courtesy of Space Imaging)	This one-meter resolution satellite image of Manhattan, New York was collected at 11:54 a.m. EDT on Sept. 15, 2001 by Space Imaging's IKONOS satellite. The image shows the remains of the 1,350-foot towers of the World Trade Center, and the debris and dust that have settled throughout the area. (Courtesy of Space Imaging)	Manhattan, New York, June 8, 2002. This satellite image of the World Trade Center cleanup in lower Manhattan was collected on June 8, 2002 by Space Imaging's IKONOS satellite. The image shows the final cleanup effort. (Courtesy of Space Imaging)

Conclusion
The availability of aerial and high-resolution satellite imagery should be particularly helpful to efforts aimed at improving the nation's security from terrorist attack. That security depends among other things on detailed, broad scale, and timely information about Earth's surface, which remote sensing can provide. Medium and high reso-lution satellite data, used in conjunction with detailed aerial imagery, make possible the creation of basic geospatial data sets for carrying out vital spatial analysis of critical infrastructure, enabling analysts to discover myriad vulnerabilities and assess the risks they pose for the nation. These same data sets allow analysts to craft solutions for improving security and enhancing preparedness for all phases of future threats to critical infrastructure-detection and characterization, preparedness, prevention, protection, response, and recovery. The following paragraphs illustrate some of the uses of remotely sensed data for protecting our nation against the threat of terrorist attack:¹²

• Detection: New digital techniques allow for "data mining", the rapid spatial and temporal comparison of both imaging and non-imaging sensor data, to craft effective and efficient threat analysis. By linking and analyzing information related both temporally and spatially, it is possible to detect potential threat patterns, distinguish likely terrorist targets, and prepare appropriate responses.
• Preparedness: Emergency response planners require current and accurate geospatial information that is readily available in interoperable databases. Up-to-date remotely sensed imagery aids planners in responding to terrorist attacks, natural disasters, and other emergencies. Emergency responders should have available current, high resolution imagery of every major facility that could be a potential terrorist target in order to assist emergency crews in case of attack. Such imagery will also assist in case of natural disaster.
• Prevention: Patterns discovered through analysis of geospatial information can provide a means to respond to terrorist threats and deter attacks. This information, when fused with additional information about borders, waters, and airspace, can help disrupt and interdict attacks.
• Protection: Remotely sensed data are particularly important for analyzing the vulnerabilities of critical infrastructures. Decision support technologies such as scene visualization and incident simulation assist in anticipating the direction and form of potential attacks and in designing protective tactics and strategies. Such technologies make it possible to view the interaction of transportation lifelines with other geographically related critical infrastructure, such as power plants, population centers, and financial centers.
• Response and Recovery: Effective response to natural and human-induced disasters requires rapid analysis of imagery and other sensor data acquired both before and after the event. Such information will enable emergency response services to clear blocked transportation routes rapidly and to reroute traffic efficiently. Likewise, recovery efforts depend on the acquisition and analysis of timely imagery and other remotely sensed data that might indicate the presence of toxic or noxious chemicals. Compatible, interoperable geospatial databases containing base information that can be rapidly updated will assist in saving lives and reducing costs.

The World Trade Center experience introduced many people previously uninformed about them to the capabilities of remote sensing technologies. It is now up to federal, state, and local officials, working with the universities and private sector firms to built on that experience by improving the interoperability of geospatial information and focusing on the development of remote sensing techniques specifically designed to assist in the fight against terrorism. Among other things, that effort will necessarily include a focus on informing decision makers about the technologies and on training additional geospatial analysts.

Authors
Dr. Ray A. Williamson, Research Professor, Space Policy Institute, The George Washington University, 2013 G Street, N.W., Suite 201, Washington, DC 20052, USA, rayw@gwu.edu
John C. Baker, Technology Policy Analyst, RAND, 1200 South Hayes St., Arlington, VA 22202, USA, jbaker@rand.org

Footnotes

¹ For an overview of the operational challenges facing the emergency responders at the World Trade Center and in other recent terrorist attacks, see Brian A. Jackson et al., Protecting Emergency Responders: Lessons Learned from Terrorist Attacks (Santa Monica, RAND, 2002), which is available at http://www.rand.org/publications/CF/CF176/.
² See the image available at http://www.spot.com/home/news/NYC-091101.jpg.
³ "Space Imaging Provides World Trade Center, Pentagon Images", Imaging Notes (November/December 2001), p. 8, which is available at http://www.imagingnotes.com.
⁴ Donna Rogers, "Images of Heroes", Photographic Processing (November 2001), available at http://www.nyc.gov/html/fdny/html/photo_unit/photo_unit_p1.html.
⁵ Kenneth Chang, "From 5,000 Feet Up, Mapping Terrain for Ground Zero Workers", New York Times (September 23, 2001).
⁶ For additional details for this program that falls within the Program Executive Office, Intelligence, Electronics, Warfare and Surveillance program, see https://www.peoiews.monmouth. army.mil/jpsd/wtc.htm, and for more details on NOAA's role, see http://www.noaanews.noaa.gov/magazine/stories/mag2.htm.
⁷ Roger N. Clark et al., "Images of the World Trade Center Site Show Thermal Hot Spots on September 16 and 23, 2001, U.S. Geological Survey, Open File Report, OF-01-405, available at http://greenwood.cr.usgs.gov/maps/ofrs.html.
⁸ Charles K. Huyck and Beverley J. Adams, "Emergency Response in the Wake of the World Trade Center Attack, Volume 3: The Remote Sensing Perspective", MCEER Special Report Series, http://mceer.buffalo.edu.
⁹ Ray A. Williamson, Stanley Morain, Amelia Budge, and George Hepner, Remote Sensing for Transportation Security, National Consortium for Safety, Hazards, and Disaster Assessment, July 2002. The full report is available at http://www.trans-dash.org, http://www.gwu.edu/~spi.
¹⁰ Bruce Cahan and Matt Ball, "GIS Ground Zero: Spatial Technology Bolsters World Trade Center Response and Recovery", GEOWorld: http://www.geoplace.com, January 2002.
¹¹ This is an issue the Federal Geographic Data Committee (FGDC) has been attempting to address in a broader geospatial context. Emergency responders experience problems in sharing geospatial data rapidly and efficiently because both the formats and the content of geospatial databases differ widely. See, for example, http://www.fgdc.gov/publications/homeland.html.
¹² Excerpted from Box 3, Williamson et al., Remote Sensing for Transportation Security.