| SATELLITE REMOTE SENSING/GIS/MAPPING: Roadmap to the Future
USGS's HILT project combines 200-year-old maps and Landsat images to offer tools in combating the impact of urban sprawl.
By Anthony P. Montesano Anton Furst's Academy Award-winning set design of a densely over-populated and overgrown Gotham City in 1989's Batman offered movie audiences a glimpse of "urban sprawl" at its worst. Amazingly, the problems faced by the unchecked growth of cities and densely over-populated metropolitan areas are only now beginning to be recognized as a having significant global impact. Within the last month, a report on the dangers of urban sprawl in the western United States even made the front page of The New York Times.
In an effort to offer the tools needed to predict the extent of urban sprawl, the U.S. Geological Survey's Human-Inducted Land Transformations (HILT) project, which contributes to the U.S. Global Change Research Program, began a temporal urban mapping with a 200-year land transformation study in the San Francisco-Sacramento, Calif. region. (That project led to a collaboration between USGS, the University of Maryland Baltimore County and NASA to develop a temporal urban database measuring the growth in the Baltimore-Washington, D.C. metropolitan region.)
"As mappers, we did not feel that enough attention was being paid to human induced changes in urban areas," said William Acevedo a USGS physical scientist based at the EROS Data Center, located at NASA's Ames Research Center. "Since we had two significant archives at our disposal - 100 years of topographical maps and 25 years of Landsat images - we felt we could effectively measure urban change in two regions over the last 200 years."
According to Acevedo, the HILT project is comprised of three main components: building a temporal database (mapping) based on collection of information, developing models which will then be used as part of a predictive maintenance program and lastly acting on the knowledge gained to prevent further deterioration due to urban sprawl. This year, modeling teams at the University of Santa Barbara and Hunter college are taking the databases developed by Acevedo and his team and turning them into models which can then be studied.
A temporal urban database provides the basic information needed to understand, model and predict regional patterns of urban growth and human-induced transformations in large metropolitan areas. The database is intended to be used by urban planners, policy and decision makers, Earth scientists and global change researchers for measuring trends in urban sprawl, analyzing patterns of water pollution, understanding the impact of development on the ecosystem and most importantly, to develop predictive modeling techniques to better forecast and manage future areas of urban growth. Developing a Database
To develop temporal urban databases for the San Francisco-Sacramento (analyzing changes from 1800 to 1990) and Baltimore-Washington, D.C. (looking at changes from 1792-1992), researchers integrated existing historic maps with remotely sensed data and related geographic information. Both studies have confirmed a dramatic increase in urban development since the end of World War II and indicate continued growth today.
While urban growth rates show no signs of slowing down, cities, which were once isolated population centers, have now become overgrown, straining at their size. Urban sprawl leads to the loss of natural vegetation and open spaces, and a decline in the size and interaction of wetlands, wildlife and farmland.
The temporal database documents were developed using a combination of cartographic interpretations, image processing and geographic information systems (GIS) techniques. The database documents old landscapes by reinterpreting a combination of historic maps and satellite images which are augmented by a combination of digital line graphs (DLG), digital elevation models (DEM), census data and local land-use maps.
For the HILT project, maps dating as far back as the late 18th century were used as the starting point of major metropolitan regions in the U.S. Transferring the information from the 200-year-old maps into a digital form required manual interpretation and annotation onto mylar overlays, which were then digitized using a GIS. Computer animation was then used to visualize the temporal changes.
A time series of land cover data was constructed to visualize the extent of change in a region over time and to calibrate a simulation model. For the San Francisco-Sacramento area, a digital database was assembled from topographic, road, and land use maps and from digital Landsat and elevation data.
Urban extent was inferred from historic maps and from Landsat data. Two time periods were mapped from USGS topographic maps for the San Francisco-Oakland-San Jose areas. The first was based on 1:62,500-scale topographic maps of the area from 1897 to 1906. The Sacramento area was mapped from a 1:125,000-scale map published in 1887 and from a 1:250,000-scale map representative of the Sacramento Valley from 1903 to 1910. For the second time period, 1:62,500- and 1:50,000-scale topographic maps published by the USGS and the Army Map Service were used.
Aerial photographs used to prepare these maps were taken between 1937 and 1940. Dense street patterns and buildings located on these maps were regarded as built-up areas. Polygons were drawn on mylar overlays around each concentration of these features to map urban extent. Urban extent around 1925 was obtained from M. Donley's Atlas of California published in 1979, which used road maps and other sources to map urban extent because topographic maps of that vintage were unavailable. These maps were published at 1:500,000 scale.
The Association of Bay Area Governments (ABAG) prepared maps of land use for the years 1954 and 1962. Aggregated urban land uses indicated the extent of urbanization. Comparisons with the other sources were difficult, however, because some low density residential areas would not have been considered urbanized when mapped from topographic maps. The USGS mapped the urbanized area for the ABAG counties in 1970 by using high-altitude aerial photographs and land use criteria.
Landsat satellite data, available since 1972, provided the most continuous and uniform data available at the regional scale. Landsat was used to create two maps. Two scenes acquired in 1974 using the 80-meter resolution multispectral scanner and two scenes acquired in 1990 using the 30-meter thematic mapper were interpreted using a digital display. A manual photo interpretation process was used to delineate the color and pattern indicating urbanized land.
Delineation of urban extent from maps, satellite imagery, and the ABAG land use maps were digitized by scanning. The digitized polygons of historic land covers and the interpreted Landsat data were then registered to a 30-meter Universal Transverse Mercator (UTM) grid developed for the area. Even though component data sets ranged in resolution from 30 meters to 100 meters, all data were registered to 30 meters to preserve the finest detail possible. Donley's maps of highway development from 1920 to 1978 were scanned; Donley derived them from source maps prepared by the California Department of Transportation. Recent highway data were obtained from 1:2,000,000-scale National Atlas digital line graphs (DLG). Elevation data were obtained from seven mosaicked 3-arc second digital elevation models (DEM). All of these digital data sets were registered to the 30-meter UTM grid to provide a consistent database for calibrating visualization and simulation models.
Since the database has a relatively high spatial and temporal resolution, to visualize the extent of change for this area over time, it was necessary to use the available historic data to interpolate.
Linear interpolation, with the urban boundary maps as reference, was used to create intermediate data at 1-year intervals. The interpolation algorithm calculates a linear distance from the starting urban boundary to the ending urban boundary. That distance is then used to assign an urban extent status to the corresponding pixel when estimating urban extent for each year.
Simple, single-frame animation techniques were used to visualize these data. The urban data sets were transmitted sequentially to a computer display in translucent color to show urban growth over time; a recent Landsat image was used as a reference base. Speed and zoom factors of the animation could be controlled to show particular areas in greater detail, providing a strong visualization of extensive land cover transformations in the area.
When the topographic data was added as an alternative base to the animation, the visualization illustrated how urbanization has been influenced by the physiography of the region.
What lies ahead for the project may be the most exciting component. As we approach the next millennium, continued urban growth seems now to be balanced with a renewed awareness of land preservation. When the San Francisco Bay area began its population growth with the Gold Rush fo the mid 1800s, early settlers could not have imagined how airports and highways would change the landscape of the region. With projects like HILT, future generations may be armed with more insight and thus a better understanding of how to preserve the best of urbanization while controlling its destructive sprawl. About the Author:
Anthony P. Montesano is a freelance writer specializing in articles about the high-tech, communications and entertainment industries. He can be reached through his company Montike Publications Inc., P.O. Box 86-4101, Ridgewood, NY 11385. Back |
