Satellite Images Studied For Clues To Solving Urban Sprawl

Texas A&M University geographer Hongxing Liu is analyzing optical, radar and thermal image data about land use in Houston. He has already used satellite technology to create more accurate maps of topography and coastlines in Antarctica. The satellite radar interferometry that reveals the slow subtleties of glacier movement involves processing repeated-pass satellite images spaced 24 days apart, while the remote sensing technique that seeks to unlock answers to urban sprawl compares a time series of satellite images acquired at different times over the course of decades.

"Radar interferometry based remote sensing technology can detect centimeter level surface motion and displacement at about 30 meters spatial resolution," said Liu, a faculty member in the College of Geosciences. "That's close enough to show movement of the ice, even at the glacially slow rates involved with giant Antarctic ice sheets."

Resolution refers to the area represented by each pixel on the satellite image. High-resolution satellite data, showing less area and more detail, is stored in computerized databases manipulated with sophisticated software to form maps composing a geographical information system (GIS).

"GIS databases help geographers combine the old science of cartography with innovative remote sensing techniques," Liu said. "We can integrate conventional maps with digital satellite images or scanned aerial photos. This type of quantitative analysis allows scientists to map and analyze physical data and socio-economic data on the same coordinate system."

According to Liu, remote sensing with satellites is much cheaper than other data acquisition techniques.

"One trip to Antarctica may cost too much, but once a satellite is in position, it can return better data to us with greater regularity," he said. "Satellite images combined with global positioning system (GPS) data can be rapidly incorporated into geographical information systems (GIS), yielding extremely detailed images.

"Active radar signals can be used to identify different ground features," he observed. "Such sensing is accomplished relatively quickly and economically, and data quality is good compared to older techniques such as on-site land surveys."

Liu's current research on land use changes in Houston represents quite an adjustment from his postdoctoral work at Ohio State University. Before he moved to Texas A&M in August 2000, Liu worked on a NASA project using digital radar imagery to detect positions of the ice margins along Antarctica's coast. Using automated resolutions of 50 meters, his team compared their data with previous information to determine ice boundary changes, which in turn indicates changes in the Earth's climate.

"We developed satellite data processing algorithms to correct geometric distortion of the continent's terrain, resulting in a radar orthoimage map," Liu said. "Radar orthoimage maps yield more details and fuller coverage than ground surveys.

"Unlike optical sensors such as Landsat, radar satellites can penetrate clouds and get a sharp image even on overcast days," he observed. "And resolutions of 25 meters are now commonplace, compared to the one kilometer resolution of AVHRR image maps. Repeat pass radar interferometry can actually let us see the ice moving."