InSAR satellite technology honed to broaden detection of earthquakes and volcanic eruptions

By Anne Strehlow

Though it moves at a speed that makes a turtle look sprightly, the surface of the Earth is in constant motion. Violent volcanic eruptions and land-dividing earthquakes are often the sole reminders of the fury brewing beneath our feet.

Since the science of predicting these powerful forces of nature has yet to become foolproof, geophysicists are continually striving to improve their ability to anticipate earthly disasters. Ironically, it is a radar satellite system orbiting 500 miles above the Earth's surface that is providing fundamental clues to understanding the dynamics of our planet so far below.

"In the case of earthquakes, we would like to know exactly how the Earth moves underground where we can't see it," said Howard Zebker, associate professor of electrical engineering and of geophysics. "In the case of volcanoes, we want to know where magma might be flowing—is it collecting in one central magma chamber or are there many underground rivers?"

Zebker, an early pioneer of the satellite technology known as Interferometric Synthetic Aperture Radar (InSAR), has been studying earthquake- and volcano-ridden regions across the globe for more than a decade. Monitoring the western United States, the volcanically active islands of Hawaii and the Galapagos archipelago, Zebker's research team has focused its efforts on expanding the applications of InSAR and honing its capability to generate more robust measurements of the Earth's movements.

In a Dec. 8 symposium of the American Geophysical Union annual meeting in San Francisco, Zebker delivered an overview presentation titled "InSAR of the Future: Seeing Through the Lens of Basic Science, Math and Technology." Graduate students Andrew Hooper of the Geophysics Department and Ana Bertran Ortiz of the Electrical Engineering Department also presented their research findings, which have made strides at overcoming the conventional limits of InSAR.

Rumbles, burps, shivers and shakes cause shifting and buckling of the Earth's crust. These minute movements are detected by four InSAR satellites circling overhead, scanning the same point on the ground once every 24 to 35 days. As the satellites' radar waves ricochet off the surface, the echoed signals are retrieved and processed, generating highly detailed two-dimensional maps ("interferograms") of the Earth's outer shell.

By comparing interferograms from the same geographic region over time, tiny changes, or deformations, in the Earth's surface can be detected down to the millimeter. Depending on the region being monitored, these deformations can result from a whole slew of terrestrial movements—slippage of the crustal plates along fault lines, flowing subterranean magma, groundwater withdrawal from aquifer systems, oil production or the creeping of glaciers in the polar icecaps.

Despite the precision of InSAR, the monitoring system is not without challenges. Thick vegetation and seasonal snowfall thwart the study of many regions of the planet. These continually morphing ground covers obscure the underlying surface and generate variable and unreliable radar signals.

Zebker and Hooper have tackled this challenge using a novel technique called InSAR Persistent Scattering (PS). A time-demanding process, PS requires teasing through reams of murky data to identify individual data points on each interferogram that remained consistent from map to map. These locations serve as benchmarks from which to align the interferograms.

By applying PS to radar readings from the lush environs of Mount St. Helens and the Sierra Negra crater of the Galapagos—terrains where use of conventional InSAR has proven difficult—Hooper was able to successfully examine crustal deformation around these volcanoes.

"We have figured out a method to analyze the radar reflections from each of the individual points on the ground across a whole series of interferograms," Zebker said. "Rather than just comparing two points and generating a single difference, we compare 30 points acquired over several years. This has enabled us to use the data from areas that we were not able to use reliably in the past."

While vegetation presents a logistical challenge for InSAR, some changes in the Earth's crust are simply missed by the orbiting satellites. Because of the lengthy time delay between scans, very subtle movements and explosive eruptions occurring with little geological warning are left undetected.

"While some volcanoes erupt rather slowly, allowing you to actually see stuff building up underneath, other volcanoes seem to increase their pressure without any observable deformation signal," Zebker said. "It all happens very quickly. So, for certain kinds of volcanic eruptions, [acquiring a radar signal only] once a month is not sufficient."

By switching the mode of one of the InSAR satellites, named Envisat, to an operation known as ScanSAR, the radar beam can "snap" five readings of a region during a single orbital pass of the satellite.

"This particular mode of the Envisat satellite allows us to scan back and forth over extremely wide areas on the ground, about 400 kilometers [249 miles] wide, increasing the number of 'looks' we have at the ground in a shorter amount of time," Zebker said.

The data collected from this nontraditional scanning mode must first be converted into a form recognizable for conventional InSAR imagery. Despite this additional step, Bertran Ortiz, who tested ScanSAR over Hawaii, will be able to detect deformations occurring in the Earth's surface on a scale of days rather than months. Her results suggest that more frequent sampling ups the odds of picking up rapid eruptions and canceling out small variations in the radar signals that cloud subtle movements.

According to Zebker, ScanSAR and PS—alongside broader technological advances in digital computing, computational power, mathematical algorithms, software and spacecraft—enable more widespread detection of earthquakes and volcanoes across the globe.

"We have gotten a lot smarter about the physics of InSAR, and so we have learned how to better exploit the information in these interferograms," Zebker said.

Anne Strehlow is a science-writing intern at the Stanford News Service.

-30-