First Impressions

  page 2Page 4

Petrakov and Chernomorets reached the site for the first time in the first week of October, two weeks after the disaster. Tutubalina was teaching and had to stay behind. What they saw stunned and saddened them. “It was terrifying,” says Petrakov. “There was this mass of black ice—a wall—blocking the Karmadon Gorge. On the surface of the ice, lakes were filling up. We had seen the pictures on the TV, and so, in one way, we were prepared. But when we arrived it was very foggy and overcast, which made it appear much more dangerous, and, well…,” he pauses while he searches for the word, “sad.”

  High-resolution Satellite Image of Debris Flow Damage in Karmadon

“The scale was so great that it really was hard to imagine how it could have happened,” adds Chernomorets. “I have been studying there [the Caucasus] for many years, seen many debris flows and avalanches in this region, but nothing to compare to that.” By the time they arrived at the site, there was no hope there would be survivors. Only 20 bodies were ever found.


After the Kolka Glacier collapsed, the Karmadon Depression filled with ice covered by black, pulverized rock. Water from dammed streams and melting ice formed lakes along the margins. The rapidly rising water was a continuing danger, threatening a sudden outburst that would cause flooding downstream. (Image copyright Digital Globe)

  Satellite Image Overview of Karmadon

“It was very dangerous at that time,” says Chernomorets. “We were trying to walk up the Genaldon, and constantly you had to watch where you stepped because the path was muddy and full of rocks that moved all the time. But you also had to look up because the walls of the valley were lined with blocks of ice and falling rock.”


The area covered by ice and debris dwarfed the hamlet of Karmadon, and the Genaldon River disappeared completely. [The outline corresponds to the detailed image above] (QuickBird Image copyright Digital Globe)

  Photograph of cobbles and unstable footing

The Safety of Satellites

Mapping the temporary lakes and watching them for signs of sudden and catastrophic release became a top priority. Chernomorets and his colleagues Ivan and Inna Krylenko helped collect topographic survey data to map the coastlines and conducted echo soundings to measure the depth of the largest one, called Lake Saniba.

“In the October 2002 trip,” Chernomorets says, “we did not yet have any satellite observations. But after we returned we received remote sensing imagery from a number of sources—photos from the astronauts on the International Space Station showing the pre- and post-disaster landscape and ASTER [Advanced Spaceborne Thermal Emission and Reflection Radiometer] data from NASA’s Terra satellite for pre- and post-disaster comparisons. Eventually we also acquired imagery from the Indian Remote Sensing (IRS) satellite, the American commercial satellite Quickbird, and the Advanced Land Imager on NASA’s Earth Observing-1 satellite.”

Throughout the remainder of 2002 and into the late spring of 2003, frequent fogs and dangerous weather conditions prevented on-site visits or aerial photography of the glaciers, so the IRS satellite data that were collected periodically by Research and Development Center ScanEx in Moscow became the key method for monitoring the glaciers and lakes. The three scientists were relieved to observe the lakes began to disappear one by one, likely draining through channels and crevasses appearing in the ice mass. By summer 2003, only three of the original 13 remained.


Boulders, pebbles, and mud covered the surface of the debris flow, resulting in treacherous footing. The pathless maze of debris was only one of many hazards that slowed exploration of the disaster area. (Photograph courtesy Sergey Chernomorets)

Time Series of Satellite Imagery of Lakes

“The satellite imagery enabled us to assess the extent of the disaster, map the boundaries of specific areas, and compile sketch maps of the whole area,” says Tutubalina. The scientists also put the satellite images to a practical use. When the danger of spring snow avalanches cleared, and they were able to return to the site in June 2003, they discovered that in the scoured and ice-covered landscape, the recent satellite images made better route maps than any pre-disaster topographic map.


The Russian researchers evaluated the risk of future danger at the disaster site using a time-series of satellite images collected in the year following the disaster. Satellite imagery was crucial throughout the late fall and winter of 2002 and 2003, when dangerous weather prevented on-site observations of the ice-dammed lakes. This sequence of images from the Indian Remote Sensing (IRS) satellites showed that the lakes (except Lake Saniba) were draining gradually through crevasses in the ice mass, and were not likely to cause subsequent catastrophic floods. (Images copyright ANTRIX, Space Imaging Inc., R&D Center ScanEx)

Print this entire article