The plate boundary along the western edge of South America produces some of the largest earthquakes anywhere on Earth. The size of those earthquakes is fundamentally related to the length of the rupture zone. The purpose of this project is to determine the long term average rupture length using suites of open tension cracks that are well exposed and preserved in the Atacama Desert of northern Chile, the driest place on Earth. To accomplish this task, cracks formed by non-tectonic processes are differentiated from those produced by plate boundary earthquakes. A research team, headed by scientists from Cornell University and Miami University, with expertise in structural geology, arid soil formation, surface exposure dating, and geophysics are visiting key localities in northern Chile to develop objective criteria by which cracks of different origins can be distinguished. By dating different, but adjacent surfaces with different crack densities, the team is determining the rate of crack production, and by studying the isotopic composition of both moist soil along the cracks and gypsum veins formed within the cracks, the role of ground water in crack formation is being determined. Finally, radar interferometry provide images of cracks that open during actual earthquakes affecting the forearc. With objective criteria in hand, crack populations identified on 2 meter resolution imagery throughout northern Chile and southern Peru are evaluated to determine which ones were formed or have been reactivated by large plate boundary earthquakes. The orientations of those cracks can be related to earthquake rupture areas by modeling the static and dynamic stresses associated with both hypothetical and real plate boundary ruptures. The major scientific outcome of this work will be the first long term evaluation of the concept of characteristic earthquakes that spans thousands to tens of thousands of events, rather than the few events that are possible to resolve with paleoseismological techniques.
This work will lead directly to an assessment of the long term, time average nature of great earthquakes that affect northern Chile and southern Peru. By inverting the crack orientation for the average slip distribution, the work can help disaster response planners to predict what areas are likely to be most affected by major earthquakes. This information is immediately pertinent to the cities of Iquique and Arica (each involving hundreds of thousands of people) which have not experienced a major earthquake since the late 1800's. With a repeat time of 100 to 150 years, this segment is currently the most dangerous part of the entire South American margin. Previous earthquakes along the Chile margin have produce tsunamis that have caused damage all around the Pacific margin. For example, the tsunami from the 1960 Chile earthquake destroyed the town of Hilo on the big island of Hawaii. Knowing the average slip distribution for the Iquique Arica segment will help build more accurate models of tsunami generation in those areas. The project is supported by the NSF Earth Sciences Division Tectonics Program and the NSF Office of International Science and Engineering.
