Have you ever experienced an earthquake? This probably isn’t something you think about often, especially if you live in southern Indiana, where earthquakes large enough to be felt (or cause any damage) are quite rare. Talk to anyone living in Japan, Chile, or even California, and the odds are that they have experienced one or more earthquakes while living there; these areas of the world are tectonically active, and earthquakes are a somewhat normal occurrence.
— Jascha Polet (@CPPGeophysics) April 16, 2016
A small number of these earthquakes have a devastating impact on the surrounding area and nearby communities. For example, the magnitude 7.0 earthquake in Kumamoto, Japan that occurred on April 15, 2016 resulted in over 70 deaths, and the total cost to rebuild due to damage will be about $24 billion. A large portion of the damage was caused by landslides that were triggered by shaking from the earthquake. An example of this devastation can be seen in the image above, where a landslide completely destroyed the Great Aso Bridge in Minamiaso, Japan.
Earthquake-triggered landslides are a worldwide phenomenon, resulting in more than 70,525 deaths from 1968-2008 (Marano et al. 2010). Earthquakes often occur in mountainous regions that have steep slopes, which is a lethal combination. Many rural communities are built in mountain valleys, and entire villages have been buried by unstable material moving down mountain slopes into valleys, after being shaken by an earthquake. Dangers like these that are caused by an earthquake, but not directly caused by earthquake shaking, are typically called ‘secondary hazards,’ and they can range from landslides, to tsunamis (large waves generated by earthquakes underwater), to fires. These secondary hazards are studied by a small portion of the scientists in the earthquake research world.
Many studies have shown that certain characteristics of an area are related to where landslides occur during an earthquake: the amount of shaking, the steepness of the ground, the type of rocks, the wetness of the ground, and the average rainfall. If we knew this information, along with locations of past landslides due to earthquakes, we could correlate these factors with locations where new earthquakes will cause landslides. This is where my advisor (Michael Hamburger) and I come in.
I study landslides that are triggered by the shaking during earthquakes, and I aim to model where they are likely to happen after an earthquake. Using a statistical model called a logistic regression, I combine the known locations of past landslides due to earthquakes with the characteristics described above to find a relationship between how the ground shakes during an earthquake and where landslides are likely to occur. This model can predict the probability of landsliding for any earthquake, so that in the future it can be used to tell where landslides are likely to happen minutes after an earthquake occurs, anywhere around the globe.
If you visit the website for the United States Geological Survey’s (USGS) Earthquake Hazards program, you’ll find a list of earthquakes that have occurred in the last 24 hours. Click on an earthquake to find a map of the shaking that occurred, as well as estimates of the deaths and economic losses that are likely due to that earthquake (see examples from Japan above). This information is used by government and relief agencies (e.g. International Red Cross/Red Crescent) who make decisions on how best to provide aid after an earthquake occurs. This is where I hope the model will someday contribute: its results could be utilized by emergency responders and relief agencies to determine where the most aid is likely to be needed in remote areas impacted by earthquakes, and it can also indicate which transportation routes are less likely to have been affected by landsliding. We may not be able to predict exactly when an earthquake will occur, but with this powerful new tool, we can more effectively respond to the landslides that occur after devastating earthquakes around the world. So, you can rest a bit easier if you do ever feel that earthquake.
If you’d like, you can find more information about this project here!
Marano, K. D., D. J. Wald, and T. I. Allen (2009), Global earthquake casualties due to secondary effects: a quantitative analysis for improving rapid loss analyses, Natural Hazards, 52(2), 319–328, doi:10.1007/s11069-009-9372-5.
Nowicki, M. A., D. J. Wald, M. W. Hamburger, M. Hearne, and E. M. Thompson (2014), Development of a globally applicable model for near real-time prediction of seismically induced landslides, Engineering Geology, 173, 54–65, doi:10.1016/j.enggeo.2014.02.002.