Research Training Program

Kristen Iriarte
College of William and Mary
Williamsburg, Virginia

Dr. William Melson, Ph.D.
Supervising Scientist
Department of Mineral Sciences

"The RTP program allowed me to expand horizons within my own discipline, while also exposing me to new fields full of possibility."

Can Rare Catastrophic Events be Predicted? The Fatal Pyroclastic Flow of Arenal Volcano, Costa Rica, August 23, 2000.

The technology of today, along with the exponentially growing population has led to human travel and occupancy in areas of the world never inhabited before. The major issues here are rare events, such as great earthquakes and certain volcanic eruptions, that now result in fatalities more frequently. Several areas currently have volcano mitigation programs involving seismic monitoring in some form, including the study volcano, Arenal. This research focuses on an event at Arenal Volcano, Costa Rica, where two individuals sustained fatal injuries in a pyroclastic flow erupted on August 23, 2000, in spite of active seismic monitoring. Arenal volcano has been continuously active since it first erupted in 1968, killing 78 individuals and destroying many homes and several square kilometers of pasture and pre-montane rain forest. In order to identify any predictive tools useful in preventing this type of tragedy, seismic data from the event were examined. Individual signals were characterized and plotted in order to create a 7-month history before the event. The August 23rd event has a distinct signal that occurred briefly in June and July, but not nearly with the length or magnitude of that of the fatal event. While the number of days with events steadily increased in the three months leading up to the event, there are no patterns that could be used to predict pyroclastic flows. Samples from the flow were examined with an electron microprobe in order to gain information on chemical content. In this study the differences were not significant enough to trigger the event, and thus the mechanism of eruption must have been simply an abnormally large volume of buoyant magma moving out of the crater. While the predictivness of the volcano using scientific data was not illustrated through this research, the methods engaged here were only some of the few available to predict and study events. Further research into maximizing early event warnings must be conducted in conjunction with onsite education as to hazardous pyroclastic flow channels and the hazards of rare events. Management must be initiated in order to prevent disasters like the August 23rd pyroclastic flow. Locals must be educated in disaster prevention, and areas that are considered dangerous must be kept off limits in spite of the sometimes spectacular views they provide of volcanic eruptions.

This research was supported by a grant from the National Science Foundation Research Experiences for Undergraduates program, Award Number DBI 9820303.

Letter of Gratitude