Goals of Haiti Trip: More Knowledge, Safer Engineering

A greater understanding of liquefaction during earthquakes and recommendations for better geotechnical earthquake engineering in developing countries are two outcomes researchers hope to achieve after visiting Haiti in early February. Assistant Professor Scott M. Olson was one of nine engineers and scientists on a Geo-engineering Extreme Events Reconnaissance team who spent a week in Haiti studying how ground failures such as liquefaction and landslides during the magnitude 7.0 earthquake on Jan. 12 contributed to human casualties and structural damage.

 

The trip was funded through the U.S. National Science Foundation and the U.S. National Earthquake Hazards Reduction Program. Also participating in the investigation were Ellen Rathje and Oscar Suncar of  the University of Texas at Austin; Jeff Bachhuber of Fugro/William Lettis and Associates; Brady Cox of the University of Arkansas; Jim French of AMEC/Geomatrix; Russell Green of Virginia Tech; Glenn Rix of the Georgia Institute of Technology; and Donald Wells of AMEC/Geomatrix.

 

The group flew to the Dominican Republic then drove into Haiti. They spent the week in tents on the grounds of a hotel near Port-au-Prince that was partially under construction and partially damaged by the earthquake. Power was available for just a couple of hours each day. They ate food they had brought with them and traveled by SUV, accompanied by a local interpreter and security personnel.

 

“The roads are just awful, and on top of that a lot of them were damaged by the earthquake,” Olson says. “It was a bumpy ride. On the way back [to the Dominican Republic], it took us 11 hours to go 200 miles.”

 

The group spent time in Port-au-Prince and other coastal towns, studying the damage, collecting soil samples and running field tests with hand-operated equipment. They observed numerous examples of poor construction quality and ground that had been developed without adequate preparation, both of which contributed to damage and loss of life, Olson said. The wharf in Port-au-Prince, which was completely destroyed by the earthquake, presented a clear case of inadequate engineering practices.

 

“It was readily apparent that the ground was much too loose to survive an earthquake,” Olson said. “It was easily foreseeable, had the port been designed properly. When these wharfs were built, they took truckloads of sand that they had dug up from the creeks and from the ocean and then just backed the trucks up and dumped them in the ocean until the level got up above the water. Then they put coarser aggregate in the upper couple of feet and paved over the top of that. But right below this coarse material is just loose sand.”

 

This made the wharf particularly vulnerable to liquefaction, in which seismic shaking transforms sandy soil into a thick, viscous liquid almost like quicksand, Olson says.

 

In addition to contributing to the death toll, engineering failures complicated and delayed humanitarian relief efforts, Olson said. The arrival of supplies was hindered by the destroyed wharf and a damaged airport control tower. The researchers can already attribute much of the damage to ground failure. While certain ground preparation methods are not feasible in developing countries because of lack of money and equipment, new practices could be developed that would prevent catastrophic loss like what Haiti suffered in January, Olson says. Based on their findings in February, the GEER team hopes to develop some recommendations on how locals could use native materials to prepare the ground before construction.

 

Field trips like these are especially important to geotechnical engineers, because soils are too complex to be modeled with a computer and respond too differently at various scales to be tested definitively in the lab. In addition, there are few well-documented cases of liquefaction, so incidences of it are valuable fodder for study, Olson says.

 

“It’s essential to get out and document how the ground performs under realistic conditions as often as we can,” he says.  “These sites that we’re studying are additional data points for our correlations and for our predictive techniques. It’s going to help us better understand how the ground—whether it be ports, or underneath buildings, or in dams, or below levees—are going to perform during future earthquakes.”