PRESS RELEASE FROM GEORGIA TECH
Community Education and Evacuation Planning Saved Lives in Sept. 29 Samoan Tsunami, Researchers Say
Community-based education and awareness programs minimized the death toll from the recent Samoan tsunami, though there are still ways to improve the warning and evacuation process, according to a team of researchers that traveled to Samoa last month. The team, funded by a National Science Foundation (NSF) grant, collected data Oct. 4 through 11 to document the impacts of the 8.1 earthquake and the ensuing tsunami that occurred on Sept. 29. They examined flow depths, runup heights, inundation distances, sediment depositions and damage patterns at various scales.
“In addition to timing – the fact that the tsunami struck in the daylight morning hours when most people were on their way to work or school – tsunami education, awareness and evacuation exercises really contained the death toll,” noted Hermann Fritz, one of the principal investigators and an associate professor of civil and environmental engineering at the Georgia Institute of Technology. “The technical solution doesn’t always work for coastlines near the epicenter with less than 30 minutes between earthquake and onslaught of the tsunami. Earthquakes with a duration of more than 30 seconds serve as a natural warning, resulting in a spontaneous self-evacuation.”
Nearly 190 people were killed in the tsunami, with the majority of deaths reported in Samoa, a country governing the western part of the Samoan Islands in the South Pacific Ocean. The two main islands of Samoa are Upolu and Savai’i. American Samoa, a territory of the United States southeast of the sovereign state of Samoa, is comprised of main island Tutuila, the Manu’a Islands, Rose Atoll and Swains Island. The Samoan government estimates the total damage from the tsunami at $147 million.
The team’s survey data – especially critical in the immediate aftermath since perishable data would otherwise be lost forever – circled all of the main Samoan islands and spanned 350 kilometers from Ofu in the east to Savai’i in the west. They learned that the tsunami impact peaked at Poloa near Tutuila’s western tip and Lepa at Upola’s southeast coast. Maximum runup heights reached 17 meters at Poloa, and inundation distances and damage were recorded at Pago Pago, more than 500 meters inland. The harbor at Pago Pago, normally well-protected from ordinary storm waves, is a classic tsunami trap vulnerable to long-period tsunami waves.
In addition, researchers noticed a marked difference between the evacuation process in Samoa and American Samoa. While most villagers in Samoa knew to rapidly evacuate after experiencing an earthquake, only a month earlier they had been told that cars help with evacuations, a deadly directive since most roads run parallel to the beach.
“Many perished trapped inside cars waiting in congested small roads or in long lines behind vehicles stopped by landslides or debris on the road,” said Costas Synolakis, principal investigator and professor of civil engineering at University of Southern California. “I have been on more than 20 tsunami field surveys, and in many ways this was one of the most surprising in terms of how carnage varied over fairly short distances. This was also the first time we noted what we suspected: misinformation kills.”
Emile Okal, a seismologist and professor of earth and planetary sciences at Northwestern University, conducted approximately 120 interviews with tsunami survivors in 70 different locations around Tutuila and Upolu. He found that most people were educated about tsunamis and knew how to react because of community-based educational programs, not ancestral stories.
“The last significant tsunami in Samoa occurred in 1917 and was very similar in seismology to the Sept. 29 tsunami. Surprisingly, no one I interviewed said they knew of family members being in a similar situation,” Okal observed. “Since the 2004 Indian Ocean tsunami and the 2007 Solomon Islands tsunami, there has been a concerted effort on the part of the local government in American Samoa to post signs and conduct evacuation drills were conducted in some Samoan communities. Many villages were completely destroyed, so I am impressed that the death toll was not larger. The bottom line is education worked.”
While Synolakis agreed that the death toll was probably minimized due to educational efforts, he said there is still a lot of progress that can be made. While working in the field on Oct. 7, the team experienced a real tsunami warning and witnessed first-hand the tremendous confusion and disorganization that followed.
“Although there are warning signs along the beaches in American Samoa, there is no information about where the evacuation routes are,” he said. “It’s also just as important to let people know when it’s safe to come back as it is to warn them. We definitely have our work cut out for us.”
The collected field data serves as benchmarking and validation of numerical tsunami models with wide-ranging applications including forecasting, warning and sediment transport. The researchers will present their findings at special sessions at the American Geophysical Union Fall Meeting in San Francisco this December. Brief publications summarizing the immediate results will follow in research journals. This survey was partially supported by the Pacific Earthquake Research Center (PEER),
Media Relations Contacts: Nancy Fullbright (912-963-2522); E-mail: (firstname.lastname@example.org); Eric Mankin (213-821-1887); E-mail: (email@example.com); Megan Fellman (847-491-3115); E-mail: (firstname.lastname@example.org).
Interviews conducted by tsunami researcher, Professor Walter Dudley of the University of Hawaii. Posted on the USGS event page:
The link here takes you to pictures taken by NOAA employee Gordon Yamasaki in Pago Pago, Samoa during the tsunami after the magnitude 8 earthquake:
A morning meeting with the government official who has been overseeing the scientific teams. The Samoan Government, UNESCO and ITIC developed “Terms of Reference” for all research groups working in Samoa to address. We got the ok to work in the country, but were asked to focus on two of the seven tasks defined:
6. Collect information on human and community vulnerability and resilience factors at work in different places: i.e., what made a particular community resilient or vulnerable?;
7. Where possible, to map the above information.
Other tasks included mapping inundation and water levels, measuring environmental and ecosystem impacts, collecting tsunami deposits, looking at damage to structures, and recording survivor stories. We were also asked to include a government official during all of our field work. We were assigned a young meteorologist in training who had been educated in New Zealand. He was very pleasant but it did mean changing our transportation plans. We had arranged a 4 person car and Samoan guide-interpreter which worked fine for the three of us. This added an hour delay while we arranged to rent a different car for the day.
Drove back to the south side of Upolo. Passed the Robert Louis Stevenson Museum on the way. Stevenson spent the last 5 years of his life on Upolo and is buried nearby on the peak of Mt. Vaea. Our first stop is Poutasi Village where 9 people died including 3 children. All villages in Samoa consist of a swath of land the runs from the coast to higher elevations in the mountains where plantations of coconuts, bananas and other crops are grown. Most people lived in houses near the coast before the tsunami, but many had sleeping fales (open air covered platforms) and other small structures in the plantation areas. Poutasi is particularly vulnerable because a river runs behind most of the coastal houses and the only access to higher ground is along a road running parallel to the beach and a bridge near the mouth of the river.
Most of the houses in Poutasi were flattened – all that remained were the concrete slab foundations. But one house right by the beach looked practically untouched and we wondered if it had been rebuilt. Most of the people who lived near the coast have moved up to the houses and fales in the plantation area and there was almost no one in what remained of the coastal part of the village. We were lucky – the man who owned the home was working on it so we were able to get the story of what happened during the tsunami and a clue to why the house had so little damage. The man’s mother lived in the home – his house was further inland. After feeling the earthquake, he ran to his mother’s home because he thought “something bad might happen”, and arrived just in time to see the water recede. He knew there was no time to get to high ground so he stationed his mother beneath the front porch on the landward side of the house and told her to hang on to the door and window frame. Just as he got her positioned, the first surge caught them. He was carried by the surge across the street and managed to grab a tree and climb up it. He could see his mother still hanging on to the house – the porch provided an air pocket for his mother. The second surge was larger, but she still managed to keep her position and both of them survived. The house was wood frame and only a few months old. He had added lots of cross bracing and the home was secured to the foundation. He had also reinforced the coastline in front of his house with basalt boulders. There was no coastal reinforcement on the adjacent home where erosion had undermined the structure.
Next we talked to a matai of Poutasi. He and his son were restoring the boulders around a family grave. In Samoa, relatives are buried in front of the home. It is the responsibility of the children to maintain the grave sites of their ancestors. The graves are also proof of ownership in a country where there are no deeds and titles. Throughout Poutasi and elsewhere in the damaged areas, the graves were the first structures to be cleaned and restored – and the sight of a newly whitewashed family tomb topped with bright flowers is a jarring contrast to the destroyed structures around it. The matai lost three of his grandchildren in the tsunami. The earthquake woke him up. His son, daughter and three grandchildren were getting ready to go to school and had gotten into the cab of a truck. He got into the back. The tsunami caught the truck and rolled it. The three adults survived but the children didn’t. After the tsunami, some people from higher up or other villages came down to the damaged areas to help find the missing. In some cases, they also helped themselves to any belongings they found in the damaged areas. This turned out to be a recurring theme – the “gaoi” (cheater, thief), who came into damaged areas as soon as the water retreated. The matai said that no one is thinking of rebuilding right now – everyone is scared and staying up in the plantation areas. They might only have guest fales in the low areas – places for people to sit and visit during the day – and have all the residents in the plantations. The Samoan village structure has several advantages in responding to a tsunami. All of the villages have land up high, everyone in the village group and anyone in the village can use the upland land as long as the matais approve. This is the first major disaster that I have visited where there are no relief camps or shelters for the simple reason that everyone has relatives they can stay with.
Today was a change islands and countries. We flew Polynesian Airlines from Pago Pago to Apia in an 18 passenger prop plane that flew low enough to not need pressurization of air conditioning. No security screening and we can carry as many liquids on board as we wish. Checking in involves getting on a scale with all your luggage so they know the total cargo weight. I had a front row seat and there was no door to the cockpit so I could watch the pilot and the controls. We were able to see some of the tsunami-damaged areas on the east end of Tutuila (the main populated island of American Samoa). Much of Tutuilla is steep and the damaged areas were confined to pockets of low lying villages near the mouths of coastal rivers. The flight is just over 30 minutes. We can see some of the damaged areas along the SE coast of Upolo, the more populated eastern island of Samoa. Even from afar, the damage looks more continuous than on Tutuila. Apia is located on the north side of Upolo.
As in American Samoa, first on our itinerary is to introduce ourselves to UNESCO and government officials to sanction our field effort. First is a meeting with Jan Steffan, theUNESCO science coordinator for the Pacific region and the interface between scientific teams and the Samoan government. He had spent two years in Indonesia before this assignment and had spent much time in Padang in the aftermath of the 2004 Andaman-Sumatra earthquake and tsunami. He knew several of the people I had worked with on the 2005 Sumatra post-tsunami survey. We explained the purpose of our trip and how our work was not just a repeat of the previous team efforts. He agreed that our expertise and focus would add to what had already been done and forwarded his approval to the Samoan Government. We have an appointment to meet with an official of the Department of Environment and Meteorology tomorrow morning for the final ok.
After the meeting, there wasn’t much time left for work. We took a quick drive over to the south side of the island and looked at some of the areas that were hit. Samoa consists of two main islands Upolo and Savai’I both of which are larger than Tutuila. The population is more thant 2.5 times larger than American Samoa. Upolo was the hardest hit by the tsunami. Unlike Tutuila which was located nearly perpendicular to the orientation of the earthquake fault that caused the tsunami, Upolo was a little to the west of the maximum wave energy. Major damage was confined to the SE coast and no damage occurred on the north coast where Apia and the most populated part of Samoa is located. If a future tsunami generating event has a slightly different orientation, the outcome for Apia could be different. One of the major needs for both Samoa and American Samoa is a credible tsunami hazard map that addresses the likely maximum inundation for tsunami sources both nearby and elsewhere in the Pacific. It is very difficult to locate evacuation sites and evacuation routes if you don’t know where the safe areas are. On September 29, many people over-evacuated, driving as high up as they could, causing traffic problems. The lack of a well defined evacuation zone also may have encouraged people to use cars to evacuate as they didn’t realize that in many cases they didn’t need to go far to be in a safe area. Hazard assessments are based on numerical modeling efforts, past tsunami inundation and water height measurements and, when available, paleotsunami deposits. All of this information needs to be pulled together to see if what happened on September 29 was the worst likely event here, or what might produce a greater impact.
We looked at the Coconut Beach Resort near Maninoa. This was a well-landscaped, high-class resort with a beautiful beach and spa that catered to international visitors. There were about 90 guests on September 29. The earthquake was felt strongly but no one connected the earthquake and a possible tsunami. Fortunately the nearby resort of Sinelei was run by a general manager who was more aware to the relationship between shaking and tsunamis. Sinelei had installed a siren and also practiced evacuation drills. They also had developed a protocol for evacuation events with a staff person grabbing a list of the clients. On September 29, Sinalei was able to quickly ascertain that two guests had not left their rooms and staff were able to find them and help them evacuate before the worst surge arrived. People at nearby resorts including Coconut Beach heard the alarm and all but one person was able to escape. One woman tripped while evacuating and was caught in the water. Coconut Beach also provided an interesting example of tree resistance to the force of the water. Smaller diameter trees were bent while the larger ones resisted the flow. Although Coconut Beach did not have plans for a tsunami, they did respond quickly after the event – and were the first to arrive at the hospital, to deposit the 12 injured guests. They were quickly rebuilding and hope to be open for a limited number of guests in February – and they plan to include information about natural tsunami warnings in guest literature.
We’ve become acclimated to life in Alega. The sound of the waves is wonderful to sleep too – although people who experienced the tsunami find it traumatic and few people have returned to living in the harder hit coastal communities even if their home wasn’t damaged.
Post event surveys are a mix of meetings, field work, interviews, dealing with set backs, and logistical planning. This morning was spent dealing with our travel plans to Samoa and getting and official ok from the Samoan government to work there. Flying from Pago Pago to Apia (a 30 minute flight) is not something you can just book online. We spent much of the morning at the airport afrranging the flight schedules for 5 people with 4 different itineraries. There are 3 carriers – one that has 5 flights a day but evidently just went out of business – their office is dark and they don’t return calls. One with pressurized cabins, AC and “large” planes that carry 30 people but they have run out of fuel and aren’t flying for the next 2 days. And finally Polynesian Air which has flying culverts but is still in business and has fuel. Apparent success – we’ll find out for sure tomorrow.
More logistics – calls to the director of the International Tsunami Information Center – aNOAA funded organization that now works through UNESCO to coordinate tsunami preparedness efforts throughout the world. For the first time, UNESCO has developed a protocol for visiting scientific teams responding to the Samoa tsunami event. It’s important for reconnaissance teams to collaborate with in-country scientists and other researchers and government officials. Over 60 international researchers have visited Samoa since the tsunami including teams from Japan, the US, Australia and New Zealand. A team from Cambridge, England is due to arrive the day I leave. Coordination is also important to avoid duplication of efforts and build on the findings of other groups.UNESCO is also committed to making sure that these teams share their information with the host country so that they benefit from the research teams’ efforts. Some groups in both this and past events have operated outside of government sanction – partly because of the need to quickly get perishable information, and sometimes because the official route can be cumbersome. In 1998 after the Papua New Guinea tsunami, I was all set to leave with a group of 5 international scientists when the PNGgovernment pulled our permission 24 hours before the plane left. It took a month to solve the problem and in the end only 2 of us were able to go. I think we’ve got our bases covered this time, but one never knows.
We spent the afternoon looking at the hardest hit areas on the western end of the island and talking to survivors. Don Vargo of American Samoa Community College provided a Ford Tundra capable of making it over the steep roads and two research assistants to drive and provide translation. First to Paloa on the NW tip of the island where only one home in the coastal portion of the village survived without damage and one person died. The earthquake occurred around 6:40 AM and children were just leaving their homes to go to school. One mother tried to get her kids into a car when she saw the water withdraw. The coastal road parallels the coast and driving would have increased their exposure. Fortunately the oldest child had studied about tsunamis and insisted they run up the hill behind their house.
A very full day of interviews and meetings. It rained heavily in the night – knocked out power and interrupted internet service for awhile. The rains pose a health risk. Both dengue fever and Leptisporis are on the rise because of the debris has created many pooling places for stagnant water. We were told yesterday of a man who was evacuated off island with swine flu, leptospirosis and dengue all at once. We spent the morning at the home of the park anthropologist we interviewed yesterday. Her home was right on the edge of the inundation zone and her pig sty was in it. She has a sow with 6 piglets and when she returned to her home after the tsunami, they had all survived but the sow had a high water dirt mark on it’s neck. All of the piglets must have swum to survive. Her village is Leone where 11 people died – it was the single hardest hit in terms of human loss. The water heights weren’t as high in Leone and the percent of damaged structures wasn’t as great as in the nearby communities of Paloa (1 death) or Amenave (no deaths) where almost all structures were erased. The difference in the casualty rate is probably mainly related to the population of the village – Leone is much larger, but tomorrow we will visit Amenave and Paloa and may have more to add to the story. A number of larger buildings appeared to have little damage and we heard several stories of people surviving on the second floors. We spent some time looking at the pattern of damage in Leone. It was irregular – flattened buildings next to ones that appeared substantially undamaged. Impact appears to have played a major role . Vehicles, telephone boles, trees, boats and large containers were transported tens of meters and the size and character of impact had a major role in the severity of damage. In our cursory look, we saw little evidence of scour – just one location in central Leone where the scour was noticeable.
We met with Don Vargo of the American Samoa Community College and several colleagues. Don and his research assistants have helped a number of visiting scientific teams and their translation help has been essential to getting accurate accounts from people who only speak Samoan. We had a lively discussion about what could be done to improve education efforts and institutionalize the lessons of this event. A top priority is a good tsunami hazards map for the territory. While there are a number of generic tsunami hazard zones posted, there is no information on how high or far people need to go. Over evacuation was common on September 29. With credible hazard zone maps it would be possible to create walking evacuation routes and evacuation areas. With evacuation routes, village evacuation drills could be held – maybe a good goal is the one year anniversary of the tsunami.
I finally arrived in American Samoa last night – after a five hour flight where I was fortuitously seated next to a large Samoan gentleman who turned out to be the matai (chief) of the village of Leone on the western end of the island. Leone was one of the areas hardest hit by the tsunami – 11 of the 34 deaths attributed to the tsunami were in his village. He was very gracious and we had plenty of time for a long conversation about what happened in his village. He had tried to push tsunami awareness after the 2004 Indian Ocean event, but many people were apathetic. On Sept 29 many people evacuated on feeling the ground shake – there had been a number of education efforts over the past year – but many people used their vehicle and some got stuck in traffic. For a number of the victims, being very large was a liability. Some people were just too big to be able to quickly walk to higher ground.
Couldn’t see much on the drive to Tisa’s Barefoot Bar where we are staying. The amount of cleanup is impressive. Some damaged boats and buildings at the back of Pago Pago harbor but most of the coast looked surprisingly ok. Tisa’s is on a beautiful little pocket beach with four fale (sleeping shacks) right on the beach. There was some flooding here during the tsunami but very little damage. I do plan to evacuate if a strong earthquakes hits in the night. Mosquito screens make our beds look regal. So far the trade winds have been pretty strong and I haven’t noticed many insects.’
First real glance at the place was around 5 AM with day dawning. American Samoa doesn’t have the big vast beaches of Hawaii, but the small beaches have pristine white sand, no people, no trash, and the Manau Island’s to the east are supposed to be great for diving and snorkeling. It’s a surprise to me that there little tourism here – to me this is a hands down winner over Waikiki. Very full day of meetings with officials from a variety of agencies to introduce ourselves, find out what was happening with their organizations and how we could collaborate with their efforts. First to FEMA who holds planning meetings every Monday, Wed. and Friday morning. We didn’t have any official introduction, just the letters Marjory had provided us with our EERI affiliation (thanks Marjorie – they really helped). We ended up going right to the top meeting with the Federal Coordinating Officer in charge – who was surprisingly interested, genuine and supportive of what we were trying to do. Before going on this trip we had been warned that the response phase was still in full swing and to steer clear of all government agencies involved with response. The actual situation is quite different. Response was over and the recovery problems were looming large – top on the list debris removal and processing and health issues related to water supplies contaminated by damaged cess pools. Debris is a problem – tsunami debris is very dirty, generally a mix of sand, vegetation, metal and building materials. Fortunately they have adequate space for debris and the landfill and scrap metal yards are in close proximity. They try to do primary sorting of metal/no metal on site so that the debris only needs to be handled once. The traditional way of dealing with woody debris and vegetation is to burn. But much of the wood is pressure treated and the burning creates additional environmental issues.
One of the group we met with was Joe Tolard, head of the Department of Homeland Securities Geospatial Analys group – he’s a whiz at GIS and space-based imaging and an alum of HSU’s international studies option in Environmental Systems masters program (he said hi to Steve Steinberg). He gave us a great number of maps and before/after space based images.
Met with the National Park folks. The park lands were barely impacted by the tsunami but park headquarters and visitor center is at the head of Pago Pago harbor where some of the highest water levels were observed. The building was destroyed along with many archeological artifacts, and all the computer data. Fortunately some of the computer data had been backed up and they were able to retrieve some data off retrieved C-drives, but some was lost. Another lesson – if its important, keep backup computer data in different locations. We spent a long time talking with the park archeologist who experienced the peak of the tsunami in Pango Plaza, one of the hardest hit areas. She is also from Leone and three of the victims were her sister-in-law and two nieces and a sister is currently in intensive care in Hawaii.
We met with territory officials in the afternoon and reconnected with the three engineers in our group. A lot of discussion on how to tackle some of the main recovery issues. Land use planning, building design requirements and cess pool – sewage treatment issues are at the top of the list. The planning team will need to present a recovery plan with recommendations soon to the territorial government and there are many touchy issues. Land ownership is complex and traditional in Samoa villages. Sewage treatment in low-lying areas is very difficult. No detailed tsunami hazard assessment has been completed here and 60 percent of residents are below the poverty line.
Final meeting of the day was with two FEMA disaster response workers. One was Jeanne Johnston, a 1946 tsunami survivor who has worked with Walt Dudley on collecting tsunami survivor stories for years and also led the State of Hawaii’s tsunami program for civil defense for a few years. She sees the lack of a full hazard assessment for Samoa as a major problem and has been surprised that few NTHMP resources have been spent here outside of some TsunamiReady funding.