On November 8, 2013, Supertyphoon Haiyan, or Yolanda as it is locally known, struck the Philippines, leaving unprecedented damage in its wake. Communities in the Eastern Visayas were destroyed by both incredible wind speeds and massive storm surge. In some coastal areas, the storm surge was as high as four meters. Sustained wind speeds reached 195 mph, and gusts measured 235 mph. To put in context, 2017’s Hurricane Maria had sustained wind speeds of 175 mph, and recent Hurricane Dorian’s sustained wind speeds reached 185 mph. Needless to say, Haiyan was one of the worst cyclones to have ever made landfall.
One of the hardest hit areas was Tacloban, the largest city on the island of Leyte. Thousands of homes were destroyed, leaving hundreds of thousands of people displaced. Immediately, the government of the Philippines, as well as the international community, began responding to the disaster. Numerous organizations and agencies began assisting with relief and eventually recovery – particularly as it relates to shelter and housing.
It is on the housing that my research is focused. After disasters, we strive to “build back better” – in other words, we try to rebuild houses and communities that are safer and hopefully, if another disaster is to occur, will not experience the same level of damage.
My research is based on a few ideas: for houses to be safer in the long-term, 1) they have to be built safely, and 2) they have to be safely maintained and/or modified. Tenet 1 is determined by the initial design and construction of the house and is, for the cases included in my study, overwhelmingly dependent on the organizations providing housing assistance after a disaster. The second tenet relies on the household occupants and what actions they take to maintain or modify their home over its lifespan. Given the constraints of doctoral research, I have posited that how households maintain their house will be influenced by how safe or unsafe they perceive their house to be. Thus, my work focuses on 1) identifying household perceptions of housing safety, 2) assessing the safety of reconstructed houses in hypothetical wind and earthquake events, and 3) comparing the findings from parts 1 and 2 to determine how and why household perceptions of safety differ from the results of engineering assessments.
To date, I have finished part 1 and am currently in the Philippines working on parts 2 and 3. On a previous trip, I collected data about a handful of houses that were rebuilt after Haiyan but need additional data in order to finish my wind and earthquake assessments. So, for the past week I have been visiting communities, taking photos and measurements of houses, and doing a lot of counting of nails.
In the time since the houses were built, some houses have seen modifications and improvements; others have seen degradation. Some temporary shelters have become permanent houses. Some houses are abandoned or were never occupied. Being here four to five years after the houses were constructed has given me a unique perspective – given the nature of humanitarian response and shelter recovery, many organizations do not have the opportunity to follow up on their projects four to five years later.
For the purpose of this post, I want to focus on two housing designs implemented in Tacloban. Both were built within a single barangay. Some were built as single-story houses and others, given the size constraints of the lots on which they were to be built, were built as two-story houses. Although the two designs had many similarities (elevated floors, gable roofs, coconut lumber frames), there was one striking difference: the two-story houses had corrugated galvanized iron (CGI) roofs, and the one-story houses had thatch roofs. Why the difference? The two-story house has fared pretty well in the elements over the past five years, but the one-story houses – the roofs have significantly deteriorated, leaving the residents to patch the roofs with tarps and other materials in order to protect themselves and their belongings from the elements.
I recognize the difficulties of providing adequate shelters to as many people as possible in the aftermath of disasters. Having never been involved in a post-disaster program, I can’t imagine the weight involved in decision-making and I don’t intend to critique the individuals who work tirelessly in these contexts. In fact, I am trying to find easy-to-implement solutions for improving housing performance. For instance, I’m investigating how adding an additional nail or two or three to each CGI panel or using screws instead of nails can improve performance. But these solutions operate under the assumption that CGI panels are being used for roof covering. If we’re providing some households with CGI roofs, why aren’t we providing others the same quality?