Mapping Recovery: One Concern in Lombok, Indonesia
Part One of our Digital Anthropology Showcase
By Trigg Hutchinson, Digital Anthropologist at One Concern
In August of 2018, the island of Lombok in Indonesia was hit by a series of earthquakes, including back to back 6.9 magnitude earthquakes two weeks apart. The series of earthquakes was centered on the northern portion of the island, a densely populated area accessible by only three roads, two of which were cut off by landslides. Initial reporting indicated that some villages suffered 100% destruction, and unfortunately these reports proved to be largely accurate.
At One Concern, our machine learning models are trained on past events and real observations, so obtaining high quality data is a top priority. In the immediate aftermath of a large scale natural disaster, our Data Acquisition team goes on high alert and we start exploring ways of generating usable data. Speed is critical, as damage can be repaired rapidly or aftershocks can cause additional damage, convoluting the data. But we have to weigh the need to move quickly against our own version of the Hippocratic Oath — no matter how fast we may want to be on scene, we will not get in the way of first responders or utilize resources needed by the affected population, so in some cases we find that it is more sensible to wait weeks or months before dispatching a team to collect data in the field. This was the case in Lombok, and we took our time organizing a ground team based in Mataram, the capital of West Nusa Tenggara province and the largest city on Lombok.
Our goal was to map as much of the island as possible in a relatively short period of time, so we reached out to Wira, a project manager from Jakarta, and recruited a team of college students from a local school in Mataram. We provided a crash course on post-disaster structural surveying and worked with Wira to begin collection across the center of the island, where damage rates were sufficiently variable to support training our systems.
Lombok is the next island over from Bali, but might as well be on a different continent — which it technically is. Between the two islands lies the Wallace Line, a deep trench representing the boundary between the Asian and Australasian ecozones. While some areas of Lombok feature the same lush jungles as Bali, the island as a whole is more windswept and grassy, especially at higher altitudes around Mount Rinjani to the north. Both islands lie just south of the equator, and are equally hot and humid. Outside of Mataram, much of the island is relatively low density, with villages interspersed between rice paddies, tobacco farms, and ponds of water spinach, a local delicacy.
As we traveled across the island from west to east, we discovered that structural and non-structural damage was much more widespread than we anticipated. We were greeted warmly in each village with offers of young coconuts, chased around by smiling children, and generally made to feel welcome — which made it that much more difficult to see the extent of damage. In almost every village, we discovered residents sleeping under blue tents provided by the national government, scared of entering their homes. And rightly so — we discovered severe cracking on most of the masonry structures, including to load-bearing walls and frames. Many of the houses we tagged could collapse at any point, and given the number of aftershocks the island continues to experience, enduring the heat under a tent was preferable to keeping cool inside a brick structure at such high risk.
As we made our way across the island, we began to notice a difference of building material and techniques in certain villages. To the far east and the south, many homes were built from wood, and these homes performed consistently better during the earthquakes — most were completely undamaged, whereas unreinforced masonry structures nearby suffered complete collapse.
While wood was historically the material of choice on Lombok, today it is predominantly found in two areas — Sengkol in the south, for cultural reasons, and Labuhan Lombok on the coast to the east, for practical ones. Residents of Labuhan Lombok raised homes on wooden stilts to prevent flooding. We discovered that these traditional stilted homes also utilize base isolators — essentially, the stilts are elevated on rocks to allow for the whole structure to roll with seismic motion, preserving the integrity of the structure through an earthquake. This is a very intriguing innovation, also present in traditional wooden buildings in Japan.
Despite such clever construction techniques and the inherent resilience of traditional wooden structures, rapid population growth and changing preferences prompted a broad shift towards brick construction. Entering brick homes in the equatorial heat, it is immediately clear why it is the construction material of choice on Lombok — the interiors of these buildings are significantly cooler than outside, without any sort of air conditioning. Moreover, brick is a much more affordable option that either concrete or wood, as the only wood suitable for construction on the island is yati, or teak, concrete is prohibitively expensive, and brick is manufactured locally. Unfortunately, on Lombok brick is rarely constrained, reinforced, or otherwise made more resilient to seismic hazards — and these buildings consequently suffered the most extreme damage.
To the north of the island, we discovered that many of the villages had suffered complete collapse, often with only traditional pondok, or Indonesian gazebos, left standing amongst the ruins of entire communities. Our travel in the north was truly heartbreaking. While the roads had been cleared of landslides and all routes reopened, it will be years before these communities recover.
After twenty days of data collection, our efforts had yielded over 25,000 structures tagged for damage across the length of Lombok. While the associated data set will be invaluable for our models and improve our ability to predict the impact of disasters in communities like Lombok in the future, every member of our team — both on Lombok and in California — understood that we needed to translate the data into positive outcomes for the population. To that end, we immediately shared the data with a number of government agencies and NGOs active in Indonesia.
For the One Concern team, this was a very particular experience. As a company, we’re committed to building community resilience, protecting lives and livelihoods, and disaster-proofing the world — but these priorities are all future-facing and ultimately abstract. There is a deep emotional response to observing the impact of natural disasters firsthand, and while collecting damage data in such a way is critical for the accuracy of our models, our team felt strongly that there is an ethical responsibility to provide support for events that have already happened, not solely mitigating impact in the future.
This data will be used to direct recovery resources to where they’re needed most. Even if only in the smallest way, we want to leave things better than we found them. As we look toward future data collection initiatives in new locations, we continue to deliberate about how we can give back to the impacted communities we work within, and will remain focused not simply on improving community resilience in the future but on improving outcomes for those already impacted.
If you enjoyed this Q&A, keep an eye out for the rest of our Digital Anthropology Showcase in the following weeks!
- Why Invest In Digital Anthropology?
- Building Resilience After the Disaster: One Concern in Kumamoto City, Japan
Want to help us build planetary-scale resilience?
Check out careers at One Concern to see how you can help.
