Mapping Haiti's uncharted communes
There's a book on my desk called Up and Down California in 1860-1864. It's about the first land surveys of the West Coast of America, through a (long) series of weekly letters from one of the expedition's crew to his brother in New York. This was back in the heyday of pioneering, romanticized by many, especially now when it seems we've charted every square inch of the planet.
One team at the Haiti Rebuilding Center is proving this not to be the case. And their methods of cartography would look strange indeed to the surveyors of yesteryear. #opensource
Design Fellow Olivia Stinson has been coordinating trips out to remote and exotic ends of Haiti for the past six months. Her teams, Haitians with GPSs and laptops are producing what these places have never had before: maps.
With a grant from the WK Kellogg Foundation, Architecture for Humanity are mapping eleven communes across southern and central Haiti: Cavaillon, Maniche, Mirebalais, Saut d'Eau, Boucan Carre, Lachapelle, Arcahaie, Aquin, St Louis du Sud, Cayes, and Ile a Vache.
The resulting maps account for basic geographic and political information made available for the first time as a resource for the country. Many of the towns, villages, roads and waterways in these communes are not currently located or cataloged on a map, and settlement patterns are currently roughly estimated through aerial and satellite photography.
After months of work, entire days wandering through villages marking coordinates, Olivia's survey team has almost completed the first few maps to be presented to communes.
I recently caught up with Olivia via e-mail. (Our Skype meetings kept being pushed off due to unexpected work/rescheduled site visit/broken-down car/bad reception...you know, a typical day in Haiti.)
Headquarters: What was your background coming into your design fellowship?
Olivia: Most of the planning work I've done in the past has been on the coordination and project management side of development, in general working in large urban areas like New York City and Jakarta, Indonesia, on infrastructure and community planning.
Often, we were using existing data to make our decisions or relying on other experts to generate analysis for reports and planning interventions. This is the first time I've been working on the data development side, and especially in rural areas of a country and one as poor as Haiti. The dearth of available information here to make planning, design and development decisions is staggering. Especially considering the desperate need of governments and organizations like AFH working in participatory planning and furnishing information to our colleagues and partners. We have to start at Square One with data collection and validation in this instance, and that's been an entirely new experience for me.
HQ: How did the project come about? Were you brought on specifically for it?
OS: The WK Kellogg Foundation hired AFH to do a mapping project so they could better understand different assets around the country and where to direct their community investments. When I started at AFH I did not know that I would get so involved with mapping and data creation, but as a planner it has been a great opportunity to learn a lot more about mapping first hand.
Surveyors prepping for another day with their GPS devices
Logging data back at the Rebuilding Center in Petionville
HQ: I can imagine. Could you talk a bit more about some of the coordination challenges, and victories?
OS: Being new to Haiti and handling a lot of my work with a mapping team in French, and running the logistics for organizing trips and data collection in a timely manner in Haiti, I felt slightly overwhelmed. We had to make some decisions about the technology were going to use, how many people it would take, and the schedule for a kind of project we hadn't carried out before.
It was very difficult to find qualified people who weren’t either hired already by another NGO, or who had basic computer knowledge, let alone mastery of more sophisticated mapping programs like GIS. Our solution was to ask advice, interview a lot and hire the smartest, most astute people we could find - then to train them. We invested a few weeks in the training so we could start working in the field and get the job done.
We decided on an open source platform called OpenStreetMap, which has become incredibly important in post-disaster Haiti. OpenStreetMap crowd-sources information about the physical environment. There is a great community of mappers here, dedicated to this program and the objectives of community ownership of information, to help spread needed data freely and quickly.
Through the trainings and field mapping work we've done, we have not only been able to help my Haitian team develop a skill set that will certainly lead to jobs for them with other organizations later, but we have collectively created crucial, open-source map information for the country at large.
HQ: What specifically was mapped? How does the new information change the landscape, as it were?
OS: The mapping has generally been for major, quantifiable community assets and physical attributes. We developed the database based on fairly simple observed qualities - we did not for the purposes of these surveys do household surveys or lots of talking to people about socio-economic information. Our principal goal for the project has been to find where things are. For the most part we mapped schools, churches, clinics, businesses, infrastructure like roads, water points and historic areas.
The GIS data we've collected can be used to do analysis now by the Kellogg Foundation. We'll be bringing the maps back to the communities we surveyed. It will be interesting to see what the reaction is considering most of these very small areas have never had maps before.
HQ: You mapped entire communes or did you focus on populated areas?
OS: We mapped the towns as well as the rural areas of all 11 communes. Because of the differences of scale, it made more sense for us to produce two maps for every Commune, one for the town and the other for the full area.
HQ: What's the difference between "rural" and "urban" process?
OS: Rural mapping is much more time consuming than mapping in town and usually yields a lot less information. For example the roads in Haiti are in pretty poor condition. They make for slow-going in a 4x4 just to find a small rural school or a church and take a single GPS point.
In town, it is possible to move relatively quickly on foot to do data collection, and of course things are much closer together. You can gather things easily and validate them later. For the rural work, you really only have one shot to get it right out in the field.
The coastal village of Aquin as mapped on Google Earth, Nov. 2011. A decent map, for approximations.
Aquin's data points are collated
Aquin today on OpenStreetMap
Aquin City final map - click to enlarge or explore the PDF
HQ: What's surprised you throughout the mapping project?
OS: What has surprised me the most is how complex data creation is and how time consuming it is in Haiti, where the logistics can be difficult at best, especially in rural areas of the country. However, it is in these rural areas that there is the most intense need for new and complete geographical data. We could study aerial photos of the areas we were mapping and see where buildings were or what a town looks like, but in order to fully understand what was actually there, we had to take the time to go into the field and mark the assets one by one.
What has been exciting is getting involved in the community of international and Haitian mappers totally dedicated to creating this data. They are active, vocal, demanding of quality and also incredibly inclusive. We have become active contributors to that conversation. It's been interesting, enlightening and gratifying to be able to part of this overall movement in Haiti for good, clear available information needed by so many.
HQ: Any other notes on impact of the project?
OS: We partnered with the National Geospatial Agency on data sharing and are actively contributing to the national database. The other benefit for AFH, my teams and even myself, is that we are fully participating within an international movement for open source mapping and data development. It has been extremely rewarding to watch the team grow and learn, and to help contribute in a quantifiable way. We have been told that the rural mapping work we are doing is extremely important and needed, from both Haitian government agencies and our NGO partners. The data and maps we've created through the process can be added to and amended at any time.