Uncharted territory: mapping the world’s forgotten cities
A ‘human genome project for cities’ is digitally mapping the developing world’s vulnerable urban areas to help NGOs and governments better respond to crisis, writes Chris Michael
In 2010, an earthquake struck Haiti. Hundreds of thousands were killed. Then things got really bad. Cholera broke out, and Ivan Gayton, of Medecins Sans Frontieres (MSF), received a call from a nun in a remote village in the middle of the Haitian forest.
“Please help,” she said. “We’ve been hit with a horrible disease we’ve never seen before.”
Gayton sent a few trucks of MSF staff to try to track down the nun in Baraderes, but they had only the sketchiest of maps. With every village they reached, they thought, “This must be it – they’re dropping like flies here.” But it wasn’t. They left behind equipment and staff, and kept driving. The road ran out. They took a boat up the coast. Eventually, they arrived at a dock, “piled with corpses like firewood”, Gayton says.
This was the nun’s village. The MSF staff did what they could. Over the next few days more than 600 deathly ill people arrived. MSF saved most of them; nearly half would likely otherwise have died.
Across Haiti only 0.9 per cent of the people treated in MSF clinics succumbed to cholera, compared with the 40 per cent mortality rate for untreated cases across the country; the charity spent US$100 million and saved countless lives. Thousands of Haitians died anyway.
Gayton likens the process of choosing where to establish clinics to making “war triage” decisions.
“In triage, some people are badly wounded, others nearly dead. We have to treat the people in the middle to maximise the number of lives saved. We can’t help everyone,” he says. “In Haiti, entire communities and areas were left to die. We got a call like the one from the nun once a week.
That’s one of the only ones we responded to. I’m convinced we missed a huge number of people.”
Part of MSF’s problem was that it was so difficult to know whether sending people to Baraderes was the best use of resources.
“Haiti was this gigantic emergency intervention and at MSF we’d committed ourselves to being the ones who’d respond – but we couldn’t do it all,” he says. “So I became quite keen to know exactly where the greatest needs were: the greatest impact for the greatest number.”
Location, location, location – this mantra has been central to epidemiology since 1854, when physician John Snow took addresses from cholera victims, put them on a map of London, traced the source of the cholera outbreak to Soho and turned off the contaminated water pump. And yet a century and a half after Snow’s breakthrough, cholera and other diseases continue to overrun the world. Part of the reason? A lack of basic maps.
In the developing world, a surprisingly large (although unknown) number of the world’s cities with populations of more than one million people remain unmapped. Development staff in those cities trade photocopies of photocopies of rough maps, scrawl the names of landmarks on post-it notes, use satellite images that lack street names or just ask locals.
In more developed parts of the world, we take it for granted that Google has mapped every street in every city. But when MSF asks patients in more remote parts for their address, the answer they receive is often of little use. Bobere, for example – is that a village, a street, a neighbourhood, a district or a province?
“Much of the time it may as well be random syllables,” Gayton says. “If there was a point source for cholera in Haiti, we wouldn’t have known where it was. We needed a map to be able to correlate the alerts we were hearing and our patients’ origins to something on the ground.”
Such a resource wouldn’t be useful in Haiti alone. If humanitarian agencies had good digital maps of places that are particularly prone to natural disasters, conflict or epidemic diseases, they’d be ahead of the curve and could respond effectively as soon as a crisis occurs.
The solution is the Missing Maps Project, an ingenious collaboration between the MSF, the American and British Red Cross, the Humanitarian OpenStreetMap Team and others to create free, digital maps for every settlement on Earth. It is nothing less than a human genome project for the world’s cities.
The method, developed by the Humanitarian OpenStreetMap Team, is deceptively simple and, crucially, it allows volunteers to take part remotely.
The first step is to take satellite images – which, perhaps surprisingly, are often made available to the open mapping community from such sources as United States government agencies and Microsoft – and plug them into the free mapping software OpenStreetMap.
Volunteers then log in from anywhere in the world and use an easy point-and-click tool to literally trace the outlines of buildings, roads, parks and rivers over a satellite image. Remove the image and voila: you have a basic, digital map.
Next, the map, which still lacks street or landmark names, is printed out and posted to volunteers in that city. These “ground troops” – anyone from students to bankers – each take a small section of the map, head out with a pencil and write down the names of streets and buildings.
Finally, the completed maps are posted back to Missing Maps HQ, in London, where volunteers fill in the names on OpenStreetMap. The result: a free, open source city map.
The first big test case was Lubumbashi, a city in the Democratic Republic of Congo that is home to 1.5 million residents. Missing Maps now aims to map the rest of the world’s poorest and most vulnerable urban areas within two years. And because the first and last mapping phases can be done by anyone with a laptop, anywhere, the whole thing can be crowd sourced.
“Finally, I can give volunteers something to do that isn’t just giving money,” Gayton says. “Many people want to help MSF, hands on, not just donating. They offer to knit socks for the kiddies and so on. I tell them, ‘No, don’t knit socks for the kiddies – the cost of getting those socks out there and distributing them just isn’t worth the potential benefit.’ “But with Missing Maps, they can actually participate in real, genuine fieldwork. That’s huge.”
Google Maps is also trying to plot cities in Africa and the developing world, in response to criticism that the company was ignoring places where there was no advertising money to be made. After all, there’s no Starbucks in a slum.
Like Missing Maps, Google has been asking people to voluntarily flesh out Google Maps, trying to harness the power of aggregated individual acts volunteered by everyday people. However, unlike Missing Maps, Google Maps is – like the rest of Google’s projects (regardless of their current openness and freedom of use) – privately owned and at any time the tech giant could choose to start charging for the service.
“The point of the project is that the maps will all be open source,” says Missing Maps coordinator Pete Masters. “It will be illegal for anyone to charge anyone to use them – meaning local people will have total access to them, not just to look at, but to edit and develop.” That idea more than any other has fired the imaginations of the people in unmapped places, says Gayton.
“It’s legally impossible for someone to steal it, to close off and own the data. It’s created by genuine volunteer labour and belongs to everyone.
And the question everyone asks me is, ‘Why?’ They couldn’t believe it.”
ULTIMATELY, THE MISSING MAPS PROJECT will need to recruit the biggest team of digital humanitarian volunteers ever conceived, as it now also tackles Ebola-torn areas.
But the benefits of digitally mapping all human settlements are by no means only about epidemiology. The possible applications for urbanism – transit planning, waste removal, housing strategy – are vast.
“Maps make us think of things like finding a restaurant on the web or navigating with a satnav,” says Harry Wood, a member of the Humanitarian OpenStreetMap Team board.
“But somebody also used a map for deciding which road the bus route should go down, and deciding where the road should be in the first place.
When you think about it, maps are a very basic fundamental knowledge tool, woven into our urban environment, and silently improving our lives in all sorts of ways.”
So it’s not just about fighting Ebola and cholera – though that is crucial, Wood says.
“When the Ebola epidemic passes, OpenStreetMap will remain as an information resource silently and indirectly helping with urban planning and economic growth in these countries.”
Just as the human genome project simply recorded genes, with scientists subsequently using the data collected to study disease and create medicines, so, too, are city maps just the building blocks for urban planning.
Whether the goal is to prevent cholera or crime, Missing Maps is a crucial first step. For much of the developing world, our current digital maps may as well say, “Here there be dragons.”
There’s a long way to go, but someone has, at last, picked up a sword.
Guardian News & Media
Preparing for the worst
The Humanitarian OpenStreetMap Team (HOT) is a non-governmental organisation that uses open source and open data sharing principles to improve the welfare of communities, 0particularly those in at-risk areas or experiencing a crisis.
Kate Chapman (right), executive director of the organisation, a public charity incorporated in the United States, says, "OpenStreetMap is the broader community of which we are a part and that is supported by the OpenStreetMap Foundation, which is UK based.
"The goal of the OpenStreetMap project as a whole is [to provide] a free map of the entire world."
An American, Chapman began contributing to OpenStreetMap in 2009, helping to fill in the details of her neighbourhood in Washington DC. "I got involved in the Humanitarian OpenStreetMap Team when it was still informal, in January 2010, after the earthquake in Haiti."
She has recently moved back to the US from Jakarta; Indonesia being one of two countries in Asia in which HOT has active projects (the other is Mongolia).
Elsewhere in Asia, HOT coordinates with existing OpenStreetMap teams. "For example, in the Philippines and in Japan there are very active communities that happened on their own," says Chapman.
"We often provide technical assistance to governments. Typically it is for disaster preparedness. We work with stakeholders to determine what needs to be mapped, what organisations and individuals are interested in helping contribute to the map, and lead workshops/training on how to contribute to and use OpenStreetMap."
The HOT team in Indonesia numbers 11 and is led by Yantisa Akhadi. "We are focusing on four target provinces - West Sumatra, East Java, South Sulawesi and East Nusa Tenggara," he says.
Those provinces have been identified by the Australia-Indonesia Facility for Disaster Reduction as being the most urgently in need of mapping.
"Rather than mapping directly, we train people how to map. A common response from people that we train is their amazement that all of the mapping effort is done manually, road by road, building by building. They often think that some kind of smart algorithm will scan satellite images and then map the roads and buildings automatically, with little human intervention."
Mark Footer
Want more articles like this? Follow SCMP Lifestyle on Facebook!