3D city models have many purposes ranging from urban planning, simulation, navigation systems, modeling, change detection, etc. They are in general used to visualize reality. The slums especially lie in the geographically (and otherwise) challenging areas and a 3D model could help with understanding some of the challenges people there face.
The idea was first presented to me by a friend, an architect Tomas Chapman, who runs a very interesting project in Johannesburg, South Africa, called Backdoors to the City (here’s an article which explains it well). He saw the building extraction that we did in Mathare and suggested that we build a 3D model. He explained me the basic principles and introduced me to Google SketchUp.
Another friend and architect April Hiebert liked the idea and decided to help work on it. She and her husband Dale Zak came to Mathare (Dale was at the time working for Ushahidi in Nairobi) to see the area and get a rough idea of how the place looks. We decided to estimate building and area heights from the satellite imagery, and complement this data with some basic measurements from the field. April then used Google SketchUp to draw the basic 3D model.
There’s still some work to be done on the model and I already have some ideas on how to use it further (I’m saving that for another blog post). In the meantime, below are some screenshots of the 3D model of Mathare.
This is a short follow-up on the blog post which I posted couple of months ago titled Doing the other 90% in Kibera. In the post I talked about the possible strategy of the Map Kibera Trust: “The Trust’s role will be a steady supplier of information and the communities, NGO’s, government etc. the implementer of activities”. Stakeholder involvement is crucial to ensure that the data is useful!
Under my mentorship, Zach – one of the mappers at the Trust – then created two maps for WSUP and a map for KWAHO. We also provided a file (an excel spreadsheet) with all the information collected.
I still believe that the Trust’s role should be the information supplier – sort of an information center within the community, and in working with KWAHO and WSUP I see a lot of potential to not only make our maps accessible to community members, but also to the organizations that are working on specific areas to improve the community.
In the past year my colleague Jamie and I have been refining our model and approach to improve upon the organization’s approach to community involvement and interactions. We’ve come a long way, and our experiences with WSUP and KWAHO demonstrate the potential for this approach.
I realize that sometimes words are not sufficient to describe the elaborate mapping projects. For this reason I posted a bunch of pictures on Flickr – and will keep adding more – with a hope to bring in the human face of the projects.
The approaches to paper mapping are well described by Mikel in his blog post called Paper Mapping in Community Meetings. What I would like to add to his points are the reasons why we conducted the paper mapping in the first place. To name a few:
Feedback. To acquire feedback from the wider community regarding the maps and the data collected by a few community members.
Understanding the place. To understand the areas where we are mapping – more accurately, to at least trying to understand complex places like Kibera. Outsiders to communities often overlook this step, and end up conducting projects in communities that they don’t really understand and haven’t taken the time to learn much about.
Acquire more information. We wanted to collect more information and double-check the information collected, find the missing objects, listen to the stories regarding particular issues, and to identify specific problems and solutions to these problems.
Monitoring the accuracy of the mapping. Through community’s input we were able to monitor the accuracy of our mapping endeavor.
These were the reasons I decided I’m going to look closely into these paper/drawn maps and try to understand what the people from Kibera were telling us. I’ve done a similar exercise before with Joshua Goldstein (as a part of Map Kibera) who was at that time working for UNICEF Innovations. We – together with the rest of the Map Kibera team, of course – had created the first, and the most comprehensive Security map of Kibera. Now it was time to focus on other problematic areas, where we also involved community members who drew on our existing map to add information and new issues. For that reason, there were, and still are, some twenty or so drawn, paper maps stored safely away in cyberspace, which cover issue topics ranging from education, health, water and sanitation. To extract useful information from these maps required some Photoshop magic. The first maps I looked into were the water and sanitation maps.
So what did we ask, and what did the community tell us?
Identifying missing resources:
Is it representative?Is anything missing or mislabeled?” were some of the questions we asked to help us see the extent of the missing data. The feedback made us realize that we missed a lot, and it made us realize we need to re-do the mapping.
Identifying problem areas:
What and where are the biggest problems regarding water and sanitation in the community? Which water points and toilets would you say are the worst and why? Are there many people who use public spaces for toilets?” were the questions which helped us identify the most pressing issues and problems. Some of the major problems identified were:
All streams are used as toilets and dumping sites
There is a health hazard because the people at times use the railway as a toilet, garbage dump & a cooking point
Misuse of the drainage system as a dumping site
Stories behind issues:
What would you say are the largest and most direct challenges to providing quality water and sanitation? What is the biggest need regarding water, waste disposal, and sanitation in Kibera? What would you want the Government and other decision makers to know about water, waste disposal, and sanitation in Kibera?” These questions helped us collect stories that lent insight into the complex dynamics of Kibera. Some stories:
Different amounts of money is required depending on which toilet you use
Solid waste management is poor
Vigilante groups demand money for the toilets which are to be constructed in their area
Maintenance of toilets is a big problem; toilets get polluted quickly
For “short calls” people use plastic tins; flying toilets (plastic bags in which people defecate and then throw out of their home) are used at night because of insecurity
Households use paper bags for waste collection; these bags are then picked up by individuals and groups who bring them to collection points for a certain amount
Where do your collect water? Where do you use the toilet? Where do you dump trash? Why?” By asking these questions we were able to observe the movements connected to the water and sanitation. Some of the findings that came out of the mobility mapping were that:
People dump their trash into rivers, and the rivers serve as collectors of most of the garbage from the slums
There are some points of data collection within the slum where the city council collects the trash and transports it out of the slum
People don’t necessarily use the closest toilets and water points but look for the most affordable ones
How would you solve some of the problems you pointed out? In what ways can you envision using this map to improve water, waste disposal, and sanitation in Kibera?” These questions encouraged people to share their views on how to solve certain problems regarding water and sanitation:
Awareness and education
Need for weekly garbage collection
More dumping sites, more toilets to be constructed
Stakeholder involvement – coordination between stakeholders
Partnership involvement – CBOs, NGOs, and landlords should come on board to make sure that each plot has a sanitary facility/usable toilet
Paper mapping or drawing exercises proved to be a very useful feedback mechanism and also a very useful tool for planning. The community’s knowledge is priceless and only the community understands and knows all the answers. There’s no point in figuring out something new when the solutions are already available and when all you have to do is just listen!
Before the late 1980s and early 1990s when some of us were so excited at what we were finding local people could do, much indigenous, local and participatory mapping had already taken place in different regions, countries and continents. Mapping and various forms of spatial representation by local people on their own have a long history, and very likely a prehistory!
More remarkable than what local people had already done in mapping and other forms of spatial representation was “our” educated professional ignorance of their mapping abilities.
A lot of ink has been spilled writing about how technology is only 10% and all the other stuff you have to do to make the project successful is 90%. These two posts talk in detail about the issue: Allocation of time: Deploying Ushahidi and Why technology is 10%. Nowadays we all agree that this is true so I’m not going to add my two cents to this discussion. What I want to write about is how Map Kibera Trust (the Trust from now on) plans to start doing the 90% in Kibera.
Let me first paint a picture of the situation at the Trust at the moment. The Trust has around 20 on and off members, who were trained in basic GPS and OSM techniques, video editing and Ushahidi platform. Because of this there’s loads of information that exists mostly in cyberspace. We believed – and it was an honest belief – that if we opened up information, people would make good use of it. But apart from a small number of individuals, mostly foreign, that have used the data for their academic research, the data stayed untouched.
The problem was that we did things the wrong way. We collected information first and then started asking people if they need it. Our approach was supply driven instead of demand driven which was nicely pointed out to us by an independent IDS research: Mediating Voices. Because of this we have now backtracked to make a new action plan for community engagement.
The question we asked ourselves is: “Now what?”
The answer is not simple and to at least start working on it Kepha and I sat down over coffee, wrote DATA on the middle of a piece of paper and asked ourselves: “What’s next?” In a short brainstorming session we came up with a general plan of community engagement in Kibera. What we realized was that the Trust is going to need help. And the help needs to come from within Kibera, from the people living and working in the community.
We decided we will start by networking and organizing community meetings at which we will present the information collected so far. At these meetings we will organize so called “peoples committees”, each representing different issues.
I will explain the work of these committees with an example concerning education:
As I said, Kepha and I started with the word “data” at the beginning of our brainstorming exercise, which is obviously not the best way to start. But making the best of the current situation, we decided that through community meetings, networking, and presentation of our maps and database of educational facilities, we will organize an “education committee”. The committee will have two branches or types of members – Trust members and Stakeholders.
The Trust members will be the link between the stakeholders and the community. Their role will be to collect and supply the information, analyze and advocate for better and new ways of information usage. We see the Trust more as a supplier of information than an implementor or the end user of this information.
Collected information will end up in the hands of the second branch consisting of community members, NGOs, local administration, private sector, legal institutions etc. Their role will be to act upon this information by writing action plans, proposition statements, determining what kind of projects should be undertaken next, involvement of government representatives and lobbying for better service provisions in Kibera or other activities.
This will be a mutual partnership between the Trust and different types of communities in Kibera. The Trust’s role will be a steady supplier of information and the communities the implementer of activities. Of course this is just a general idea but we hope it will get something rolling.
So will the people want to be a part of something like that?
I believe the answer is Yes! I’ve seen people excited when they saw the data, the maps, and the videos. Organizations need information – facts – in order to do their work or to address certain issues. I’ve seen people talking at community meetings, contemplating how to use the data to plan activities or who to engage when information was presented to them in an understandable manner. In Mathare, where we began by talking about data to community groups, we found a large demand for data by community leaders, and groups. It’s something about having facts, a proof, in your hands that makes you fill with possibilities, with hope that you can actually do something and move from just talking about things to actively doing them. For once I’m optimistic.
Our biggest focus during this pilot project in Mathare was on four villages: Mathare No10, Thayu, Mabatini, and Mashimoni. This is because our local partner (Community Cleaning Services) operates in the area and also because we wanted to test our capacity – figuring out how much can be done in the limited amount of time.
Apart from mapping toilets, water points and open defecation in the area; building extraction from the satellite imagery; and through ground truthing building classification, we also mapped most of the open drainage lines.
These are some of the results:
All four villages cover 209848 m2 out of which open defecation areas cover 4014 m2 or 2%.
We collected 8.35 kilometers of open drainage lines in the area.
There are 25 points where toilets are located, altogether with 95 units (toilets). These are all in, or in the immediate vicinity of, the four villages, and most of them are public toilets.
These toilets serve almost 4000 users.
Hand washing facilities and towel bins are almost non-existent.
There are 30 water points in the area, all operational, and mostly privately owned.
The second phase of Water and Sanitation mapping is over and with it the pilot in Mathare, which lasted for 4 months. You can read all about the events and the processes leading up to this point on our previous blog entries. I will call this an intermediary analysis because the work is by no means over, but we’re at a good stopping point to reflect and set the way forward.
Here’s the data we collected regarding toilets, water points and open defecation areas, and what it tells us (these numbers reflect the data which we collected during our pilot):
We collected 144 points of interest where toilets are located in the whole of Mathare. These points have 373 individual units attached to them. These are mostly publicly accessible points, meaning we only took points which are public and, for example, didn’t walk into people’s homes.
Most of these toilets are privately owned, with 50% of the toilets having known ownership. The other 50% are owned by different organizations, government, etc. and are considered public.
The majority, 50%, of the toilets with known type are Asian Style, followed by 26% pit latrines, and 11% hanging toilets. The rest are trench and English toilets.
Most of the toilets are located in the public within the communities; this is because we collected mostly public toilets in the first place. I’m happy to say that 90% of collected toilets are operational.
Most toilets are connected to sewage (60%), and have available piped water (54%), and are usually cleaned, either by the operator (38%) or the caretaker (33%). Many toilets have almost non-existent hand washing accessories (37%) and non-existent towel bins (0,7%).
Couple of visualizations of the upper data displayed on maps:
We collected 167 water points.
60% of these water points are privately owned. As with toilets, the rest are owned by different organizations, government etc. and are considered public.
Most of these water points are piped or tap water (75%), while others are water tanks.
As with toilets, most of the water points are located in the public within the communities.
98% of collected water points are operational and have connection to water pipes.
A visualization of the upper data displayed on a map:
Open defecation areas
Open defecation areas cover 16144 m2 of Mathare (3 km2) and their placements vary throughout the slum. They are mostly located in the areas where there are majority corrugated iron sheet houses, which usually don’t contain toilets.
I believe that the reasons to map open drainage in the slums are well known and are obvious to most people. Everyone who’s ever been to the slums knows that open drainage presents a huge health hazard to the people living around it. In combination with poor or non-existent water and sanitation systems, open drainage presents a recipe for disaster which is always present and can erupt at any time and wreak havoc amongst the populations. If I quote a few experts on the dangers that open drainage poses:
“Potholes in the streets, pools of stagnant water, and waste gushing from bathrooms and kitchens provide breeding sites for malarial mosquitoes and other spreaders of disease.” (Nwaka, Geoffrey I. “The Urban informal Sector in Nigeria”)
“In slum areas, a minor flood is not just inconvenient. It is life-threatening. With open drains and overflowing latrines even a small flood means that children have to wade through raw sewage.” (Willem Alexander, HRH, Chair of UNSGAB. “World Water Day Op Ed”)
In an article on the health status of people of slums in Nairobi, Kenya the authors conclude that: “Environmental conditions can have major influences on health status. Therefore, environmental improvements are important in the improvement of health status.” (Gulis et. Al. “Health status of people of slums in Nairobi, Kenya”)
As my friend Simon, a long-time resident of Mathare and our Map Mathare Coordinator, put it on the Mathare Valley blog: “One does not need to be a scientist to know what would happen in case of disease outbreak. The improvement of proper water distribution and repairing the broken water pipes coupled with constructing good drainage system is the key to fighting common illness in the slums.”
What I can add to all of this is that in order to improve or at least start improving living conditions, you need to first know and understand the situation on the ground. By mapping open drainage areas, our mappers are providing the first detailed geographical information about how open drainage is distributed throughout Kibera and Mathare, thus providing necessary information to anyone hoping to solve the problem.
This I believe answers the question of why and brings me to the next question: How?
Obviously open drainage creates “appalling living conditions,” but maybe not so obviously, it creates appalling mapping conditions too! Open drainage winds and curves between houses, often sinks below them, intersects, flows, stands, overflows, merges with the content of broken sewage lines (sewage lines burst and sewage comes above ground, mixing with the content of open drainage), and creates pools of standing water. Open drainage is the main collector of garbage: garbage in open drainage areas often clogs drainage systems, thus creating floods during rains. All of this mesh slowly flows downwards, usually towards rivers that pick it up and carry it further downstream, polluting the environment sometimes hundreds of kilometers away.
In order to understand the scale of the problem in Mathare and Kibera we went on an investigative – or, if I put it in army terms, reconnaissance – mission with our teams to see the problem first hand. Our participants surveyed the problem, and their input helped us to better understand what we’re up against. We walked through the slum evaluating different drainage lines and problems we faced in mapping them, Sebastien (our new volunteer) and I then facilitated the mappers to decide on which points of interest we should collect and in what order.
Findings from the field regarding open drainage include:
Open drainage has many sources and it forms for many reasons (resulting from topography and from human actions)
It comes in many forms regarding the content and structure (garbage, water, sewage, earth trench, concrete trench, etc.)
It has main channels with many tributaries
We came to realize that drainage systems in slums (Kibera, Mathare) basically work as micro river systems! This made us understand that it has to have an outlet, and both Kibera and Mathare have outlets in the form of rivers that run through them.
All of this helped us made our action plan:
All the drainage flows towards the rivers, so we started collecting data along the rivers.
The next step is to collect data for the main channels, which lie along the main paths.
After this, we focused on the smaller paths in the slum, collecting points of interest along the way. This includes the start and end points for “following” a drainage line, intersections of different lines, man-holes (along sewage lines), and broken sewage lines. At intersections, we’re collecting the directions of different intersecting lines
Throughout this process, the use of building extraction (when possible) helps us sketch the drainage lines
After the extensive data collection we’ll use the satellite imagery and building extraction to help us determine the exact positions of drainage lines.
So, what’s next?
Just knowing and understanding the situation will not make it better. This means that our process cannot end with mapping, but will take the map and the information we collect a step further to actually make the information useful and create an impact. The next step is bringing together community members, stakeholders, NGOs, local administration and government representatives and expose the problem with a method I like to call: “A punch in the face” – where we present the problem, with maps and other media, in all its vastness, and hopefully get some heads thinking and acting upon the information which is presented. The map lets stakeholders present problems to responsible actors in a comprehensive and difficult to ignore manner, making the issues more concrete than just words by providing detailed geographical representations of their scope and characteristics.
In order for this last step to be successful, local ownership, insights, and understandings of both the issues at hand and how mapping and presenting information can help bring change, is crucial. Creating this local ownership amongst key stakeholders – namely the local Map Mathare and Map Kibera teams – has to start from the very beginning of the process. We are working hard to create this local ownership from the beginning in Mathare and to bring a greater sense of ownership to the Kibera teams.
Map Kibera and Map Mathare are currently involved in vast mapping operations regarding Water and Sanitation in both Mathare and Kibera, two of the biggest slums in Nairobi. These projects are big as the issues they are trying to address and will present test whether exposing vast quantities of information about a particular issue truly has the ability to influence change. As I often say: Collecting data, as complicated and hard as it may be, is the easiest part here.
The first phase of detailed water and sanitation data collection and editing is over. In two weeks our teams collected and edited 262 points including: 89 points with toilets, 108 water points and 65 open defecation areas.
What we learned from the numbers:
The smaller number of toilets does not show the real picture as there can be many toilets at one point (we collected the number of units per point, but this is not reflected in the 89 figure).
The number of open defecation areas is alarmingly high – there is almost the same number of toilet sites as there are open defecation areas – which clearly points to a lack of toilets.
The number of toilets and water points will likely stay low because of the specific structure of Mathare: half of its buildings are high-rise apartment buildings that usually have toilets and water connections on every floor , which brings us to the next point:
In order to get the real picture of the water and sanitation in Mathare, we need to figure out how to collect the data from these apartment buildings (it will need to be an inclusive approach, connecting community members, stakeholders, administration, government, etc.)
There was some confusion here and there because of the massive amount of papers which contained the data of different points (something to think about in the future – how to get rid of the huge stocks of paper), but otherwise the data entry itself went well.
Although the Ground Truthing of the shapefile which was sent to us by AAAS was done in three days (read more here and here), fixing the shapefile itself took a bit longer. In the process, we created three shapefiles.
The main shapefile contained the closest resemblance to the situation on the ground. Our teams have checked and corrected on the ground the shapes of most of the buildings, written down the types of the buildings, and classified them. From this exercise, 4 attributes were added:
a) Type means the type of building, which is either Brick, Corrugated Iron Sheet, or Wood Stalls
b) Name is name of the building (if it has a name)
c) Designated is the designated use of the building, describing what the building is used for. It can be many things, including: school, house, business (like bar, cyber cafe, garage, hardware shop, market place, etc.), church, toilet, etc.
d) Action is what kind of action was carried out on the shape of the building – there are two options: 1) Added if the building was missing and was added and 2) Modified if the building was either moved, merged, or in any other was modified
The third shapefile contains deleted shapes (buildings) because they are either not there, are incorrect (there are many shapes representing one building), or don’t represent permanent structures (like temporary tents in the Chief’s camp).
Our teams have collected around 750 points in the first run of data collection and are now it the second phase of detailed Water and Sanitation data collection. The idea is to eventually attach the points they collect to building extraction for the whole of Mathare.