Ground Truthing Mathare

The American Association for Advancement of Science (AAAS) donated us a WorldView-2 image (read more) with 6% cloud cover, taken on 30th of June 2010. It covers the areas of Mathare Valley, Korogocho, Kariobangi, Baba Dogo, as well as part of Kasarani, Dandora, and some surrounding areas. Altogether, the imagery is of a broader eastern part of Nairobi, containing the areas with the highest concentration of people in the city.

Satellite image of Mathare

The image is of course useful in numerous ways, but the first thing we want to use the image for is to create a building extract for the whole area of Mathare! The AAAS showed a great interest in helping us with this issue and is indeed helping us now. The deal was that we set out a test region to begin with so one of their analysts could analyze that area. We would then take a look at it and go to the field to ground check their mapping. If that worked out satisfactorily, they would expand the extraction further to do the whole area.

As said, we first set out the test region for them to analyze. Our focus was on Mathare 10, Thayu, Mabatini, and Mashimoni. We picked the area because it is situated very centrally in Mathare, has different variety of buildings (tin and brick) and is also an interest area of our partners, Plan Kenya and CCS. A day after the area was set we got the first results – the first automated building extract of the area.

Building extraction on a test area

A quick look already revealed some of the faultinesses of the automated process like overlaying shapes, crooked shapes of objects, shattered uniform objects and merged diverse objects, digitized/mapped non-existing objects, etc.

Some of the faultinesses of the automated process

Our next “job”, which is currently under way, is ground truth-ing the results. This basically means checking every object in the field and comparing it to the created extracts, than confirming or rejecting certain objects, classifying them, repairing the shape files, digitizing collected (repaired) information and sending it back so they can reset their algorithms.

Ground Truthing

The biggest challenge is how do we do that practically with limited equipment and resources? We knew from the start that we’ll get the best results if the people who come from the area are the ones ground truth-ing their own area. This is also important because this kind of work basically takes you into the compounds and backyards of houses and you obviously don’t want strangers walking on other people’s backyards.

For the purpose of data collection I created forms, printed them out, bought some pencils, sharpeners and erasers and called Francis and Jackson from Mashimoni to take me around their neighborhoods and give me the feedback – they are two of the mappers who will be the ones leading the project on the ground later, which is why their opinion is crucial.

Francis and Jackson in action

During the exercise we focused on specifics like:

  • We numbered each house in the test area, classified them and added different types of buildings (tin, brick, etc.)
  • We repaired shapes of the buildings extracted
  • We deleted redundant shape files
  • We wrote down names of all known landmarks

The guys also gave me feedback regarding the forms:

  • Prepare a bigger map of the whole test area and outline the smaller area on it for better recognition of the area
  • Add names of the roads to the forms for better recognition of the area
  • Make darker outlines of extracted buildings on the forms
  • Add names of landmarks for better recognition of the area

After the ground truth-ing of the area the AAAS will (hopefully) expand the extraction to the whole of Mathare. And then the fun really begins!

Base map of Mathare is complete!

We’ve completed the base map of Mathare. It took us 17 days during which 15 mappers (on average each day) would map and edit for an average of 5 hours per day. That accumulates up to 1250 man hours. Of course the exact figure of man hours is much lower mainly due to a limited number of computers and equipment which resulted in many of the mappers just strolling along or observing the creation of the map from afar. We have mechanisms prepared for this type of instance, as we want everybody interested to learn and contribute. One of these mechanisms, which we implemented in the last two weeks of mapping, is rotating venues.

If I throw some statistics at you, the base map has – or, if I put it differently, mappers have collected, edited and digitized:

  • 750 points
  • 41.3 km of roads and paths
  • 24 Mathare villages digitized, altogether an area of 3006444 m2 3.0064444 km2
  • 138 buildings digitized, altogether  an area of 58322 m2 or 0.058322 km2
  • Other areas digitized (walled areas, fields, football and recreational pitches, natural areas, etc.)comprise 360602 m2 0.360602018 km2

Below is the base map where the first map (Map 1) presents only the points collected and the second map (Map 2) presents the types of resources collected (represented by points), such as hospitals, toilets, water points, schools, religious institutions, etc.

Map 1
Map 2

This belongs to Mathare!

Mike: “Karibu Huruma. You want some tea?” Me: “Is it safe here?” Mike: “Don’t worry, you’re very safe here!”

Huruma

In order to encourage participation of as many residents of Mathare as possible, we’re continuing with a rotating venues approach. We are trying to get different groups to host us so we can be closer to the communities, where the participants then explain in detail to all interested parties, from elders to children to grandmothers, what they’re doing, why we’re there and why there are some people walking around with strange telephones (GPS units). We also want to enable our participants to walk to the trainings and not spend their money on public transportation.

Working from different parts of Mathare has another importance: on-site discussion around the biggest needs of the particular community leads to on-site recognition of the particular problem. This then leads to determining what exactly they want on the map, what they would need the map for, and therefore what the focus should be. All this information will be crucial in making good thematic maps which will show the biggest and most immediate needs of the particular area.

Trainings in Huruma

Resident: “People rarely come to these areas; we don’t get any attention like Kibera does; NGOs don’t come here and don’t care about this place; Mathare is very poor; there is almost no development. We say United Nations is for the people who work there, not for us. Aren’t you afraid?” Me: “Should I be?” Resident: “Not if you’re with this crew. That’s very good what you’re doing!”

I still feel afraid every time I step out of the cab with 4 computers and 10 GPS units that someone will knock me over the head with a rock. However, they never let me out of their sight. Even when I go to the toilet someone is standing in front of the door. I feel they are keeping a close eye on us and the programs. They’ve had too many disappointments with projects that failed or didn’t live up to their expectations. We understand these issues and are working hard in Kibera to recognize our mistakes and make sure that we live up to the expectations of our team there (more on this coming soon).

So Jamie and I, with advice from Sammy from Plan Kenya and Simon from Ngoza Njia – Community Development Center, approached Mathare by taking a massive amount of responsibility from our shoulders and handing over most of the decision making to Mathare residents. It’s them who decide what they want on the map, for what purpose they want to use the map, what they want to document with video, where the next venue should be, who to invite to public discussions, how to engage other people to participate, how to make these tools useful to as many as possible etc. From the start we presented our programs as Mathare’s own programs, something no one can dictate or tell them what to do with it. We can already see it and they’re taking ownership. The conversation is often like this, resident: “So you’re going to map all the CBOs in the community?” Me: “I don’t know, you tell me!”

Trainings

Last week (24. Jan – 30. Jan) we worked in 3 locations in Mathare. The mappers were focusing on the eastern part of Mathare – Huruma, Kiamaiko, New Mathare and Mathare Nort. We were hosted by  Ngoza Njia – Community Development Center situated between Huruma B and Kiamaiko. From there the teams spread out and mapped the hell out of Huruma. We collected approximately 200 points in two hours of field work.

On Thursday we were hosted by Vision Youth Group in Huruma – Ngei1 area, where we edited the data collected on the previous day.

Vision Youth Group

On Friday we digitized over satellite imagery as a part of learning different techniques in mapping. We were hosted by Community Transformers in Mathare No10. I’m particulerly proud of Huruma girls (and one man) who digitized a big chunk of Huruma (Kaimaiko area) and were listening to every advice with their full attention.

Huruma Girls (and a man)

Video had some technical problems when trying to edit videos in Mathare No1o at Community Transformers on Wednesday and Thursday. But with the passion and help from a videographer Nathaniel Canuel they will make it next week.

And there is exiting news for the Voice of Mathare: We’re getting there!

I should finish by saying: “No I’m not afraid. As long as Mathare folks know that this is their program nothing will happen to me when I’m roaming through the streets with all that equipment” But just in case, they are watching closely, I can feel it!

Mathare is still there after the New Year

It’s been three weeks since the New Year when everybody in Kenya (at least who can afford) travels to shambaland. We’ve been doing trainings for the past two weeks – altogether 6 days of mapping and 2 days of video.

The map needed (and still does) more information in order to become a good base map for further thematic mapping which will eventually fill in the remaining empty spots (remember: Rome wasn’t built in a day and the map won’t be either – it’ll take 4 months). Here’s the map by the way.

Our main goal is not the map itself! It’s the empowered youths and community members equipped with knowledge and data about their community. As Simon said for Al Jazeera: ”By knowing what we have we will be in a position to engage our government!” So, in order to better control and supervise the masses of the would-be mappers and reporters from Mathare and our trainers-trainees from Kibera, we divided Mathare into 3 parts. West is villages 1, 2, 3A, 3B, Kosovo and 4B, Central is Mathare No 10, 3C, Thayu, Mabatini, Mashimoni and Mathare 4A, and East is Huruma with all sub entities, New Mathare, Kiamaiko and Mathare North. Yea, it’s a big place!

Trainings

The next thing we did was listen to the ever-present, ever-wise voice of the community – Simon Kokoyo (who, in order to amplify the community’s voices, started a blog which is written by Mathare residents called www.matharevalley.wordpress.com) – who said we should rotate or move our venues all over Mathare in order to get the community’s support and make it easier for people to walk to the trainings.

Community Members

The second week of trainings we held mapping trainings at Mathare Environmental Conservation Youth Group/Staken at Mlango Kubwa in village 1 which is run by another well known and respected community member Kaka and at Community Transformers. Video trainings were held one day at Community Transformers and one day out in the field.

Kaka

To help plan the video trainings we invited two established people, both well known in Mathare: Videographer Nathaniel Canuel and a reporter Wairimu Gitau. They’ve structured the trainings in a way to maximize the effect and tell the stories relevant to Mathare. They’ve divided members into 4 teams and talked about the ideas for the videos to be made. They’ve selected 4 ideas which could be done in a day of shooting and they’ve come up with:
– Lack of toilets
– Road accidents or bad road conditions
– Lack of playgrounds for children
– Broken sewage lines
Next week (24. Jan – 30.Jan) they will talk about which footage was good and which was bad and the folks will learn how to edit the data selected. Can’t wait!

Nathaniel taking a nap exhausted after the whole day of video footage collection

Our third program Voice of Mathare is still under construction (web stuff).

Exciting times in Mathare! There’ll be a storm of activity in Kibera soon, so stay tuned!

Map Mathare – Second week of training

The second week of data collection and trainings in mapping and video editing techniques is behind us. We decided we’ll hold map editing separately from video editing because of the overwhelmingly large numbers of people who showed interest in being trained.

The data editing in OSM session was held on Tuesday, 14.12.2010. Turnout was good as always, although some people from one or two villages did not turn up (I guess they’ll do the work on some other occasion). We only edited points as it would take too much time if we tried to edit the tracks as well. The tracks will be edited separately.
Observations:

  • People had different skills as usual
  • Drawing tracks is a bigger challenge than editing points
  • We have to separate people into smaller groups, so we’ll have to rethink our strategy (trainings will have to happen on different days, in morning and afternoon hours, all three programs separately – this is because we need to create smaller teams so the trainings are more efficient)

All was good in the end, some of Mathare got mapped and hopefully some people got excited.

Video and voice trainings were on Wednesday, 15.12.2010. Around 15 people showed up for video in the morning session. No one came for the afternoon session (probably because everybody came in the morning). Kibera News Network guys did a good training and managed to produce two short video reports about A Bicycle Repair Man and Mathare Sewage.
Nobody showed up for voice (probably because people weren’t directly called and most of them have been to the mapping and video trainings). So Sande (second from top to bottom) from Voice of Kibera and I talked about merging mapping and voice trainings. First the trainees would receive training in GPS data collection and editing and after every session they would receive Voice training as well – since it’s a kind of mapping anyway, that way we’ll recruit more members to Voice.

The first two weeks were sort of a sample of what we’re offering. The real trainings and work will start in the year of 2011. We’ll have to figure out how to deal with large numbers. I sense they might shrink with time, but we’ll see.

For you to enjoy, here’s the move edited by Joe from Kibera News Network and Jeff Mohammed from Mathare about Kibera teams training Mathare: Map Kibera presents Map Mathare.

See you in the New Year. Exciting things are coming up!

Or as Jeff would say: Everything is cool, buda!

Jeff Mohammed

Digitizing Village Boundaries of Mathare

The biggest slums in Nairobi (like Kibera and Mathare) usually consist of different villages. That is why before we started working in Mathare we had to determine where these villages are located.

This was/is important for many reasons:

  • Determining the area. By locating the villages of the slum we outlined the area of Mathare.
  • Planning purposes. We use these villages to plan how different teams will operate. The idea is that each team maps its own village, bringing me to the next important thing:
  • Each village needs to be represented. A person coming from a certain area knows that area best.
  • And the last but not least – security. There is no way that a person, specially coming from outside, can freely walk the streets of Mathare.

For the purpose of determining the village boundaries we printed out the satellite imagery of Mathare, which was donated to us by the American Association for the Advancement of Science.

Mathare Printed Satellite Imagery

After that I gave the satellite imagery to Simon Kokoyo form COOPI and Reality Tested Youth Program. He walked around the slum and talked to elders, who helped him determine the village boundaries of the slum by drawing them on the printed satellite imagery.

Drawn villages of Mathare

I uploaded the satellite imagery into JOSM and digitized the boundaries according to the drawn image.

Digitizing in Java OpenStreetMap Editor

And HERE are the villages of Mathare and Mathare slum area in OpenStreetMap – the borders have already changed couple of times 🙂