Open Environment Data

These sensors pick up bits of information that cameras and microphones never could: they smell the air, they taste moisture, and they feel the sting of pollution. Though a camera can give you a general sense of how a place looks, these sensors can tell you very precisely about all the complex networks — from weather to social groups — that make a place what it is.

Six weeks from idea to prototype – now we are waiting for Kumbha Mela to start in order to collect for the first time ever environmental data in realtime from a pop-up city in India.

First time when the data came in from our sensor box

Here is how the entire story started.

Six weeks ago I’ve sent the link of data canvas to a friend in Delhi who is running a data analytics company, a brief note attached: “Do this in Delhi” and the PR will be yours!”. The idea of the data canvas in short: Install sensors all over the city and measure in real time the city’s environmental data (air quality, noise, light, temperature, CO2 ….) and visualize it. It was laid out as an art project. And it was sponsored by Seeed in Shenzen, a company I had just visited. This was actually what caught my attention. Seeed provided the Arduino boards and the sensors for the project.

My friend, M2 and his business partner, immediately got the idea and jumped on it.
And then the story evolved.

I connected M2 with Seeed, they figured out what kind of sensors would be the best for getting an overall idea about the quality of the enironment in a city like Delhi, I went to Shenzen MakerFaire two weeks later and ordered the sensors while I was there. Another week later M2 and Srinivas Kodali (an open data expert) applied for the MIT Kumbhathon, they got accepted and they went. I was there as a mentor.

M2 and Srinivas introduced the idea and immediately four hardware students from Pune joined the team.

With MIT Kumbhathon the focus had changed a bit – we were much more practical and focussed now because all of a sudden we had an ideal test market – the Kumbha Mela – right in front of us. During this huge religious gathering more than 20 million people will come Nashik. This would give us the chance to measure environment data under “regular” circumstances, but also when the city suddenly pops up to 10 or 20 times its size. What would change? Are there any changes at all? How ould environmental data correlate with health data?

For the very first time we will be able to measure this kind of data (we have 9 different sensors) in real time. This will enable authorities and government official to achieve alerts when an alert case occurs and it will provide the baseline for predictions for any future pop-up situation in any Indian city.

During the MIT Khumbathon the newly founded team of volunteers finished the first 6 prototype boxes, we got the datastream running and the very first sensor box is already sending data. There are a few bugs which need to be fixed, but the system is up and running. The website and an app will provide realtime visualization within the next 2 weeks. In mid August the team is planning to return to Nashik and to implement and install another 45 sensor boxes.

Our first prototype from the inside

During the MIT Kumbhathon we also defined our open data strategy and outlined how we are planning to move forward:

Our goals

  • To achieve a better understanding of how environmental paramenters change during pop-up situations.
  • To gather a sandbox full of data which make pop-up cities more predictable.
  • And – since data is publicly available – to make citizens and its authorities smarter. Smart citizens for a smart city.

Our business modell

  • All data will be published in a well structured format under a creative commons non-commercial share alike licence.
  • Commercial use and specific analysis and visualization will be chargeable.

Our organizational structure

  • Currently we are a loose structure of volunteers committing our time to this project.
  • Midterm ideally the organizational structure is transfered in some kind of trust or society where individuals and companies can become members.
  • We want an advisory board with a diverse range of people (data & environment scientist, MIT Media Lab, authorities, ….)

Our partners

  • Other MIT Kumbhaton projects (cloud steering and epidemic tracker)
  • Nashik officialsand authorities as well as government
  • Outside MIT Kumbhathon: Seeed, juxt smart mandate, Janwar Castle

The plan with “my” Janwar Castle is to assemble the 45 boxes with the kids in Janwar and teach them slowly how to do this. It will be th ebeginning of the first maker space in rural India, right next to the first skateboarding park in rural India;-)  The bamboo house which will be the home of the young makers will be ready by the end of August … another milestone in Janwar. I believe a good way to link urban development with rural India – for the benefit of all!

The entire project is set up as a sandbox – meaning we are providing the data pool, we structure the data and provide it in an easy to use way – all this for free as long as feeds into a non commercial use-case. We lay the ground and open up to the collective intelligence of all how to make sense and use out of this data.

After the Kumbh we will think how to move forward.

The learnings will help us a lot – I am sure.


These are the parameters we measure:

Carbon Dioxide
Air Quality
Dust (PPM)
Multi Gas (Ammonia, NO2, CO)