In a post dating back to the start of July, I wrote about the Where Do You Think You’re Going digital transport event at Hoult’s Yard in Newcastle. One of the particular things which had interested me at this event was a five minute talk by someone called Margaret Bell from Newcastle University, who was talking about the Transport Observatory.

As I said at the time:

This looked at mapping actual traffic data, and air quality/pollution data and modelling what happens to these flows under different conditions. This enables those planning transport to investigate what would happen if certain changes were made (what would happen if you added a bus lane here, what impact on pollution would it have if we changed the order of these traffic lights and so on).

The Transport Observatory was a fascinating use of data, and I’m planning to get in touch with Margaret to find out a bit more about this work…

Building Blocks Of The Digital Economy

And I did get in touch with Margaret, and she did send me some more information. Unfortunately, what with one thing and another it has taken me quite some time to get around at looking at this, but I’ll have a shot now.

Basically, there are four universities involved in the programme — including Imperial College as well as obviously Newcastle University — and between them they have installed 100 sensors monitoring levels of five different pollutants.

The scientists will also model pollution clouds in 3D, by attaching sensors to traffic lights and street lamps. They aim to understand how it forms, lingers and dissipates in high-emission zones. The team hopes this will lead to insights about whether, for example, poor signalling causes traffic congestion that contributes to reduced air quality in the area.The Engineer: Pollution Project

Once the models have been shown to be correct, this will allow modelled run-throughs of pollution maps with some of the variables changed — for example, would adjusted traffic signalling or priorities make a difference?

Different sensors were developed by the different universities — for example the Newcastle sensors are not as big as those designed by Imperial College, but have a significantly longer battery life: once put in place they can remain in situ for 6 months before needing to be recharged. Live data is being relayed from 30 sensors in Newcastle, 15 in London, 50 in Leicester and 2 in Palermo. Sensors were also previously deployed in Gateshead during early testing and development.

I think this is a wonderful idea: if we can map and model pollution, it gives us the ability to understand more about it and look at methods to minimize pollution and/or ameliorate the hazardous effects. Obviously there are some drawbacks: a lot of the official documentation seems to have been written in that strange language academese:

Modelling is used by the traffic manager to design an appropriate control strategy to alleviate pollution hotspots. Often several options are considered (such as queue cascading or re-location to less sensitive areas of the built environment or open spaces to facilitate dispersion) and the most suitable for the specific problem is implemented. Crucial to the accuracy of the air quality prediction is the validation of
the traffic micro-simulation model.Pervasive sensor applications: Model Validation and Assessment

…and you can’t help feeling that the project is called Mobile Environmental Sensing System Across Grid Environments simply so they can use the acronym MESSAGE, but from what I’ve seen of the data, and the modelling — the capture and processing of real time pollution information and the pretty graphics showing it — it is going to be an amazing tool.

Sadly, for the moment this data is not generally available, but I sincerely hope that the MESSAGE project will look at — once they’re satisfied that their models are fine — making the data available online in a very graphable, mashable format. And from a personal point of view, the sooner the better — if we can get access to real time pollution information and graphs, that’s brilliant. If the modelling part of the system can’t be included at the same time, never mind — but let’s get this data out there and get it shared.

Timetric show how useful data comparisons can be — for example, if you want to know the CO2 pollution per capita, take a look at this:

I knew the United States was producing a lot of CO2, but China seems to be mentioned as a big polluter, but if you compare the per-person tonnes of CO2 produced, there really is no contest: per person per year, the United States emits around 19.8 tons of CO2; the UK 9.2, China 4.2 and Belize 3.0. Admittedly, there are more people in China but why should they feel the need to cut their CO2 output when each Chinese person is producing less than a quarter of the CO2 of each US citizen?

And Timetric shows the power, the shareability, the embedability of what can be done with data. Please let’s ensure that other data that is collected — including the MESSAGE data — is shared and made available in this way.