Citizen science and air quality monitoring in Tokyo

As you may have recently heard, the city of Beijing recently went off the scale for established measurements of the amount of particulate matter in the air.  As of the 13th of January the U.S. embassy, the only publicly reporting monitoring station in the city, measured a concentration of 886 micrograms per cubic meter (To give an idea of how high a level this is, consider that the average concentration of particulate matter in Los Angeles county is 13.9 micrograms per cubic meter.).
Issues such as this are the driving force behind Safecast Air, which seeks to develop and distribute open-source air quality monitors so that participants can record local air quality data to a public database.  This type of work, which will allow for the public to engage in scientific research, is the type of work members of LA Makerspace engage in as citizen scientists.  As both an educator and member of LA Makerspace I also work to tie my teaching in with this type of work.  In teaching high school science I also have students who work with me to run these projects, Safecast Air included.  Our work will allow for us to both supplement existing air quality monitoring networks, such as the AQMD in southern California, as well as to allow people in cities with little or no current monitoring efforts to begin to establish a tangible environmental data set for their own neighborhood.

Currently, those of us involved in developing Safecast Air, are in Tokyo for a weeklong hackathon in order to develop and test the equipment needed to make a replicable air quality monitor for public use.  To this end both myself and one of my students, Noa, went out to test some of our sensors at the Shibuya crossing in Tokyo.  Here’s a video of one of our recent tests.

The device we were using monitors temperature, humidity, the concentration of Carbon Monoxide, as well as the concentration of Volatile Organic Compounds (VOCs – essentially unburned hydrocarbons like methane and gasoline ).  The sensors we’re using are designed to change their electrical resistance in response to a change in the concentration of a particular gas.  For example, the resistance of the VOC sensor will drop as the concentration of VOCs rises in the air.  Calibrating these sensors involves converting these readings, often in combination with temperature and humidity measurements, into a physical value for gas concentration.  What we have so far in running our monitor are these resistance changes.  Below is a graph taken from the VOC sensor on the air monitor from about twenty minutes of standing by the Shibuya crossing:
Here’s the graph:

What is important about this graph is that we see sharp drops in resistance approximately every 120 seconds.  These sharp drops in electrical resistance on the VOC sensor corresponds to a sharp rise in the concentration of VOCs at the intersection, which happens at the beginning of every traffic cycle when cars suddenly accelerate.  In the first few seconds of acceleration the dozens of cars at the intersection are emitting higher amounts of unburned gasoline, which the sensor detects as a VOC.  These drops also correspond to the traffic cycle at the Shibuya crossing, with 60 seconds of car traffic followed by 60 seconds of foot traffic.  As the Shibuya crossing is the busiest intersection in the world it is reassuring that our monitor can see this periodic behavior in the concentration of a known contaminant.
As we continue this work we will be adding on sensors to monitor the concentration of Nitrogen Dioxide, Ozone, and particulate matter to the monitor, allowing for the first replicable prototypes to report temperature, humidity, the concentration of particulate matter, as well as the concentration of four common atmospheric contaminant gases.
We’ll keep you posted.