BreezoMeter can now report on air quality up to an incredible micro-resolution of 5-meter air quality data. We’ve been asked many times how we’ve achieved this, so we thought we’d explain things for you in more detail!
Physical Sensors Don’t Provide the Whole Picture
The majority of air quality data providers rely on physical monitoring stations, which means the data they provide is entirely dependent on the number and location of physical stations.
In most cases, only a subset of sensors is available in each station – these do not necessarily monitor all pollutants and report with an average delay of many hours. As a result, they present only a partial reality of actual air quality. Often, these air quality data providers make it look like air pollution varies at the state or city level. In truth, air pollution varies much more than this – even from street to street and hour to hour.
Mapping Air Pollution Between the Sensors
To accurately represent the spatial representation of air pollution, you need to calculate the air pollution dispersion between the monitoring stations.
BreezoMeter achieves this by taking into account multiple sources of air pollution information like traffic patterns, millions of connected cars, satellite data, weather, active fire’s smoke models, meteorological information, models and land cover in addition to the information provided by monitoring stations.
On top of this, we deploy highly-sophisticated algorithms and machine-learning techniques. All this then goes through a strict QA process before reporting to ensure the highest level of accuracy.
The end result is the most granular, high-resolution and most accurate real-time air quality information available today.
BreezoMeter’s Air Quality grid: 5 meter air quality data
You can imagine our air quality grid as a kind of woven blanket laid over the world: Each tiny square is made up of tiny grid points that are just 500 meters or less away from each other.
Visual representation of BreezoMeter’s 500m Air Quality Grid
In itself, the representation of air pollution at the granularity of 500 meters is a huge achievement. BreezoMeter is the only provider that can deliver air pollution at this level and it requires huge computational power: Almost half a billion geographical grid points are calculated on an hourly basis and for each grid point, we report on up to 12 different pollutant types and support 61 different local Air Quality Indexes (AQIs). This adds up to more than 6 billion pollutant concentrations and AQI calculations every hour!
Understanding Our New 5 Meter Air Quality Resolution
Our air quality grid is essential for understanding our new 5 meter granularity. Until today, when users queried the air quality at their location, they would receive information based on the closest grid point. In order to progress our data resolution from 500 meters to 5 meters, we added a new Traffic Vectors layer to our model based on live and micro-local traffic information.
This new Traffic Vectors layer means that in addition to calculating a particular lat/long query, we also check whether or not there is a traffic jam at that location. If a traffic jam is registered, the reporting will take the specific traffic pollution into account, if there is no traffic jam, the information will continue to deliver information based on the query’s closest grid point.
For the first time, the addition of this new traffic vectors layer means we can calculate information within the squares of the grid itself – in other words, it’s a lot more granular:
How Do We Calculate Traffic Jams?
Every 12 minutes we calculate the air pollution emitted by 10 million traffic jam sections around the world and report this information for 30,000+ cities worldwide.
To perform these calculations, we partner with several different traffic data providers who deliver real-time information such as the number of car lanes at a destination, average speed, the severity of reported traffic jams, and more. For each section, we use unique machine learning algorithms that take into account the parameters and geographical location to calculate the pollution emission of each traffic jam to its local surroundings.
Visually Representing 5 Meter Air Pollution Data
To ensure our updated air quality resolution was clearly represented in our heatmaps, we applied a new color scale to make it easy to distinguish between different streets, landmarks, and roads in the city.
The end result is a much more granular representation of the dynamic reality of air pollution in the city and 3D-like visuals of a city’s true landscape.