Contact
Email addressInvalid email
Phone number
Message
Send
Log in
Sign up
Account recovery
Email address
Password
Incorrect username or password
Email address
Invalid email
Enter your email and we will send you an email with instructions on how to reset your password.
Email address
Invalid email
Forgot password?
Log in
Get started
Back
Send
The message was sent
We will try to respond quickly
We will try to respond quickly
Yes
Message was not sent
Close
Invalid email...
Close
Sending...
OK
Thanks! Registration letter was sent to your email
OK
Close
The technology behind Aerostate products
The diverse data sources and cutting edge tech that help us deliver the most reliable air quality solutions
Our products help users make life-changing decisions and enrich their product with the air quality data. The core value of all the technological efforts of Aerostate is data accuracy and reliability. This page describes the science and technology behind our products. It overviews the data and algorithms we use to provide our users with the most accurate air quality forecast and analytics.
How it works
Data mining
Satellite imagery
We use imagery taken from US and EU satellites that track weather, air pollution and Earth surface.
MODIS, Landsat, Sentinel, EUMETSAT, GOES and many other missions and satellites are crawled and stored within Aerostate backend.
These data is the fuel for the air quality analytics.
Air quality measurements
Aerostate technology is data agnostic. It integrates governmental and industrial observation sites as well as the personal
air quality sensors. The measurements are cross-calibrated and used to train AI algorithms and refine the forecast.
Industry emissions
As the baseline solution we use CARMA database and then enrich it with our proprietary data.
The data about the additional industry sources is mined from OpenStreetMap and the emission rates are reconstructed
using the inverse modeling techniques.
Traffic
Data feed from global traffic providers is downscaled by the AI algorithms that utilize the city structure to calculate
the number of the cars in the streets and the average flow velocity. These values are then converted to the emission
rates after the local emission policies are applied.
Weather
Air flows, temperature and turbulence conditions affect the pollutants dispersion in the planetary boundary layer.
We use data from global weather providers and refine it with the finely tuned mesoscale atmospheric models.
Data storage and processing
All the information gathered by the system is stored and preprocessed in the Map-Reduce fashion in order to be served either to the atmospheric models or to the AI algorithms.
Atmospheric models
In Aerostate we have developed the atmospheric physics and chemistry model based on the open-source Weather Research and Forecast (WRF)
model. Our commits to WRF allow it to utilize broad range of new input data and calculate the atmospheric flows and parameters
more precisely based on the processing of the data within the urban environment.
We combine global weather data providers, satellite imagery and local measurements in order to construct optimal boundary and initial conditions for the model calculations.
We combine global weather data providers, satellite imagery and local measurements in order to construct optimal boundary and initial conditions for the model calculations.
ML and AI
We use machine learning to refine the forecast, simulate the atmospheric flows and
dispersion, downscale the atmospheric parameters and traffic data. We utilize all the
modern AI machinery: gradient boosting, long short-term memory
The product
All the previous steps are accomplished to deliver the best quality product for our users.
We serve millions of API calls daily and support various Aeroplan installations for different
locations and industries. Aerostate API provides the users with 72 hour forecast of 10 air
pollution spieces. Aeroplan helps decision makers to monitor and manage the air quality
using the data and UI we provide.
Live air pollution map
Try it for free
Get started