Global air quality forecasts
Air quality impacts our health and quality of life. We develop analytical services to mitigate the impact of air pollution on people’s health. We combine air pollution satellite data with other data sets like weather and population density to create hourly air quality covering the world.
Symptom tracking and dose-response functions
The forecasts are displayed on a mobile app to end-users. The users can also track their own symptoms to air pollution and pollen. We combine the registered symptoms with our forecasts using machine learning to develop dose-response functions for individual users. The users can then better understand their symptom triggers and take actions to avoid exposure to these.
Many users with respiratory issues are affected by a combination of air pollution, temperature, humidity and pollen. To help these users and people with pollen allergy, we are developing a pollen model to quantify and forecast pollen levels.
Users and their needs
Our key customers are people with non-lethal diseases triggered by air pollution and pollen. Globally there are at least 1,5 billion people affected by asthma, rhinitis, bronchitis or pollen allergies.
Major user groups:
- People who suffers from periodic asthma, sneezing or coughing and parents with children suffering from asthma/respiratory issues (global coverage)
- Need precise forecasts and advice to plan their daily actions, the goal being to reduce exposure
- Need to understand what triggers their symptoms
- Need forecasts and advice to plan travel
- Could use profiling of locations to decide on where to buy a house
- People allergic to one or more pollen types (Europe, possibly North America)
- Need alerts well before pollen season starts in their location, to start medication in time
Want forecasts to better plan their days (exposure and medication)
Service/ system concept
App - Key capabilities to users
- Air Quality (PM2.5) hourly forecast for 7 days (Global)
- Hourly pollen forecasts and 1-week warning for pollen season start (Europe)
- Logging and analysis of individual symptom triggers
API data sales
- Pollen and air quality forecasts (output from models)
Space Added Value
Air quality models
- From Sentinel-2 and Sentinel-5P: Particulate matter (PM2.5)
- Will be used to calculate air pollution levels and calibrate our air quality forecast models
- Aiming to use vegetation index from Sentinel-2 to determine chlorophyll levels. This in turn predicts the pollen release.
- Few pollen models are currently available on the market, the satellite approach will give us a potential global coverage and competitive advantage.
- Set up the infrastructure for continuous download of air quality data from satellites and ground stations
- First version of the model for estimating air quality globally is operational
- Investigating SILAM model for use in pollen forecast models. Explorative data investigations (phenological data, vegetation/chlorophyll index from satellite).
- Recording and reporting on forecasted pollen levels in Europe
App and web development
- Third version of app in final testing before launch
- Web page with weekly blog posts operational
- User interviews with target customer groups have been conducted
- Market surveys and competitor analysis performed
App screenshot, Air quality forecasts
Calculated hazel pollen levels per 21.01.2020