Estimated surface air-pollutant concentrations (L4)
Description
Estimated air pollutant concentrations are converted from GEMS L2 total column aerosol, NO2 measurements into surface air-pollutant concentrations, which directly impact human health, by incorporating AI techniques and diverse data sources such as meteorological data.
Especially, to better reflect the most recent trend of air pollutant, this AI-based method adopted a real-time learning model that uses the previous 30-day data and accordingly regulates the sampling and weighting ratios of low and high concentrations.
※ ESC has released estimated surface PM10 and PM2.5 concentrations since 30 December 2021. Additionally the imagery of estimated surface NO2 concentration is to be released from 20 December 2022.
Applications
Estimated surface air-pollutant concentration distribution can contribute to overcoming the spatial coverage constraints of ground-based air quality monitoring networks by providing air quality information of areas with no or few monitoring sites in operation.
Estimated surface air pollutant concentrations tend to be sensitive to errors of diverse auxiliary variables such as total column AOD, NO2 and weather forecast model(UM).
In addition, the uncertainties in estimated air pollutant concentrations can increase if in-situ monitoring sites are rarely distributed.
In this connection, it’s difficult to monitor marine regions, especially, since they lack in monitoring networks and have different characteristics from the land.