Key points are not available for this paper at this time.
Exposure to abnormally high concentrations of particulate matter and ozone can cause severe harm to human health, highlighting the need for real-time satellite monitoring to enable rapid responses and timely warnings. However, the existing methods for on-orbit diagnostics under resource constraints are limited. This study presents SoC-POM (System-on-Chip for Particulate Matter and Ozone Monitoring), a real-time, on-orbit artificial intelligence algorithm embedded in satellites and designed to detect anomalous concentrations of PM 2.5, PM 10, and O 3 . Based on tests with the Himawari-8 and Himawari-9 satellites, we demonstrate that SoC-POM achieves an average latency of 5.5 min, breaking through the hourly processing barrier while maintaining high accuracy, with correlation coefficients of 0.78, 0.76, and 0.81 for PM 2.5, PM 10, and O 3, respectively. This novel approach enables real-time monitoring of abnormal particulate matter and ozone levels and demonstrates the potential for the timely analysis of health exposure and its dynamic changes, marking a sustainable advancement in air pollution alerts and public health.
Chen et al. (Tue,) studied this question.