AAAR 37th Annual Conference October 14 - October 18, 2019 Oregon Convention Center Portland, Oregon, USA
Abstract View
Integrating Low-cost Sensor Networks with Fixed and Satellite Monitoring Systems for Enhanced Accuracy, Reliability, and Applicability
JIAYU LI, Huang Zhang, Chun-Ying Chao, Chih-Hsiang Chien, Chang Yu Wu, Cyuan-Heng Luo, Ling-Jyh Chen, Pratim Biswas, Washington University in St Louis
Abstract Number: 338 Working Group: Air Quality Sensors: Low-cost != Low Complexity
Abstract A reliable data source for air quality communication, pollution mapping, and exposure estimation is particulate matter (PM) mass concentrations measured in conventional monitoring stations. A high spatiotemporal density of monitoring stations is essential to achieve a better understanding of PM transport on a regional and global scale and to accurately establish the impacts of such PM, for example, health effects. However, due to the cost and operational complexities, a limited number of such real-time PM monitors can be deployed. PM mass concentration can also be retrieved from aerosol optical depth (AOD) data collected by remote sensing, but these datasets are usually compromised by weather conditions and aerosol optical properties. Monitoring station networks are normally located in populated areas, while remote sensing can cover sparsely populated areas. To increase the measurement density, a network of low-cost PM sensors is a promising approach. With the advent of low-cost PM sensors and advances in data analytics, we have proposed integration of information from multiple measurement approaches. In this study, we demonstrate this approach by synergizing the data from 75 monitoring stations, 2,363 AirBox low-cost sensors, together with Terra remote sensing data for the main island of Taiwan. A machine learning method is utilized to select the useful data from the massive AirBox datasets. Ordinary kriging is used to create a visual PM distribution map. The AirBox and remote sensing datasets are calibrated with data from collocated monitoring stations. The maps created from these three data sources demonstrate an approximate 30-fold synergistic improvement in the spatial resolution of PM mapping, with minimal bias. This method will greatly assist the validation of PM transport models and enhance the accuracy of exposure estimations. With this method, we identified a special pollution event during a Typhoon approached Taiwan.