Leveraging an open-access low-cost sensor network and an open-source R-package to observe changes in air quality both locally and globally before, during, and after the implementation of COVID-19 related measures

Abstract:

To curb the spread of COVID-19, a “Safer at Home” order was implemented in March of 2020 in Los Angeles, which resulted in drastic changes to the routines of residents and businesses and therefore also changes to emissions. The implementation of COVID-19 related measures provided a unique opportunity to examine the potential for low-cost air quality sensor networks to provide insight into not only air quality trends but also the effects of certain actions. Given their accessibility and capacity to provide high temporal and spatial resolution data, low-cost sensor networks may be able to provide useful information regarding the results of interventions and policies intended to improve air quality. The Air Quality Sensor Performance Evaluation Center (AQ-SPEC) at the South Coast Air Quality Management District leveraged the open source R package “AirSensor”, developed through a partnership with Mazama Science, and the open-access data available from the PurpleAir sensor network, to explore the data from approximately 500 sensors in the South Coast Air Basin and approximately 5000 sensors globally. In this presentation, we will discuss how regional PM2.5 trends observed pre, during, and post the implementation of the official “Safer at Home” order in Los Angeles compare to the trends from previous years. We will also discuss how PM2.5 trends pre, during, and post implementation of the official order vary at the local level and how techniques such as time series clustering can support the analysis and interpretation of data from large sensor networks. We will provide an overview of approaches to QA/QC and procedures for processing data that can help to enhance the reliability of data from low-cost sensor networks. Finally, we will share what we can learn from this sensor network at the global level.