Title Information
Type of Resource (primo): images
Abstract: Common air pollutants are known to vary over a wide range of spatial and temporal scales, including intra-urban. Analyzing air pollution data in the frequency domain enables an understanding of the layered periodic processes that together form the overall signal. When conducted on a suite of monitoring stations in a single geographic area, spectral analysis yields information about both the spatial and temporal scales of the underlying processes; thus, it can be utilized to differentiate between short- and long-term variability. Breathe Providence is a hyperlocal network of 24 low-cost air monitors located in Providence, RI, USA measuring common air pollutants (PM2.5, CO, O3, NO, NO2, and CO2) at a high temporal frequency. Here we use spectral analysis and Breathe Providence monitoring data to explore the contributions of local sources and weather- or seasonality-driven effects on pollution patterns in Providence. Using the Fast Fourier Transform, 12 months of PM2.5 and CO data from 16 Breathe Providence monitors are decomposed into high-, medium-, and low-frequency bins whose respective contribution to overall variability is quantified. Correlations between pollutants and wind speed and direction are performed in each frequency bin in order to explore pollution sources and relationships with meteorological factors. Results demonstrate that in Providence, low frequency variability (such as caused by regional transport and seasonal influences) appear to drive PM2.5 concentrations, while a combination of local and regional sources appear to drive CO. Meteorological correlations indicate an increasing association of higher CO and PM2.5 levels with lower wind speeds as the period of pollutant pattern lengthens. This effect is more pronounced with CO and demonstrates the influence of stagnant conditions concentrating locally-produced pollution. Correlations between pollutants in each frequency bin improve west-to-east across Providence, which could reflect the homogenization of pollution patterns in the downwind direction. This work shows that spectral analysis of dense networks of low-cost sensors can provide valuable insights into air pollution controls across different spatial and temporal scales. In the future, these insights will be used to develop predictive nowcast and forecast models. "
Name
Name
Name
Origin Information
Subject (Local)
Subject (Local)
Subject (Local)
Subject (Local)
Subject (Local)
Subject (Local)
Subject (Local)
Subject (Local)
Genre: posters
Note: funding: Funding provided by Clean Air Fund
Access Condition: use and reproduction: All rights reserved
Access Condition: rights statement (href="http://rightsstatements.org/vocab/InC/1.0/"): In Copyright
Access Condition: restriction on access: All Rights Reserved
Identifier: DOI: 10.26300/fg9q-n852