AAAR 37th Annual Conference October 14 - October 18, 2019 Oregon Convention Center Portland, Oregon, USA
Abstract View
Utilizing Hygroscopicity of Aerosols to Develop Corrections for Low Cost Air Quality Sensors
SAHIL BHANDARI, Brandon Feenstra, Ashley Collier-Oxandale, Wilton Mui, Vasileios Papapostolou, Andrea Polidori, South Coast Air Quality Management District
Abstract Number: 880 Working Group: Air Quality Sensors: Low-cost != Low Complexity
Abstract Low cost sensors (LCS) provide the opportunity for sampling at higher spatiotemporal resolution while minimizing costs. The ability to map the large variability in individual personal exposure makes LCS especially relevant to air quality agencies. However, LCS need to undergo multiple data corrections in an attempt to meet the performance standards of EPA-approved methods instruments. Purple Air PA-II is an optical PM sensor, with thousands of devices deployed across the five continents. The sensor uses proprietary algorithms to convert measured particle counts to particle mass. Recent work on sensor-specific laboratory calibrations did not account for climate-controlled and species dependent hygroscopicity of aerosols.
Theoretical work by Petters and Kreidenweis (2007), validated by laboratory experiments, addresses this issue using a physically meaningful, species-specific, and size-dependent single parameter model that quantifies the hygroscopicity parameter kappa for species and their mixtures. Over the period 2016-2018, the Air Quality-Sensor Performance Evaluation Center (AQ-SPEC) at the South Coast Air Quality Management District conducted a long term study to evaluate the performance of PA-II sensor against the FEM GRIMM EDM 180 and 24-hour gravimetric measurements from the EPA Chemical Speciation Network (CSN).
Using results from this field evaluation, we utilize estimated kappa from GRIMM to quantify collection efficiency and volume equivalent diameter in each bin measured by the LCS. Here, we compare this effective kappa to that obtained from an external mixture assumption of source-apportioned speciated gravimetric mass from CSN. We further compare the size-species resolved kappa to size-only estimates and mass-only estimates.
Utilizing these corrections, we can better quantify PM2.5 mass measurements based on number counts in LCS with proprietary algorithms. These PM2.5 measurements could be used in source apportionment models and their performance quantified relative to measurements from regulatory and research grade instrumentation.