AAAR 37th Annual Conference October 14 - October 18, 2019 Oregon Convention Center Portland, Oregon, USA
Abstract View
Improving Estimates of Ground-Level PM2.5 by Application of High Spectral Resolution Lidar
XINYI LING, Nicholas Meskhidze, Kyle Dawson, Matthew Johnson, Barron Henderson, Sharon P. Burton, Chris Hostetler, Richard Ferrare, NC State
Abstract Number: 111 Working Group: Aerosols, Clouds and Climate
Abstract Remote sensing is an effective means of monitoring aerosol properties. Retrievals of aerosol optical depth (AOD) have been used to improve model simulations of PM2.5 concentration and to infer ground-level PM2.5 over regions where in situ monitoring is not available. However, recent studies have revealed that the AOD-PM2.5 relationship can vary from place to place and is strongly affected by aerosol vertical distribution and meteorological variables. Despite advances in active remote-sensing techniques which allow for vertical profiles of aerosol extinction to be used when deriving AOD-PM2.5 relationships, current retrievals provide limited information on the chemical composition of aerosols. Here we present a novel approach in which model-predicted concentrations of aerosol chemical species and surface PM2.5 are corrected using High Spectral Resolution Lidar (HSRL)-derived vertical extinction and aerosol types. This is achieved by using a new algorithm called Creating Aerosol Types from CHemistry (CATCH). By using this algorithm, the relative contributions of PM2.5 chemical components are inferred for measurements acquired during the DISCOVER-AQ Baltimore-Washington, D.C. campaign. The HSRL retrievals of aerosol extinction and types are then used for improving the EPA's CMAQ (Community Multiscale Air Quality) Model simulations of surface PM2.5 concentration and chemical composition. The new approach is examined by comparing the data for the prior and posterior CMAQ-predicted aerosol component concentrations and PM2.5 with ground measurements from EPA’s AQS stations.
Results show an increase in R2 values (from 0.30 to 0.65) and reduction of RMSE (from 13.77 to 7.04 μg/m3) for posterior estimates compared with unconstrained simulations. Furthermore, this new methodology allows for the estimation of PM2.5 concentration and chemical speciation by using HSRL retrievals of aerosol extinction and types alone. Data derived through the combination of the CATCH algorithm with the HSRL retrievals agree well with ground measurements.