Particulate matter air quality assessment using integrated surface,satellite, and meteorological products: 2. A neural network approach

摘要

In recent years, sparse, surfac-based air quality monitoring has been improved byusing wide-swath, satellite-derived aerosol products. However, satellites are sensitive tothe entire aerosol column, not only the aerosol near the surface that impacts human health.In part 1 of this series, we used multiple regression to demonstrate how inclusion ofmeteorological analyses can help constrain the surface level proportion of the aerosolprofile and improve the estimate of surface PM2.5. Here, instead of multiple regressiontechnique, we describe an artificial neural network (ANN) framework that reduces theuncertainty of surface PM estimation from satellite data. We use 3 years of MODISaerosol optical thickness data at 0.55 pm and meteorological analyses from the rapidupdate cycle to estimate surface level PM2.5 over the southeast United States (EPAregion 4). As compared to regression coefficients obtained through simple correlation(R = 0.60) or multiple regression (R = 0.68) techniques, the ANN derives hourlyPM2.5 data that compare with observations with R = 0.74. For estimating daily meanPM2.5, the ANN techniques results in correlation of R = 0.78. Although the degree ofimprovement varies over different sites and seasons, this study demonstrates the potentialfor using ANN for operational air quality monitoring.