Establishing Robust Correction Schemes for Improved and Reliable ARM-AOS Aerosol Optical Data Products

Abstract

Aerosol light absorption and scattering of solar radiation play an important role in the earth’s atmosphere in terms of direct and semi-direct radiative forcing.  Optical parameters of importance to the US Department of Energy (DOE) climate models include absorption and scattering coefficients, single scattering albedo (SSA), absorption Angstrom exponents (AAE), and the asymmetry parameter (g). These parameters depend on aerosol size, shape and composition (refractive index), and are spectrally sensitive in the shortwave region. Additionally, these parameters have a complex dependency on the emission source, especially for carbonaceous aerosols.  The DOE Atmospheric Radiation Measurement (ARM) user facility has deployed aerosol observing systems (AOS) containing several filter-based instruments to measure and constrain aerosol optical properties and related parameters at multiple sites worldwide. For measurement of aerosol light absorption, the AOS includes filter-based instruments (particle soot absorption photometer and tricolor absorption photometer) that infer particle-phase aerosol absorption coefficients at nominal red, green, and blue wavelength bands from the attenuation (ATN) of light passing through a particulate filter on which aerosols are deposited. Measurement of aerosol scattering is done in situ using nephelometers. By combining inferred absorption coefficients from filter-based ATN measurements and in situ scattering coefficients, value-added products (VAPs) such as SSA, AAE, and g are derived.

Filter-based measurements of aerosol light absorption are strongly influenced by filter type and specific filter characteristics that determine potential absorption enhancement due to multiple scattering from filter medium and deposited particles, location of particles in/on the filter medium, angular distribution of scattered light, and particle morphology changes upon deposition. Currently, two correction algorithms are implemented–Bond/Ogren and Virkkula–for inferring particle-phase optical properties from ARM filter-based measurements. Both correction algorithms were formulated for reference materials (and not real-world/ambient particles) and suffer from unquantified errors and artifacts. Consequently, application of these correction algorithms for processing of ARM raw filter measurement datasets and VAPs limit the use of ARM data products by the community toward achieving DOE Atmospheric System Research’s  strategic goal of “utilizing ARM observations and process-level models to develop, evaluate, and ultimately improve the parameterization of aerosol-radiation processes in climate models”.

This study is focused on developing robust process-based correction schemes for accurately determining aerosol optical properties using ARM filter-based measurement techniques and bounding the associated errors. Wavelength-dependent correction factors, based on first principles, will be established for aerosol emitted in urban and biomass burning environments. New filter-aerosol correction schemes and process-based understanding resulting from this study will be applied to ARM absorbance datastreams collected during ARM field campaigns and long term measurements in fixed observatories. Differences and improvements assessed in comparison to Bond/Ogren correction schemes will be comprehensively investigated and reported. The proposed work will enable explicit quantification of uncertainties in retrieved and derived ARM-AOS data products essential for more faithful representation of aerosol optical properties in DOE’s regional and global climate models.