10th International Aerosol Conference September 2 - September 7, 2018 America's Center Convention Complex St. Louis, Missouri, USA
Abstract View
Effects of Multiple Scattering by Fresh Soot Aerosols on Open-Path Optical Diagnostics of Atmospheric Plumes
BRADLEY CONRAD, Matthew Johnson, Jeremy Thornock, Carleton University
Abstract Number: 1566 Working Group: Instrumentation
Abstract Combustion-generated soot (often termed black carbon when quantified optically) has been identified as the second-strongest direct radiative forcer in our atmosphere. Globally, soot emission sources include residential combustion, biomass burning, agricultural waste burning, transportation, and gas flaring. Apart from the transportation sector, most soot emissions are from open-atmosphere combustion processes that are inherently difficult to quantify. This has impeded development of accurate global soot emissions inventories. However, non-invasive, open-path optical diagnostics of soot-laden atmospheric plumes are a promising measurement approach that is intrinsically well-suited to this challenge.
Sky-LOSA (i.e. line-of-sight attenuation of skylight) is one example of an open-path optical technique for soot measurement, which has been successfully applied to the measurement of soot emission rate from flares in the oil and gas (OG) industry (e.g. Conrad and Johnson 2017). In this approach, time-resolved grayscale images of a plume are analyzed over a narrow bandwidth in the visible spectrum. The observed transmittance of the soot-laden plume is coupled with two-dimensional velocity data, computed with image correlation velocimetry, to derive the mass emission rate of soot through an artificial control surface in the image plane. The enabling element of the sky-LOSA measurement is the correction of the observed plume transmittance for the in-scattering of skylight and sunlight by the plume, to quantify path-integrated soot concentration at each pixel. In the current algorithm, this is accomplished by assuming only single-scattering of ambient light by the plume. In the limit of an optically thin plume, this is inherently valid; however, as the optical density of the plume increases, interparticle multiple scattering events bias the correction of the observed transmittance for skylight and sunlight.
This work seeks to quantify the effect of multiple scattering by soot on the accuracy and detection limits of open-path optical diagnostics, with a focus on the sky-LOSA technique. Large eddy simulations of gas flares typical of the upstream OG sector are simulated using the massively-parallel ARCHES software under the Uintah computational framework. Gas phase concentration fields of high spatiotemporal resolution are scaled to provide realistic turbulent fields of soot number density. These data are then coupled with calibrations of the sky-LOSA camera under a Reverse Monte Carlo method to assess skylight and sunlight measurement bias by higher-order scattering events. The utility of a multiple scattering correction to Beer-Lambert law is discussed and results are used to infer the relation between multiple scattering effects, plume geometry, and vortex street metrics.