Satellite Aerosol Amount and Classification Retrieval Over Bright Surfaces at High Latitude: A Novel Perspective - SALVI

The temperature in the Arctic has increased almost twice the global average over the past 100 years. The climate effects of aerosol have large uncertainties at high latitudes. In spring and winter the Arctic can be affected by recurring haze due to long range transport of pollution (residential combustion, flaring) and in the summer by smoke from boreal regions. Further observational and modelling studies are required, especially at high latitudes, to understand the climate impact of both natural and anthropogenic aerosol. Satellite remote sensing of aerosol is challenging due to the need to account for the aerosol size distribution, chemical composition, shape, and vertical distribution. In addition the reflectivity of the underlying surface and cloud contamination affect the satellite measured radiation. The SALVI project aimed to remotely sense aerosols at high latitude over snow and ice-covered surfaces. A comprehensive test framework was developed by combining radiative transfer modelling of the SLSTR and OLCI satellite sensors with realistic input from modelling of a peat fire in Greenland 2017. Simulated SLSTR and OLCI radiances were explored for fire signatures using the new capabilities of the SLSTR and OLCI sensors. Fire signatures were found over non-snow covered ground, but not over snow-covered ground. Thus, disabling detection and retrieval of aerosol over snow-covered ground. During the development of the test framework the following additional findings were made: 1) Soot particles form during combustion of carbonaceous materials. When freshly emitted, they are typically fractal-like aggregates. We found that soot particles sampled after evaporating of cloud droplets, are significantly more compact than freshly emitted and interstitial soot, confirming that cloud processing, not just exposure to high humidity, compacts soot. 2) Tar balls are atmospheric particles that are abundant in slightly aged biomass burning smoke. We found that aggregates have a single scattering albedo up to 41% and 23% higher than that of individual tar balls at 550 nm and 350 nm, respectively. The top of the atmosphere simple forcing efficiency over dark surfaces shows large variabilities with an increase up to 53% for tar ball aggregates compared to individual tar balls. 3) To estimate the effect of mineral dust in the Arctic realistic radiative forcing (RF) calculations were made for the year 2012. It was found that mineral dust deposited on snow accounts for nearly all (99%) of the bottom of atmosphere (BOA) RF. The high-latitude dust sources contribute about 52% of the annual Arctic BOA IRF and about 39% to the top of the atmosphere RF. More information is available at https://salvi.wp.nilu.no/.

SALVI has provided new insights to the importance of realistic treatment of aerosol composition and shape on radiative forcing and detection of aerosols. This includes the understanding that both the top of the atmosphere radiative forcing and the aerosol retrievals are largely affected by high albedo. Thus, the other investigated approximations are of less importance for high albedos. For lower values of surface albedo, assumptions in mixing and shape largely affect the retrieval of aerosol properties. As part of SALVI, a comprehensive model framework for simulation of optical satellite instruments with realistic input, has been developed. The developed tools may prove useful in future studies on related topics, e.g. improving smoke detection by separating the signal of spherical particles (e.g. mist) from the signal of soot resulting from combustion. SALVI has improved the international cooperation between DNV GL, NILU and Michigan Technological University (MTU).

The SALVI project responds to the NRC Earth Observing system objective to increase monitoring capability, and pollution tracking from local and long-range transported aerosols in the northern and polar areas. SALVI achieves this by developing a novel satellite retrieval of the aerosol optical depth (AOD), aerosol type (AT) and aerosol layer height (AH), customized for the northern areas. SALVI will utilize the Sea and Land Surface Temperature Radiometer (SLSTR) and Ocean and Land Colour Instrument (OLCI) on board the Sentinel-3 satellite. Specifically SALVI will address: 1. The accuracy of particle optical property models assumed in the retrieval algorithms, as AOD, AT and AH retrievals depend critically on the models used. 2. The retrieval over bright (snow/ice) surfaces of AOD, AT and AH, due to difficulties in separating the contributions to the satellite measured radiance from surface reflectance and the back-scattering by aerosols. SALVI novel methodologies include: 1. Novel observationally based aerosol optics parameterization, using data from various field and laboratory campaigns, targeted to parameterize aerosol chemico-physical, optical and spectral properties. 2. Dual sensor retrieval algorithm, which combines SLSTR dual view with the OLCI higher spectral coverage, to improve separation of atmospheric and surface contribution to the measured signal. SALVI will improve upon the current state of the art, where aerosol retrievals use simplified aerosol optical properties treatment and simplified parameterization of surface bidirectional reflectance. SALVI directly addresses the programmatic objectives of the NRC Romforskning program, and benefits ongoing research activities at UiO and NILU, trough synergetically contributing to several projects with focus on climate change at high latitude, land atmosphere interactions and monitoring programs.