Aerosol single‐scattering albedo over the global oceans: Comparing PARASOL retrievals with AERONET, OMI, and AeroCom models estimates

Abstract

The aerosol single-scattering albedo (SSA) over the global ocean is evaluated based on polarimetric measurements by the PARASOL (Polarization and Anisotropy of Reflectances for Atmospheric Sciences coupled with Observations from a Lidar) satellite. For the first time, global ocean SSA and Absorption Aerosol Optical Depth (AAOD) from this instrument are shown and evaluated against other observations (the Aerosol Robotic Network, AERONET, and the Ozone Monitoring Instrument, OMI). The observational data sets compare reasonably well, with the majority of the colocated points within 0.05 of the AERONET measurements. PARASOL shows that SSA is characterized by high spatial and seasonal variability, also over the open ocean far from the inland emission regions. The near global coverage in the visible spectral range provided by the PARASOL retrievals represents a unique opportunity to evaluate aerosol optical properties simulated by global aerosol models, as performed in the Aerosol Comparisons between Observations and Models (AeroCom) framework. The SSA (AAOD) estimated by the AeroCom models is generally higher (smaller) than the SSA (AAOD) retrieved from PARASOL. On the other hand, the mean simulated aerosol optical depth is consistent or slightly underestimated compared with observations. An overestimate of the aerosol scattering, compared to absorption, by the models would suggest that these simulate an overly strong aerosol radiative cooling at top of atmosphere, over most of the ocean surfaces. This implies that aerosols have a potentially stronger direct and semidirect impact within the atmosphere than currently simulated. © 2015. American Geophysical Union. All Rights Reserved

Comparison of AOD, AAOD and column single scattering albedo from AERONET retrievals and in situ profiling measurements

“…For the Level 1.5 data they apply the additional requirement that the solar zenith angle must be ≥ 50 • . Lacagnina et al (2015) find quite good agreement (within ± 0.03) for AAOD and note that larger differences between PARASOL and AERONET retrieval occur at higher AOD conditions, possibly due to less-homogenous aerosol (i.e., plumes).…”

“…Several different approaches of varying complexity have been developed to better represent absorbing aerosol for cleaner conditions. Some of these approaches utilize SSA at high AOD to estimate AAOD at lower AOD conditions (e.g., Bond et al, 2013;Wang et al, 2014), while others utilize Level 1.5 retrievals with the added uncertainty that entails (e.g., Lacagnina et al, 2015;Mallet et al, 2013). Based on the analysis presented here, we cannot say how to best estimate SSA or AAOD from AERONET retrievals for the low-AOD conditions prevalent around much of the globe.…”

“…Some studies have utilized AERONET Level 1.5 retrievals of absorption-related aerosol properties in order to avoid being limited to the high AOD levels required by Level 2 data (e.g., Lacagnina et al, 2015;Mallet et al, 2013). These studies note that Level 1.5 data include more relevant AOD values but that there are accompanying higher uncertainties in the retrievals for absorption-related properties.…”

“…Mallet et al (2013) use Level 1.5 data to evaluate the spectral dependence of aerosol absorption. Lacagnina et al (2015) utilize both Level 2 and Level 1.5 AERONET data in their comparison with PARASOL satellite retrievals of SSA and AAOD. For the Level 1.5 data they apply the additional requirement that the solar zenith angle must be ≥ 50 • .…”

“…Based on the correlograms, AERONET retrievals were merged with the in situ profile data when the retrievals were within ±3 h of the end of the in situ profile. This is the same time range constraint used to compare AERONET and PARASOL SSA values (Lacagnina et al, 2015). Additionally, Fig.…”