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

Elisabeth Andrews,Bjorn Samset,John A Ogren,Stefan Kinne

doi:10.5194/acp-17-6041-2017

Abstract

Abstract. Here we present new results comparing aerosol optical depth (AOD), aerosol absorption optical depth (AAOD) and column single scattering albedo (SSA) obtained from in situ vertical profile measurements with AERONET ground-based remote sensing from two rural, continental sites in the US. The profiles are closely matched in time (within ±3 h) and space (within 15 km) with the AERONET retrievals. We have used Level 1.5 inversion retrievals when there was a valid Level 2 almucantar retrieval in order to be able to compare AAOD and column SSA below AERONET's recommended loading constraint (AOD > 0.4 at 440 nm). While there is reasonable agreement for the AOD comparisons, the direct comparisons of in situ-derived to AERONET-retrieved AAOD (or SSA) reveal that AERONET retrievals yield higher aerosol absorption than obtained from the in situ profiles for the low aerosol optical depth conditions prevalent at the two study sites. However, it should be noted that the majority of SSA comparisons for AOD440 > 0.2 are, nonetheless, within the reported SSA uncertainty bounds. The observation that, relative to in situ measurements, AERONET inversions exhibit increased absorption potential at low AOD values is generally consistent with other published AERONET–in situ comparisons across a range of locations, atmospheric conditions and AOD values. This systematic difference in the comparisons suggests a bias in one or both of the methods, but we cannot assess whether the AERONET retrievals are biased towards high absorption or the in situ measurements are biased low. Based on the discrepancy between the AERONET and in situ values, we conclude that scaling modeled black carbon concentrations upwards to match AERONET retrievals of AAOD should be approached with caution as it may lead to aerosol absorption overestimates in regions of low AOD. Both AERONET retrievals and in situ measurements suggest there is a systematic relationship between SSA and aerosol amount (AOD or aerosol light scattering) – specifically that SSA decreases at lower aerosol loading. This implies that the fairly common assumption that AERONET SSA values retrieved at high-AOD conditions can be used to obtain AAOD at low-AOD conditions may not be valid.

Highlights

We first present comparisons of aerosol optical depth (AOD), aerosol absorption optical depth (AAOD) and Single scattering albedo (SSA) from the in situ measurements at BND and Southern Great Plains (SGP) with AERONET retrievals. This includes (1) direct comparisons of each in situ profile with contemporaneous AERONET retrievals, after which the BND and SGP comparisons are put in the wider context of a literature review of similar direct comparisons of in situ and AERONET AAOD and SSA; (2) seasonal comparisons of AOD, AAOD and SSA from Phase II AeroCom model results, AERONET retrievals and in situ measurements for BND and SGP; and, (3) discussion of these results in the context of biases in determination of AAOD
A survey of the literature suggests that even at higher loading (AOD440 > 0.4) AERONET SSA retrievals tend to be lower than SSA values obtained from vertical profiling flights, discrepancies are within the reported uncertainty bounds down to ∼ AOD440 > 0.3
Since the observed discrepancy in SSA cannot definitively be attributed to either technique, the idea of scaling modeled black carbon concentrations upwards to match AERONET retrievals of AAOD should be approached with caution

Summary

Introduction

The amount and location of absorbing aerosol in the atmosphere are critical for understanding climate change (e.g., Hansen et al, 1997; Ramanathan and Carmichael, 2008; Bond et al, 2013; Samset et al, 2013). Ramanathan and Carmichael (2008) note the effects of absorbing aerosol (which they termed black carbon (BC)) on atmospheric heating rates, precipitation and weather patterns. (Note: the terminology used to refer to absorbing aerosol is imprecise (Petzold et al, 2013; Andreae and Gelencsér, 2006) and encompasses the terms describing chemistry, e.g., BC, and terms describing optical effects, e.g., absorption. The amount and location of absorbing aerosol in the atmosphere are critical for understanding climate change (e.g., Hansen et al, 1997; Ramanathan and Carmichael, 2008; Bond et al, 2013; Samset et al, 2013). Ramanathan and Carmichael (2008) note the effects of absorbing aerosol (which they termed black carbon (BC)) on atmospheric heating rates, precipitation and weather patterns. The measurements reported all refer to light absorption.) The vertical distribution of BC can influence its effect on climate (e.g., Haywood and Ramaswamy, 1998; Samset et al, 2013; Ramanathan and Carmichael, 2008). Uncertainty in the value of SSA due to uncertainties in the amount of absorbing aerosol can even prevent determination of the sign of aerosol forcing on local to regional scales. Bond et al (2013) assessed BC as the second-most-important global-average warming species (top-of-atmosphere forcing +1.1 W m−2; 90 % bounds: +0.17 to +2.1 W m−2) after CO2 (in Bond et al, 2013, the direct effect of BC is 0.71; 90 % bounds: +0.09 to 1.26 W m−2)

Methods

Results

Discussion

Conclusion