Aerosol Fine Mode Fraction Research Articles

Evaluation of Moderate‐Resolution Imaging Spectroradiometer (MODIS) Collection 004 (C004) aerosol retrievals at Kanpur, Indo‐Gangetic Basin

We have compared the spectral aerosol optical depth (AOD, τλ) and aerosol fine mode fraction (AFMF) of Collection 004 (C004) derived from Moderate‐Resolution Imaging Spectroradiometer (MODIS) on board National Aeronautics and Space Administration's (NASA) Terra and Aqua platforms with that obtained from Aerosol Robotic Network (AERONET) at Kanpur (26.45°N, 80.35°E), India for the period 2001–2005. The spatially‐averaged (0.5° × 0.5° centered at AERONET sunphotometer) MODIS Level‐2 aerosol parameters (10 km at nadir) were compared with the temporally averaged AERONET‐measured AOD (within ± 30 minutes of MODIS overpass). We found that MODIS systematically overestimated AOD during the pre‐monsoon season (March to June, known to be influenced by dust aerosols). The errors in AOD at 0.66 μm were correlated with the apparent reflectance at 2.1 μm (ρ*2.1) which MODIS C004 uses to estimate the surface reflectance in the visible channels (ρ0.47 = ρ*2.1/4, ρ0.66 = ρ*2.1/2). The large errors in AOD (Δτ0.66 > 0.3) are found to be associated with the higher values of ρ*2.1 (0.18 to 0.25), where the uncertainty in the ratios of reflectance is large (Δρ0.66 ± 0.04, Δρ0.47 ± 0.02). This could have resulted in lower surface reflectance, higher aerosol path radiance and thus lead to overestimation in AOD. While MODIS‐derived AFMF has binary distribution (1 or 0) with too low (AFMF < 0.2) during dust‐loading period, and ∼1 for the rest of the retrievals, AERONET showed range of values (0.4 to 0.9). The errors in τ0.66 were also high in the scattering angle range 110°–140°, where the optical effects of nonspherical dust particles are different from that of spherical particles.

Comparisons of remote sensing retrievals and in situ measurements of aerosol fine mode fraction during ACE‐Asia

We present sunphotometer‐retrieved and in situ fine mode fractions (FMF) measured onboard the same aircraft during the ACE‐Asia experiment. Comparisons indicate that the latter can be used to identify whether the aerosol under observation is dominated by a mixture of modes or a single mode. Differences between retrieved and in situ FMF range from 5–20%. When profiles contained multiple layers of aerosols, the retrieved and measured FMF were segregated by layers. The comparison of layered and total FMF from the same profile indicates that columnar values are intermediate to those derived from layers. As a result, a remotely sensed FMF cannot be used to distinguish whether the aerosol under observation is composed of layers each with distinctive modal features or all layers with the same modal features. Thus, the use of FMF in multiple layer environments does not provide unique information on the aerosol under observation.

Comparison of Moderate Resolution Imaging Spectroradiometer (MODIS) and Aerosol Robotic Network (AERONET) remote‐sensing retrievals of aerosol fine mode fraction over ocean

Aerosol particle size is one of the fundamental quantities needed to determine the role of aerosols in forcing climate, modifying the hydrological cycle, and affecting human health and to separate natural from man‐made aerosol components. Aerosol size information can be retrieved from remote‐sensing instruments including satellite sensors such as Moderate Resolution Imaging Spectroradiometer (MODIS) and ground‐based radiometers such as Aerosol Robotic Network (AERONET). Both satellite and ground‐based instruments measure the total column ambient aerosol characteristics. Aerosol size can be characterized by a variety of parameters. Here we compare remote‐sensing retrievals of aerosol fine mode fraction over ocean. AERONET retrieves fine mode fraction using two methods: the Dubovik inversion of sky radiances and the O'Neill inversion of spectral Sun measurements. Relative to the Dubovik inversion of AERONET sky measurements, MODIS slightly overestimates fine fraction for dust‐dominated aerosols and underestimates in smoke‐ and pollution‐dominated aerosol conditions. Both MODIS and the Dubovik inversion overestimate fine fraction for dust aerosols by 0.1–0.2 relative to the O'Neill method of inverting AERONET aerosol optical depth spectra. Differences between the two AERONET methods are principally the result of the different definitions of fine and coarse mode employed in their computational methodologies. These two methods should come into better agreement as a dynamic radius cutoff for fine and coarse mode is implemented for the Dubovik inversion. MODIS overestimation in dust‐dominated aerosol conditions should decrease significantly with the inclusion of a nonspherical model.