Aerosol is a suspension of solid or liquid particles in a gas. These particles are distinguished from cloud droplets by their location on the Köhler curve — aerosols exist below critical supersaturation, above which particles will grow through condensation until they reach 100% relative humidity. They have a very wide range of sources, both natural and anthropogenic, particularly in the lower troposphere . These include direct emissions such as wind-blown dust, sea spray, and combustion products as well as condensed gas particles such as dimethyl sulphide and other organic compounds emitted by vegetation and industrial processes.
The range of different sources results in an equally vast range of particle sizes, chemical compositions, and microphysical properties . Aerosol particles generally have relatively short atmospheric lifetimes — a few hours to days — before they are lost through contact with the Earth’s surface (dry deposition) or are rained out (wet deposition). These factors combine to produce a great diversity of aerosol loading and properties that exhibit high temporal, vertical, and spatial variability in amount, size, and composition. This plethora of different behaviours makes the remote identification and inventory of aerosol loading in the atmosphere a substantial observational challenge .
Improved knowledge of atmospheric aerosol is of vital importance to our understanding of climate change as both the direct and indirect forcings of aerosols are significant and highly uncertain. Indeed, the IPCC  lists the indirect and direct radiative forcing of aerosol as the two most uncertain climate forcings. A better understanding of the global distribution, properties, and evolution over multi-decadal timescales of atmospheric aerosol is a key requirement to improving our understanding of the climate system, its natural variability (including major volcanic eruptions), and anthropogenic effects.
The estimation of aerosol properties is fundamentally very difficult. Any satellite retrieval of aerosol properties evaluates an ill-posed set of equations relating a measurement to the atmosphere observed and that system is very non-linear. The sensitivity to aerosol is dependant on the quantity of aerosol observed and its characteristics as well as the properties of the underlying surface, the presence of clouds and trace gases, and instrument-specific characteristics such as the spectral range, polarization, and viewing angles. To solve these equations, assumptions and auxiliary information are required to reduce the number of unknowns to match the number of measurements. Different instruments are used for aerosol retrieval, only a few of which were designed for this purpose (e.g., POLDER, MODIS, MISR). For various reasons other instruments have been and are being used for aerosol retrieval despite being less than ideal (e.g. ATSR, MERIS, OMI, etc.). The use of different instruments, algorithms, and assumptions on a highly non-linear system generates a wide variety of results, even when considering the same original data.
Regardless, it is important to assess how the loading of aerosol and its properties have changed over time. EODG considers the era of satellite obervations since 1982 and how to make the optimal use of the information available (which changes as satellites are launched, fail, and replaced with more advanced instruments). We are especially concerned with producing a data record that is suitable for climate studies — simultaneously global in coverage, stable against calibration artefacts and the auxilliary information available, sufficiently long-term to observe trends, and physically consistent with well-characterised uncertainties.
It is equally important to further develop algorithms where they are weakest. Thick aerosol and thin cloud are often indistinguishable such that each is a contaminant and significant source of uncertainty in the traditional estimation of the other. We consider the joint retrieval of aerosol and cloud to investigate that ambiguity and provide data users with a fair assessment of the quality of any given retrieval. We also perform laboratory and in situ measurements of aerosol properties. For significant aerosol events, such as volcanic erruptions, this can better constrain the ill-posed problem and produce more accurate retrievals.
 Seinfeld, J. and Pandis, S. (1998). Atmospheric Chemistry and Physics. Wiley, New York, first edition.
 Bohren, C. F. and Huffman, D. R. (2004). Absorption and Scattering of Light by Small Particles. Wiley-VCH, Morlenbach, Germany.
 Intergovernmental Panel on Climate Change (IPCC), Stocket, T. F., Qun, D., Plattner, G.-K., Tignor, M. M. B., Allen, S. K., Boschung, J., Nauels, A., Xia, Y., Bex. V., Midglet, P.M. (2013). Climate Change 2013: The Physical Science Basis. Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, New York.