Measurements of light scattered from the Earth's limb have long been utilized for aeronomy of the upper layers of the atmosphere. More recently the advantages of such measurements from space based platforms have been used to retrieve profiles of particulates and trace gas concentrations in the stratosphere and upper troposphere. Two instruments are currently making dedicated limb scatter measurements in the ultraviolet, visible and near-IR spectral regions: the Optical Spectragraph and InfraRed Imager Suite (OSIRIS) on board the Odin space-craft and the SCanning Imaging Absorption spectroMeter for Atmospheric CartograpHY (SCIAMACHY) on board Envisat. These will be joined by the Ozone Mapping Profiling Suite-Limb Profiler due to be launched on the NPOESS Preparatory Project (NPP). With these measurements, it is possible to infer vertical profiles of aerosol extinction and extend the 21-year record of aerosol extinction profiles established by Stratospheric Aerosol and Gas Experiment II (SAGE II) data into the next decade.

We demonstrate how these extinction profiles can be derived in a manner similar to that used by Naudet and Thomas, but here with an optically thin radiative transfer model (RTM). The limitations of this simple, but effective model, are explored. At a wavelength of 1 micron, the error due to use of this RTM are < 5% throughout the upper troposphere and stratosphere. Larger errors can be produced by uncertainty in instrument pointing and stray light contamination. For stratospheric aerosols, errors resulting from incorrect assumptions about the size distribution can comparable to instrument related error terms. However, multi-wavelength measurements can be used to refine the assumed size distribution and thus significantly reduce this error term. Observations of similar air masses, but different scattering angles can also be used to reduce this important error term.

This model is applied to data from contemporary limb scatter instruments and compared to extinction values from the well trusted SAGE II data set. Examples of individual profiles and multi-year time series are shown.