| << | New Instruments and Algorithms | >> |
Abstract
The inversion of optical Raman lidar data, to retrieve microphysicalaerosol properties, is challenging, since solutions depend non-linearly on measurement errors. However, the retrieval of integral properties of the aerosol size distribution, such as total volume or surface area density, depend much less on measurement errors than the size distribution itself, and can be predicted with useful accuracy using a Principle Component Analysis (PCA) based inversion technique.
The PCA kernels are sensitive to complex refractive index, which can be
used to account for aerosol type variation in the optical data. This
is especially important in the troposphere, where the refractive indices
of the ambient aerosols are usually unknown. We show the accuracy that
can be expected from this type of analysis.
PCA is a stable method, easy to implement and fast. It is
therefore a useful complement to existing inversions. It can also be
used as an extra constraint for other, more involved inversion
techniques using regularisation and constraints.