High-Performance Aerosol Retrieval Algorithm – GRASP


Traditional aerosol retrieving algorithms
deduce only about 10 parameters. The Generalized Retrieval of Aerosol and Surface
Properties, short GRASP, is an algorithm that infers nearly 50 aerosol and surface parameters,
including the particle size distribution, the spectral index of refraction, degree of
sphericity and absorption – the most important aerosol property for climate models. GRASP works under different conditions, also
including bright surfaces like deserts, where traditional aerosol algorithms struggle with
the reflectance that overwhelms the signal of aerosols. GRASP is a highly versatile algorithm that
allows input from a wide variety of satellite and surface measurements. In order to achieve high accuracy GRASP is
based on  a highly optimized numerical fitting procedure, and a detailed model of atmosphere,
aerosol and surfaces. All computations are done on-line without
the use of traditional look-up tables. The advantages of this approach have been
confirmed by an independent study. GRASP uses an innovative multi-pixel approach
that takes advantage of two extra constraints  coming from observations:
Aerosols have limited horizontal variability, and
surface properties change slowly with time. GRASP is based on very positive and well-recognized
heritage. The roots of GRASP are based upon aerosol
retrieval developments implemented for AERONET – a worldwide network of over 200 radiometers
run by NASA that are acknowledged as highly-valued aerosol products in a very large number of
scientific publications. We think: “When spending millions for a satellite
instrument, you should insist on processing your data with the most accurate aerosol retrieval
algorithm.” What do you think?