Discrete representation and resampling in limb-sounding measurements

The processes of discretization, interpolation, and resampling are frequent ly used in data analysis. Here the formalism of functional spaces is used a s a framework for the description and characterization of both the measurem ent operation and these subsequent processes. The tools provided by this fo rmalism are applied to the problem of resampling of atmospheric volume mixi ng ratio vertical profiles obtained with limb-sounding measurements. In par ticular, a resampling method that uses the conservation of the vertical col umn as a constraint is presented and compared with other methods. The effec ts of the resampling process in terms of error propagation and loss of vert ical resolution are also evaluated. (C) 2001 Optical Society of America.