Accurate interpolation of Spectral Response Functions of Ocean Colour Monitor

摘要

Precise mathematical representation of spectral response functions (SRF) of Earth Observation sensors plays a key role in calibration exercises and also for out-of-band signal compensation in the operational phase of these sensors. With narrow spectral bandwidths and high radiometric resolutions built with ocean colour sensors (Moderate resolution imaging spectrometer-MODIS, Medium resolution imaging spectrometer-MERIS, Sea viewing wide field of view sensor-SeaWIFS, Ocean colour monitor-OCM) in order to extract small signal from the ocean upwelling radiance in the high background atmospheric scattering, more accurate representation of the SRF assumes significant importance. The SRF measurements are typically carried out at about 1nm intervals in the in-band spectral region and at about 10nm intervals in out of band regions. In this paper we investigated a detailed interpolation analysis of the SRF of Ocean Colour Monitor flown on-board Indian Oceansat-2 mission. Several methods have been reported in the literature. Of these, it is found that the OCM SRF data is best fitted with B-spline interpolation method, and hence was adopted in the present study. Criteria for resampling rate was worked out and spectral interpolation covering all the spectral regions was carried out. The interpolated data was analyzed in detail for band specific parameters (peak wavelength, centre wavelength and full width half maximum). We found that cubic B-spline based interpolation function is the optimal method for representing the SRF covering all the spectral regions in terms of arriving at a smooth function to compensate noise in the laboratory recording process, and also providing accurate interpolated values in the gaps between samples.