Ion noise filtering software for SSULI ground data assimilation

Abstract

The SSULI (Special Sensor Ultraviolet Limb Imager) is a low-resolution hyperspectral far and extreme ultraviolet limb-scanning imager designed to monitor ionospheric and thermospheric airglow. SSULI has a spectral range from 80 to 170 nm, and a nominal resolution of 2.1 nm (at 147 nm). The instrument is scheduled to fly aboard all Defense Meteorological Satellite Program (DMSP) Block 5D3 weather satellites. The first SSULI instrument was launched in fall 2003, aboard the DMSP F16 flight, and has been collecting data since December 2003. The second SSULI flight aboard DMSP F17 began in fall 2006. Early in the missions, both instruments began to observe intermittent but significant periods of noise across the entire instrument passband, beyond the expected ion noise associated with sub-auroral latitudes and the South Atlantic Anomaly. The morphology and intensity of the noise correlates strongly with environmental conditions such as spacecraft potential. In order for the ground processing software to extract individual emission features from the measured spectra, the data must be filtered for quality and the noise must be characterized on short time scales and introduced as additional basis functions for use with the Multiple Linear Regression (MLR) feature extraction algorithm. New algorithms, in the form of an Ion Noise Filter, have been developed for use with the MLR. The techniques used in the Ion Noise Filter are discussed and examples of the successful extraction of spectra are demonstrated.