Assimilation of nontraditional datasets to improve atmospheric compensation

Detection and characterization of space objects require the capability to derive physical properties such as brightness temperature and reflectance. These quantities, together with trajectory and position, are often used to correlate an object from a catalogue of known characteristics. However, retrieval of these physical quantities can be hampered by the radiative obscuration of the atmosphere. Atmospheric compensation must therefore be applied to remove the radiative signature of the atmosphere from electro-optical (EO) collections and enable object characterization. The JHU/APL Atmospheric Compensation System (ACS) was designed to perform atmospheric compensation for long, slant-range paths at wavelengths from the visible to infrared. Atmospheric compensation is critically important for air-and ground-based sensors collecting at low elevations near the Earth's limb. It can be demonstrated that undetected thin, sub-visual cirrus clouds in the line of sight (LOS) can significantly alter retrieved target properties (temperature, irradiance). The ACS algorithm employs non-traditional cirrus datasets and slant-range atmospheric profiles to estimate and remove atmospheric radiative effects from EO/IR collections. Results are presented for a NASA-sponsored collection in the near-IR (NIR) during hypersonic reentry of the Space Shuttle during STS-132.