Evaluating impacts of imaging spectrometer calibration on mineral identification and mapping using airborne data collections in Alaska, USA, and Khandahar, Afghanistan

Calibration of spectrometer data to reflectance is important to obtain accurate and robust results in identifying surface materials. Among the aspects that affect calibration are: sensor characterization, including channel wavelength position and bandpass, radiometric accuracy, and atmospheric correction. For the detection of surface minerals in soils and bedrock, inaccuracies in any of these aspects generally decrease the degree of mineral discriminations that can be made, for example, separating carbonates, serpentines, chlorites, and amphiboles. Because of their design, pushbroom spectrometers can have variable wavelength position and bandpass in the across-track direction. Well-calibrated airborne imaging spectrometer data collected using HyMap were used to examine the effects of inaccuracies in Hyperion wavelength position on mineral identifications and to explore empirical methods for calibrating data to reflectance. On steep terrains, at high latitudes, it is difficult to atmospherically correct imaging spectrometer data to surface reflectance. Using HyMap data collected in Wrangell-St. Elias National Park, Alaska, a simple atmospheric correction, assuming a single elevation and flat terrain, was compared to a complex atmospheric correction accounting for pixel-by-pixel variations in elevation and sensor viewing and illumination geometry.