Footprint size and cloud-clearing error relationships derived from NAST-Ⅰtop of the atmosphere radiance measurements

摘要

We discuss the relationship between instrument footprint size, field-of-regard sample density, and cloud clearing technique on measured top of the atmosphere radiance error under partly cloudy conditions. The cloud clearing technique (N~*) uses the linear relationship between observed radiance and the amount of cloud in a field-of-view. We extrapolate radiance observed for two adjacent fields-of-view possessing differing cloud amounts to the cloud free value (i.e., zero cloud). Options include techniques to compensate for "black" or "gray" clouds, where a single channel N~* may not provide adequate spectral correction. Spectrally dependent error statistics are developed from pardy cloudy samples of varying footprint size and sample patterns. Data were collected by the NPOESS (National Polar-orbiting Operational Environmental Satellite System) Aircraft Sounding Testbed-Interferometer (NAST-I) flying on the NASA Proteus or ER-2 high altitude research aircraft, and include tropical, temporal and arctic flight sections. Analysis shows that larger sounder footprints contain more cloud contamination and higher cloud clearing errors; these errors can be significantly reduced by techniques that utilize high-spectral and -spatial resolution coincidently collected radiance measurements from sensors like MODIS. Data also indicates that full area sampling results in smaller cloud clearing errors than small footprint sampling on a wider spaced grid.