HR: 16:50h
AN: SA44A-04    [Abstracts]
TI: Automated Detection and Tracking of Equatorial Plasma Depletions Using Ground-Based Optical Imagers
AU: * Pedersen, T
EM: todd.pedersen@hanscom.af.mil
AF: Space Vehicles Directorate, Air Force Research Laboratory, 29 Randolph Road, Hanscom AFB, MA 01731 United States
AU: Carrano, C
EM: charlie@radex.com
AF: Radex Inc., 3 Preston Court, Bedford, MA 01730 United States
AU: Griffin, J
EM: griffin@radex.com
AF: Radex Inc., 3 Preston Court, Bedford, MA 01730 United States
AB: Optical imaging is one of the few means available for determining space weather parameters simultaneously over large areas, but tropospheric cloud cover presents a significant barrier to operational use of data from ground-based optical instruments. Distributed sensors experiencing different tropospheric conditions but with overlapping fields of view in the upper atmosphere are one possible solution to the cloud cover difficulty, while intelligent processing of imager data to discriminate between clouds and upper atmospheric features is another potential means of providing reliable data output from only a single instrument. We evaluate and discuss a variety of processing algorithms developed or adapted for the purpose of detecting and tracking equatorial plasma depletions in all-sky imager data under realistic conditions including significant cloud cover. Our most successful technique thus far relies on discrimination between depletions and other image features based on their signatures in velocity and correlation space rather than physical coordinates. In addition to allowing identification and tracking of the depletions, accurate knowledge of the velocity allows multiple frames of image data to be processed coherently in the reference frame moving with the depletions. This processing can virtually eliminate cloud effects up to 50 percent cloud cover. With externally provided velocity information (such as from a spaced-antenna scintillation system, for example) or an improved velocity algorithm, useful data can be obtained at even greater cloud cover fractions. A similar motion-based technique can also be applied to the background star field, allowing stars to be easily distinguished from pixel noise and hot pixels for rapid automatic identification of image regions affected by clouds without the need to identify, locate, or track any specific stars.
DE: 2415 Equatorial ionosphere
DE: 2463 Plasma convection
DE: 2494 Instruments and techniques
DE: 0310 Airglow and aurora
SC: SPA-Aeronomy [SA]
MN: 2004 AGU Fall Meeting