Title | Solar eclipse as a source of satellite image contamination in multi-scene clear-sky composites |
| |
Author | Trishchenko, A P ;
Ungureanu, C |
Source | Canadian Journal of Remote Sensing vol. 42, 2016 p. 730-738, https://doi.org/10.1080/07038992.2016.1249563 |
Year | 2016 |
Alt Series | Earth Sciences Sector, Contribution Series 20160171 |
Publisher | Informa UK Limited |
Document | serial |
Lang. | English |
Media | paper; on-line; digital |
File format | pdf |
Subjects | extraterrestrial geology; Nature and Environment; eclipses, solar; satellites; satellite imagery; reflectance |
Illustrations | diagrams; formulae; satellite images |
Program | Climate
Change Geoscience Risk Analysis |
Released | 2016 10 20 |
Abstract | The solar eclipse is a relatively rare event, but it could cause significant impact on satellite products if not accounted for properly. The solar eclipse effect on daily and 10-day clear-sky composites
is described here. It is shown that the standard scene identification and compositing algorithm developed at the Canada Centre for Remote Sensing (CCRS) for Moderate Resolution Imaging Spectroradiometer (MODIS) imagery fails to identify pixels
affected by a solar eclipse. As a result, the solareclipse-affected imagery is almost entirely included in the composite product. A similar effect is observed in the standard NASA MOD09/MYD09 daily composite reflectance products. We describe here the
details of the eclipse-affected imagery, the reason why clear-sky compositing criteria failed, and propose some modifications to the standard algorithm, such as reflectance normalization and scene identification procedures, to overcome the solar
eclipse contamination problem in multiscene image composites. |
Summary | (Plain Language Summary, not published) The solar eclipse is a relatively rare event, but it may cause significant impact on satellite products if not accounted for properly. The solar eclipse
effect on daily and 10-day clear-sky composites is described here. It is shown that the standard scene-identification and compositing algorithm developed at the Canada Centre for Remote Sensing (CCRS) for Moderate Resolution Imaging Spectroradiometer
(MODIS) imagery fails to identify pixels affected by a solar eclipse. As a result, the solar eclipse affected imagery is almost entirely included in the composite product. A similar effect is observed in the standard NASA MOD09/MYD09 daily composite
reflectance products. We describe here the details of the eclipse-affected imagery, the reason why clear-sky compositing criteria failed, and propose some modifications to the standard algorithm, such as reflectance normalization and scene
identification procedures, to overcome the solar eclipse contamination problem in multi-scene image composites. |
GEOSCAN ID | 299225 |
|
|