Verification of Cloud Analyses used to Support Overhead Imagery Collection

• Main Author: Cleary, Robert J
• Other Authors: NAVAL POSTGRADUATE SCHOOL MONTEREY CA
• Format: Text
• Language: English
• Published: 2012
• Subjects: Meteorology; *CLOUDS; *WEATHER FORECASTING; BARENTS SEA; SATELLITE IMAGERY; SOUTHWEST ASIA; THESES; WWMCA(WORLD WIDE MERGED CLOUD ANALYSIS); CDFS II(CLOUD DEPICTION FORECAST SYSTEMS II); CLOUD FORECASTS; II; CLOUD ANALYSIS; CLOUDSAT; 2B-GEOPORF; Barents Sea
• Online Access: http://www.dtic.mil/docs/citations/ADA561828; http://oai.dtic.mil/oai/oai?&verb=getRecord&metadataPrefix=html&identifier=ADA561828

We verified the USAF World Wide Merged Cloud Analysis (WWMCA) against observations from the Cloudsat atmospheric sounder. We analyzed WWMCA data for 2010 for two regions that are of high interest to the national intelligence community and that differ in their meteorological characteristics. The two regions covered were: (a) much of southwest Asia; and (b) much of western Russia and the Barents Sea. We analyzed WWMCA performance according to four criteria: (1) type of cloud event (Definite-Cloud, Probable-Cloud, and No-Cloud); (2) geographic region; (3) time of day; and (4) time of year. We measured WWMCA performance using contingency table metrics and found marked differences in performance for the four criteria. In particular, WWMCA tended to perform better in analyzing: (a) No-Cloud and Definite-Cloud events than Probable-Cloud events; (b) the lower latitude region than the higher latitude region; and (c) persistent cloud events than variable cloud events. Our Heidke skill scores indicated that WWMCA performance was, in general, moderately better than that of a random set of analyses. Overall, WWMCA performance was problematic, given that WWMCA is a near real-time analysis product and is designed to initiate short lead time cloud forecasts used by the intelligence community.