Publications

Amani, M; Mahdavi, S; Bullock, T; Beale, S (2020). Automatic nighttime sea fog detection using GOES-16 imagery. ATMOSPHERIC RESEARCH, 238, 104712.

Abstract
Accurately detecting sea fog is important for oil and gas operations in the Grand Banks, Newfoundland and Labrador (NL), Canada. Although the Grand Banks is one of the foggiest places in the world, there is no remote sensing technique specifically developed for fog detection in this region. Therefore, an automatic approach was proposed in this study to detect Nighttime Sea Fog (NSF) and distinguish it from clear sky and ice cloud. To this end, Geostationary Operational Environmental Satellite system-16 (GOES-16) imagery along with several ancillary datasets were employed. Selecting the Optimum Threshold Value (OTV) for identifying NSF in satellite images acquired at different times was also extensively discussed. The NSF maps obtained by the proposed method for 25 advection fog events in the study area were compared to surface-based weather observations (i.e., visibility data) and the National Oceanic and Atmospheric Administration's (NOAA) global fog/low cloud products. The Probability of Detection (POD), False Alarm Rate (FAR), Hanssen-Kuiper Skill Score (KSS), and Equitable Threat Score (ETS) were 0.80, 0.08, 0.72, and 0.57, respectively, demonstrating the potential of the proposed NSF detection algorithm. Additionally, the results showed that the proposed method was better at discriminating Grand Banks fog than NOAA's algorithm in terms of both visual and statistical accuracies.

DOI:
10.1016/j.atmosres.2019.104712

ISSN:
0169-8095