![]() Similarly, Zhang and Yi ( 2013) suggested a monthly-dependent dynamic threshold algorithm combining real time brightness temperature (BT) from the Moderate Resolution Imaging Spectroradiometer (MODIS) IR channel (11 μm) with climatological monthly SSTs. With this approach, Park and Kim ( 2012) separated sea fog from other clouds by using the difference between infrared cloud top temperature and sea surface temperature (SST), obtained from the Multi-functional Transport Satellite-1 Replacement (MTSAT-1R) and the Advanced Microwave Sounding Radiometer for EOS (AMSR-E), respectively. The temperature difference between the infrared cloud surface and the ground is relatively small due to the fact that fog occurs right above the surface of the sea and land. To separate fog from stratus effectively, Ellrod and Gultepe ( 2007) proposed an additional threshold combining BTD observed from satellite instrument, with shelter temperatures from surface observing sites. This method, however, cannot differentiate between fog and stratus because they have similar particle size and altitude. Because this method is simple and highly effective for fog detection, it has been widely applied to polar and geostationary satellites (d’Entremont 1986 d’Entremont and Thomason 1987 Saunders and Kriebel 1988 Bendix and Bachmann 1991 Ellrod 1995 Lee et al. This has led to the primary use of the brightness temperature difference (BTD) between the shortwave infrared (SWIR) and infrared (IR) channels to identify fog (Hunt 1973 Eyre et al. However, during nighttime, only infrared channels are available. Due to the highly reflective and homogenous characteristics of the surface of fog, a visible (VIS) channel with high resolution is very effective in distinguishing fog from others. Geostationary satellites in particular have great potential for monitoring the development of weather phenomena as they continually observe the same area with spatial resolution of a few kilometers, with coverage of one quarter of the Earth’s surface area. ![]() ![]() The use of satellite measurements helps to overcome the temporal and spatial limitations of ground measurements (Ahn et al. Although numerous ground observations have been conducted to reduce the losses due to sea fog, it is difficult to understand the overall distribution of sea fog because observation sites are limited to coastlines and islands (Cermak and Bendix 2007, 2008). In Korea, sea fog is a crucial issue because the Korean peninsula is surrounded by sea on three sides. 2007, 2009), highlighting the significance of fog monitoring.įog develops over both land and sea. Furthermore, the loss of life and damage to property caused by fog are comparable to those caused by tornados and hurricanes (Whiffen 2001 Gultepe et al. The formation of fog contributes to numerous traffic accidents and delays caused by low visibility (Ahn et al. The algorithm has been evaluated using the Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) vertical profile products, which showed promising results within a local domain with probability of detection (POD) of 0.753 and critical success index (CSI) of 0.477, respectively.įog consists of suspended droplets or ice crystals that reduce visibility to less than 1 km parallel to the surface near the ground (Gultepe et al. This allows for the application of optimal threshold values for fog detection by considering the characteristics of a specific domain. It identifies distinguishing features of the data by organizing and optimizing the data. The unsupervised learning method employed in this paper is the expectation–maximization (EM) algorithm that is widely used in incomplete data problems. In order to improve this, the unsupervised learning approach, which allows a more effective interpretation from the insufficient information, has been utilized. Although this method using predetermined thresholds is very simple and effective in detecting low cloud, it has difficulty in distinguishing fog from stratus because they share similar characteristics of particle size and altitude. The threshold values were previously determined from climatological analysis or model simulation. Previous algorithms generally employed threshold values including the brightness temperature difference between the near infrared and infrared. The algorithm is based on data sets that combine brightness temperatures from the 3.7 μm and 10.8 μm channels of the meteorological imager (MI) onboard the Communication, Ocean and Meteorological Satellite (COMS), with sea surface temperature from the Operational Sea Surface Temperature and Sea Ice Analysis (OSTIA). ![]() This paper presents a nighttime sea fog detection algorithm incorporating unsupervised learning technique.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |