Progress in Aircraft Icing Diagnoses from Satellite Data and Transition to Weather Forecasting Operations

The presence of supercooled liquid water (SLW) in clouds is a significant threat to aircraft due to ice buildup on the airframe. This can alter the ability of the aircraft to maintain proper lift and control leading to serious aviation accidents. To determine the SLW icing threat to aircraft, an icing diagnosis algorithm was developed to utilize cloud properties retrieved from satellite data including cloud phase, liquid water path, and droplet size. The algorithm is designed to identify and categorize icing in all cloud conditions. Multi-layered clouds with significant vertical separation and clouds over snow-covered surfaces remain the most problematic areas for quantifying the icing threat. The satellite icing product includes information on the probability and intensity during the daytime. Less information is available at night due to poor sensitivity to optically thick cloud microphysics when using infrared-only data, thus the only determination currently possible is a simple flag indicating icing, no icing, or unknown conditions. The vertical extent of the icing layer embedded in the cloud is also available at all times. Most validation work has focused on the continental United States (CONUS) due to the availability of numerous pilot reports (PIREPS) of icing. With support from the GOES-R proving ground and NASA, regional and global satellite icing products are now flowing routinely to the NOAA National Weather Service (NWS) Aviation Weather Center (AWC) and the Alaska Aviation Weather Unit (AAWU) for demonstration and evaluation. This paper reports recent work validating the GOES and MODIS satellite icing products over Alaska and initial evaluations from aviation weather forecasters. Recent progress to improve the product, particularly over snow and ice covered surfaces and in multilayer conditions, are also reported. These are made possible with the use of GOES-16 which for the first time provides high spectral, spatial, and temporal resolution satellite data coincident with the widely available icing PIREPS over the CONUS which are critical for algorithm formulation and evaluation.