Satellite icing algorithms for CAWS support

Today was a busy day for the CAWS folks during the experiment. Not only was there a plethora of turbulence due to the strong low lumbering through the east CONUS, the environment was also prime for icing. It was here that the GOES-R icing algorithms were able to provide situational awareness.

Before looking at GOES-R products, forecasters examined a more classic satellite tool to identify areas of icing concern (below).

20160224 1915 UTC IR imagery with icing color scheme highlight areas of 0 to -20C; overlaid with surface analysis and icing CAWS (blue dashed). Turbulence CAWS also added (dashed yellow).

Areas highlighted in the pinks and purples indicated possible areas of icing. With a synoptic scale system, the potential for areas of icing is wide. Next forecasters examined the satellite derived supercooled liquid water.

20160224 1615 satellite-derived supercooled liquid water droplets

Recall, the CAWS area of interest was Memphis and the surrounding region. Satellite-derived SLD did show supercooled water droplets in the area. Note, though, the darker reds in the immediate vicinity of Memphis. While this would indicate higher SLD values, in actuality this appeared to be larger values caused by cirrus cloud contamination. However, over all the product did quite well in this area and was corroborated with a number of MOG icing PIREPs.

Lastly, the FIT was examined:

20160224 1615 Flight Icing Threat index imagery with PIREPs overlaid

Immediately noticeable is the widespread reds and pinks, or higher FIT values. It is important to note this is the time of year when satellites enter eclipse periods, i.e. suffer from backscatter effects. This image was snapped at 1615 UTC, a time during which these effects are noticeable. In this case, it caused the effective radius used in the generation of the FIT to be increased significantly, causing the product to be overdone for several scans.

However, for the area surrounding Memphis, the product was still found useful. The 1615 and 1645 UTC frame aside, there were consistently higher FIT index values. Again, several PIREPs corroborated that MOG icing was occurring in this area.