During the evening of May 20, 2017, a fog layer developed off of the Pacific Northwest coast and streamed into the Strait of Juan de Fuca. The NWS Seattle office sent out this great tweet showing the evolution of the fog deck in GOES-16 visible imagery.
After sunset, of course, visible imagery can no longer be used. Additionally, IR imagery alone is not ideal for detecting and tracking liquid water clouds (low clouds and fog). Instead, the 3.9 um – 10.3 um channel difference highlights liquid water clouds well. Figure 1 below depicts the continued evolution of fog after sunset on the 20th into the early morning hours of the 21st just prior to sunrise. The darkest gray (low negative) color represents low clouds and fog (liquid clouds), the lightest gray to white (positive) color represents high/ice clouds, and the medium grays (near 0) are clear sky.
Figure 1: 21 May 2017 GOES-16 3.9 um – 10.3 um Fog Difference. Full resolution: https://satelliteliaisonblog.files.wordpress.com/2017/05/20170521_fog_nw2_anno1.gif
For low clouds and fog (liquid water clouds), cloud top emissivity at 3.9 um is lower than for the same cloud at 11 um. Therefore at night, low clouds and fog will appear cooler at 3.9 um than at 11 um. So by taking the difference, low clouds will stand out as relatively low negative values.
During the day, the sign is flipped due to reflectance and scattering of solar radiation at 3.9 um. So 3.9 um will become warmer than 11 um. The difference will be positive for low clouds and fog during the day.
The 3.9 um – 10.3 um difference will be positive for high clouds day/night. It is ideal to use visible imagery during the day to monitor low cloud and fog evolution
– Bill Line, NWS
“The GOES-16 data posted on this page are preliminary, non-operational data and are undergoing testing. Users bear all responsibility for inspecting the data prior to use and for the manner in which the data are utilized.”
Satellite Liaison Blog 
