Visibility determination from WV-imagery

[rese_support]

We have a couple questions:

center latitude is about 25 degrees north, SW Florida coastal region. 

in 30 Jan 2017 we have 4 WV2 scenes that cover our study area. the scenes were collected within 25 seconds of each other. 

the final mean visibility's of the right 2 scenes are ~26 and 22, and the 2 on the left are ~35 and 32. the right pair pretty well matches as does the left pair; however the right and left do not match. 

can we set the visibility to one scene value? the right lower covers land entirely without clouds so we think we should set the remaining 3 scenes to its visibility (22). all other inputs are the same for all the scenes. 

our second question has to do with the 16 Feb 2018 WV3 scene. the final mean visibility four all scenes (one strip) is ~81.3. this seems high even for the winter.  what do you think? 

As you have seen, visibility determination from WV-imagery is not as stable as one would like it to be... 
In order to force the correction to a given visibility, you can enter the visibility from the best scene as negative number in the *inn file of the other scenes. If it is given this way, no overrides will happen during the processing.