The item that caused me to write to this blog is the one on target selection.
Select Target/feature associated with strongest gradient
I have discussed Warm Front Shields in my previous posts. One of the features I have noticed in this CM (Conceptual model) is that it does not have strong gradients, in its most identifiable form in an IR image (i.e. without TA model assistance). It is one big fuzzy blob. It was such a system over the Netherlands (3 December, 1996) where I thought the EUMETSAT motion vectors should have been of very good quality because the animation showed a very smooth movement of clouds, all in one direction.
The surprise for me was that there were very few high quality vectors,