Question regarding CALIPSo cloud fraction by types

[Marston]

Hi,

I'm trying to scientifically vet my COSP1.4.1 cloud types: LCC, MCC, HCC, TCC.
This is building on a previous question and here I just have a small question.

It is perhaps counter-intuitive but should not the TCC be the larger or equal to the largest cloud fraction seen in LCC, MCC, and HCC?
Here's a small sample of a sliver of the model domain where the max cloud fraction is HCC, MCC, LCC, and TCC are measured:
chunk 1
LCC = 2.0
MCC = 86.0
HCC = 92.0
TCC = 27.66

chunk 2
LCC = 100.0
MCC = 100.0
HCC = 76.0
TCC = 42.0

Here the max TCC is much smaller than then max HCC. While these maxes may not be necessarily colocated, TCC should me the equal to or greater than the max in all of HCC, MCC, and LCC.
Or am I missing something here?

No this is not always the case and sometime TCC is equal to the max of all the other cloud layers, and sometimes it is higher, but not consistently so. This problem shows up in the daily and monthly means
where TCC is significantly lower that the the maximum of the other cloud types.

This behaviour also persists whether I use the model Reff or COSP default.

Has anyone come across this problem before?

Appreciate your thoughts on this.

Best,
/M