The idea is to find a figure of merit to compare an averaged profile obtained via N repetitions of RAMP (N single daily profiles which are then averaged to a single one) and one obtained via N+1 repetition.
Lets call average profile obtained with N repetitions a(N) and the figure of merit is a function which takes a(N) as argument.
Given this figure of merit, one could then say "running N+1 times is pointless because FOM(a(N+1)) does not differ much from FOM(a(N))"
This could lead to speed up in whole year profiles calculation
The idea is to find a figure of merit to compare an averaged profile obtained via N repetitions of RAMP (N single daily profiles which are then averaged to a single one) and one obtained via N+1 repetition.
Lets call average profile obtained with N repetitions a(N) and the figure of merit is a function which takes a(N) as argument.
Given this figure of merit, one could then say "running N+1 times is pointless because FOM(a(N+1)) does not differ much from FOM(a(N))"
This could lead to speed up in whole year profiles calculation