Single vs Multi site run

[trotsiuk]

@checcomi I wonder if at the first step it will be more appropriate to have only the single site version of the the fortran code. Then we have an R routine that will loop through sites (even those it might slow down a bit the analysis).

The reasons I'm thinking about it is:

• there will be multiple parameters for each species, and if we have e.g. 10 sites that all have different species it will be complicated to appropriately prepare the input for fortran
• similar with the site data, if multiple species are involved
• As mentioned in Month assignment if startMonth != 1 #13 for different sites we might have a different time periods, and would be interesting to retrieve the data at different time intervals, thus to putt all together might be challenging.

[florianhartig]

I second that - use fortran only for the code that is crucial for the runtime - loops can more easily be done in R, this also allows easier parallelization