split coral cover data from `loadSiteData()`

[ConnectedSystems]

Initial implementation expects initial coral cover data to be part of the site_data inputs.

These were represented as a column of data for each site in a CSV. This should be split out so that site_data is only concerned with spatial data.

Splitting initial coral covers out will also better support multiple starting conditions.

[ConnectedSystems]

Initial work on this has been started

See ai.loadCoralCovers() in ADRIA.m in the Brick_modelling branch

[KRNA01]

@ConnectedSystems, working to update Rose's mat file for initial coral covers for Brick.

Looking at line 120 in ADRIA.m x = baseCoralNumbersFromCoversAllTaxa(prop_cover_per_site{row, :}); Do you still want to go through this function, or are you happy with the mat file format Rose produced?

Further, am I right that we want start covers for each species and size class rather than start numbers?

[ConnectedSystems]

It would be best if we could standardized to one (preferably open) format. Saves us from having to write and maintain these conversion functions.

So yes please, the .mat file is preferred for now, with a view to move to NetCDF or similar later.

Not sure what start numbers mean, but I take it you're referring to the number of representative species/size class groups (i.e 36). If so, then yes please!

[KRNA01]

Okay. Will produce a mat file for all 567 sites (or what Barbara boils it down to). Year will be 2026. Okay if this is a cube of the format sites, by coral group (6) by size class (6) Or do you prefer an array of sites by 'species' (36)

[ConnectedSystems]

I would prefer to keep things in line with the data sets we have already (at least for now) so by "species" (36).

We should discuss later at some point (for next major version of ADRIA) if this will need to be changed.

[KRNA01]

Thank you - will do.

[ConnectedSystems]

Thank you!

[KRNA01]

New data for initial covers for Brick cluster

Left Rose's file in place but renamed with suffix 'old'. File can be removed

[KRNA01]

PS: this is in the brick_modelling branch

[ConnectedSystems]

Hi Ken,

~Would you be able to do the same thing for Moore? I'm envisioning that it will have a single initial coral cover scenario for 2 taxa with the rest filled with 0s.~

~Not pressing at this stage but if we want to run Moore again with the latest version of ADRIA.~

I should also ask, what order are these sites in? These datasets would be much easier to work with if we also store the corresponding reef_siteid

EDIT: Scratch the request for Moore coral covers - I can create something that will address this.

[ConnectedSystems]

Answered my own question! Thanks anyways

[KRNA01]

Apologies, just realised I hadn't answered this. Answers are

1. Yes of course, and
2. Agree