Almost complete, but very rough, draft of introduction. Might be too long? Also, possibly too heavy on modeling and too light on ecology?
Methods are in much rougher shape. Would like to discuss overall organization and priorities (e.g. what, if any, content should be added or dropped/moved to supplement)
Haven't touched results or discussion
Moving forward, goal is to work on this in smaller increments, but much more regularly (ideally, daily)
Model execution -- 1 of 8 combinations finished, but I accidentally killed the others. Restarted remaining 7 combinations, with much smaller ensemble sizes (20 members, instead of 100 to get results faster; I'll run more ensembles pre-publication). When the time comes, we may want to consider temporarily investing extra $$$ to do more runs in parallel (doubling our current allocation should be ~$150/month).
Processing outputs -- Wrote code to do this as efficiently as possible, but it still takes a long time because of how ED's outputs are now structured (many small files). Goal is to have some of these results read Monday or Tuesday.
Updates
This week