coreytcallaghan / bird_diversity_MoB

working draft of analysis for a paper currently under review
1 stars 0 forks source link

get beta diversity #12

Closed coreytcallaghan closed 3 years ago

coreytcallaghan commented 3 years ago

I think this is what we were after:

image

Thoughts?

Looks like S_n is where something interesting is going on!

coreytcallaghan commented 3 years ago

Or...

image

dmcglinn commented 3 years ago

Cool! So beta_Sn is the metric that more closely reflects just species aggregation (but it can also reflect issues of low coverage). WTF about the big change in beta_Sn from linear to non-linear fits- that seems strange given how steep the red line was in the linear fit - I would have expected a flat line given the non-linear pattern.

What n-value was used to calculate beta_Sn? Also gamma is always fixed although grain is varied - is that correct?

If we take the non-linear pattern as being the truth then it suggests for all metrics that across grains beta diversity is lowest at intermediate GHM - almost close to the random expectation (of 1.0) at least for beta_Sn. Given your other results on alpha and gamma diversity (which I can't seem to find now on the git repo) I think this is because alpha peaks at intermediate GHM but gamma stays relatively flat across GHM. The key take home is that species do not appear to be showing a lot of microsite preference at intermediate GHM and are a rather random sample of the species found in their region

coreytcallaghan commented 3 years ago

Awesome. This is super helpful interpretation.

This is good. The rest of the results are coming...

See #18 as I really need some advice there before I settle on figures and such!

My current thinking is that these figures (like above) will be presented for one grain size as the 'key results'/answer to the main objective and then look at grain size in figure 3. But we can chat more about that soon!

coreytcallaghan commented 3 years ago

Not sure about n-value or gamma. I think it is whatever the default would be in get_mob_stats