Recording species counts

[phylojenie]

@superornitho So just to clarify, we are only recording the first screenshot that a particular species appears in, correct? So if Nemo appears in SS_01-01-01_GBR, we record that screenshot as his only occurrence for the GBR region?

But for example if there are two different green crabs in two different screenshots, we would record two occurrences of that species and list the two different screenshots we saw them in?

I'd like to get away from loading the spreadsheet with multiple screenshot names per column and instead record numbers and location data for easier analysis. So for example we could have:

Location	GBR	EAC	T
Species X	3	1	0

Hope this makes sense, let me know what you think. As it stands the data sheet will be very hard to parse because we're not separating the observations by location or recording the count numbers.

[superornitho]

Yes, that is a better idea, I will follow this structure from now on. I was recording every species in every screenshot in the prospect that we would like to record species interactions, but I think this will take too much work and we won’t easily be able to assess the validity of these interactions. The literature on this subject is quite scarce I believe. Have a nice day, Étienne

Le 25 mai 2020 à 15:16, phylojenie notifications@github.com a écrit :

@superornitho https://github.com/superornitho So just to clarify, we are only recording the first screenshot that a particular species appears in, correct? So if Nemo appears in SS_01-01-01_GBR, we record that screenshot as his only occurrence for the GBR region?

But for example if there are two different green crabs in two different screenshots, we would record two occurrences of that species and list the two different screenshots we saw them in?

I'd like to get away from loading the spreadsheet with multiple screenshot names per column and instead record numbers and location data for easier analysis. So for example we could have:

Location GBR EAC T Species X 3 1 0 Hope this makes sense, let me know what you think. As it stands the data sheet will be very hard to parse because we're not separating the observations by location or recording the count numbers.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/DataDrivenEcologicalSynthesis/VirtualBiogeography/issues/37, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOJL7PXNKHAR47DPPXJVKGTRTK72HANCNFSM4NJXJ6GA.

[phylojenie]

Good point about the species interactions, however I think we can use our defined locations (Great Barrier Reef, Ocean Trench) etc. as a stand-in for counting species interactions. For example, if a clownfish and a blue tang are both recorded in the GBR screenshots, we can count them as interacting.

I will update the spreadsheet with the recording system I am using and we can both fill it out using this system.

[superornitho]

Yep, totally agree on that!

Le 25 mai 2020 à 15:25, phylojenie notifications@github.com a écrit :

Good point about the species interactions, however I think we can use our defined locations (Great Barrier Reef, Ocean Trench) etc. as a stand-in for counting species interactions. For example, if a clownfish and a blue tang are both recorded in the GBR screenshots, we can count them as interacting.

I will update the spreadsheet with the recording system I am using and we can both fill it out using this system.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/DataDrivenEcologicalSynthesis/VirtualBiogeography/issues/37#issuecomment-633692174, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOJL7PW3LC6ZJSBZX6KZHATRTLA2PANCNFSM4NJXJ6GA.

[phylojenie]

@superornitho Take a look at the new spreadsheet format when you have a chance. I have added a column for each location, and we can record the total number of each species at a given location. If you like this format, we'll go ahead with it!

[superornitho]

Hi all, I don’t see how counting the number of occurence in each « biome » would be useful in our analysis. Also, I don’t want anyone to try counting how many mackerels there are in some scenes. I would consider something like:

LOCATION GBR EAC … Species1 Screenshot#of first occurence [blank if species not present] Species2 [blank if species not present] Screenshot#of first occurence

Le 25 mai 2020 à 15:30, phylojenie notifications@github.com a écrit :

@superornitho https://github.com/superornitho Take a look at the new spreadsheet format when you have a chance. I have added a column for each location, and we can record the total number of each species at a given location. If you like this format, we'll go ahead with it!

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/DataDrivenEcologicalSynthesis/VirtualBiogeography/issues/37#issuecomment-633693339, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOJL7PQMUH7HZ4LHA4E2IE3RTLBL7ANCNFSM4NJXJ6GA.

[phylojenie]

I see your point, however tracking the occurrences at each point will give us a better idea of how to match the species to their range maps; for example if one species is observed at both the GBR and in the Sydney Harbour, we could then check to see if their range extends that far in real life.

Also, estimating counts of each species was going to be one of the ways we checked for the accuracy of the abundance of each species.

[superornitho]

Sorry if my wording was unclear. I agree with you on this first point "tracking the occurrences at each point will give us a better idea of how to match the species to their range maps; for example if one species is observed at both the GBR and in the Sydney Harbour, we could then check to see if their range extends that far in real life."

As for calculating the abundance of each species in each area, I pondered how this could be useful. We will maybe get 1-5 individuals of each species in the whole film (+ some outliers like the mackerels), so I don’t think extrapolating these abundances to this whole part of the ocean makes a lot of sense. Also, I am not sure if GBIF occurrences can offer an unbiased estimate of the local abundance of species. I think that the areas that are moe accessible will have more occurrences and I don’t think that this can be reliably associated with the abundance of the species in the area.

Le 25 mai 2020 à 16:12, phylojenie notifications@github.com a écrit :

I see your point, however tracking the occurrences at each point will give us a better idea of how to match the species to their range maps; for example if one species is observed at both the GBR and in the Sydney Harbour, we could then check to see if their range extends that far in real life.

Also, estimating counts of each species was going to be one of the ways we checked for the accuracy of the abundance of each species.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/DataDrivenEcologicalSynthesis/VirtualBiogeography/issues/37#issuecomment-633703649, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOJL7PRPU3TNLDD76BTSTYLRTLGLHANCNFSM4NJXJ6GA.

[phylojenie]

@superornitho I see your point! I suppose we could look and see if the ratio of species to species scales up to match the occurrence records? For example, if we could 5 butterfly fish and 1 eagle ray, will the occurrence data show 5 butterfly fish observations for every 1 eagle ray?

However since we have decided to use range maps rather than occurrence data, that point is somewhat obsolete...so perhaps we should just record presence/absence of a species at each location? 1 for present and 0 for absent?