search

forc-db / Global_Productivity

Creative Commons Attribution 4.0 International
2 stars 0 forks source link

NPP:GPP analysis is redundant with NPP:Rauto #123

Closed teixeirak closed 3 years ago

teixeirak commented 3 years ago

@beckybanbury , NPP:GPP analysis is redundant with NPP:Rauto, as @hmullerlandau pointed out. Let's remove one from Fig. S3 and from mention in the main text.

beckybanbury commented 3 years ago

@teixeirak @hmullerlandau I think it would be better to remove NPP:GPP then, so that all of the ratios are true ratios - do you agree?

bpbond commented 3 years ago

(Butting in.) I don't disagree, but why is NPP:GPP not a true ratio? 😕

beckybanbury commented 3 years ago

Sorry that's probably the wrong wording - I meant that NPP is a component of GPP so not independent, whereas NPP:R auto are both component parts of GPP. With the other ratios we decided in the end to just look at how the components vary wrt each other rather than the bigger flux.

teixeirak commented 3 years ago

First, I confused myself by putting Reco rather than Rauto in this issue (it is Rauto in the paper). Fixed here.

teixeirak commented 3 years ago

@teixeirak @hmullerlandau I think it would be better to remove NPP:GPP then, so that all of the ratios are true ratios - do you agree?

Actually, in this case I'd probably go with NPP:GPP. NPP/GPP=CUE, which we mention a few times in the paper and is the more common way of referring to ratios among NPP, GPP, and Rauto. (It could be argued the other way, though, and I'm open to either if you or @hmullerlandau have strong feelings otherwise.)

hmullerlandau commented 3 years ago

Yes I agree. Appropriate models differ for ratios vs fractions.

On Wed, Dec 9, 2020 at 1:07 PM Becky Banbury Morgan < notifications@github.com> wrote:

Sorry that's probably the wrong wording - I meant that NPP is a component of GPP so not independent, whereas NPP:R auto are both component parts of GPP. With the other ratios we decided in the end to just look at how the components vary wrt each other rather than the bigger flux.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/forc-db/Global_Productivity/issues/123#issuecomment-741951973, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADVG5NYX4RU23LU6Z4MAWNLST64GBANCNFSM4UJNLG7A .

-- Sent from Gmail Mobile.

beckybanbury commented 3 years ago

@teixeirak Yes, I've just realised that we added CUE on request of the reviewer for the reason that it's a commonly used metric, so makes sense to remove NPP:R auto here

hmullerlandau commented 3 years ago

I disagree. We can draw inferences about CUE even if we are analyzing the ratio of NPP to Rauto. After all, if NPP to Rauto changes, then obviously CUE changes. Analyses of fractions (CUE) are fundamentally different from analyses of ratios (NPP:Rauto), and we've settled on a good method for analyzing ratios. We should stick with the ratios, and interpret them in terms of changes in CUE.

On Wed, Dec 9, 2020 at 1:41 PM Becky Banbury Morgan < notifications@github.com> wrote:

@teixeirak https://github.com/teixeirak Yes, I've just realised that we added CUE on request of the reviewer for the reason that it's a commonly used metric, so makes sense to remove NPP:R auto here

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/forc-db/Global_Productivity/issues/123#issuecomment-741970398, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADVG5N7OZEKMIHVRHMROCGTST7ADXANCNFSM4UJNLG7A .

-- Helene Muller-Landau hmullerlandau@gmail.com +507 6471-5214 (cell)

beckybanbury commented 3 years ago

I disagree. We can draw inferences about CUE even if we are analyzing the ratio of NPP to Rauto. After all, if NPP to Rauto changes, then obviously CUE changes. Analyses of fractions (CUE) are fundamentally different from analyses of ratios (NPP:Rauto), and we've settled on a good method for analyzing ratios. We should stick with the ratios, and interpret them in terms of changes in CUE.

Yes okay, this seems good to me. I agree that I'm happy with the methods for the ratios so I'd prefer to stick with them, as it would be more complicated to change the methods than the interpretation. It should be relatively easy to interpret more specifically in terms of CUE - I'll edit this. Thanks everyone!