search

EcoClimLab / ForestGEO-tree-rings

Repository for analysis of tree-ring data from 10 globally distributed forests (Anderson-Teixeira et al., in press, Global Change Biology)
2 stars 2 forks source link

summary picture #136

Closed teixeirak closed 3 years ago

teixeirak commented 3 years ago

From R1: "given the complexity of the results, it would be very useful to report a summary picture of the main results by biome/ecoregions and level of naturalness of the studied stand (old-growth, seminatural forest ecosystems or secondary forests, reforestation/afforestation if present)"

Offhand, I'm not sure what to do about this...

teixeirak commented 3 years ago

Here's a sketch of what I think I'd like to do:

image
teixeirak commented 3 years ago

@srusso2 , we've emailed about a summary figure, so I wanted to flag you here to get your opinion.

crollinson commented 3 years ago

I really like this. I think this will be a really intuitive way to help summarize.

srusso2 commented 3 years ago

Likewise! I think this is a great way to summarize the results from all of the analyses. I suppose you can group the arrows so all the ups are together, all the downs are together, etc. That way, it's easy to see what the dominant trends are.

teixeirak commented 3 years ago

Glad you like it, @crollinson and @srusso2 !

hemlockous commented 3 years ago

For what it is worth, I also like this. Justin

teixeirak commented 3 years ago

Great, thanks @hemlockous!

teixeirak commented 3 years ago

Update - here's a start:

image

This uses drawio and is editable here: https://app.diagrams.net/#HEcoClimLab%2FForestGEO-tree-rings%2Fmaster%2Fdoc%2Fmanuscript%2Ftables_figures%2FSummaryFigure.png.

teixeirak commented 3 years ago

Here is a version that might be final: image

Comments welcome.

crollinson commented 3 years ago

This is awesome! I think it's going to make this paper become a go-to citation for the complexity of multivariate responses across species and environmental gradients. It also very clearly shows that linear environmental/climatic responses should NOT be the default!! I'm actually surprised at how consistent some of the climate responses are and how messy DBH is, but it makes sense, particularly when we remember that we are parsing DBH and year effects where traditional dendro conflates them.

In terms of questions/feedback:

Probably all solvable with the caption, but those were my at-a-glance reactions.

teixeirak commented 3 years ago

Thanks, @crollinson! Glad you like it.

Therea re footnotes explaining both your questions, but are they too hard to notice/ not intuitive?

hemlockous commented 3 years ago

I think this looks great! I have the same confusion as @crollinson with the sometimes overlapping arrows. But other than that, I think it is awesome!

hemlockous commented 3 years ago

Just saw your comment in response to Christy. I understand now. I wonder if it would be better to move the footnotes into the caption where most people go to figure out a figure?

teixeirak commented 3 years ago

Perhaps move just the single- asterisk footnote (on overlapping arrows) to the caption? The ** and *** notes seem to me like they might be better as footnotes, but I'd also be okay moving them to caption.

hemlockous commented 3 years ago

Yea I agree only the * footnote. I see other other overlap pong arrows outside of the DBH column so I think that would be better for the caption.

——— 50% of my typos are because of this phone... don’t ask about the other half.

On Aug 4, 2021, at 10:50 AM, Kristina Anderson-Teixeira @.***> wrote:

 This message was sent from a non-IU address. Please exercise caution when clicking links or opening attachments from external sources.

Perhaps move just the single- asterisk footnote (on overlapping arrows) to the caption? The ** and *** notes seem to me like they might be better as footnotes, but I'd also be okay moving them to caption.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/EcoClimLab/ForestGEO-tree-rings/issues/136#issuecomment-892723237, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AU73K2KGYXW6L4IF7BGGKWDT3FHSDANCNFSM5BAY665Q. Triage notifications on the go with GitHub Mobile for iOShttps://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Androidhttps://play.google.com/store/apps/details?id=com.github.android&utm_campaign=notification-email.

crollinson commented 3 years ago

Yeah, agreed. I think I overlooked that most of those were in the DBH column and that's where the * was, so caption would work better for me since it's a general explanation that applies to a lot of the figure rather than specific sites or instance. Agreed with the other two staying as footnotes as they're more explanation than figure interpretation.

teixeirak commented 3 years ago

Okay, thanks @hemlockous and @crollinson . Here's the new version:

image

srusso2 commented 3 years ago

Hi, I agree with what’s been said about the figure, and it does look terrific.

A few comments -temperature is mis-spelled -Shouldn't it be precipitation and not water? Soil moisture depends on soil type and precipitation. -maybe it’s not showing up on my phone but you might make the lines between forest types more prominent than the lines between sites - so could make dotted between sites or thicker between forest types.

Sabrina E. Russo Professor, School of Biological Sciences & Center for Plant Science Innovation University of Nebraska-Lincoln, USA

From: Kristina Anderson-Teixeira @.> Sent: Wednesday, August 4, 2021 10:10:43 AM To: EcoClimLab/ForestGEO-tree-rings @.> Cc: Sabrina Russo @.>; Mention @.> Subject: Re: [EcoClimLab/ForestGEO-tree-rings] summary picture (#136)

Non-NU Email

Okay, thanks @hemlockoushttps://urldefense.com/v3/__https://github.com/hemlockous__;!!PvXuogZ4sRB2p-tU!StEutpA5kMwLfqPGfrWSTI85tfe7cKtTfa4RUTS2_0NOuNkHxjs5UuCfltPYS4o$ and @crollinsonhttps://urldefense.com/v3/__https://github.com/crollinson__;!!PvXuogZ4sRB2p-tU!StEutpA5kMwLfqPGfrWSTI85tfe7cKtTfa4RUTS2_0NOuNkHxjs5UuCfuZlZg-o$ . Here's the new version:

[image]https://urldefense.com/v3/__https://user-images.githubusercontent.com/6355854/128206056-ed1fbcb3-8c28-41a1-8869-a9aef78f2f28.png__;!!PvXuogZ4sRB2p-tU!StEutpA5kMwLfqPGfrWSTI85tfe7cKtTfa4RUTS2_0NOuNkHxjs5UuCfCvXMecA$

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://urldefense.com/v3/__https://github.com/EcoClimLab/ForestGEO-tree-rings/issues/136*issuecomment-892740225__;Iw!!PvXuogZ4sRB2p-tU!StEutpA5kMwLfqPGfrWSTI85tfe7cKtTfa4RUTS2_0NOuNkHxjs5UuCfhV4qroA$, or unsubscribehttps://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/AHHXIUGJ73ROSOCAF4EWLBTT3FJ7HANCNFSM5BAY665Q__;!!PvXuogZ4sRB2p-tU!StEutpA5kMwLfqPGfrWSTI85tfe7cKtTfa4RUTS2_0NOuNkHxjs5UuCfOxd7NVk$. Triage notifications on the go with GitHub Mobile for iOShttps://urldefense.com/v3/__https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675__;!!PvXuogZ4sRB2p-tU!StEutpA5kMwLfqPGfrWSTI85tfe7cKtTfa4RUTS2_0NOuNkHxjs5UuCfQxBOku0$ or Androidhttps://urldefense.com/v3/__https://play.google.com/store/apps/details?id=com.github.android&utm_campaign=notification-email__;!!PvXuogZ4sRB2p-tU!StEutpA5kMwLfqPGfrWSTI85tfe7cKtTfa4RUTS2_0NOuNkHxjs5UuCfJuFRW6M$.

teixeirak commented 3 years ago

Thanks, @srusso2! I've implemented these changes. (FYI, I've used "water" as a more general term in place of "Precipitation" because we did test for the effect of streamflow at Niobrara, but precip ended up as a better variable, so precip works in this context."

Here's the new, "final" version:

image

I think we can close this issue now.

ValentineHerr commented 3 years ago

Re-opening this issue as I have a few comments:

  1. I don't see, in the legend, what the difference between small solid dots and big solid dots is.
  2. I think I am confused between the meaning of the dotted arrows as described in the image legend vs in the text legend.
  3. in the text legend I would say "The arrows' shapes approximate responses..."

For 2, I wonder if the dotted arrows that represent model without Year should be of a different color. for example Grey. that way if they are dotted or solid reflects their significance and you could add the grey color in the image legend (still keeping the meaning in the text legend).

ValentineHerr commented 3 years ago

I wonder if we could add a shaded bar underneath "Precipitation" and "Temperature" that would represent the months from p.Jan to c.Aug and the darkness of the bar would represent the number of sites that have that month within their best window?

If you like the idea I can try to make it.

BUT, looking into this, I now wonder why we have windows past current August? Ha. It is because the reference month is December for BCI and HKK. I think we forgot to mention that in manuscript.

teixeirak commented 3 years ago

I wonder if we could add a shaded bar underneath "Precipitation" and "Temperature" that would represent the months from p.Jan to c.Aug and the darkness of the bar would represent the number of sites that have that month within their best window?

If you like the idea I can try to make it.

BUT, looking into this, I now wonder why we have windows past current August? Ha. It is because the reference month is December for BCI and HKK. I think we forgot to mention that in manuscript.

Hmmm, interesting idea! I think this could only make sense if we were to do a tiny bar for each site, which could be really helpful, but perhaps too much to digest? I don't feel that it's necessary, so I'd drop it unless it's something you'd really like to include.

teixeirak commented 3 years ago 
1. I don't see, in the legend, what the difference between small solid dots and big solid dots is.

The bigger dots were a mistake. I just fixed that.

teixeirak commented 3 years ago 
3\. in the text legend I would say "The arrows' **shapes** approximate responses..."

done.

teixeirak commented 3 years ago

For 2, I wonder if the dotted arrows that represent model without Year should be of a different color. for example Grey. that way if they are dotted or solid reflects their significance and you could add the grey color in the image legend (still keeping the meaning in the text legend).

It's true that I didn't differentiate between cases where the effect was in the best model but p>0.05, and cases where the response was different. That seemed too complicated, but perhaps it makes sense to go back and fix that. I could adjust the arrow head type-- something like this:

image

Does that seem good to you?

srusso2 commented 3 years ago

Hi – These are great ideas, but I think it’s hard to represent all of the detailed results in the figure.

In my view, the figure is meant to be a summary, not necessarily a symbolic depiction of every last detail of the results.

I think if there is too much detailed symbolism operating, it might take away from the main message of the figure because people will focus on wondering what this arrow head or that arrow head or color means and so on.

One could put a little disclaimer in the text legend to this effect – that the figure is a summary of the major take-away results, and does not show all of the detailed results for every site and species – for that, see Tables XXX and Figures XXX …that sort of thing.

That said, I am fine with whatever you decide.

This is really a lovely paper, and I appreciate all of the attention to this figure – it’s a great summary and addition to the study.

Best, Sabrina

From: Kristina Anderson-Teixeira @.> Sent: Wednesday, August 11, 2021 9:16 AM To: EcoClimLab/ForestGEO-tree-rings @.> Cc: Sabrina Russo @.>; Mention @.> Subject: Re: [EcoClimLab/ForestGEO-tree-rings] summary picture (#136)

Non-NU Email

I wonder if we could add a shaded bar underneath "Precipitation" and "Temperature" that would represent the months from p.Jan to c.Aug and the darkness of the bar would represent the number of sites that have that month within their best window?

If you like the idea I can try to make it.

BUT, looking into this, I now wonder why we have windows past current August? Ha. It is because the reference month is December for BCI and HKK. I think we forgot to mention that in manuscript.

Hmmm, interesting idea! I think this could only make sense if we were to do a tiny bar for each site, which could be really helpful, but perhaps too much to digest? I don't feel that it's necessary, so I'd drop it unless it's something you'd really like to include.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://urldefense.com/v3/__https:/github.com/EcoClimLab/ForestGEO-tree-rings/issues/136*issuecomment-896864802__;Iw!!PvXuogZ4sRB2p-tU!XTlb4Cx2FXad_pcFc8YLqGcN1Fg6L9IR5jn2j5KK9nBXuIvqtskwuzb0Nrkq9Eg$, or unsubscribehttps://urldefense.com/v3/__https:/github.com/notifications/unsubscribe-auth/AHHXIUG7Q5RDSRHHNVL2HNLT4KAZHANCNFSM5BAY665Q__;!!PvXuogZ4sRB2p-tU!XTlb4Cx2FXad_pcFc8YLqGcN1Fg6L9IR5jn2j5KK9nBXuIvqtskwuzb0ziwGXa8$. Triage notifications on the go with GitHub Mobile for iOShttps://urldefense.com/v3/__https:/apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675__;!!PvXuogZ4sRB2p-tU!XTlb4Cx2FXad_pcFc8YLqGcN1Fg6L9IR5jn2j5KK9nBXuIvqtskwuzb06HXYk-s$ or Androidhttps://urldefense.com/v3/__https:/play.google.com/store/apps/details?id=com.github.android&utm_campaign=notification-email__;!!PvXuogZ4sRB2p-tU!XTlb4Cx2FXad_pcFc8YLqGcN1Fg6L9IR5jn2j5KK9nBXuIvqtskwuzb0_M7fTnI$.

teixeirak commented 3 years ago

Thanks, @srusso2 .

I think the bars showing T/ precip months would be too much.

I am working on changing the arrowheads to differ for models with or without year, which adds one thing tot he legend, but I think is a meaningful distinction.

teixeirak commented 3 years ago

Here's the modified version:

image

This was worth the work, as it highlighted the fact that when year is accounted for, RW always initially increases with DBH for shade-tolerant species. Pretty cool! I've made a few edits to the text to capture this, and need to add a footnote to table 1 (#143 ).

I think we can close this issue again