hypotheses and glossary

[lizzieinvancouver]

See also issue #25 ... List of all hypotheses (team?): @FrederikBaumgarten @AileneKane @kavs-P -- @FrederikBaumgarten will schedule meeting ASAP/make it happen ... they will consult with @alanaroseo

Please just work on here, or add new tex file to repo!

[AileneKane]

Question for @lizzieinvancouver, @FrederikBaumgarten and @kavs-P : should we include in our table the things you've already compiled in issue #22?

[FrederikBaumgarten]

@AileneKane and @kavs-P : I would say yes since the table should be complete on its own. Also good to try to break it down to the essential.

[FrederikBaumgarten]

I thought for every hypothesis we ideally would have the following:

• name
• short description
• something on how well this is testable
• how well this has been addressed so far
• origin (author, year)
• perhaps how relevant we think it is Even if we can't find all this (or judge) it might be worth to think about.

Let me know if you have other ideas

[lizzieinvancouver]

@FrederikBaumgarten @rdmanzanedo had this amazing figure from last year which might be something to adapt/think about ...

[AileneKane]

I have uploaded our full hypothesis table ("Grephon.hypotheses.csv"), which includes 10 main/broad hypotheses ("group_final" column) as well as how these groups and hypotheses align with the hypotheses listed in the above figure. In addition, I uploaded a summary table with the number/identity of papers within each mina hypothesis group ("hyp_summarytab.csv"). Some work is now needed to adjust the hypotheses across the figure and our table, so that they better align (e.g., the figure does not contain C-sink limitation, which is one of the more common hypotheses).

[FrederikBaumgarten]

cool, very nice!

[lizzieinvancouver]

I have a couple queries that I am writing down here so I remember:

• Is PP (photoperiod) control over growth really different than population adaptation to GS length? I wonder if PP control could be considered a special case of population adaptation where you know the mechanism.
• Does 'effect of growth rate not equal to growth duration' mean the extended season is basically just extra time when it's cold so it does nothing? Or when the effect of increased temperature on biological rates is more important than a longer season?

[lizzieinvancouver]

@FrederikBaumgarten @AileneKane @kavs-P ... scratch idea of updating @rdmanzanedo figure for the hypotheses here ... hoping @rdmanzanedo or others would have creative ideas of how to visualize some that don't match to previous figure:

Letters shown here:

[FrederikBaumgarten]

I have a couple queries that I am writing down here so I remember:

* Is PP (photoperiod) control over growth really different than population adaptation to GS length? I wonder if PP control could be considered a special case of population adaptation where you know the mechanism.

-> good point! But couldn't you argue for every cue like this? Temperature sensitivity in spring seems also to be a case of population adaptation (at least some provenances show distinct differences in GDD until BB). Do you feel we should pool even more Hyps together?

* Does 'effect of growth rate not equal to growth duration' mean the extended season is basically just extra time when it's cold so it does nothing? Or when the effect of increased temperature on biological rates is more important than a longer season?

-> I meant your second suggestion, but not restricted to temp. In other words: Conditions affecting growth rate play a more important role for total growth than growing season length per se

[lizzieinvancouver]

@kavs-P during meeting added: Species differences could have different curves for different species with some moving or changing more than others?

[rdmanzanedo]

An updated figure after the nice discussions with all (and the numbers and sketch by lizzie). Ideally on the right side we will add the numbered references of each paper in the final document :)

[AileneKane]

Thanks for doing this @rdmanzanedo! It is looking good! I have some questions/suggestions:

1. The numbers of of studies and names of studies do not match my notes and the updated table from our last meeting. Please check the hypothesis table that aligns each study with a hypothesis and the hypothesis summary table that counts the number and name of study associated with each group
2. For example, "longer growing season != more growth" should have 12 studies, I believe? that is a big difference from the 3 that are listed, so please check the above linked tables and let me know where you think the mistake was made (i.e., which studies are mislabeled) OR correct the figure here to match the numbers in the hypothesis summary table
3. I'm not sure what the 5-6 studies are for "Population adaptation limits responses" - i like the name and drawing for this, but i think it is referring to the 2 studies we have labeled "population specific responses"?
4. Also missing is "carbon fertilization" (2 papers) but maybe we decided not to include that in this figure?
5. I like the idea of having the references listed on the right, in a column next to the number of studies (see the hypothesis summary table for the refs

[kavs-P]

This is probably my fault (if Ruben used the sketch I sent as a base for this). And in that case, Oops!
I took the numbers from Lizzie's sketch so they need to be updated. And I also didn't include the carbon fertilization one. Since we talked about maybe just adding it in as a sentence in the image caption. Unless of course it's referenced more heavily in the main text

[lizzieinvancouver]

@AileneKane @kavs-P @rdmanzanedo Thanks for your work on this! I know that I did lump some things and may have changed names though I regret don't remember now how I did it.

[rdmanzanedo]

@AileneKane I see, indeed, I used the numbers in the sketch of Kavya and didn't really check with the other one (I stupidly assumed they were very similar). The number of studies are easy to adapt (I have also a version with a second column to put the references, but I will update that one with the correct references in the submission version).

we have some discrepancies there with some but the graphs and the structure of the figure is probably very easy to adapt. I will change this figure following these points next week and then we can maybe quickly confer (us or in the next grephon) to finalize it?

(yeah, I think we said that carbon fertilization would be added if it is actually a big section on the text, but that otherwise we would add in the caption that some papers studied the role of increased carbon availability and that that could interact with all the described patterns). :)

[AileneKane]

Thanks @rdmanzanedo , for updating the figure and thanks @kavs-P and @lizzieinvancouver for clarifying where the numbers came from! I'm open to changing the wording/groupings as long as we know where the studies are coming from. Let's review this figure with the whole group, if possible, (and perhaps the hypothesis summary table?) and ask everyone to check that the papers they read are associated with one or more hypotheses that make sense. After that we can shift, as needed.

[lizzieinvancouver]

@AileneKane @kavs-P @rdmanzanedo I am hoping to have a new draft of the ms to the group in a couple days, so suggest we wait until after that to decide how to edit the figure (we can also edit the manuscript to match the figure, or a little of both). Thank you again for all your help!

[lizzieinvancouver]

@AileneKane @FrederikBaumgarten As we finalize the figure and table, I need to make sure it lines up with grephontable ... which I decided to delete bruening2017 from. Someday I'd like to write in the paper in Sweave:

Total time increasing growth was by far the most common hypothesis for why longer seasons should increase growth, across our systematic review of growth $\times$ growing season length studies, with \Sexpr{hypnum} of \Sexpr{studynum} total studies including it (see Supplement for review details).

For now I am writing 19 of 59 ....

[rdmanzanedo]

@AileneKane @lizzieinvancouver that sounds great to me. Looking forward to read the Msc! 😄

[AileneKane]

We need to rework this to match whatever with c-source and sink in the text!

[rdmanzanedo]

@AileneKane @kavs-P Re-adapted the figure but 2 small doubts:

• Did we decided to add in the end carbon fertilization graph or not? My impression is that we decided against, but I may be wrong.
• In the hypothesis table there is no mention of the "more temp=more growth rate", which confuses me? are they maybe merged with another one? In the original notes we had 5 studies in this category.

Happy to change anything else if you notice

[AileneKane]

@rdmanzanedo

• i believe that in our last grephon meeting we decided to avoid carbon fertilization, as you say.
• I believe that in one of our earlier hypothesis table/figure meetings we decided to merge in "T control over growth rate" (which I think is the one you are referring to) with- "longer growing season != more growth" OR perhaps you are referring to "Higher temp = more growth (temp limitation)" which got lumped into "earlier-longer-more growth"?
• You can see all the papers and hypotheses, along with their various groupings/iterations here: https://github.com/lizzieinvancouver/grephon/blob/main/data/Grephon.hypotheses.csv
• Perhaps you saw lizzie's email in which she suggested merging "longer growing season ! = more growth" and "earlier != more growth"- I would be supportive of this. Could we make the new hypothesis title be "earlier/longer growing season ! = more growth"?

[rdmanzanedo]

@lizzieinvancouver @AileneKane

• great about carbon fertilization (let us know if this changes lizzie)
• I am not sure I was clear. I meant the 'more temp = more growth', which is in teh GSL+temp part. Having a check at your the table indeed it seems like we merged it with 'Earlier/longer GSL != more growth'. I am still usnure whether it is good to merge those two mechanistically, but I will remove and we can easily bring it back if we want.
• Yeah, saw the mail! Answering here to maybe not have duplication: I totally agree on merging the species-populations (done)
• I am not super supportive of merging 'longer != more growth' with 'earlier != longer', they seem very different to me and they seem like they would be tested and supported by very different evidence. Feels like it is a bit of the confusion we were criticizing to not make clear the differences between timing and final output. That being said, I see the logic on merging them and I won't fight this very hard (specially if we discuss it in the figure caption or in the text).

[rdmanzanedo]

With all the changes and merging the numbers from the summary table it would be:

(I am keeping all intermediate versions editable, so don't worry about asking more changes or bringing things back :) )

[rdmanzanedo]

btw, the shifting allocation one we have is about increased underground (I believe from the table), would that make sense to put root production or something like that in the graph, rather than reproduction?

[lizzieinvancouver]

@rdmanzanedo Thank you! This looks amazing. I wanted to flag two things to check -- especially for those who read the papers for the hypotheses:

1. I added the split between delta GSL alone and delta GSL + increased temp -- does it work and is it accurate across how the hypotheses are treated in papers (generally)?
2. Do the predictions fit with the how the hypotheses are treated in papers (generally)?

[rdmanzanedo]

Adding letters and small changes following @kavs-P 's recommendations

[lizzieoverleaf]

@rdmanzanedo and @kavs-P Thanks for your work on this!

@kavs-P I am adding your email here so everything is together:

Here is my first pass on referencing the hypothesis figure in the text. In the attached pdf I highlighted current mentions of the figure and places where it should be referenced. I also imagined that each hypothesis was labeled a-f (should we do this?) just as a shorthand for referencing the hypothesis more specifically in red text. And added some comments.

I think most of the hypothesis already have representation in the text but here are some observations:

Earlier/longer GLS != increased growth — I don’t really see a place in the text where this is explicitly mentioned in the way it’s shown in the figure. Maybe in the internal constraints section, there could be something about growth limits too beyond the solstice stuff? Or is this the solstice stuff?
Population and species specific responses — this one seems well covered in the species-specific, population and individual level sections
Shifting allocation (increased other function) — Right now this is part of the species-specific, population and individual level sections which makes sense but maybe we can write a sentence or two saying that this could generally lead to breakdown of the expected gslXgrowth relationship beyond just leading to species/population differences? Maybe also add text about belowgrownd vs above ground growth? 
Earlier GSL = increased risk — In the text, this one is referred to in biotic external factors but not for like frost damage. Maybe we should add a sentence or so in the external drivers paragraph for climate impacts and link it to the figure?
Increased temp = increased drought — in the text this is referred to more from the perspective of precipitation/vpd/external climatic factors other than temp. Just something to note

It would be great to have a second (or third or fourth) pair of eyes on it too!

I went back to the Grephon.hypotheses.csv to try to decide how to handle this and some of the issues raised above. Here's my take:

1. I think we should cut CO2 fertilization as both entries are flagged with: remove this?authors believe increased growth/productivity is due to muktiple factors in addition to longer/warmer growing season
2. Earlier/longer GLS != increased growth ... first, I feel good combining earlier and longer as the reasons underneath them are really varied (photoperiod control, general sink limitation', effect of growth rate not equal to growth duration, temperature control on growth and shift of whole pheno sequence) but I think do all fit with internal constraints. We could relabel this toEarlier/longer GLS != increased growth (internal constraints)' to make the connection easier for me to write, though I'd really appreciate someone from the hypothesis team checking my work here ....
3. I can add some above/belowground text (thanks @rdmanzanedo for the new ref).
4. I can add frost/check how we use it.

[lizzieoverleaf]

Also, @AileneKane @kavs-P @FrederikBaumgarten re-posting my query:

I wanted to flag two things to check -- especially for those who read the papers for the hypotheses:

I added the split between delta GSL alone and delta GSL + increased temp -- does it work and is it accurate across how the hypotheses are treated in papers (generally)? Do the predictions fit with the how the hypotheses are treated in papers (generally)?

[lizzieinvancouver]

@FrederikBaumgarten Here's the general phrasing if you can try to make it all match up! See: Controllers on growth × growing season length relationships

External, including temperature and moisture control and species interactions (competition etc.) Internal including:

• Universal limits, see text includes Zohner solstice
• Species-specific
• Population- & individual-level limits

[FrederikBaumgarten]

@lizzieinvancouver @AileneKane @rdmanzanedo @kavs-P Wow, I needed a couple hours to go through this issue, catching up and trying to come up with a solution. We are expecting a lot from this figure (maybe too much). Trying to bridge the text, the hypothesis but also conceptualize more broadly.

Here are my suggestions:

• Split between "delta GSL" and "delta GSL + Temp" is not logical to me, because a longer GS is also caused by increased temp. But we could separate between temperature effects that causes GSL to change (e.g. "GSL (arrow up)") and temperature effects that changes the growth/development rates during the GS (eg. "T (arrow up) during GS"). This would also solve the issue with figure f that only shows increased temp effects without GS lengthening (which I think is good). If we agree on that we can probably simplify the names of the hypothesis in the figure by removing the terms "longer GS", because everything on that branch is related to that. Currently there is a mix of mentioning GSL and effects on growth.

• Fig. a: This corresponds to all external factors currently limiting the GSL and associated growth. I would rephrase to "Earlier/Longer GS = (arrow up) growth".

• Fig. b - e are internal. For you Lizzie so you can mention them in the text.

• Fig. b: perfect. this corresponds to all internal/endogenous constraints limiting growth (e.g. photoperiod control, determinism, fixed growth target etc.). Suggest to change to "Earlier/Longer GS". The most common response curve is not shown in this figure: growth is shifted towards spring (without extension in autumn). Either we split those up and make a new fig. (gather hyps termed "shift of whole pheno sequence") or mention carefully in the text.

• Fig. c: perfect.

• Fig. d: great!

• Fig. e: What about frost in autumn. Should include also risk of a GS extension in the fall. Perhaps rephrase to "(arrow up) risk = (arrow down) growth". But not sure here. Given our search terms its not surprising that we haven't found any studies here. Or do you think this is still valid since no study made any connections to increased frost or herbivory damage?

I hope this helps. @rdmanzanedo: Sorry for some more change requests. But it seems to me on a good way. (when I started I was not sure about that...:)). Let me know if my comments need some more explanations. I like your figures! BTW: What is AUC change?

[lizzieinvancouver]

@FrederikBaumgarten Thanks for this! I suspect AUC means 'area under curve' ...

All changes sound good to me; I will try to think more on e.

[lizzieinvancouver]

@FrederikBaumgarten I am finally back at this paper and hoping to spiff it up shortly. To do this I need to merge the hypotheses with the other data we extracted ... However two issues:

1. zu et al 2021 does not exist in Grephon ... what paper is this?
2. eckes-shephard2020 does not exist in the hypothesis table ... was it excluded for some reason?

Also, @AileneKane -- thanks for your help on the summary for the figure (in hypothesest.R). Can you confirm you excluded the studies not in GREPHON? I noticed some are listed in the original table....

[lizzieinvancouver]

Note to self: deleted `glossary' from title of this issue today.

[FrederikBaumgarten]

@FrederikBaumgarten I am finally back at this paper and hoping to spiff it up shortly. To do this I need to merge the hypotheses with the other data we extracted ... However two issues:

1. zu et al 2021 does not exist in Grephon ... what paper is this?

-> hm interesting, I cannot find it. I probably found on my search for additional hyps outside our paper collection. But since there is no hypothesis attached to it we can discard it!

2. eckes-shephard2020 does not exist in the hypothesis table ... was it excluded for some reason?

-> Eckes-Shephard2020 was assigned to Kavja - so not sure why she excluded it. I checked the paper. It's a tricky one as they don't formulate a clear hypothesis but what they basically test how water constraints in the soil control sink (growth) activities. Hence this papers comes into the hypothesis of environmental factors controlling growth. I believe this would end up in bin "more temp = more drought (drought limitation)" similar to the entry of Drew2018. Hope that helps!

Also, @AileneKane -- thanks for your help on the summary for the figure (in hypothesest.R). Can you confirm you excluded the studies not in GREPHON? I noticed some are listed in the original table....

[AileneKane]

@lizzieinvancouver Yes, I excluded the studies not in GREPHON from the table. The group_final column is what we ended up with. Please let me know if there are additional changes I should make or if you notice any mistakes!

[lizzieinvancouver]

@FrederikBaumgarten Thanks. Do you think zu et al. 2021 could be Zhu et al. 2021? ("Different response of earlywood vessel features of Fraxinus mandshurica to rapid warming in warm-dry and cold-wet areas") -- that paper is in the main GREPHON table but missing from hypotheses (and there's no info in the table so I cannot figure out if it si the right paper)?

We're also missing Zohner 2020 ('Increased growing-season productivity drives earlier autumn leaf senescence in temperate trees'). Could you let me know what to add for that?

Actually @AileneKane might know what zohner2020 should be?

[lizzieinvancouver]

@FrederikBaumgarten @AileneKane Also, three papers have no entries ... does that mean they gave no reasoning or something else?

   paper_id wording_figure
37  brand2022               
38   chen1998               
39 zuetal2021  

[FrederikBaumgarten]

@FrederikBaumgarten Thanks. Do you think zu et al. 2021 could be Zhu et al. 2021? ("Different response of earlywood vessel features of Fraxinus mandshurica to rapid warming in warm-dry and cold-wet areas") -- that paper is in the main GREPHON table but missing from hypotheses (and there's no info in the table so I cannot figure out if it si the right paper)?

Yes indeed! I remember we kicked it out since it was more a xylem anatomical study and didn't present a clear hyp. If we must fit one it would be the "more temp = more drought (drought limitation)" hypothesis

We're also missing Zohner 2020 ('Increased growing-season productivity drives earlier autumn leaf senescence in temperate trees'). Could you let me know what to add for that?

Actually @AileneKane might know what zohner2020 should be?

[FrederikBaumgarten]

@FrederikBaumgarten @AileneKane Also, three papers have no entries ... does that mean they gave no reasoning or something else?

 paper_id wording_figure
37  brand2022     

-> I didn't reviewed this one, but checked it now and it seems clearly to belong in the hypothesis group of "more temp = more drought (drought limitation)"

38 chen1998

-> no reasoning was given, hence no hypothesis attributable

39 zuetal2021

[AileneKane]

@lizzieinvancouver For Zohner2020, I would call it "Earlier/longer GLS != increased growth" though it is a slight variant that might be more precisely described as "Earlier/longer GLS != increased growth, ATLEAST NOT AS MUCH AS PREVIOUSLY PREDICTED" due to interactive climate effects. These "interactive climate effects" include effects from warmer autumn temperatures, which reduce chilling, and photoperiod effects. Is that enough detail? Please let me know if you prefer to put this information somewhere else besides just in this issue

[lizzieinvancouver]

@FrederikBaumgarten Just to confirm: I am listing NO hypothesis for chen1998 and for zhuetal and brand2022 I am listing: more temp = more drought (drought limitation) ... is that correct

[lizzieinvancouver]

Please let me know if you prefer to put this information somewhere else besides just in this issue

Thanks @AileneKane. This is good for now, I am entering your answers (with notes that I entered them via this issue) in Grephon.hypotheses.csv so I think we're all set.

[FrederikBaumgarten]

@FrederikBaumgarten Just to confirm: I am listing NO hypothesis for chen1998 and for zhuetal and brand2022 I am listing: more temp = more drought (drought limitation) ... is that correct

yes, thats correct!

[lizzieinvancouver]

@rdmanzanedo I finally believe that I have updated numbers, you can check them out here but here is the take home from them for this figure:

a) 17 (earlier-longer-more growth) b) 15 (longer growing season != more growth PLUS earlier!= longer growing season) c) 6 (species-specific responses and population-specific responses) d) 1 (shift in allocation) e) still zero! f) 9 (more temp = more drought (drought limitation))

Carbon fertilization... we decided not to include (conversation as of 22 Nov 2023). Though I wonder if cannot be included in an a-f hypothesis? @AileneKane @FrederikBaumgarten Could we fit into another hypothesis?

Also, @rdmanzanedo can you make @FrederikBaumgarten suggested changes (see 11 Dec 2023 notes above)? I think he is on a European timezone so you can connect with him if you have questions/concerns.

[lizzieinvancouver]

@AileneKane We have TWO summary tab files from the hypothesis work: summarytab2 ... which is updated when I run hypothesisest.R and summarytab which I think we'll really want for the supp but does not get updated. Could you track down where it is created and make it so it is updated as I update data/Grephon.hypotheses.csv" ? Thank you!

[AileneKane]

@lizzieinvancouver I don't see any difference in what's contained in summarytab2 ... vs. summarytab What am I missing? Could you share what you want from summarytab that you're not seeing in sumamerytab2? I saved summarytab2 under a different name, initially, so that i could compare the new version to the old without overwriting it, but i don't think summarytab.csv is necessary to keep any longer....

[lizzieinvancouver]

but i don't think summarytab.csv is necessary to keep any longer....

That might be the issue then! I will check later and just delete whichever one is not getting updated.

[AileneKane]

@rdmanzanedo and @lizzieinvancouver I have some suggestions for figure 1, with the main goal of better alignment across the figure and the text. My suggestions and questions are below- I'm happy to set up a meeting to discuss if helpful!

1. I have gone back and forth about this, but I think it might be helpful to add "internal" and "External" to the figure and separate out the hypotheses for "internal" (population and species specific responses, shifting allocation) and "external" (everything else, right?) controllers on growth × growing season length relationships (see the top photo below for one idea on how to do it- there is likely a better way!)
2. I think it would help the reader to explicitly say what the expected growth response to increased GSL would be under each hypothesis, perhaps by adding a column after the "# of studies" that days "predicted growth response" or something (see lower photo below)
3. Do we need to have "peak growth" labeled? this does not seem necessary to me
4. I think we should add the "carbon fertilization" studies to the figure, since this is mentioned in the text (I believe this would be an external controller)
5. Make sure the number of studies are updated and aligned to the text

[rdmanzanedo]

Hi everyone, sorry for the late reply, finally found some time to this figure.

Replying and adding the new figure:

@FrederikBaumgarten great!, added most of points as I understood them (see them below, let me know if I missed them in any way. The two points I did not change are: Panel b. I don't really see it like that, the idea is that that figure captures extending in either spring or autumn not translating in a total change, so the peak has to reduce, if the early advances but the peak doesn't change then we have a net increase (it is what we had in one of the previous versions before merging (see the comment in this threat of october 26, 2nd figure). I would say putting a figure like that one but with lower peak would unnecessarily take from the longer GS we have in the name of that category. I agree the shape of the figure was unclear, I tried to make it simpler to maybe convey this better? Let me know if that is not the case. Panel d. I agree with your point about the potential of later frost part, but I think adding it would make the figure even more crowded (and honestly, there are already no studies linking earlier phenology with reduced growth via frost/pests, so isn't it better to be more specific in the text about spring/fall risks (totally agree with the other wordings and have changed it accordingly

@AileneKane love the idea of the expected growth response column, I added it! I like the idea of internal, external differentiation and I tried to rearrange to fit that but I am not sure it is clearer or better (also how would we name them? internal mechanisms, external mechansims, isn't earlier phenology internal? happy to chat about it and it sounds interesting but I am not sure this version would be clearer than what we had for our intentions, in any case, see the figure draft below). Also, the idea of peak growth was to try to see metrics of differentiation, e.g. a and b would have same response in onset but different peak growth and AUC. Also, many dendrometer studies use that peak growth parameter. Happy to delete it if we think it muddles our message.

@lizzieinvancouver I thought originally in having another column to the right with the actual studies, do we want this here or a talbe in the supplementary will be enough? I tend to like to see the papers listed right away in main material but I understand it may become too large of a figure.

Option A:

Option B: