Tables, figures, and outline for results/discussion

[mcgregorian1]

Table 1 - hypotheses and predictions (Introduction)

• [x] fill in table

Table 2 - list of the species, alphabetical (Methods)

• [x] include species means of each leaf hydraulic trait (print rp value)
• [x] include DBH mean and range of cored trees

Table 3 - tested_traits_all (variables) (Results, after second heading hypotheses)

• [x] format in excel and read in as image

Figure 1 - time series climate conditions (Results, drought descriptions)

• [x] move to new location

Figure 2 - height profiles (Results, first heading hypotheses)

• [x] fully format [priority]

• [x] Figure 3 - use TWI as base graph, put cored tree locations on top of that. [priority]

SI

• [x] Table S1 - height regression equations
• [x] Table S2 - candidate variables
• [x] Table S3 - top_models_dAIC
• [x] Figure S1 - PLA and TLP with height and TWI

Results

Descriptions of Droughts

• physical characteristics of drought: duration, min PDSI (and when occurred), see slette et al 2019 to see how they suggest reporting each one
• sum nb.series for each scenario, mult by perc.neg (pointers df), define thresholds (e.g. 75% of all trees experienced the growth reduction of 40% or more)
  • make table for this
  • citing table 3 (the coefficient for individual year) - maybe mention how 77 and 99 have neg coefficients, indicating potential height effect: as time goes on, trees are more affected by drought anyways because they grow taller
  • this is followed by Figure 2

Hypothesis results

• one header for each main question
• present results for each one, walk through via order and logic of Table 1
• no citations here

Discussion

• first paragraph is summary of main findings, not re-presenting the results
  • "we supported this hypothesis, but not this one" (following same order as Table 1 still)
  • direction of responses seems to be relatively consistent but the individual responses vary
• tie things in (see here
  • main thing is that we're now better understanding what confers vulnerability or resilience on trees during drought - how forests respond to future droughts/climate change
• "we filled the gap" = 1-2 sentences
• limitations at our site:
  • aren't able to analyze historical forest community, nor trees that were killed by these droughts [aka we don't have data for individuals that were most severely affected] (though we found that there's little variation in climate sensitivity for trees that were cored dead vs cored alive).
  • p50/p80
  • We used crown position despite its uncertainty. However, height and crown position change relatively slowly so it shouldn't be that far off.
  • We did not use crowding index because it has much more uncertainty
• limitations for extrapolating other sites:
  • forests are different from place to place, but we've seen how in forests around the world, forests tend to suffer more. We've identified some facets for why this happens. (cite other studies)
  • has been observed elsewhere that individuals in more moist habitats are more susceptible to drought (cite other studies)
  • the species may be different from other sites but the hydraulic traits are general across species;
• next paragraph saying that our study advances understanding of droughts, e.g. bennett et al 2015 doesn't go into mechanisms but we do
  • and we show that height is more important than exposure, but doesn't eliminate the effect of it
  • further contribution is we show that two leaf hydraulic traits - relatively easy to measure - are good predictors of drought response, can be helpful for scaling up (e.g. from our site to eastern deciduous biome)
  • biophysical mechanisms are things that should be seen as universal but relative importance of each can vary within each drought
  • first study to show how these traits affect woody growth response to drought check with Lawren to confirm
  • science is better now. by advancing our understanding of the mechanisms for individual-level responses to drought, this opens door for better predictions (elaborate from above). "This is absolutely critical to predicting forest responses to future droughts, which we're likely to see more of in the future as a result to climate change. Forecasting forest responses to these droughts is a huge and important challenge", science is better.

[mcgregorian1]

Additional questions to consider:

Table 3

• [x] we didn't decide what we wanted to do for reporting the coefficients of categorical variables (crown position and ring porosity)

Table S2

• [ ] Do we want different header names?

Table S3

• [x] how do we want to present Table S3? Would it be better as an image of an excel file, since it's so large? Did we want to keep all coefficient values here?

[teixeirak]

Regarding Table S3, can you please point me to the file?

[mcgregorian1]

S3 is the top_models_dAIC file. Currently I'm only loading it into R as is

[teixeirak]

* how do we want to present Table S3? Would it be better as an image of an excel file, since it's so large? Did we want to keep all coefficient values here?

see issue #40.

[teixeirak]

Figure/table aspects here are basically all dealt with. Any remaining, shift to issue #48.

Before we close this issue, we need to move the outline to the manuscript file.

[mcgregorian1]

@teixeirak I'm not going to move the outline to the manuscript file until I know you're not working on it.

[teixeirak]

I'm working on Table 1 right now. You can go ahead.

[mcgregorian1]

Ok, it is done and pushed to github