mindthegap-erc / carbonatestratpalevo Goto Github PK

Implement functions get_AIC_scenario, get_AIC_time, and get_AIC_no_of_sampl_loc in code/make_plots.R

renv::restore() tries to compile Matrix from sources, but crashes. Attached the log - as much as I could capture from the termninal. This follows the step where you agree to install missing packages. R 4.3.1
It is resolved by changing the version to 1.6-5 in renv.lock but I am yet to check if that breaks anything down because of other packages that won't load either

Rewrite section "Carbonate Platform B" in results section after adjustment of sea level curve.

• Redo figures 1 & 2 based on captions in manuscript. Raw figures are on GH under figs/matlab

As figure 6, but for the tests that cover the time interval of scenario B

Make a figure with 2 rows (one for each scenario) and 5 rows (one for each distance from shore)
Each subfigure is a histogram of the time elapsed between successive samples taken in the stratigraphic domain. The y axis is number (dimensionless), the x axis is time difference (myr). Make sure the x axis limits are the same for each subplot and start at 0

This cannot work because the data folder is empty

1. Name figures according to their name in the manuscript _raw (e.g. Fig1_raw.pdf, FigS2_raw.pdf)
2. Redo Figs. 4-7. Specifically 6 & 7 based on new data
3. Adjust figure size to Journal requirements (https://bmcecolevol.biomedcentral.com/submission-guidelines/preparing-your-manuscript#preparing+figures)
4. Reduce temporal resolution for plots of trait evo in time domain, (maybe 200 sampling points, distributed over the 2 / 2.56 Myr)
5. Check if for stasis, figures become more clear with only one lineage. If yes, plot only one lineage for stasis
6. Make axis labels smaller & titles larger
7. Rearrange order of subfigures in Fig. 5 & similar figures (ADM & time domain 1st row, dist. from shore in 2nd & 3rd row) - Maybe there is a better/more aesthetic arrangement?
8. Automate labeling of subfigures (A, B, C, ...)

Preservation of stasis, Brownian motion, and weak & strong Brownian drift at different distances from shore in scenario A and B. See supplementary materials folder in onedrive for captions

• x axis lablel "Stratigraphic height [m]"
• y axis label "Trait value"
• reduce whitespace at bottom
• Letter same size as title
• correct whitespace in title

Make figure displaying spatial variability in preservation. Use strong Brownian drift in scenario A

Start sketching the author contribution statement in the manuscript. Have a look at https://www.elsevier.com/authors/policies-and-guidelines/credit-author-statement and the guidelines by the journal: https://www.biomedcentral.com/getpublished/editorial-policies#authorship

Make a fists sketch of the figure with stratigraphic completeness and hiatus duration vs. basin position.
Make one plot for each scenario, name it scenarioX_strat_info.pdf
The plot should contain:

• x axis: ranging from 0.1 to 15, distance from shore [km]
• left y axis: stratigraphic completeness as percent
• right y axis: hiatus duration in myr. Start with plotting median, max, and 1st and 3rd quartile of the hiatus distribution. Use a log scale if if necessary

When ms is in final stage

• Deposit large data files on osf, specifically .mat, .Rdata files
• Add code to automate data download

Figure should compare preservation of strong Brownian drift 2 km offshore in scenario A (left) vs 6 km offshore in scenario B (right). Both locations have comparable completeness, but scenario A has longer gaps

renv.lock requires fs 1.6.2 and R attempts to compile it from source and fails, but changing the required version of fs to 1.6.4 resolves the problem

Make figure showing variable effects of strat. architectures on preservation of modes. Use scenario A, 6 km offshore

• Reduce whitespace at the bottom
• x axis labels consistent "Distance from shore [km]"
• left y axis labels "Stratigraphic completeness [%]"
• right y axis labels "Hiatus duration [Myr]"
• Change header to "scenario X"
• Make letter same size as header
• cut off x axis at 13 km
• Add legend to top left corner, listing
  • Completeness
  • maximum hiatus dur.
  • 1st quartile hiatus dur.
  • Median hiatus dur.
  • 3rd quartile hiatus dur.

• Make capitalization consistent (Trait value -> Trait Value, Time domain -> Time Domain etc.)
• Make axis labels consistent (e.g. in fig. 4 parts are "Stratigraphic Height [m]", others just "Height [m]")
• Reduce font size of titles to avoid cut-off of titles (e.g. fig. 4 bottom right)
• Shift letters indicating sub-plots a bit to the right to reduce ambiguity
• fig 6 & related figures:
  • move km from ticks labels to axis labels
  • reduce line width on box plots
  • make y axis consistent from 0 to 1
  • move legend out of plot windows on the right
  • adjust capitalization of titles

Add dashed line to figures 4,5,and 6 at AIC = 0.9 to indicate good model fit