Skip to content

cnmodel - #338

Open
stineb wants to merge 190 commits into
masterfrom
cnmodel
Open

cnmodel#338
stineb wants to merge 190 commits into
masterfrom
cnmodel

Conversation

@stineb

@stineb stineb commented Aug 12, 2026

Copy link
Copy Markdown
Collaborator

cnmodel branch is currently up to date with main and contains:

  • full coupled C and N cycling (see Stocker & Prentice, 2024 biorxiv)
  • started implementation of latin hypercube sampling and posterior weighing for calibration

stineb and others added 30 commits December 26, 2021 22:00
@stineb
implemented soil temperature
3db883b
@stineb
added more output variables for cnmodel
51f599c
@stineb
first successful compilation with npp implemented
b1c9087
@stineb
nothing really
6b5b0d2
@stineb
added file
a93e795
stineb
implemented nuptake, made it compile
d33fbbe
stineb
starting with turnover
1cbdf42
stineb
added file
9ca4192
stineb
implemented turnover, compiles ok.
70c95e0
@stineb
implemented grharvest. compiles ok.
72a8829
@stineb
work in progress on littersom
df4f721
@stineb
implemented littersom and got it to compile
e70a678
@stineb
work in progress on ntransform
4b17860
@stineb
work in progress on cnmodel branch
9f8f88c
@stineb
made it compile with ntransform
2185cce
@stineb
successfully compiling after implementing allocation
b757d76
@stineb
implemented long list of parameters
d3a7aa2
@stineb
implemented phenology-turnover-vegdynamics link
8cd1c2f
@stineb
first successful run (don't actually now if it's doing anything sensi…
85d5d5c 
…ble, but at least it runs).
@stineb
mass balance tests, several bugfixes
897cd41
@stineb
more bugfixes. now running an entire year, but with strongly increasi…
1bcda2b 
…ng LAI and plateauing leaf C - weird.
@stineb
first kinda-succussful trial run, ignoring C balance tests and severa…
cbe0337 
…l debug simplifications
@stineb
added npp to output
2799bd9
@stineb
still struggling with N balance from littersom
5b4dc6c
@stineb
got it running meaningfully, now looking into grain filling
5869149
@stineb
fuckup
28247e3
@stineb
nasdfasldfjasl;kjdfakl;sjdf;lkajf;lasjdf;klasjfdk;lasfjl;
be967dc
@stineb
shitfuck
5db4f1e
@stineb
Revert "nasdfasldfjasl;kjdfakl;sjdf;lkajf;lasjdf;klasjfdk;lasfjl;"
af65fd6 
This reverts commit be967dc.
@stineb
reverted fuckup
af720f9
stineb added 11 commits April 25, 2024 18:54
@stineb
added N allocation and resorption to output
a028513
@stineb
avoided writing soil c spinup output to subdirectory. writing to pwd …
a7b065a 
…instead.
@stineb
hard-coding soil moisture stress
92f9813
@stineb
nothing
d624148
@stineb
Merge pull request #331 from stineb/cnmodel
bfb539e 
Cnmodel
@stineb
added example workflow for C-only model run
9656ef1
@stineb
demo for C-model and evaluation, start of fAPAR intervention
fbc3575
@stineb
MAJOR: merged main into cnmodel and resolved all merge conflicts. WAR…
378a4de 
…NING: so far tested only for a C-only run.
@stineb
Added LAI data download for CH-Oe1 and evaluation in vignette example…
6bd495a 
…_cmodel.Rmd.
@stineb
implemented calibration for CN-model using latin hypercube sampling, …
6e137f3 
…parallel runs, and posterior weighing.
@stineb
Implemented calibration of delayed-start temperature modifier (cold-h…
761818d 
…ardening).
@stineb

stineb commented Aug 12, 2026

Copy link
Copy Markdown
Collaborator Author

(A)I started implementing something based on latin hypercube sampling (LHS) and committed. vignettes/example_cmodel.Rmd contains a demo but the posterior weights are too unevenly distributed across parameter combination samples. Potentially part of the solution but also other open points are:

  • Drawing more than 1000 samples (eventually need to run this on the cluster with large CPU nodes)
  • Some iteration to the search, densening sampling near optimum
  • Solution to the large autocorrelation of errors in fluxes
  • Come up with a step-wise calibration (P-model only, then C-model, then CN-model) taking posterior of one step as prior of the subsequent step.
  • Multiple targets calibration also for LHS

This is now in cnmodel branch.

@fabern fabern mentioned this pull request Aug 12, 2026
Closed
5 tasks
stineb and others added 17 commits August 17, 2026 12:30
@stineb
Refined but not consolidated calibration implementation. Now limited …
04eeb37 
…functionality due to Fortran-driven STOP behaviour.
@stineb
unbalanced posterior weights for calibration of daily data
9bfa3a5
@stineb
Made annual soil C and N text file output optional (defaults to FALSE…
e2d480b 
…); implemented calibration to multiple GPP time series features in calibration demo vignette (calibration_cmodel_ch_oe1.Rmd), unresolved: calibration of error for each feature.
@stineb
separated two vignettes for calibration demo
b02d21d
@stineb
multiple-target calibration demo implemented but unresolved: poor res…
1a61e0f 
…ult for LAI
@stineb
added info about the approach and its relation to the published liter…
bdf4e17 
…ature
@MilanFischer
Use prescribed fAPAR in CN-model GPP
e98fcfa
@MilanFischer
Add prescribed LAI to CN-model forcing
1ad8bda
@MilanFischer
Add inverse LAI to leaf carbon mapping
3d9f697
@MilanFischer
Add resource-constrained prescribed LAI
eb2ca12
@MilanFischer
Make prescribed fAPAR and LAI configurable
ae6c5bb
@MilanFischer
Make CN-model output safe for parallel execution
4cff458
@MilanFischer
Fix CN-model interface argument order after rebase
eea3f51
@MilanFischer
Restore soil diagnostic file error handler
afb7348
@stineb
Merge pull request #340 from MilanFischer/spatial-cnmodel
b678596 
Add prescribed fAPAR and LAI support to the CN-model
Milan Fischer
Reduce cluster export overhead in calib_sofun_lhs
b5faaa0
@stineb
Merge pull request #341 from MilanFischer/cnmodel
d54ab99 
Reduce cluster export overhead in calib_sofun_lhs
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@stineb @MilanFischer