search

plantphys / gsti

A project focused on the development of generalized spectra-trait models for the prediction of leaf photosynthetic capacity. This includes models focused on the prediction of Vcmax, Jmax and dark respiration as well as leaf nitrogen, leaf mass per area (LMA), leaf water content (LWC).
GNU General Public License v3.0
8 stars 2 forks source link

Create new 1pt fit option to allow the inclusion of 1pt Vcmax as evaluation data #34

Closed serbinsh closed 1 year ago

serbinsh commented 2 years ago

Long term goal

JulienLamour commented 2 years ago

I have added the function to do that, f.fit_One_Point() in fit_Vcmax.R. I havent tested it yet since I dont have one point Vcmax data. If you have a dataset with one point data and reflectance please add it and I ll try

serbinsh commented 2 years ago

There should be an open Arctic dataset from Utqiagvik, do you remember which one @regnans ?

JulienLamour commented 1 year ago

Can you add your one point black rock dataset Ken? In the Davidson_et_al_2022 folder. You can try on your own or send it to me. As you prefer @DavidsonKen. There is a Davidson_et_al_2022 folder for it

regnans commented 1 year ago

I'm late to the party here (sorry), but I don't think we have paired 1-point and reflectance data from the Arctic. We have some 1-point data from 2018 here, https://doi.org/10.5440/1506965 and reflectance data from the same species, but from earlier years here https://doi.org/10.5440/1437044

serbinsh commented 1 year ago

@regnans @JulienLamour yes we do have 1pt and spectra - this effort is based on my previous work linking all of the spec+trait datasets together from alaska here https://github.com/serbinsh/SSerbin_NGEEArctic_Spectra_Trait

I dont recall now why I asked you that and if I was wondering if it had migrated to the portal as I was accessing the data from my local copy.

The connections between Arctic datasets can be found in my curate data R scripts in here: https://github.com/serbinsh/SSerbin_NGEEArctic_Spectra_Trait/tree/main/R_scripts/Curate_datasets

2016: https://github.com/serbinsh/SSerbin_NGEEArctic_Spectra_Trait/blob/d34954e47d5542152960eabedd9191b775eac092/R_scripts/Curate_datasets/Curate_leaf_gasex_datasets.R#L177

DavidsonKen commented 1 year ago

@JulienLamour Okay working on uploading the BRF data now. I will reach out via email if I run into any issues and will let you know here once the data are on GitHub

regnans commented 1 year ago

@serbinsh I guess I misunderstood what you were asking for. We do have paired spectra with full ACi curves. I thought you were referring to the specific "1 point method" measurements we made, for which we do not have paired spectra.

DavidsonKen commented 1 year ago

@JulienLamour BRF data is now in the "Davidson_etal_2023" folder. Will have to update things once I have an article and DOI for the data (internal at EDI right now). Smooth sailing all around, your 1pt code worked well. One thing you could add is a way to specify the value for Rday if available, rather than assuming it is 0.015*Vcamx. But then again you may want to stick with a common assumption all 1pt fittings to keep in consistent.

Also if you are going to fully implement 1pt, you may want a separate processing pipeline (e.g., no need for QAQC curve plots). Happy to work on documentation for that if you want

JulienLamour commented 1 year ago

@DavidsonKen Hi Ken, thanks! I made small changes to your code so it is similar to the others (changed dataset folder name, removed the QAQC as it was not necessary as you mentioned, and corrected the std error in the fit one point function that was wrong). Thanks for putting the site info even if they were a discrepancy between the doc and the datasets!! I will work on that today.