Repository navigation
PFT/landuse/urban tools: availability, documentation, training #2206
Replies: 26 comments 18 replies
|
@lawrencepj1 shared with us by email the latest pdf of his documentation. I'm open to uploading here if you all think it makes sense. |
|
Thanks @slevis-lmwg yes please upload and make available as a work in progress |
|
Meeting 2023/11/16 @lawrencepj1 @ekluzek @mvdebolskiy @slevis-lmwg Peter points out:
So Peter proposes standardizing steps 2 and 3 initially. Peter is organizing step 2 on the cgd machines here: Peter is organizing step 3 on cheyenne and needs to move stuff to /campaign before cheyenne goes away. This is a sample namelist used by the C-program (each directory here has a diff. namelist): For /glade/p data we decided (Erik, Peter, and Sam L will meet Dec 6 to do the following):
|
|
Updates: |
|
README from Land Use Data Prep CTSM52 Land Use Data Prep Workflow This is step two in the bigger picture of generating CTSM raw PFT/CFT data.
Use tools/runcreateluh2modisearthstatextraptreerawfiles CTSM52_MODISCurrent CTSM52_POTVEGMax 2005
Generate the Descriptor files in rawdata: Generate the Descriptor files in surfdata:
Generate the Descriptor files in rawdata: Generate the Descriptor files in surfdata:
In surfdata: |
|
Moving tools from Peter to ESCOMP:
Peter will present on CESM version for CCIS with datasets created with the previous version of mksurf tools.
Newer versions of landuse tools are not on github. (for scenarios). |
|
Notes from 17th May 2024:
|
|
Notes 18th September 2024
|
|
October Agenda: |
|
Descriptor files tool have been moved from |
|
Updated README file with all tools using namelist files rather than command line arguments. For @slevis-lmwg @ekluzek @mvdebolskiy CTSM52 Land Data Prep Workflow This is step two in the bigger picture of generating CTSM raw PFT/CFT data. Historical Descriptor File Generation
tools/runcreateworkclm5extrapallvcfpftrawdata worknamelist/CTSM52_MODISCURRENTEXTRAPVCFPFT_generateallrawdata.txt Check Extrapolation netcdf files in worksurfdata for PFT composition (TREE_EXTRAP,HERB_EXTRAP,GRASS_EXTRAP) tools/runcreateworkclm5surfdata worknamelist/CTSM52_MODISCURRENTEXTRAPTREEVCFPFT_generatesurfdata.txt
Generate the Descriptor files in rawdata: Generate the Descriptor files in surfdata:
Generate the Descriptor files in rawdata: Generate the Descriptor files in surfdata:
tools/runcreateluh2descpftlaisurfdata luh2descnamelist/CTSM52_LUH2HISTDESCPFTLAIS_generatedata.txt Future SSP Time Series Descriptor File Generation - Assumes that Historical Descriptor Files have been generated.
|
|
Meet with @slevis-lmwg @lawrencepj1 and @wwieder. Mostly confirming the CMIP7 process for ctsm5.4 and CESM3.0 release.
Non-CMIP7 work will be outside of the standard workflow. On a branch or in a separate project. Reforestation, or bioenergy work for example. |
|
Meeting with @lawrencepj1 @ekluzek 2025/2/28 When the new version becomes visible in /project/ccis/lawrence (today it's not), Peter will put it on github. Today's process... On a cgd machine Peter made a demo directory with an older version because the new is not visible right now: cd /project/tss/lawrence/ctsm52landusedataprepdemo/ctsm52landusedata025
# In the future we will obtain the above with git clone of the dataprep tool. Then...
cd bin
make
# Next follow the instructions in the README (this part is step 2).
# Now copy the .nc outputs to /glade and work on derecho (this part is step 3).
# The user would git clone the landuse data tool now.
cd /glade/campaign/cesm/development/lmwg/landuse_source_tools
# Notice here /MksrfDataTools is the old mksurfdata_map refurbished to work on derecho
cd CTSM52LandUseDataTool/bin
make
cd ../../../landuse_source_data/CTSM53RawData/globalctsm53histMKSRFDeg025_demo
/glade/campaign/cesm/development/lmwg/landuse_source_tools/CTSM52LandUseDataTool/bin/ctsm52landusedatatool globalctsm52histTRENDY2023namelist.txt # <-- the namelist is the only file you need to update, then you may want to run with qcmd --
# When cloning, the user will get a demo namelist.txt file and demo look-up tables
# Next is step 4, i.e. the mksurfdata_esmf (or mksurfdata_map) step to generate fsurdat/landuse filesNext meeting March 28 to do all this from github and to discuss CMIP 7 (see invite). |
|
@lawrencepj1 will add new tools to github, for biofuels and reforestation/deforestation. We will put them in ESCOMP, but not fully support them like CTSM. We will meet with @olyson to talk about dynamic urban datasets. Will meet in August for this. We discussed it and want to make sure we keep this capability as Keith continues to phase into retirement. Dynamic lakes -- we thought about it, but don't have a plan for it. SSP datasets for CMIP7 will be coming out as a ctsm6.1/cesm3.1 @lawrencepj1 will create CMIP6 SSP scenario data extensions to CMIP7 historical harmonized with LUH3 and year 2024. This will be available for people to use. And a way to show that it will work for the SSP version update. So a non-supported option that people can use by swapping in those datasets for the CMIP6 SSP datasets that WILL be in the release. TRENDY -- we will verify with @adrifoster that we will create datasets at hcru_hcru resolution. @lawrencepj1 is creating the rawdata that @slevis-lmwg will use in mksurfdata_esmf for the datasets needed. This needs to be 1700, and 1700-2024 transient. |
|
The one issue for the release is #3300 to make sure it's in the CTSM checkout. |
|
Land Use and Urban Data Tools Meeting. 08/06/2025 Peter, Sam L., and Erik Land Use Data Tools: Peter has migrated the CLM5 Land Data Tools from CGD /project/ccis to: The tools have individual projects that progress through the workflow:
Each data directory has a set of tools:
Current projects include cmip6, smoothed, ccis, trendy2025. Keith is very interested in joining the next meeting to discuss the inclusion of urban tools which have been developed with CLM5 through CTSM 5.4. Today also made progress in #3300. |
|
@mvdebolskiy we were just meeting, and we wondered the status of the conversion of the tech note from word to RST? I know you worked on this a long time ago and maybe didn't finish. And I know you are also busy with the NorESM release. But, we just wanted to know where you were at with it, so we can plan on what to do next. If you converted some of it, and developed a method to do the conversion we could take up doing that. |
|
We had a shorter meeting today. @olyson is out today. So will meet next week. We talked about the documentation. And had the idea of the User's Guide being a Jupyter notebook. This might be an easy way to get it started, and also be useful for users to demonstrate some of the code. The downside is that Notebooks need to stay up to date with the code, and need conda environments and such to stay working. So there is work in maintaining Notebooks and making sure users can use them on any platform (container to handle this?). Or we could just say -- it's only supported on NSF-NCAR JupyterHub? Could we just use the containers that @briandobbins made for the CESM Tutorial for this? Brian also supported the Tutorials for CTSM and tower sites. Also notes that Myst markdown is good for papers and documents. It was used in the SEA ISS conference for the conference proceedings. It has good support for references and figures in formal papers. |
|
Meeting 2025/9/10 with @olyson @lawrencepj1 @wwieder @ekluzek @slevis-lmwg Urban tools status CLM5 CTSM5.3 CTSM5.4 |
|
Erik wrote 2026/4/1, and I added from meeting with Peter (on the same day):
Peter: CMIP7 future scenarios now available:
|
|
Met today. Talked about biofuels needed for future scenarios. Will need to add a namelist item or change to the parameter file that doesn't collapse it to tropical corn. And add datasets and testing for it. I'll make an issue for this. I had thought we had a testmod that exercised this with a surface dataset -- but we don't. I didn't see it in previous versions either. We also noted that it would be good to have the tech note for LUH done by the time the release branch is done. But, we should shoot for having it done by the CTSM tutorial in November 16th. We also noted that #3300 should be done, but doesn't need to be done until the release branch as well. |
|
Met today with @lawrencepj1 and @slevis-lmwg. We mostly talked @slevis-lmwg using Claude to switch the CLM-U from Word to RST. A tricky thing was changing the math equation format so that the conversion would work. A note on biofuels @lawrencepj1 has it working in cesm2.1 so ported to clm5.0.X and those changes can be brought into the latest code. The addition beyond the current code is for handling the pools which is needed when you couple to CAM for CO2. |
|
Thanks @slevis-lmwg and @ekluzek I appreciate the assistance and guidance Sam. If I get into trouble I will reconsider your offer. Best |
|
@lawrencepj1 @slevis-lmwg and I met. The plan for next meeting is to start on the Word to RST conversion during the meeting. We talked about a new idea to distinquish between bare ground that can allow vegetation to grow and bareground that is impervious or rocky and doesn't allow vegetation to grow. It seems the best way to handle it would be to add a "rock" landunit type that has depth to bedrock of zero. It's difficult to distinquish it with Satelitte data products, but there are things that could be used. The soil texture dataset might have this for example. There's also somewhat of a "chicken or egg" problem in that for example above treeline it's rocky, but part of that is because vegetation can't survive there in order to break down the rock into soil. |
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
TODOs as of Feb 2024.
mksrurfdata_esmfto explain what is expected to be in the mksruf_ files for landuse.@lawrencepj1 @mvdebolskiy @ekluzek
All reactions