I am trying to run a single-point NEON tower simulation using CTSM-Tutorial-2022 and specifically using Day0d_NEON_fromScratch.ipynb. I have access to Cheyenne but I couldn't locate the CTSM source code in any of the directories in Cheyenne. I tried the ucar jupyter hub and cloned the https://github.com/wwieder/CTSM-Tutorial-2022 but this doesn't contain any (/CTSM/cime/scripts). How can I proceed then? Please help me in this regard.
How to Locate CTSM Source Code in Cheyenne
- Thread starter Makduma Badhan
- Start date
You need to do the "homework" steps of the Tutorial. Which are found here...
github.com
CTSM-Tutorial-2022/Day0a_GitStarted.ipynb at newTutorials · wwieder/CTSM-Tutorial-2022
Materials for CTSM Mini Tutorial 2022. Contribute to wwieder/CTSM-Tutorial-2022 development by creating an account on GitHub.
github.com
Hi @erik.Thanks for the reply. I could run the single-point simulation in Cheyenne using the tutorial.
Hi Oleson,
Thanks for your quick response. I could run the CLM in Cheyenne. Now I am trying to run a single-point simulation for Harvard Forest Site (NEON site). I want to see how the changes in temperature and precipitation will affect the other variables (GPP, Nitrogen Deposition, Soil Moisture, etc.) of that NEON Site. For performing this:
1. Do I need to create my own atmospheric/meteorological forcing dataset or is there any script available in the model code where I can do some code modification? (If I am right otherwise correct me)
Say, for example, I want to see how 1 degree Celsius temperature increment in the HARV site will affect the soil moisture.
2. If I want to see a similar effect, but in this case I want to keep my meteorological forcing same as default, I want to change the vegetation type for the site, how can I perform this?
I am a new user of CLM, I read the documentation but didn't find anything I can understand at this point. Your suggestion and guidance will be really helpful.
(1) You'll need to develop your own scripts to modify the existing atmospheric forcing.
(2) You can change plant functional types (PFTs) in the surface dataset. This is the file pointed to by "fsurdat" in your lnd_in file. The PFTs are defined by the variable PCT_NAT_PFT in that file. The definitions of the PFTs can be found in this table in the technical note:
(2) You can change plant functional types (PFTs) in the surface dataset. This is the file pointed to by "fsurdat" in your lnd_in file. The PFTs are defined by the variable PCT_NAT_PFT in that file. The definitions of the PFTs can be found in this table in the technical note:
Hi Oleson,
Thanks. Your info was very helpful. I have some confusion regarding the modification of the PFT. Can you please clear those?
1. I had a look at my "fsurdat" file. I opened it with python and it shows me as follows:
- PCT_NAT_PFT
(natpft, lsmlat, lsmlon)
float64
...
array([[[ 0.]],
[[ 0.]],
[[ 0.]],
[[ 0.]],
[[ 0.]],
[[ 0.]],
[[ 0.]],
[[ 0.]],
[[ 0.]],
[[100.]],
[[ 0.]],
[[ 0.]],
[[ 0.]],
[[ 0.]],
[[ 0.]]])
- PCT_NAT_PFT
2. Do I need to change another component variable here?
3. If correct, how can I incorporate my modification in the model simulation next time? Any documentation/discussion forum link or tutorial will be helpful.
Thanks again!
1. Yes
2. No, unless you want to change other site characteristics (e.g., soil texture)
3. Point to your new surface dataset using fsurdat in your user_nl_clm in your new case directory. I would recommend watching some of the CESM Tutorial materials, in particular, the namelist and code modifications lecture:
www.cesm.ucar.edu
2. No, unless you want to change other site characteristics (e.g., soil texture)
3. Point to your new surface dataset using fsurdat in your user_nl_clm in your new case directory. I would recommend watching some of the CESM Tutorial materials, in particular, the namelist and code modifications lecture:
Tutorials | Community Earth System Model
www.cesm.ucar.edu
Hi @oleson,
Thanks again for your help in understanding the model better step by step. I have the following questions for stepping into the next level of my work. I am looking forward to your suggestions:
1. Regarding the modifying of PFT by editing 'fsurdat' data:
I downloaded the default 'fsurfdat' from Cheyenne and opened it in Python as I showed above. I changed the array in the .nc file from Broadleaf evergreen shrub [[100.]] to C3 arctic grass [[100.0]] using python and then uploaded it back to Cheyenne. I pointed the my new surface dataset using fsurdat in my user_nl_clm in my new case directory and ran the simulation. I am wondering if the modifying 'fsurfdat' was in the correct way?
2. I followed https://github.com/wwieder/CTSM-Tut...orials/notebooks/Day0d_NEON_fromScratch.ipynb before. But I can't access it now. Can you please provide me with the right link?
3. By default (I used the tutorial link mentioned in point 2), my simulation stored monthly history files for 2 years (2018-2019). I want to run this case for longer (10 years instead of 2), where should I make the changes? Can you please guide me on this?
Thanks again for your help in understanding the model better step by step. I have the following questions for stepping into the next level of my work. I am looking forward to your suggestions:
1. Regarding the modifying of PFT by editing 'fsurdat' data:
I downloaded the default 'fsurfdat' from Cheyenne and opened it in Python as I showed above. I changed the array in the .nc file from Broadleaf evergreen shrub [[100.]] to C3 arctic grass [[100.0]] using python and then uploaded it back to Cheyenne. I pointed the my new surface dataset using fsurdat in my user_nl_clm in my new case directory and ran the simulation. I am wondering if the modifying 'fsurfdat' was in the correct way?
2. I followed https://github.com/wwieder/CTSM-Tut...orials/notebooks/Day0d_NEON_fromScratch.ipynb before. But I can't access it now. Can you please provide me with the right link?
3. By default (I used the tutorial link mentioned in point 2), my simulation stored monthly history files for 2 years (2018-2019). I want to run this case for longer (10 years instead of 2), where should I make the changes? Can you please guide me on this?
1) Yes. I assume the model ran and you got the results you expected.
2) I'll ask someone about this.
3) I'm not sure if you can run this longer than 2 years, it would depend on whether atmospheric forcing data is available or whether you want to cycle back on the existing atmospheric forcing. To run for 10 years, ./xmlchange STOP_N=10 (this assumes that STOP_OPTION=nyears already).
2) I'll ask someone about this.
3) I'm not sure if you can run this longer than 2 years, it would depend on whether atmospheric forcing data is available or whether you want to cycle back on the existing atmospheric forcing. To run for 10 years, ./xmlchange STOP_N=10 (this assumes that STOP_OPTION=nyears already).
Hello, as I wrote in my email, I think the tutorial you were look at before has changed names
CTSM-Tutorial/NEON_cases_fromScratch.ipynb at NEON-singlePoint · NCAR/CTSM-Tutorial
If you clone the NEON-singlePoint branch locally (on cheyenne) you'll have your own working copy of the tutorial to work from and you don't have to worry about us changing things under you.
We can't run NEON cases beyond 2021, the last year when NEON provides input data for us. Soon we'll have data through 2022, and potentially to current day (with ~ 1 month lead time), but this work is still in progress.
We're also developing a tutorial to run with anomaly forcing out to 2100 under SSP scenarios. This tutorial is a work in progress, but you can check back on the CTSM-Tutorial github page for updates.
We'll be offering more tutorial throughout the summer, including at the ESA conference to be held in Portland in Aug.
Finally, in the future, please just write to this bulletin board. I realize you're under a deadline, but it makes more work and confusion for us when you reach out across multiple platforms and emails for help with the same question.
CTSM-Tutorial/NEON_cases_fromScratch.ipynb at NEON-singlePoint · NCAR/CTSM-Tutorial
If you clone the NEON-singlePoint branch locally (on cheyenne) you'll have your own working copy of the tutorial to work from and you don't have to worry about us changing things under you.
We can't run NEON cases beyond 2021, the last year when NEON provides input data for us. Soon we'll have data through 2022, and potentially to current day (with ~ 1 month lead time), but this work is still in progress.
We're also developing a tutorial to run with anomaly forcing out to 2100 under SSP scenarios. This tutorial is a work in progress, but you can check back on the CTSM-Tutorial github page for updates.
We'll be offering more tutorial throughout the summer, including at the ESA conference to be held in Portland in Aug.
Finally, in the future, please just write to this bulletin board. I realize you're under a deadline, but it makes more work and confusion for us when you reach out across multiple platforms and emails for help with the same question.
Thanks, @wwieder, for your suggestion and help regarding the issue. I will follow this bulletin board for the next steps and at the same time I am really sorry for the inconvenience caused by my multiple times inquires.