Suggestions for community development activities

[rosiealice]

This forum is hosted on GitHub as it has many features that are desirable (open website without signup, taggable discussion threads) and everyone using CLM is by definition already a GitHub user. we can consider whether other platforms (slack, reddit, discord) might also prove useful. Please add thoughts here.

[EvaRiPo]

I'm new to CLM, so I'm not 100% sure if these suggestions are good, but for now, I think this would help me progress learning CLM:

1. A tutorial guide for people outside of the US, or at least an update to the 2019 and 2022 tutorials to fit in place of the Cheyenne machine.
2. Another thing that would be very useful could be a theses repository (master's and doctorate) that use CLM, traditional theses. Because they summarize the background and guide how the studies were conducted. When I read the newspapers, they assumed that I know how to use CLM.
3. Another idea is to have a summarizing table contains the main important process that you can model and the available composets for each case, the data type you will need, the variables and main equations, time necessary for the spin up, and so on. I understand that this is very difficult due to the flexibility of the CLM, but at least the most important ones, to have a big picture. For me the main problem is that there is too much information and I get lost reading.

[yelizy]

Hi Eva, @EvaRiPo

It might be a steep learning curve to keep up with CLM. Here in Norway, we have several research projects and courses require the use of CLM which might be a real struggle for the new PhD students. In the last years, we tried to document our CLM experiences (some examples might be outdated). They are similar to the tutorials from NCAR but on Norwegian machines. We also have a big effort to create a modeling platform which allows the users to run CLM on a laptop or a virtual machine via a user interface and hopefully 'simplified' software infrastructure. I personally think that it has a great documentation which is mainly written by another Eva @evalieungh :) I hope this is helpful for you.

Creating a thesis repo sounds like a good idea. I think many people would benefit from this.

[rosiealice]

Hi @yeliz, thanks for all these links. Maybe a discussion of the platform could be a good topic for a future call?

[yelizy]

Thanks @rosiealice ! I think this is a good idea. Perhaps @evalieungh and @lasseke would also be interested in further discussions about the platform.

[lasseke]

Hi @EvaRiPo! If you are struggling with the technicalities and data formats, the Land Sites Platform (which uses FATES as the vegetation model) could indeed be a good place to start! As mentioned, it provides fully set-up single-site use cases that you can run on your laptop. It also includes some starting points for visualizing the inputs and outputs, mainly focusing on alleviating the technicalities around these things (working with NetCDF files, etc.). Just be aware that the user interface, API, and Docker containerization add an extra layer of complexity when your goal is to work with the model on a deeper level - but you could still use the code in the repo as a starting point.

I fully agree with your suggestions in 3. - that's mostly stuff we don't yet cover in the platform and it would be very useful!