Documentation: Add Autosubmit examples to all use cases

Hi @abatalla @tkariyat,

Opening this issue to start updating the startR use cases to use Compute() + Autosubmit. Of course, you will need to wait until all the machines are up and running again.

The first step is for each of you to create an Autosubmit experiment to use in your examples. Each one of you should create one, then write down the experiment id (EXPID). You can create it on the bsceshub03 by running:

module load autosubmit
autosubmit expid -H nord4 -d <Description>

You only need to do it once, and you can reuse the same expid for the different use cases.

After this, you need to add the Autosubmit Compute() call to the example, based on the already-existing ecFlow calls, to reproduce the same results. Here is an example of a Compute() call using Autosubmit:

res <- Compute(wf, 
               chunks = list(lat = 2, lon = 2), # you can change this to the number used in the example
               threads_compute = 4, # you can change this to the number used in the example
               threads_load = 2, # you can change this to the number used in the example
               cluster = list(
                 queue_host = 'nord4',
                 expid = "a83l", # change to yours
                 hpc_user = "bsc032762", # change to yours
                 r_module = "R/4.1.2-foss-2019b",
                 CDO_module = "CDO/1.9.8-foss-2019b",
                 autosubmit_module = 'autosubmit/4.1.13-foss-2023b-Python-3.11.5',
                 cores_per_job = 4, # you can change this to the number of cores used in the example
                 job_wallclock = '01:00:00', # you can change this to the wallclock used in the example
                 max_jobs = 4, # you can change this to the number used in the example
                 special_setup = 'nord4'
               ),
               workflow_manager = 'autosubmit',
               autosubmit_suite_dir = "/home/Earth/vagudets/startR_local_autosubmit/", # Change to yours
               autosubmit_server = NULL,
               wait = TRUE # Change this to whatever is used in the example
               )

This is the list of use cases that need to be changed:

• 2.1. Function working on time dimension - Ariadna
• 2.2. Function using attributes of the data - Theertha
• 2.3. Use function CDORemap for interpolation - Ariadna
• 2.4. Use two functions in workflow - Theertha
• 2.5. RainFARM precipitation downscaling - Ariadna
• 2.6. Use external parameters in atomic function - Theertha
• 2.7. Calculate the ensemble-adjusted Continuous Ranked Probability Score (CRPS) - Ariadna
• 2.8. Use CSTools Calibration function - Theertha
• 2.9. Use a mask to apply different methods to different gridpoints - Ariadna
• 2.10. Apply an existing mask on data - Theertha
• 2.11. Two datasets with different length of target dimensions - Ariadna
• 2.12. Transform and chunk spatial dimensions - Theertha
• 2.13. Load irregular grid data and interpolate it in the workflow - Ariadna
• 2.14. Get margin dimension indices in startR workflow - Theertha
• 3.1. Weekly ECV Subseasonal Hindcast Verification - Ariadna

The use case files are found inside inst/doc/usecase/ in the repository.

Once the new startR release is on CRAN, we can start working on this and coordinate to create a branch for this development. If you have any questions, feel free to ask them here.

Cheers,

Victòria