lenka_notes.md

Lenka's notes:

• 12 Nov: setting up for the "synoptic diffusion project" using the full dynamics model, plan:

  • held - suarez setup
  • held - suarez setup + damping of all waves (following walker and schneider 05)
  • held - suarez setup + damping of some wavenumbers

• 30 Nov: segmentation error when using vert diff up/down so switching to the full idealized Caltech FMS

• 14 Dec:

  • add explicit truncation using the input namelist
  • the wavenumber selection by divergence damping etc will have to wait (hs_syn+_diff_tr_old)

• 30 Jun:

  • running slab ocean wavenumber experiments
  • default environment messed up - use srun to trigger from compute node for now
  • num_fourier = (no four modes - 1); make sure \geq ntasks

• 5 Jul:

  • several analysis jobs failed - faulty node: hpc-23-12 (slurmd: execve error -> specify a nodelist,e.g. in slurm script, need to add #SBATCH nodelist=hpc-22-15)
  • the above fails are also suffering from analysis having the wrong permissions (chmod)