I’m simulating online particle trajectories with the ORCA2 configuration, and I run into problems when working with large particle numbers. My aim is to simulate particle trajectories for 250,000 particles over a period of 5 years. The simulation runs fine with 1 particle up to 5 years. It also runs for 5 years with 250,000 particles when they are all initiated at the same location. But when I run >100 particles released at individual locations, the computation freezes at a time step usually around 13-14 months (depending on the number of initiated particles). No error message is to be seen in any of the output files, and the simulation remains frozen until the wall time runs out and the job is cancelled.

I’m using the stable version 3.6, compiling with: makenemo -m GBAR -j1 -v3 -r ORCA2_LIM -n ORCA_FLOATS add_key "key_floats key_xios2"

I have plenty of disc-space, and there should be no errors in my namelist- and xml-files. I’m quite new to NEMO and ocean modeling, and have run out of ideas for how to manage this problem, so I would very much appreciate any advice!

Thanks,

Maria