model gets stuck after 1.5 year.

Question

Hi Chris,

I am getting into problems when I try simulating runs longer than one year. The model slows down and gets stuck after some time (generally after 1.5 years), and I get a strange log message (see below). This occurs both on the local workstation and cluster.

Step:   10642/3008448 SimDate: 2003-12-30 03:10:00 (2003364.1319)  Time: 16:35:42

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

Step:   10643/3008448 SimDate: 2003-12-30 03:13:20 (2003364.1343)  Time: 16:35:42

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000    

 coeffs   3.750000       20.00000

Answer 1

Hi Behnam,

The log message is just an accidentally left in debugging printout of the blueice_vis_ext_k and blueice_ir_ext_k values.

Does the model crash or is it still using CPU?

Comments

Thanks for your reply. The model doesn't crash and still keeps using CPU as it should. But the model runs extremely slower than earlier time steps. The first 5 months run in 4-5 hours, but to get to the 9th month it takes more than 1.5 day and then it seems to get stuck at some unknown point
The lake level doesn't change and the lake is fed by a constant groundwater in/ouflow without any surface flow

---

Hi Chris,

The long simulation and getting stuck is exactly because of the ice model. I just turned off the ice model and everything runs smoothly and fast, without getting stuck or hanging both on Windows and Linux (cluster). On Windows the ice model makes also the HydrohubLite hang.