CFL ERROR with ERA-5 dataset with increased e_vert

[Anie_Lal]

Hi all,

I am getting cfl error towards the end of a 5-day simulation with WRF-V4.2.2 when using ERA-5 0.25 IC/BC. But the same works for FNL-0.25 data set.

I tried the following:

1. dt = 12, 10, 8 (D1=6km and D2 =2km)
2. w_damping = 1,
3. smooth_cg_topo = .true.,
4. epssm = 0.2, 0.5, 0.9,

I successfully gave run (say EXP01) with ERA-5. But in EXP02 its throwing error continuously.

The only difference between EXP01 and EXP02 is the eta_levels:

eta_levels for EXP01:
eta_levels = 1.000, 0.999, 0.998, 0.997, 0.996,
0.995, 0.994, 0.993, 0.992, 0.991,
0.990, 0.989, 0.988, 0.987, 0.986,
0.983, 0.980, 0.977, 0.974, 0.971,
0.965, 0.958, 0.951, 0.944, 0.936,
0.929, 0.922, 0.910, 0.899, 0.880,
0.865, 0.807, 0.765, 0.719, 0.672,
0.622, 0.571, 0.520, 0.468, 0.420,
0.376, 0.335, 0.298, 0.263, 0.231,
0.202, 0.175, 0.150, 0.127, 0.106,
0.088, 0.055, 0.040, 0.020, 0.000
eta_levels for EXP02:
eta_levels = 1.0000, 0.9994, 0.9986, 0.9975, 0.9962,
0.9951, 0.9944, 0.9932, 0.9922, 0.9913,
0.9905, 0.9892, 0.9885, 0.9875, 0.9861,
0.9835, 0.9806, 0.9774, 0.9744, 0.9715,
0.9655, 0.9585, 0.9515, 0.9445, 0.9365,
0.9295, 0.9225, 0.9105, 0.8995, 0.8805,
0.8655, 0.8075, 0.7655, 0.7195, 0.6725,
0.6225, 0.5715, 0.5205, 0.4685, 0.4205,
0.3765, 0.3355, 0.2985, 0.2635, 0.2315,
0.2025, 0.1755, 0.1505, 0.1275, 0.1065,
0.0885, 0.0555, 0.0405, 0.0205, 0.0000

The model is running fine till 4rth day and gives either cfl error or segmentation at the same time step every time. Please find the namelist and rsl error file attached.
I understand this is due to model instability. But unsure of how to solve this. Kindly help me resolve this issue.

Looking forward to any kind of help/advice.
Link to rsl files:

Smash

fromsmash.com

Thank you

 

[kwerner]

Hi,
When you say the only difference in EXP01 and EXP02 is the eta_levels, does that mean you are simulating the same exact dates, domain, using the same input data, etc? Or is this for different dates and/or domain?

 

[Anie_Lal]

Yes for the exact same dates, domain, physics and input.

 

[Anie_Lal]

Hi,
I tried with different spin ups also, but getting the same error (cfl / segmentation) on the same day. Please help me solve this. Is there anything that I am missing here.
Looking forward to any kind of help.

Thank you

 

[kwerner]

If everything is the same except the eta_levels, then the problem is with the eta_levels you've provided in the failing simulation. Because the model provides reasonable eta_levels without users needing to specify them, we typically recommend using the default values; however, since your other simulation works properly with your assigned levels, I would advise using that set, instead of the ones that aren't working. In case it's useful, take a look at this FAQ that discusses vertical levels.

 

[Anie_Lal]

Okay. Thanks