Turbulent Structures

Topics about the HYSPLIT dispersion model.
Post Reply
Posts: 24
Joined: June 22nd, 2020, 3:05 pm
Registered HYSPLIT User: Yes

Turbulent Structures

Post by kjmohr »


I was wondering if anyone knows how HySLIT generates smaller resolution structures within the Plumes when the meteodata file's resolution is much higher.
Example: If I am using ERA5 meteodata at 27 km grid resolution and I run HySPLIT with a 1 km resolution, it appears small structures form within the deposit and the cloud. As I continue to increase the simulated resolution up to 100 meters I continue to see more of these smaller structures.

I was wondering if this is an artifact of HySPLIT running turbulent calculations with the meteodata (currently running the default turbulent regime, Kanthar-Clayson)? I am trying to clarify that the structures I am seeing within the deposit and cloud are "true" results and not an artifact due to calculations.

Here are two images with what I am asking. The first image is at 1 km resolution and the second is at 500 m resolution. The deposit begins to show higher concentration pockets within the deposit, we were wondering if these could be due to Kelvin-Helmholtz instabilities, but we think that the resolution is too coarse to actually see those structures with a 27 km grid resolution of the wind file.

These structures tend to show up in multiple meteodata sets including GDAS (which is what was used for the attached images).

Any information on this would be greatly appreciated.


500 meter res
500 meter res
1 km res
1 km res
Posts: 284
Joined: February 8th, 2016, 12:56 pm
Registered HYSPLIT User: Yes

Re: Turbulent Structures

Post by alicec »

I'm fairly sure what you are seeing is just an artifact. I think what you are seeing is that the release of the mass is actually not continuous.
The model can only create new particles at a time step. So the particles are being released at every time step, but you are in a situation where there is not a lot of turbulent mixing and the wind speed is high enough that the releases at the different time steps remain separated in space. You see this separation when you define a very fine concentration grid.

Look at your MESSAGE file and see what kind of time steps the model is using during your release period. If the time steps are large, then you can reduce them by setting DELT in the SETUP.CFG file. The smallest time step that you can set is 1 minute (DELT=1). See how that goes.

Also look at the output of the particle output file, PARDUMP. That will make it fairly obvious if you are getting distinct lines of particles.

Let me know if this helps. And if you have more questions please attach your CONTROL and SETUP.CFG files.
Post Reply