ROMS particle trajectory setup

[lyzheng]

Hello all,

I am running ROMS particle trajectory. The particles are initially released near the bottom and I would like to limit particles move without change of sigma-coordinater. From the input file, the option "T" is used to controlled float trajectory type. When T=1, it is neutral density 3D Lagrangian particles (I assume this is a fully three-dimensional particle trajectory). When T=2, it is an isobaric (constant depth) floats. I don't really know what this means for T=2
case. For example, the particle is initinally released at 10 m below the surface, does it mean that the particle is limited at the depth of 10 m below the surface? If this is correct, how about when the particle moves to shallow coastal water with depth less than 10 m?

Your explanation is greatly appreciated. Thanks.

--lyzheng

[kate]

Having a new option to preserve the sigma value of a float would be really easy - just don't advect it in the vertical. You would have to look at step_floats to see how the advection is currently being handled and make changes there.

I'd be very surprised if the neutral density option was in there yet.

The particle that's fixed to be at 10m is not able to go into shallower water. If it is below the bottom, it can't get a valid velocity to advect it somewhere more sensible.

[lyzheng]

Kate,

Thanks for your reply. I check the step_floats.F. First I want to find where the variable "FLOAT_VWALK" (vertical random walk) is defined, but I can't find it in all ROMS3.0 package. Do you know where the variable "FLOAT_VWALK" is defined? Thanks!

lyzheng

[kate]

It's one of the cpp options. You define it wherever you define FLOATS.

[lyzheng]

Hi, Kate,

Ok, in my case, I don't define vertical random walk. At the input file of the float trajectory, we provide initial locations of real depth (meter) below the surface in vertical. I am not sure at ROMS float trajectory subroutine, it first converts this depth to s-coordinate then using vertical velocity omega to calculate vertical movement distance, or it first converts omega to real vertical velocity w, then calculating the net vertical displacement at physical plane. Thanks.

--lyzheng

[kate]

I once knew the float code, but things have changed since then. As far as I can tell, the units are:

Code: Select all

 track(izgrd,nfp1,l)=REAL(k,r8)+...

i.e., The izgrd value will be between 0 and N for the vertical component.