open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi

I'm trying to run a 3D model with temperature and salinity. The model runs fine with just hydrodynamics but problems occur with the temperature and salinity tracers.

I have a model with an open sea domain and three river inputs. I've set the initial conditions with warm (27.5C) salty (35.5) water with colder (15C) fresh (0.5) water coming in the 3 rivers. As I don't have proper spatially varying temperature and salinity conditions, the model is set with constant values and will run the model sufficiently long for stratification to develop for the initial conditions for a later run.

However, if I prescribe just tracers coming in the three rivers I also suddenly get fresh cold water coming in the open sea boundary (even though not coded to do that) The prescribed value for the open sea boundary is that of the initial conditions. I use the tracer code TBOR of 2 for closed boundary and 5 for prescribed in the .cli file. If I code the open sea with either a free tracer (4) or prescribed (5) I fix the open sea boundary but I no longer get flow through the rivers, the tracer value only appears on the boundary nodes.

If I run with just one tracer, say salinity, the model runs fine.

Any suggestions would be greatly appreciated.

Regards

David

Hello,

Normally you should set prescribed (5) for tracers from the rivers and also 5 from tracers at the open boundary in the sea, giving the value that is in the open sea. Then the program will change itself the 5 into 4 when the flow goes out to the sea, discarding the prescribed value.

This is how it should work, so if you have a problem with 2 tracers and not with one we would need the case to see if there is an interference with TPXO or another problem.

With best regards,

Jean-Michel Hervouet

Thank you for you response Jean-Michel.

Yes that is what I thought. I have tried a number of combinations of options, numerical schemes etc but I still have problems. I was incorrect in thinking it worked with one tracer and not 2.

I have the boundary file set as you suggested, but I find that there is no tracer transport into the river flows. The value is prescribed at the boundary nodes but does not enter the model domain (see image 1). I can see that the flow is coming into the domain but the tracer value does not change from the initial condition (see image 2). I notice in the the output that there is no calculation of diffusion of the tracer:

---

ADVECTION-DIFFUSION OF TRACERS
MURD3D_POS OPTION: 14 9 ITERATIONS
DIFFUSION OF TRN1
GRACJG (BIEF) : 0 ITERATIONS, RELATIVE PRECISION: 0.7445018E-07
MURD3D_POS OPTION: 14 9 ITERATIONS
DIFFUSION OF TRN2
GRACJG (BIEF) : 0 ITERATIONS, RELATIVE PRECISION: 0.1614900E-06

---

Bizarrely, if I set the initial condition for the whole domain slightly lower than the prescribed tracer value (salinity of 30 instead of 35) than I do get tracer flowing into the model. However, i get unrealistic salinity driven currents on the open boundary, so I'd rather not try and trick Telemac into doing what I need!

All comments and suggestions welcome.