TELEMAC-MASCARET Forum: TRACER LOSS FROM MY DOMAIN (1/1)

Welcome, Guest

TOPIC: TRACER LOSS FROM MY DOMAIN

TRACER LOSS FROM MY DOMAIN 11 years 6 months ago #8478

Comite	Hi all, i have steady hydrodinamic state and i start a tracer trasport simulation. I don't have tracer decay but i have tracer loss from my domain. Why? How can i correct it? Thanks. Best regards. Massi
	The administrator has disabled public write access.

TRACER LOSS FROM MY DOMAIN 11 years 6 months ago #8480

jmhervouet

Hello,

Causes for tracer loss in 2D may be :

CONTINUITY CORRECTION :

must be set to yes, for exits with prescribed depth (in such points continuity is replaced by imposing the prescribed values)

TREATMENT OF FLUXES AT THE BOUNDARIES

it must be 2;2;2... etc. for all boundaries, otherwise the correct imposed valu is set, not the correct flux.

TYPE OF ADVECTION

like 1;5;3

the third number is the advection scheme for the tracers. Conservative schemes are 3,4 and 5 (if no tidal flats) or 13,14 (if tidal flats)

13 or 14 will only work with mass-lumping and TREATMENT OF NEGATIVE DEPTHS = 2

Last cause : the diffusion, which is not conservative if we use

OPTION FOR THE DIFFUSION OF TRACERS : 1 (default, but recommended as option 2 has a division by depth that will crash on tidal flats). Try :

COEFFICIENT FOR DIFFUSION OF TRACERS : 0.D0 to see the influence of this point.

All this should bring some improvement,

With best regards,

Jean-Michel Hervouet

The administrator has disabled public write access.

TRACER LOSS FROM MY DOMAIN 11 years 6 months ago #8487

Comite	Hi Jean-Michel, thanks for the precision. I tried and only change that improve the situation in my case is TYPE OF ADVECTION with 14, but without mass-lumping and TREATMENT OF NEGATIVE DEPTHS = 2, but with OPTION FOR THE TREATMENT OF TIDAL FLATS = 2. Is it correct or is the computation wrong in something? Best regards. Massi
	The administrator has disabled public write access.

TRACER LOSS FROM MY DOMAIN 11 years 6 months ago #8492

jmhervouet

Hello,

The tracer mass-lumping is copied from the mass-lumping on depth, this is a condition to have a mass conservation proof, so you are with mass-lumping.

OPTION FOR THE TREATMENT OF TIDAL FLATS = 2

is not really mass-conservative (it is only in steady state, because the adding and removing of elements creates errors of mass conservation).

Actually scheme 14 really needs positive depths so I would prefer :

OPTION FOR THE TREATMENT OF TIDAL FLATS = 1
TREATMENT OF NEGATIVE DEPTHS = 2

With best regards,

Jean-Michel Hervouet

The administrator has disabled public write access.