[MITgcm-support] noise near the bottom

[Yangxin He]

Hi there,


I am running a simple 2d oceanic model with surface tide with diurnal frequency propagating up and down a slope. There is noise near the bottom of the slope, especially the shelf break area. I have attached a snap shot of the noise near the shelf break, where the y-axis is the horizontal velocity right above the slope and x-axis is the number of points.

The noise seems unphysical(?), though its not too bad, I wonder is there any way I can do to remove the noise?


Here is my set-up:

My deep water depth H=2km and shallow water depth h=200m. I have 80 points in the shallow water region and 400 points in the whole water column. The horizontal resolution dx=46m. I am using pp81 for the vertical viscosity and diffusion and Smagorinsky for the horizontal viscosity.

I have tried increasing the resolution both x and z direction, reducing dt, smoothing out the slope more, increasing the viscosity etc. None of these will reduce the noise except reducing hfacMin from 0.1 to 0.01. My data file is attached.


# ====================

# | Model parameters |

# ====================

#

# Continuous equation parameters

 &PARM01

 tRef=0.,

 sRef= 0.,

 viscAz=1.E-5,

 viscAh=0.,

 no_slip_sides=.FALSE.,

 no_slip_bottom=.FALSE.,

 diffKhT=2.E-1,

 diffKzT=1.E-5,

#new parameters for smag

 viscC2Smag=2.0,

 f0=6.5E-5,

 beta=0.,

 eosType='LINEAR',

 tAlpha=2.E-4,

 sBeta =0.,

 rhonil=1028.0,

 rhoConst=1028.0,

 gravity=9.81,

 rigidLid=.TRUE.,

 implicitFreeSurface=.FALSE.,

# exactConserv=.TRUE.

 nonHydrostatic=.FALSE.,

# minimum cell fraction. This reduces steppiness

 hFacMin=0.01,

#- not safe to use globalFiles in multi-processors runs

 globalFiles=.TRUE.,

 readBinaryPrec=64,

 writeBinaryPrec=64,

 writeStatePrec=64,

 saltStepping=.FALSE.,

#parameters for kl10

 implicitDiffusion=.TRUE.,

 implicitViscosity=.TRUE.,

#change advection shceme

 tempAdvScheme=33,

 staggerTimeStep=.TRUE.,

 &


# Elliptic solver parameters

 &PARM02

 cg2dMaxIters=1000,

 cg2dTargetResidual=1.E-13,

 cg3dMaxIters=400,

 cg3dTargetResidual=1.E-13,

 &


# Time stepping parameters

 &PARM03

 nIter0=0,

 nTimeSteps=447121,

 deltaT=12.,

 abEps=0.1,

 pChkptFreq=5365440.,

 chkptFreq=0.,

 dumpFreq=5365440.,

 monitorFreq=2500.,

 monitorSelect=2,

#obcs forcing

 periodicExternalForcing=.TRUE.,

 externForcingPeriod=486.,

 externForcingCycle=89424.,

 &


# Gridding parameters

 &PARM04

 usingCartesianGrid=.TRUE.,

 delXfile='delXvar',

# delX=2000*500.,

 delY=100.E3,

 delRfile='delZvar',

 &


# Input datasets

 &PARM05

 hydrogThetaFile='T.init',

 bathyFile='topog.slope',

 vVelInitFile='vgeo.bin',

 &


Thanks


Yangxin

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.mitgcm.org/pipermail/mitgcm-support/attachments/20190730/84d36c36/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: noise_5184.eps
Type: application/postscript
Size: 26596 bytes
Desc: noise_5184.eps
URL: <http://mailman.mitgcm.org/pipermail/mitgcm-support/attachments/20190730/84d36c36/attachment-0001.eps>