[MITgcm-support] obcs eta

[Michael A. Spall]

I am running some nonhydrostatic calculations in a channel with
the obcs package and surface cooling.  I specify an initially 
geostrophically
balanced state and impose u, v, T on inflow (western boundary)
and use Orlanski on outflow (eastern boundary).  The obcs appear to
work well, at least the fields look smooth approaching the outflow
boundary.  However, the eta field changes by 10's of meters over only
a few days, sometimes upward, sometimes downward.  The
gradients in eta look good, and balance the geostrophic flow.
I get the same result when I do a hydrostatic version of the run.
It also seems impervious to changes in convergence parameters
and horizontal/vertical resolution.

I can't find how to specify eta on the inflow boundary.  How does the
model calculate eta on the boundaries?  Is there a way to ensure that
mass is conserved in the domain?  I defined ALLOW_OBCS_BALANCE,
which sounded like it was supposed to balance mass, but this
doesn't help.

My data file and OBCS_OPTIONS.h files are below.  Any suggestions
welcome.

Thanks,
Mike


========================================================


# ====================
# | Model parameters |
# ====================
#
# Continuous equation parameters
 &PARM01
 tRef=25*1.,
 sRef=25*35.,
 viscAh=1.0,
 viscAz=1.E-5,
 no_slip_sides=.TRUE.,
 no_slip_bottom=.TRUE.,
 diffK4T=0.E5,
 diffKhT=1.0,
 diffKzT=1.E-5,
 diffK4S=0.E4,
 diffKhS=1.0,
 diffKzS=1.E-5,
 f0=1.e-4,
 beta=0.E-11,
 tAlpha=2.E-4,
 sBeta =0.E-4,
 rigidLid=.FALSE.,
 implicitFreeSurface=.TRUE.,
 exactConserv=.TRUE.,
 eosType='LINEAR',
 hFacMin=0.05,
 nonHydrostatic=.TRUE.,
 readBinaryPrec=64,
 bottomDragLinear=0.E-4,
 tempAdvScheme=33,
 useSingleCpuIO=.TRUE.,
 &

# Elliptic solver parameters
 &PARM02
 cg2dMaxIters=1000,
 cg2dTargetResidual=1.E-13,
 cg3dMaxIters=50,
 cg3dTargetResidual=1.E-8,
 &

# Time stepping parameters
 &PARM03
 niter0=0,
 endTime=864000.,
 deltaT=40.0,
 abEps=0.01,
 pChkptFreq=0.0,
 chkptFreq=0.0,
 dumpFreq=86400,
 monitorFreq=86400.,
 &

# Gridding parameters
 &PARM04
 usingCartesianGrid=.TRUE.,
 usingSphericalPolarGrid=.FALSE.,
 dXspacing=200.,
 dYspacing=200.,
 delZ=25*20,
 &

# Input datasets
 &PARM05
 bathyFile='topog.bin',
 surfQfile='Qnet.forcing',
 hydrogThetaFile='T.init',
 uVelInitFile='U.init',
 vVelInitFile='V.init',
 pSurfInitFile='Eta.init',
 &
===========================================
C $Header: /u/gcmpack/MITgcm/pkg/obcs/OBCS_OPTIONS.h,v 1.5 2005/10/10 
05:53:49 mlosch Exp $
C $Name:  $

C CPP options file for OBCS package
C
C Use this file for selecting options within the OBCS package
C
C OBCS is enabled with ALLOW_OBCS in CPP_OPTIONS.h

#ifndef OBCS_OPTIONS_H
#define OBCS_OPTIONS_H
#include "PACKAGES_CONFIG.h"
#include "CPP_OPTIONS.h"

#ifdef ALLOW_OBCS

C Enable individual open boundaries
#undef ALLOW_OBCS_NORTH
#undef ALLOW_OBCS_SOUTH
#define ALLOW_OBCS_EAST
#define ALLOW_OBCS_WEST

C This include hooks to the Orlanski Open Boundary Radiation code
#define ALLOW_ORLANSKI

C Enable OB values to be prescribed via external fields that are read
C from a file
#define ALLOW_OBCS_PRESCRIBE

C This includes hooks to sponge layer treatment of uvel, vvel
#undef ALLOW_OBCS_SPONGE

C balance barotropic velocity
#define ALLOW_OBCS_BALANCE

#endif /* ALLOW_OBCS */
#endif /* OBCS_OPTIONS_H */


===========================================

-- 
==================================================

Michael A. Spall
Department of Physical Oceanography
W.H.O.I. MS #21
360 Woods Hole Road
Woods Hole, MA   02543

mspall at whoi.edu
(508) 289-3342 (office)
(508) 457-2181 (fax)

=================================================