Search found 26 matches

by kurapov
Thu Aug 03, 2017 3:12 am
Forum: Meetings/Workshops
Topic: Coastal Modeling, Prediction & Observation Session - 2018 OS
Replies: 0
Views: 1599

Coastal Modeling, Prediction & Observation Session - 2018 OS

Dear colleagues, We are happy to offer again our successful session - please consider submitting an abstract (submission now open with a deadline of 6 September). Go to session OM001 (ID: 26761) under "Ocean Modeling": https://agu.confex.com/agu/os18/preliminaryview.cgi/Session26761 Advances in Coas...
by kurapov
Wed Jun 03, 2015 9:53 pm
Forum: ROMS Problems
Topic: On restart, ROMS cannot see 3rd file in the multi-file list
Replies: 3
Views: 2049

Re: On restart, ROMS cannot see 3rd file in the multi-file l

This does help a lot.. and hopefully solves the problem in my case...

Thank you for checking!
by kurapov
Wed Jun 03, 2015 8:34 pm
Forum: ROMS Problems
Topic: On restart, ROMS cannot see 3rd file in the multi-file list
Replies: 3
Views: 2049

On restart, ROMS cannot see 3rd file in the multi-file list

I am running a coastal application in ROMS providing wind velocities (and other forcing) as multiple files (one file per year, 2008, 2009, 2010, 2011, 2012). Because of the wall clock limit on the supercomputer I can run it for about 3 model months at a time. It runs fine (with restarts roughly ever...
by kurapov
Tue Oct 12, 2010 12:13 am
Forum: Job Opportunities
Topic: Post-doctoral position, COAS
Replies: 0
Views: 1627

Post-doctoral position, COAS

Research Associate (Postdoctoral), College of Oceanic and Atmospheric Sciences, Oregon State University, to work on modeling, data assimilation, and analysis of ocean processes that determine shelf-open ocean exchange off Oregon. The successful applicant will participate in an NSF-sponsored research...
by kurapov
Fri Sep 17, 2010 1:03 am
Forum: ROMS Problems
Topic: Intel’s new i7 980x CPU gives disappointing speedup
Replies: 48
Views: 24351

Re: Intel’s new i7 980x CPU gives disappointing speedup

Ban, -- Thanks for interesting statistics. Did you run ROMS test in OpenMP or MPI regime?
-- Alex
by kurapov
Wed Jun 03, 2009 12:07 am
Forum: ROMS Problems
Topic: OMP_NUM_THREADS or NCPUS (2) greater than available cpus (1)
Replies: 1
Views: 1288

OMP_NUM_THREADS or NCPUS (2) greater than available cpus (1)

I am trying to run a roms application as OpenMP. I use ROMS version 3.1 and compile it as OMP using a pgi compiler on a Sun station (linux, 4 proc) and obtain executable oceanO. I set env variable OMP_NUM_THREADS to 2 and set the number of tiles to 2 in the roms .in file. I am trying to run the exec...
by kurapov
Fri Oct 03, 2008 6:13 am
Forum: ROMS Discussion
Topic: barotropic pres gradient
Replies: 2
Views: 1552

Re: barotropic pres gradient

bzhang: your comment makes sense. Thanks!
by kurapov
Thu Oct 02, 2008 11:30 pm
Forum: ROMS Discussion
Topic: barotropic pres gradient
Replies: 2
Views: 1552

barotropic pres gradient

I've been trying to do fancy momentum term balance analysis, doing this by post-processing of ROMS output. I've translated the roms pressure grad term in Matlab and stumbled upon the following detail. In the code, the barotropic pressure gradient, e.g., at level N, is apparently twice the value that...
by kurapov
Fri Sep 26, 2008 1:29 am
Forum: Job Opportunities
Topic: Research associate, coastal ocean data assimilation
Replies: 0
Views: 1365

Research associate, coastal ocean data assimilation

CLASSIFIED ADVERTISEMENT Research Associate (Postdoctoral), College of Oceanic and Atmospheric Sciences, Oregon State University. The successful applicant will lead research in areas of development and implementation of algorithms for variational data assimilation in a nonlinear model of circulation...
by kurapov
Fri May 30, 2008 12:41 am
Forum: ROMS Bugs
Topic: KELVIN case typo
Replies: 9
Views: 5304

Re: U = sqrt(g/H)*exp(-f*y/sqrt(g*H))*cos(omega*time).

Anybody looking into that, check dimensions. Is U a velocity (m/s)?
sqrt(g/H) has units of (1/s). Does the wave aplitude have to be sqrt(g*H)?
by kurapov
Sun Mar 23, 2008 5:50 pm
Forum: ROMS Discussion
Topic: roms-3.1 checkout
Replies: 1
Views: 918

roms-3.1 checkout

Hi, I've tried to checkout the tagged roms version, roms-3.1, which is shown as revision 167 in the roms-svn repository (https://www.myroms.org/projects/src/browser/tags). I run, from the linux machine: >> svn checkout https://www.myroms.org/svn/src/tags/roms-3.1 MyDir The download has been successf...
by kurapov
Tue Feb 26, 2008 5:14 pm
Forum: ROMS Discussion
Topic: open boundary condition problems
Replies: 6
Views: 3501

Good poi`nt, about the diagnostic. Dig it and keep us updated.
by kurapov
Mon Feb 25, 2008 5:36 pm
Forum: ROMS Discussion
Topic: open boundary condition problems
Replies: 6
Views: 3501

Double-check your open boundary files. We've seen the statements like "I've checked the boundary files and they were fine". And then, later "sorry, there was a NaN at one point in one snapshot". NaN does not appear for no reason. Something has been set NaN. You obtained your open boundary files by i...
by kurapov
Fri Jan 18, 2008 5:32 pm
Forum: ROMS Discussion
Topic: Surface boundary condition
Replies: 5
Views: 2708

Use a better resolution. Try to resolve the boundary layer. Or, for the sake of experiment,
increase Akv by an order of magnitude.
by kurapov
Fri Jan 18, 2008 12:47 am
Forum: ROMS Discussion
Topic: Surface boundary condition
Replies: 5
Views: 2708

The Ekman layer thickness is delta=sqrt(2*Akv/f) (see Pedlosky'87, 4.3.16) Akv=1.e-3 m2/s f=1.e-4 1/s => delta= 4m So the sharp change you see in u,v in the upper 4 m is Ekman veering. Besides, sustr=Akv du/dz at the surface. You are probably trying to estimate it at the points below the surface. Th...
by kurapov
Wed Nov 14, 2007 2:09 am
Forum: Job Opportunities
Topic: Research Associate (Postdoctoral), COAS, Oregon State Univ.
Replies: 0
Views: 1273

Research Associate (Postdoctoral), COAS, Oregon State Univ.

The College of Oceanic and Atmospheric Sciences (COAS) at Oregon State University announces the availability of a full-time position for a Research Associate (Postdoctoral). We seek a research colleague who will join our coastal modeling and data assimilation group to work on variational data assimi...
by kurapov
Fri Oct 19, 2007 7:12 pm
Forum: ROMS Discussion
Topic: boundary conditions for tracers in pre_step3d
Replies: 0
Views: 988

boundary conditions for tracers in pre_step3d

Can anybody comment on why boundary conditions are implemented for tracers in pre_step3d.F? To my understanding, pre_step3d initializes time stepping of interior variables, u(:,:,:,nnew), v(:,:,:,nnew), and t(:,:,:,nnew,:). Then, why lateral boundary conditions are implemented for the intermediate t...
by kurapov
Tue Oct 09, 2007 8:44 pm
Forum: ROMS Discussion
Topic: W for tracer advection
Replies: 0
Views: 1197

W for tracer advection

Here is a question about the computation of W before time-stepping tracer fields. Let's consider a b/c time step, from time n to n+1. To time-step tracers, mass fluxes are needed at the intermediate time n+1/2 (is my interpretation correct?). Horizontal fluxes are updated in step3d_uv, and vertical ...
by kurapov
Thu Aug 16, 2007 4:31 pm
Forum: ROMS Discussion
Topic: Bulk fluxes, surface-relative winds and gustiness
Replies: 12
Views: 6164

The wind speed scale is meters per sec, while the ocean surf currents are 1-2 orders of magnitude as small. Subtracting the ocean currents from wind currents would change the resulting atm. flux only by 1-10% (the difference would be esp. small in most important cases, when winds are strong). What d...
by kurapov
Mon Jul 09, 2007 9:34 pm
Forum: ROMS Installation
Topic: different result obtained with different versions of pgi
Replies: 5
Views: 2158

The underlying continuous equations (w/out dissipation) with any type of local boundary conditions used in ROMS constitute an ill-posed problem (ref: Oliger and Sundstrom, SIAM J. Appl. Math, 1978). That may mean for us that ROMS with open boundary conditions (and low enough dissipation - like coast...
by kurapov
Sat Jun 16, 2007 1:08 am
Forum: ROMS Problems
Topic: barotropic wave in the channel (no forcing, no dissipation)
Replies: 9
Views: 4027

Sasha and Stef, Thanks for your valuable comments. I agree, it should really make sense to have b/t time step small enough to resolve the fast b/t wave in time. Here are results of experiments w/ decreased DT. The number shows the decrease in magnitude over two days (max zeta(t=48h)/ max zeta(t=0)):...
by kurapov
Thu Jun 14, 2007 7:20 pm
Forum: ROMS Problems
Topic: barotropic wave in the channel (no forcing, no dissipation)
Replies: 9
Views: 4027

DT=240 sec
NDFAST=30
Maximum barotropic Courant Number = 1.77177877E-01

Channel depth H=100 m,
so wave_phase_speed=sqrt(gH) = 31.2 m/s

wave_length_of_interest = channel length =100 km (=1.e+5 m)

2*pi * dt * wave_phase_speed / wave_length_of_interest = 0.47

Any comments?
by kurapov
Thu Jun 14, 2007 6:49 pm
Forum: ROMS Problems
Topic: barotropic wave in the channel (no forcing, no dissipation)
Replies: 9
Views: 4027

Colleagues, Thank you for your informative and thoughtful replies. I'll check on these and let you know. Meanwhile, can you answer the simple question? In the 3D case, does rhs3d (specifically, prsgrd) compute the total pressure (- g d zeta/ dx + (g/rho0) int_{z}^{zeta} [d rho / dx] dz') or only its...
by kurapov
Wed Jun 13, 2007 8:20 pm
Forum: ROMS Problems
Topic: barotropic wave in the channel (no forcing, no dissipation)
Replies: 9
Views: 4027

I verified that in the 3D test v is the same at every level. Also T and S are not changed.
In the output, svstr=0, bvstr=0, AKv=0, AKt=0.
by kurapov
Wed Jun 13, 2007 6:45 pm
Forum: ROMS Problems
Topic: barotropic wave in the channel (no forcing, no dissipation)
Replies: 9
Views: 4027

barotropic wave in the channel (no forcing, no dissipation)

I have tried to use roms (Feb. 2007 version) to reproduce the barotropic linear wave solution in the periodic channel. I presumed this could be done both in the 2D set-up (#undef SOLVE3D) and in the 3D set-up (#define SOLVE3D) if unstratified water is used in the second case. Surprisingly, the 2D an...
by kurapov
Mon May 14, 2007 3:49 pm
Forum: ROMS Discussion
Topic: Vertical velocty at the wall (WALL boundary condition)
Replies: 2
Views: 1470

Re: Vertical velocty at the wall (WALL boundary condition)

Re: I consider 3D rectangular basin with uniform wind forcing on the top in E-W direction. I put periodic boundary conditions in E-W direction and WALL boundary conditions on the North and South. So I put no slip condition GAMMA2 = -1.0d0 in the .in file. And the result was that the horizontal along...