Ocean Modeling - Science topic

You can follow the user manual step by step and ask ChatGPT more questions. QvQ

No doubt that the Blue OCEAN has significant effects on organizational performance. You may refer to the following article:

- Habibur Rahman and Sabbrina Choudhury (2019) The Influence of Blue Ocean Strategy on Organizational Performance, Global Disclosure of Economics and Business, Volume 8, No 1.

Hi,

For my curiosity, I downscaled an EMIC with RegCM 10 years ago. However, I don't keep that code with me now.

What did was investigating one of the recommended meteorological input to RegCM, for example NNRP, and then preparing your global output exactly like NNRP using CDO and NCO and run like NNRP input.

Hope this helps.

Cheers,

Kishore

In case of mass transfer, the ratio of flux absorbed by surface to flux passing over surface. Either the surface won't absorb any fraction of flux or it will absorb all flux passing through the surface. The lower bound is 0 and upper bound is 1 with mass transfer.

The rapid divergence of high resolution models from the actual state of the system is very well known and can be explained by the rapid error growth observed mainly for a fine scale part of the spectrum. The relative behavior of the error for the coarse and fine scales was described by Lorenz in his well known presentation on predictability.

The summary using Lorenz's original wording is as follows:

“1.Small errors in the coarser structure of the weather pattern – those features which are readily resolved by conventional observing networks – tend to double in about three days”

“2. Small errors in the finer structure – e.g., the positions of individual clouds – tend to grow much more rapidly, doubling in hours or less”.

How can I calculate total temperature increase/decrease over the last 10 years ?

Dear Kulthoum,

since the binary format is machine-dependent (big-endian vs little-endian) and may depend on the model's used programming language (row-major order vs column-major order), I can only guess what you have in mind. I'm also not sure if you work under Windows or Unix/Linux?

A very simple tool, which can be called from Matlab via the command line (see Matlab command system, for instance) is ncks as part of the netCDF Operators, or NCO, suite. It is easily available for Unix/Linux machines; I do not have an idea regarding Windows.

The ncks call is:

you will find details about this specific command at http://nco.sf.net/nco.html#binary . The Matlab call could be

status = system(ncks -O -b binary.dat -p Input_NetCDF.nc Dummy_NetCDF.nc)

afterward, you can delete the also created file Dummy_NetCDF.nc. Anyhow, I would check several times the binary conversion because of the machine dependence to avoid surprises. You may equip the model code with the capability to read NetCDF directly (recommended).

Happy modeling!

In CROCO and ROMS_AGRIF, you can get the height of the boundary layer as an auxiliary output variable

auxiliary_history_fields: rho Omega W Akv Akt Aks Visc3d Diff3d HBL

^^^^^

and use that as a reference to integrate the tracer of interest.

If you already have ocean circulation outputs, you can post-process it by using Bluekenue software (check the manual). Tecplot, Fudaa and Paraview are also able to do that. If you have to perform simulations, TELEMAC-2D and TELEMAC-3D are very appropriate numerical models (http://www.opentelemac.org/) or even SCHISM mentioned aboved.

Ok, thank you

-Huisman J., Sharples J., Stroom J.M., Visser P.M., Kardinaal W.E.A., Verspagen J.M.H., Sommeijer B. (2004). Changes in turbulent mixing shift competition for light between phytoplankton species. Ecology ,85, 2960–2970. doi:10.1890/03-0763

-Peters F. (2008). Diatoms in a future ocean--stirring it up. Nat. Rev. Microbiol. ,6, 407; author reply 407. doi:10.1038/nrmicro1751-c1

-Valenti D., Denaro G., Spagnolo B., Conversano F., Brunet C. (2015). How diffusivity, thermocline and incident light intensity modulate the dynamics of deep chlorophyll maximum in tyrrhenian sea. PLoS One ,10, 1–31. doi:10.1371/journal.pone.0115468

-Bopp L., Monfray P., Aumont O., Dufresne J.-L., Le Treut H., Madec G., Terray L., Orr J.C. (2001). Potential impact of climate change on marine export production. Global Biogeochem. Cycles ,15, 81–99.

-Toggweiler J.R., Russell J. (2008). Ocean circulation in a warming climate. Nature ,451, 286–288.

-Huisman J., Pham Thi N.N., Karl D.M., Sommeijer B. (2006). Reduced mixing generates oscillations and chaos in the oceanic deep chlorophyll maximum. Nature ,439, 322–5. doi:10.1038/nature04245

-Boyce D.G., Lewis M.R., Worm B. (2010). Global phytoplankton decline over the past century. Nature ,466, 591–6. doi:10.1038/nature09268

Thanks Sean M Ridge for clarification. What I understand from users guide,

to get the unit suggested by you ie,

<wt(0)> in (ms-1 K,), the surface flux in W /m**2, it is to be divided by (3980*1025).

Is this correct?

Hi Delei,

Be aware that model biases tend to differ regionally and that for oceanographic applications that are sensitive to curl simple bias correction might not be sufficient. Furthermore, model biases tend to be a function of (1) the size of a low pressure system and model resolution, often with larger biases for smaller and stronger systems, and (2) proximity to land, where model smoothing combines land and ocean winds. In some cases, e.g., NCEP1, modeled winds can be much greater in high latitude storms because of limitations related to resolution.

Why do you think the results indicate that the method is incorrect?  What are the results that make you think this?

If you are processing moving-vessel ADCP data, here is the most robust method ---Velocity Mapping Toolbox, developed for ADCP.

You can find more information in this paper, and you could download it from the  USGS Hydroacoustics Web pages.

Regards,

Hao

thank you(it would be definitely do-able using GrADS maxloc() function. You may want to further polish with some maskout() function to provide any criterion of your liking)

it is very good. thank you very much

Thank you all! I didn't look at this page for a long time and now I found some new answers. Actually, what I meant is not a geothermal heat flux, but heat exchange between seabed and ocean waters in shallow areas, where 1) seasonal temperature variations are significant, and 2) part of solar radiation passes down to the bottom and makes it warmer.

Have some ideas too..

Thank you Guillermo.

 

In this paper they assimilate data from an on-board ADCP. The data is "averaged" to remove small scale features not represented by the model.

My case, however, is a little different since the ADCP is moored.

The model is able to reproduce the high frequency I am interested on. Given the processed data is provided hourly, there should not be any "mandatory" a priori filtering for the assimilation... 

Dear Nicolas,

You can also use a Bland Altman diagram by ploting the series of differences as a function of the mean of the series and compute the limit you admit for the differences (e.g. +/- one sigma). One limit of the method is the gaussian distribution of the difference series

Hoping it can help you, Regards, G.C.

Abbas,

It would be better to ask the relevant institution that concerns about the records of that region at least they might guide you to your goal.

M. Basel

Thanks Daniel!

If you want to model in small basin, SWAN is much more simpler to implement.

No, you can't. Here is why: 

- radiation stresses apply to the total momentum (which is the sum of "current" moment + wave momentum). 

- bernoulli head + vortex force only apply to the "current" momentum. 

The vortex force compensates for the change in wave momentum due to refraction over currents: the wave momentum goes out of convergence zones, while the current momentum goes in... but the sum of the two is zero. 

LCs appear because of this piling up of "current" momentum in the convergence zones. You do not see it if you work with the total momentum. This is a little discussed in Garrett (1976) although not the main point of his paper.

Hello,

I'm doing a similar application with heat model 5 (ocean) with varying meteo parameters. Can you tell me which parameters did you put in the .tem file?

Thank you!

Magda

I do not think we know what is happening at 70 m/s... The sea is white, the air is full of spray...

From the great webpage of Colorado Sea level Group: "the ocean basins are getting slightly larger since the end of the last glacial cycle. GIA is not caused by current glacier melt, but by the rebound of the Earth from the several kilometer thick ice sheets that covered much of North America and Europe around 20,000 years ago. Mantle material is still moving from under the oceans into previously glaciated regions on land. The effect is that currently some land surfaces are rising and some ocean bottoms are falling relative to the center of the Earth (the center of the reference frame of the satellite altimeter). Averaged over the global ocean surface, the mean rate of sea level change due to GIA is independently estimated from models at -0.3 mm/yr (Peltier, 2001, 2002, 2009; Peltier & Luthcke, 2009)." Check it, ypu will probably find a good answer: http://sealevel.colorado.edu/content/what-glacial-isostatic-adjustment-gia-and-why-do-you-correct-it

To go deeper, my suggestions is to read Peltier, as Arvind suggested you: Peltier, W. R. (2004). Global glacial isostasy and the surface of the ice-age Earth: the ICE-5G (VM2) model and GRACE. Annu. Rev. Earth Planet. Sci., 32, 111-14, although he is more phocused in the modelling on every paper he publishes, and I do not know if you are so interested on theses details.

Nevertheless, do not only take into account these isostasic models. If you are dealing with local data, check tide gauges time series, and especially, compare them with global measn sea level rise time series (Church and White, 2012) and altimetry satellite data, to check what is going on in your area. Peltier model is clear for higher lattitudes but it is still unclear about mid latitudes.

Dear Jayakrishnan

The best in your case is to compute yourself the heat content from analyses or reanalyses. Because there are lots of available (re)analyses as indicated by Roxana, sometimes with important differences, with various mesh size and vertical levels; then because you can choose the depth of your heat content instead of standard depths. Then you can compare heat contents between the different available (re)analyses.

Among the different products you can find on the web:

gridded oceanic data: EN3 (Ingleby and Huddleston, 2007), EN4 period 1950-present; 42 levels

Oceanic analyses: ARMOR3D (Guinehut et al (2012) 24 levels, 1993-2010; CORA (Cabanes et al., 2013): 59 levels, 1990-present

oceanic reanalyses: GLORYS (Ferry et al., 2012) 75 levels, 1993-2009; GODAS (Saha et al., 2010), 40 levels, 1980-present; ORAS4 (Balmaseda et al, 2013), 42 levels, 1958-present; NEMOVAR1 COMBINE (Balmaseda et al. 2010), 42 levels, 1958-2008.

Hope this can help you!

Best regards,

G.C.

Thanks to all. I finally found an answer in Cole et al. (JGR 1998): "Cyclic loading and creep response of aligned first-year sea ice" which is generally consistent with Wadhams (JGR 1973), in proposing a reduced B coefficient for creep. This is now in version 5.10 of WAVEWATCH III (to be released in a few months) ... and was used for this paper:

But yes it is in general a function of ice temperature and thickness and floe sizes are very important parameters too. 

I mean open boundary zones, and yes I´m using tidal forcing. 

Thank you Ariadne Tsambani

Thank you sir.... than you for your information....

If you want to try looking at some model data you could try:

Check these out and if you have problems downloading or visualizing the data message me and I will give you some matlab code that could help.

Cheers,

You can also consult the famous Holland (1980)'s paper: MWR 108(8) 1212-1218 which can give you good indication for wind profiles in hurricanes. Such profiles are largely used in the literature.

Best

G.C.

You can do the differentiation in space domain or wavenumber (k) weighting in spectral domain. Basically:

space domain: slope s= dh/dx

spectral domain: S_slope(k)=k^2 S_elevation(k)

S_slope is slope spectrum, S_elevation is elevation spectrum, h is elevation,d x is sample spacing, wavenumber k is related to wavelength L by k=2pi/L.

The spectral approach gives you statistical comparison, and is probably much more preferred for your case of comparing lidar and WaMoS measurements (or for data sets from different sensing systems in general.)

try

Dear Rayhane,

in the C grid, tracers like temperature T are defined in the middle of a box, while the zonal velocities are defined in the middle of the eastern and western side of the box, and meridional velocity is defined in the middle of the northern and southern side of the box, i.e. approximately like this

  x------V------x------V------x

  |                    |                    |

  U        T       U        T        U

  |                    |                    |

  x------V------x------V------x

when looking down. Now, in the zonal Momentum equations you have terms like

d/dt U = -f*V

If you want to write down this equation for one specific U-value on the grid (say the one in the center of my little sketch) you therefore need the meridional velocity at the same grid point, which you don't directly have. You only have V values at the four surrounding V points. So for calculating the Coriolis effect you have to first form an average velocity at the U point from the four surrounding V points. This introduces some error.

Does that help?

Cheers, Christoph

Hi Guoquiang,

yes the short answer is no, but there are people like Tamura, Kudryavtsev, and me trying to get there ... it will be a long journey. At present the model parametrization start to be pretty bad after 3 times fp (in particular in the directional distribution).

The good news is that we are starting to get data:

and there are plenty of remote sensing and underwater acoustic data that can be used to better constrain the models in that range, e.g.:

Please contact Professor Diogo Bolster in the Civil Enginering Department at the Univ. of Notre Dame. He has 3 students who work on a somewhat related problem. One of them, Tomas C.  Aquino, has given gave 3 lectures proposing 3 different approaches to the problem, in groundwater and surface river flows containing solutes. One is using upscaling and fractional calculus, one involving Peak and Tail Scaling in hydrological tracer tests, and one involving Breakthrough Curves to predict the arrival of solute concentration over large length and time scales. Another student, Nicole Sund, used small time asymptotics and the aid of a Spatial Markov model to predict the solute transport in complicated flow fields. 

Contact Point: bolster.5@nd.edu 

Please see the report arXiv:1503.04313v1.pdf

Hello,

I use World Tides (Boon, 2004) for both tidal and current analyses. It works under the Matlab environment. In order to get about the meaning and use of inferred constituents you may refer to the TASK user guide (attached). I hope this may help.

You are very right to be involved with global warming, because there is so much happening in the weather. Yet, no theories!

I am concerned about the rise in humidity worldwide, obviously the oceans are evaporating at an increasing rate (have become unstable). Is your ocean index to do with this phenomenon?

English is not my tongue, but I can help "polish" some text,if you would like.

best wishes

There are a number of papers by Roco and Shook in 1980s, where they developed  models for friction and turbulence between different sediment laden layers. Although these were for pipelines, some aspects of these can be applied in your cases.

On top of tides, winds and density, if you are referring to near-shore circulation, the effect of gravity waves through Stokes drift velocities and wave setup (water piling against the coast) can be very important... but it depends on the spatial scale, as Francesco said, and the particular characteristics of the area you are focusing on.

This is a broad research theme. I would recommend the paper by Uchiyama et al (2010) for  the wave currents interaction in the near-shore.

You could also look at the WOCE Pacific Atlas - see http://www-pord.ucsd.edu/whp_atlas/pacific_index.html   This is a graphical representataion of the data referred to by Loic in the previous answer.

My co-authors and I provide a partial review of the impact of changing glacial winds on the Agulhas Retroflexion during the LGM in this paper. We discuss the impact of variation in regional winds on marine production and coastal currents off West Africa. The results should help you to hypothesize changes you may see in the Indian coastal currents for comparison with your study.

You can also review the chapter from my dissertation work on the California Current during the LGM for comparison. Primary production can be enhanced by several processes in addition to the Fe hypothesis noted in the prior answer. For example, increased offshore Ekman transport, or increased Ekman pumping via enhanced wind stress curl will increase the nutrient flux to the surface ocean, thereby potentially increasing primary and by inference secondary production.

D Nof, V Zharkov, J Ortiz, N Paldor, W Arruda, E Chassignet, The arrested Agulhas    retroflection, Journal of Marine Research 69 (4-6), 4-6, 2011

J Ortiz, A Mix, S Hostetler, M Kashgarian, The California Current of the last glacial maximum: Reconstruction at 42 N based on multiple proxies, Paleoceanography, 12, 191-206, 1997

As the prior answers suggest, it's a compromise.  Structured grids are simpler to code, so if a significant constraint is programming time, this is the way to go.  Unstructured grids allow you to place the resolution where you need it (more easily), so if computing time is the main limit, this is the way to go.

In deciding whether you're more limited by programming time or computing time, don't underestimate the programming time.  It's awfully easy to do so.  Programming time includes not just the initial getting to the point of having meaningful answers in this particular model, but in developing all the programs and scripts for the care and feeding of the model (e.g. translating forcing fields to the model's grid -- easier with structured grids), and in maintaining all of your system's codes (again, can be more complex with unstructured grids).  The maintenance, care, and feeding of a model system is about 10x the programming time of the core model in my experience.

Hello Atteleth, 

I personally don´t think you should use the advection/dispersion module from MIKE 21, which takes into account decay of the modeled substance. What you can do is do a MIKE 21 HD simulation and then in plot composer add particules. This particules will move according to the hydrodynamics in your modelling and they can give you an idea of how the propagule will distribute in your study area. An alternative is to use the ABM from MIKE 21, which allows you to determine how the propagule will propagate, but since they are only floating, then I guess there is no point on complicating your study!

I hope this helps. Cheers.

Usually along-track data are assimilated, since

maps contain additional errors given by

the interpolation/mapping procedure to generate

the maps. Also this procedure can introduce spatial

correlations of the errors and perform

some gap-filling, i.e. assimilating the maps is risky

because you don't really know where the observations were.

We assimilite along-track and as far

as I know also other groups (OPA-Var, etc.).

to study the upweling you need to have a 3D model. U can get Ekman transport with the help of wind stress. but the offshore Ekman transport causes vertical velocity in subsurface water, which is the is measure of upwelling.

Marc,

Two people I know with good knowledge on WaveWatch 3 are Erick Rogers and Jim Dykes. Erick.rogers@nrlssc.navy.mil and james.dykes@nrlssc.navy.mil

Gregg

Oops, the attachmen tis coming now...

Thank you for all the comments.

The reason I filter out the high frequency is to remove processes that my grid and model configuration (eg.: forcings) are not capable of representing. Since I am using tides in the simulation, I have to add this information back to the data for the assimilation.

The impact of the HF radar radial velocities assimilation very important. The exclusion of this data from the assimilation process leads to ~50% larger errors in the SLA. This becomes even more important if we focus on the surface currents.

I have completed some experiments and am now finishing a paper on the subject. To give you an idea of the conclusions, the assimilation of raw HF-radar data gives better results for coarser grids - reducing the rms errors and time shifts in relation to the data. If we use a finer grid, the improvement is less important.

Cheers

Joao

Hi Mohammed Salim, We have assimilated along track SLA using Ensemble Optimal Interpolation technique. My suggestion is that before you jump into the data processing part, first you decide which technique you want to use and then make step by step algorithm. Then you will get an idea of the format of the datasets to be used.

Using gridded data does not add actual information compared to along-track data. To the contrary: gridded products are actually too smooth, the small scales have been removed by the interpolation method. In addition the error covariances of gridded data are full and unknown, while those of along-track data are almost diagonal and simple. Therefore I advocate strongly to use along-track observations. (Of course this also depends on the scheme -- if you use a simple scheme such as nudging, it might be moe effective to go for gridded data, but an advanced scheme can handle sparse data without trouble.)