Hello, I would like to know how to perform a near and far field modeling with OpenFlow Flood of the brine discharge (coming from a desalination plant) in the sea.
Hello Jorge,
A brine discharge can be simulated by OpenFlows FLOOD. To do it you'll need to setup a 3D hydrodynamic simulation by creating a new domain using MOHID Water numerical engine (see more information here). Then create a new simulation.
Being a 3D simulation, the setup of the model will greatly depend on the area you're trying to simulate and the processes governing hydrodynamic circulation in that area. For simulation brine discharges, where buoyancy plays a decisive role, your model should be setup to simulate tide and 3D wind and density driven currents. This is required as the initial dilution and shape of brine plume will depend directly from it.
So you'll require realistic initial and boundary conditions for water level, currents, temperature and salinity. For water elevations at the open boundary you can use a global tide solution as described here. For the currents, temperature and salinity it depends on the area you're simulating, but using a global/regional circulation solution (EU Marine Copernicus, NOAA; etc) is rtypically recommended. The use of nested models methodology may be required when doing this type of applications. For 3D applications the use of an appropriate treatment of vertical turbulence is of extreme importance, so you'll need to activate module GOTM your simulation. Also the vertical resolution of your model must be adequate to be able to represent the vertical density gradients especially near the bottom, as the brine plume will be transported near the bottom.
As this is a complex simulation (not related with tide) FLOOD user interface for configuring each module does not contain all the necessary options, so you'll need to use the text editor to setup each module. To do this follow the instructions here.
Once you have your 3D hydrodynamic simulation implemented and validated, you can then setup the brine discharge. There are several ways of doing it depending on the characteristics of the discharge. Is the brine discharge made via a submarine outfall or a surface discharge? If it's a submarine outfall you will want to activate module Lagrangian which simulates using particle tracking methods the dispersion of the plume (in the near field and far field). In the Lagrangian module input file. you can specify specific commands to compute the near field namely COMPUTE_PLUME : 1 and the path to the submarine outfall configuration file (JET_DATA_FILE : ..\General Data\Outfall\Outfall.dat) where you define parameters such as outfall length, position and orientation, number of ports and orientation, etc. You'll need to specify the characteristics of the discharge (flow, temperature and salinity). Then run the simulation and check the outputs from the near field model (dilution, shape and position of the equilibrium point of the plume in the far field, etc). You can then use the Eulerian transport module (Water Properties and DIscharges module) to configure your discharge in the far field. This is required as it is a negatively buoyant plume
I understand this is a very generic answer to your question. If you provide more details regarding the type of application you're trying to do I'll be glad to help.
Kind regards,Luis
I was reviewing some youtube videos and to create the domain, several photos must be uploaded to the Digital Terrain Model section, these photos correspond to the place where I want to model? should I upload many photos of the same place? Beforehand thank you very much
Hi Andrés,
Indeed the module exists but it was developed a long time ago and meanwhile I believe it is not compliant with the current version of HYCOM. I don't have any sample of the ConvertToHDF5Action input data file.
One way you could convert HYCOM outputs to HDF is by using the generic netcdf to hdf5 conversion module. See the documentation related with the "CONVERT NETCDF CF TO HDF5 MOHID" operation in the wiki article regarding ConvertToHDF5.
Hello Luis!
Thank you for your previous answers. I had a doubt about the calibration, as I should model near and far field independently, I should do one calibration for near field and then another for far field?. In addition, I wanted to know what parameters I recommended to be introduced so that I could do a good calibration and validation.
I thought for the near field to use:
currents (Copernicus), salinity measurements of the receptor medium (copernicus).
While for the far field I thought to use:
Tidal level (FES2014), currents (copernicus), temperature and salinity measurements of the receiving medium (copernicus), wind and atmospheric pressure data in the area of interest
In advance, thank you very much!
It will be difficult to give a detailed answer here as your question is regarding modelling methodology.
Just to clarify, the near field model requires as input the submarine outfall characteristics. This are normally known (if the submarine outfall already exists) or designed.
In order to run a near field simulation you'll have to run it attached to a far field simulation. For calibration purposes you may use schematic or coarse resolution data, but in fact a high resolution 3D hydrodynamic model in the vicinity of the discharge is recommended. To have a good representation of the hydrodynamics, depending on the site of the discharge, you'll need to simulate tide, 3D currents (due to tide, wind and density driven currents, turbulence, wave-current interaction, etc).
In conclusion, before running any submarine outfall simulation, you'll need to set up the hydrodynamic model, calibrate, validate it. Only after that you should focus on plume dispersion aspects.
Thanks to your previous answers I have a general idea of what I should do, but I am confused as I am missing the detail of how to carry it out.
I understand that calibration is a process of adjusting the model parameters, initial and boundary conditions to reproduce with a degree of accuracy the field measurements. However, I am a bit confused to calibrate the program. I think that once the new workspace is created, the domain is defined and the bathymetry of the area is entered, the simulation can be run. Add the current data (copernicus) and run the simulation again.
After that I was thinking of adding the parameters of the subsea outfall and again running a simulation. Already with this I would have the near field modeling with Lagrangian approach.
After this, for the far field simulation I should add the wind, tidal and current forcings with Eulerian approach.
Am I on the right track?
Where can I find more information on how to perform the calibration?
As an aside, I was doing it all over again and wanted to check the generate a digital terrain section and it tells me that the page is not available.
communities.bentley.com/.../generating-digital-terrain
Thank you very much in advance
Hi Jorge,
You're on the right track, but wind, tide and currents, should be part of the hydrodynamic solution before you run the near field model. The near field model requires that you setup the hydrodynamic model correctly first. You should only run the near field model after that is ready.
Regarding the digital terrain model generation, please find the information in the article How to create a digital terrain model in OpenFlows FLOOD.
Kind regards,
Luis