[MITgcm-support] How to force river runoff data in a regional domain

[Stefano Querin]

Hi Kunal,

As far as I can say, I would suggest option 3 (OBCS along the northern side, in your specific case).
Since you are using high spatial resolution, you could somehow mimic the shape of the riverbeds and simulate (as realistically as possible) both the thermohaline and momentum contribution of the main rivers.
We adopt this configuration in our forecasting system in a small area of the Mediterranean Sea (http://medeaf.inogs.it/adriatic <http://medeaf.inogs.it/adriatic> - river discharges are still climatological, at the moment...): in the maps, the small "sticks" along the coastline are the final segments of the various riverbeds (19), close to the mouth.
You could try to implement something similar by imposing prescribed/measured values of V, T and S (if you use northern OBC, then U=0). Then, V times the area of your riverbed section is your discharge (in m^3/s).
If you also have data on the depth of the riverbed and very small dz in the first layers, you could also mimic the salt wedge and similar processes (see, for example, Figure 5 in https://www.mdpi.com/2077-1312/9/5/469 <https://www.mdpi.com/2077-1312/9/5/469>).
With this approach, river segments are part of the bathymetry: you simply have to apply the OBCS at the upstream end of each "channel". Of course, velocities can be very high (up to a few m/s...), so you must pay particular attention to time stepping and CFL condition.

Hope this helps...

Cheers,

SQ


> On 28 Jul 2021, at 12:09:56, kunal madkaiker <kunal.madkaiker02 at gmail.com> wrote:
> 
> Dear MITgcm users,
> 
> I am doing regional simulations in the Northern Indian Ocean. Stratification in Bay of Bengal is controlled by salinity and freshwater flux since many rivers like Ganges, Brahmaputra empty into this basin.
> 
> I am not able to properly simulate my temp, salinity and currents into the Head bay region of BoB as I am unable to produce an equatorward flow in my domain. Freshwater increases in this region during month of JJAS due to melting of Himalayan glaciers.
> 
> I have gone through previously asked similar questions about river runoff and came to know the following possibilities:
> 
> 1. To feed at surface through data.exf
> 
> 2. Feeding a 3-D run-off input using addMassFile
> 
> 3. Through OBCS/ RBCS.
> 
> My domain is regional, with small dz and high spatial resolution.
> Please let me know which option is the best suited and reliable, since I not only want to improve my temp/ salinity but also to rectify the circulation in head bay region. I would be using Dai et al. (2009) river runoff data.
> All suggestions are welcome. Attaching figure of my model domain.
> 
> Thanks and Regards,
> Kunal
> <Both_coast_bathymetry.png>_______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.mitgcm.org/pipermail/mitgcm-support/attachments/20210728/906c8b15/attachment.html>