Hello everyone,
I have succesfully run the ESTUATY_TEST tutorial case (https://www.myroms.org/wiki/ESTUARY_TEST_CASE), and the results look good. The parameter set gives a highly stratified estaury with a persistent salt wedge structure with two-layer exchange. After some tidal cycles, the salinity field reached a quasi-steadt state. The freshwater input from the river side is balanced by the upper-layer freshwater output at the ocean side.
Things became complicated when I tried to use ROMS to model weakly and partially stratified estuaries (I increased the value of q_west in ana_m2obc.h). At the ocean-side boundary, the whole water column is seawater for most of the tidal cycle, and no freshwater is exiting the domain from the upper layer as in the highly stratified case. As a result, as the simulation progresses, due to the continuous freshwater supply from the upstream river, the whole estuary will be filled with freshwater. No quasi-steady state can be reached. This is of course unwanted because, finally there is no longitudinal salinity gradient in this case for me to investigate physical processes in tidally energetic estuaries, such as strain-induced periodic stratification.
Could anyone give me some hints on how to reach a quasi-steady state for weakly and partially stratified estuaries? Would this even be possible? Thank you in advance.
Best regards,
Jay
weakly and partially stratified estuaries using ROMS
Re: weakly and partially stratified estuaries using ROMS
i remember setting this up. there was some trial-and-error to keep the salt where i wanted it.
there is a balance between bathy, Qriver, and mixing (basically the tidal forcing).
So as you adjust that balance, the salt field will evolve. For that particular very idealized case (constant dh/dx) the balance is sensitive. If you change the mixing, then you will probably have to change the Qriver to keep the salt in the estuary.
The test cases are just test cases- they may not be appropriate for all realistic conditions. Find an estuary in the real world that has conditions that you are looking at, then modify the test case to be that way.
You will see that in real life the geophysical characteristics of the estuary help to set its behavior.
I have another grid from Hans Burchard for a more trumpet like shape. we distribute that in coawst. it may provide a more realistic setting.
there is a balance between bathy, Qriver, and mixing (basically the tidal forcing).
So as you adjust that balance, the salt field will evolve. For that particular very idealized case (constant dh/dx) the balance is sensitive. If you change the mixing, then you will probably have to change the Qriver to keep the salt in the estuary.
The test cases are just test cases- they may not be appropriate for all realistic conditions. Find an estuary in the real world that has conditions that you are looking at, then modify the test case to be that way.
You will see that in real life the geophysical characteristics of the estuary help to set its behavior.
I have another grid from Hans Burchard for a more trumpet like shape. we distribute that in coawst. it may provide a more realistic setting.
-
stevenmiguelfigueroa
- Posts: 22
- Joined: Mon Apr 20, 2020 12:49 pm
- Location: Chungnam National University
Re: weakly and partially stratified estuaries using ROMS
Hi Jay.
When running an idealized estuary, the depth and width convergence will affect the result. For a long prismatic channel, the tides will tend to be more progressive, and you can get Stokes drift and Stokes return flow dynamics.
So you can be aware that the initial estuarine shape will affect the tide result, which is important in my opinion, especially if morphodynamic evolution isn't accounted for (not letting morphodynamics adjust from initial conditions).
The trumpet shape COAWST test case has a quite strong convergence at the mouth (convergence length 3 km if I recall) leading to a prismatic channel.
In that one, the estuary is about 100 km long, and depending on your boundary condition on the landward side, the landward side can be somewhat influenced.
Longer estuarine channel may avoid some impact of the landward boundary condition on the estuary.
I'm not sure the initial salt condition you are using. Running the naive initial condition of ocean salinity and having the freshwater form the estuary from the beginning can take quite a while.
I found using an initial salinity gradient speeds it up. Also, applying a weak, background along-shore coastal current (~5 cm/s) can avoid bulges of freshwater forming in the estuary (if you are including a shelf region).
That was a method used in Hetland and Geyer (2004): https://journals.ametsoc.org/view/journ ... b_body=pdf
I published two papers which used COAWST to model idealized estuaries of several types (strongly stratified, partially mixed, periodically stratified, well-mixed).
I'll link them here in case they are useful:
paper 1: https://agupubs.onlinelibrary.wiley.com ... 21JC017829
paper 2: https://www.frontiersin.org/articles/10 ... 35501/full
For my case, I used the estuarine parameter space to guide what tide and river I needed to make periodic stratification. It took some trial and error.
Depending on whether your estuary has progressive or standing tides will change how periodic stratification occurs in my expectation.
- Steven
When running an idealized estuary, the depth and width convergence will affect the result. For a long prismatic channel, the tides will tend to be more progressive, and you can get Stokes drift and Stokes return flow dynamics.
So you can be aware that the initial estuarine shape will affect the tide result, which is important in my opinion, especially if morphodynamic evolution isn't accounted for (not letting morphodynamics adjust from initial conditions).
The trumpet shape COAWST test case has a quite strong convergence at the mouth (convergence length 3 km if I recall) leading to a prismatic channel.
In that one, the estuary is about 100 km long, and depending on your boundary condition on the landward side, the landward side can be somewhat influenced.
Longer estuarine channel may avoid some impact of the landward boundary condition on the estuary.
I'm not sure the initial salt condition you are using. Running the naive initial condition of ocean salinity and having the freshwater form the estuary from the beginning can take quite a while.
I found using an initial salinity gradient speeds it up. Also, applying a weak, background along-shore coastal current (~5 cm/s) can avoid bulges of freshwater forming in the estuary (if you are including a shelf region).
That was a method used in Hetland and Geyer (2004): https://journals.ametsoc.org/view/journ ... b_body=pdf
I published two papers which used COAWST to model idealized estuaries of several types (strongly stratified, partially mixed, periodically stratified, well-mixed).
I'll link them here in case they are useful:
paper 1: https://agupubs.onlinelibrary.wiley.com ... 21JC017829
paper 2: https://www.frontiersin.org/articles/10 ... 35501/full
For my case, I used the estuarine parameter space to guide what tide and river I needed to make periodic stratification. It took some trial and error.
Depending on whether your estuary has progressive or standing tides will change how periodic stratification occurs in my expectation.
- Steven