River runoff problem

General scientific issues regarding ROMS

Moderators: arango, robertson

Post Reply
Message
Author
javadi
Posts: 115
Joined: Mon Jan 27, 2020 5:55 pm
Location: UoT

River runoff problem

#1 Unread post by javadi »

Dear all

I run the model but I have faced a problem. I have 4 rivers and define monthly discharge with salinity=3 and Temperature=12.

Here is the location of river points in my grid.

As my river comes from left to right, I define river transport as positive as mentioned here https://www.myroms.org/wiki/River_Runoff.

But after running, the model in the river runoff position shows more salinity compared to other points(top-left).

I cannot understand why?

The only thing that can be the wrong i,j? or transport?



please help me with your interesting comments
Attachments
Capture12.JPG
salt.JPG
salt.JPG (34.6 KiB) Viewed 14175 times

jcwarner
Posts: 1172
Joined: Wed Dec 31, 2003 6:16 pm
Location: USGS, USA

Re: River runoff problem

#2 Unread post by jcwarner »

it could be i, j, or transport.
make sure the Q is positive for flow to the right.
look at the u values and ensure the rivers are coming from the correct locations.
those locations should be masked points. the river flow will overwrite the zero values at the masked locations.

-j

javadi
Posts: 115
Joined: Mon Jan 27, 2020 5:55 pm
Location: UoT

Re: River runoff problem

#3 Unread post by javadi »

Hi

Thanks for your reply!


I have checked u velocity. It is here. I guess that the flow is towards the left.

The transport is positive so this can not be the problem.

With regard to i,j, I consider i=15 and j=295 in my file(for example). is it right? the result is based on this approach!

now should i define i=16 and j=295 or i=14 and j=295?
Attachments
Capture25.JPG
56.JPG

jcwarner
Posts: 1172
Joined: Wed Dec 31, 2003 6:16 pm
Location: USGS, USA

Re: River runoff problem

#4 Unread post by jcwarner »

https://www.myroms.org/wiki/River_Runof ... _Eposition
is a good description of the river locations.
it needs to be coming out of the land masked cell.

User avatar
jivica
Posts: 169
Joined: Mon May 05, 2003 2:41 pm
Location: The University of Western Australia, Perth, Australia
Contact:

Re: River runoff problem

#5 Unread post by jivica »

A long time ago Rich Signell wrote very useful Matlab function called "map_rivers.m" (I can see you are using Matlab).
If you can't find it in the ROMS matlab repo or somewhere on the net, I will try to dig it out.

Hope this example (you are showing on the figures) is not something you want to use seriously, you need to define proper river inputs/model forcing/ grid etc...(this is the Adriatic Sea and there are publications with many studies related to ROFI and whole circulation is veeeery sensitive to all that...)

Good luck,
Ivica

User avatar
wilkin
Posts: 875
Joined: Mon Apr 28, 2003 5:44 pm
Location: Rutgers University
Contact:

Re: River runoff problem

#6 Unread post by wilkin »

The function roms_plot_river_source_locations.m in my roms_wilkin tools https://github.com/johnwilkin/roms_wilkin will plot the sources in Matlab with symbols to indicate the flow direction.
John Wilkin: DMCS Rutgers University
71 Dudley Rd, New Brunswick, NJ 08901-8521, USA. ph: 609-630-0559 jwilkin@rutgers.edu

javadi
Posts: 115
Joined: Mon Jan 27, 2020 5:55 pm
Location: UoT

Re: River runoff problem

#7 Unread post by javadi »

Hi

Thanks!


I have a dilemma about one point.

In my case, I have 40 layers.

I should define V_shape (s_rho)from 1 to 40 or 0 to 39?

Another dilemma would be the least number would be for the bottom!

Best,

Javad

User avatar
wilkin
Posts: 875
Joined: Mon Apr 28, 2003 5:44 pm
Location: Rutgers University
Contact:

Re: River runoff problem

#8 Unread post by wilkin »

river_Vshape has the properties ...

Code: Select all

        double river_Vshape(s_rho, river) ;
                river_Vshape:long_name = "river runoff mass transport vertical profile" ;
                river_Vshape:requires = "must sum to 1 over the s_rho coordinate" ;
so in your case dimension s_rho = 40. But whether you count from 0 to 39 or 1 to 40 depends only on the App you are using to read/write it. To distribute the river uniformly over all depths you want river_Vshape = 1/40 (i.e. = 0.025) for all values so it satisfies the property must sum to 1 over the s_rho coordinate.

If you want to weight the source toward the surface instead of being uniform, you make the weights larger toward higher values of s_rho because s_rho = 0 is the bottom and s_rho = (here 40) is the surface.
John Wilkin: DMCS Rutgers University
71 Dudley Rd, New Brunswick, NJ 08901-8521, USA. ph: 609-630-0559 jwilkin@rutgers.edu

javadi
Posts: 115
Joined: Mon Jan 27, 2020 5:55 pm
Location: UoT

Re: River runoff problem

#9 Unread post by javadi »

I try to define a profile thus I weigh in different layers. Here is it

I considered s_rho=1 to s_rho=40 is not right?


Do you mention that I should s_rho=0 to s_rho=40? It will be 41 layer?
Attachments
v_shape.JPG
v_shape.JPG (14.97 KiB) Viewed 13972 times

User avatar
wilkin
Posts: 875
Joined: Mon Apr 28, 2003 5:44 pm
Location: Rutgers University
Contact:

Re: River runoff problem

#10 Unread post by wilkin »

Here s_rho is just an index for the vertical dimension. You say you have 40 vertical layers, i.e. you have N = 40 in ROMS. If you are working in Python then the index is from 0 to 39, if you are in Matlab the index is from 1 to 40.

That plot is an oddly varying profile for river_Vshape. You do understand what river_Vshape does, right? It partitions the inflow river_transport into the vertical cells. This doesn't have a huge impact on the solution because ocean dynamics quickly adjust the profile.
John Wilkin: DMCS Rutgers University
71 Dudley Rd, New Brunswick, NJ 08901-8521, USA. ph: 609-630-0559 jwilkin@rutgers.edu

Post Reply