More on rivers, ROMS2.0 to 2.1?

[evan]

Hi

The recent discussion about rivers has been of interest to me as I have just been adding a river to my configuration (using Biscay) of the Gulf of Cadiz. I have understood from the discussion that given my application - processes within the Gulf - and grid cell width ~4 km, depth 10 m at river mouth, I want my input to be in the top cell (or cells?). However I am not sure how I specify this - I believe it's Vshape in the NetCDF file, but what value will give me the top cell (N=23)? And are there any other parameters I need to set/be careful with?

A week ago, using ROMS2.0, I successfully ran the model with river data (I used Vshape=1). However I have just changed to ROMS2.1 (no problems compiling) and the identical configuration now blows up very abrubtly after ~5 model days. When I use an analytical river it runs fine, so it seems that there is something about my river data that ROMS2.0 is ok with, but not ROMS2.1. Does anyone have any suggestions about this?

Ok, I'll leave it at that for the moment.

Evan

[evan]

Following up on my previous post, I believe I have isolated the problem to the way I am using river_Vshape. Can anyone tell me what is the correct way to set this variable - given that I want the flow to be in the top cell (of 23)?

Thank you

Evan

[abever]

We set river_Vshape to be a 'number of rivers' by 'number of vertical levels' array.

for i=1:Nrivers
% 20% of transport in each of top 4 levels,
% then 10%, 5%, 3%, 1%, 1% below
% N=20 - the number of vertical layers
River.vshape(i,N )=0.20;
River.vshape(i,N-1)=0.20;
River.vshape(i,N-2)=0.20;
River.vshape(i,N-3)=0.20;
River.vshape(i,N-4)=0.10;
River.vshape(i,N-5)=0.05;
River.vshape(i,N-6)=0.03;
River.vshape(i,N-7)=0.01;
River.vshape(i,N-8)=0.01;
end


Also, I think I just read that you must have river temperature and salinity specified in version 2.1.

Aaron