Hi Eliézer Cláudio,

The rivers reach which joins to the sea you will be using the sea water level as the boundary condition. In this case in HEC-RAS you define the boundary condition as Known W.S and there you give the sea water level.

But for the river reaches 'A' which joins to another downstream reach 'B' the downstream water level is not fixed for your reach 'A'. The water level will change based on the discharge in the reach (as you mentioned - rating curve). In this case in HEC-RAS you will be defining the boundary condition with the junction name where the reach 'A' meets 'B'.

Hope it helps.

Daksiya. V

If subcritical, we usually give a known discharge as the upstream b.c while specify a depth as the downstream b.c..

You may use

1. A velocity/flow boundary at upstream and a pressure/water elevation at down stream;

2. A pressure/water elevation at upstream and a pressure/ water elevation at downstream.

I understand now.

But if I use the Surface Water Known as downstream boundary conditions, I will not have to calibrate restulados for this season downstream because I am already giving the exact amount. I can only compare the obserados and simulated results for the other stations upstream.

However, when I use the Rating Curve calibration, I can compare the observed and simulated results for all seasons.

That is my question actually: I buy the observed and simulated results for the downstream station that I'm using as a boundary condition or only to the other upstream?

There are two points I would like to make:

1. You calibrate a model with the data provided for the boundary condition for one status and for one or few points (if you would like to optimise a calibration parameter in the domain) in the domain, and then, verify the model at other points with the same boundary condition, or, at the same point with alternative boundary conditions.

2. Using a rating curve should not affect the results for a steady flow, as you are running the model for a certain discharge, hence a certain corresponding water elevation results. However if the discharge of your study is variable with respect to time, then the flow is unsteady, in which case you need to apply the variable water elevation as the boundary condition.

To summarise for a steady flow, you need to have a water elevation for the specific discharge you are running the model for at downstream boundary, and a couple of stations (say at least three) in the domain, where you can use one pint for calibration and the others for verification.