Problems accounting for ET & infiltration

Hi Chris

PLEASE would you give me some advice with this problem:

I am running an unsteady HECRAS application of a RiverCad project for a wetland. I have successfully calibrated the model for using the following boundary conditions:

1. Inflow hydrograph for the main river.
2. Lateral inflow hydrographs for tributaries.
3. Uniform lateral inflow hydrographs to characterise the rainfall contributions (rain falling on inundated areas).

.... the problem is to account for losses due to ET,infiltration and the ponding of water that doesn't drain from the catchment. I have a way of estimating the time-series of these losses, but how to include them in the model - If i include them as negative values, uniformly distributed along the length of the floodplain, there are periods when my estimates are near the flow in the river, or greater, and the model goes unstable due to near zero or negative flow rates. Isn't there any way in HECRAS to prevent this? I can't find anything.

The only way I have found is to extract all the losses from the inflow.. BUT they are quite large at times (taking all flows from the system along the length of the floodplain), and so I end up with a very inaccurate inflow hydrograph and concommitant stages at the upstream boundary --- improving towards the outflow, where it they should be correct if my estimates are good.

I really need a better way of extracting the losses, but can't figure how to do so along the length of the floodplain with creating instabilities.

As a point of interest, I have found that the minimum setting for the hydrographs in the boundary condition window does not work if one links the data to the HEC-DSS - it only appears to work for manually typed-in data.

I would really appreciate it if you could shed some light on this difficulty of mine.

Regards
Drew.

Dr Drew Birkhead
Streamflow Solutions
Johannesburg
South Africa.

Dear Dr. Drew Birkhead :

I think that you are hitting a limitation of the HEC-RAS model.

However, one item you can try is to specify lateral pumps in your model. You can define pumps along the river, setting the pumps to turn on and off, as well as pump at various rates as the water surface elevation rises. In that way, you can simulate the evaporation and infiltration at various rates depending upon how much water is flowing at a cross section. This solution is not perfect, but will allow you to work around this limitation.

If you have any additional questions or need any further information, please contact me again.

Sincerely,

Chris E. Maeder, M.S.
Senior Technical Engineer
chris.maeder@bossintl.com

Thanks Chris

I will give this a try - seems a better way of doing it.

Regards
Drew.