Overland flow model doubts

[Eliana Medeiros Ferreira da Silva Gallo]

Hello.

I have used OGS for groundwater flow model and it was fine. But, now I should couple it with overland flow model. I could not find a tutorial for overland flow model with OGS as there is one for groundwater flow model with OGS. Do you know any tutorial for overland flow model with OGS? 

At first sight, I have the following questions:

1) As benchmark, there are  "Wool_quad", "govin_quad", "govin_line", "gian_tri" and "gian_quad".  I don't know which one to follow. What are the assumptions for each case? 

2) Why are there no boundary condition file for "gian_tri" and "gian_quad"?

3) I look it up the file format explanations in OGS website, but they don't clarify the terms used for overland flow model. What are the parameters defined for surface friction in medium properties file? and the "new" (corresponding "NUM_TYPE") in process file?

4) What is the ".rfd" file?

Thank you a lot for your time.

Cheers,

Eliana

[Johannes Boog]

Dear Eliana,

3) NUM_TYPE refers to the definition of numerical methods. I gues that this keyword has ben deprecated.

4) *.rfd file allows you to define time series) that can be used, for example,  as boundary conditions in the *.bc file. You define a time series by setting the key word #CURVES followed by time steps (t) and corresponding values ( f(t) ). you can define several time series (curves) in one file. the curves are automatically numered starting with 1. make sure to set the time steps according to your time stepping scheme defined in *.tim.

Example:

*.rfd:
//---------------

#CURVES
// t    f(t)

0    0

100 150

200 0

300 0

#CURVES

0    88
100 11

200 66

#STOP

//----------------

You refer to these time series in the *.bc file by setting the keyword $TIM_TYPE followed by the corresponding curve number.

Example *.bc bloc:

#BOUNDARY_CONDITION
 $PCS_TYPE
  MASS_TRANSPORT
 $PRIMARY_VARIABLE
  Tracer
 $GEO_TYPE
  POINT POINTVAin
 $DIS_TYPE
  CONSTANT 1.0     // define mulitplier, the  values of the curve are muliplied by this value
  $TIM_TYPE

   CURVE 1       // define  curve number to use

#BOUNDARY_CONDITION
 $PCS_TYPE
  MASS_TRANSPORT
 $PRIMARY_VARIABLE
  Tracer
 $GEO_TYPE
  POINT POINTVAin
 $DIS_TYPE
  CONSTANT 2.0     // now you would double all the values of your curve
  $TIM_TYPE
   CURVE 1       // define  curve number to use

ogs automatically interpolates the function values if your time stepping scheme in smaller then defined steps in #CURVES.

FYI: The *.fct file allows to define user functions in general (x, f(x) ), which can be called by names within *.bc blocs or blocs of other input files. This maybe more convenient if you have a lot of functions/time series and do not want to get irritated by too many curves numbers.

Best,
Johannes

Am Donnerstag, 10. Mai 2018, 09:35:54 MESZ hat Eliana Medeiros Ferreira da Silva Gallo <eliana...@gmail.com> Folgendes geschrieben:

--
You received this message because you are subscribed to the Google Groups "ogs-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ogs-users+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Eliana Medeiros Ferreira da Silva Gallo]

Thanks a lot!

4) Only a feedack. In overland flow benchmark, the .rfd file is used to define the $TIME_STEP for the sorce term file simulating the precipitation.

As far as I could notice, the "gian_tri" and "gian_quad", both could refer to the example in the book "Thermo-Hydro-Mechanical Processes in Fractured Porous Media", for the surface flow on a tilted V-Catchment.

Concerning the mesh for the overland flow model:

5) Should I generate it just like as it is in the tutotrial "OpenGeoSys Tutorial Computational Hydrology II"?

Again thanks.

[Johannes Boog]

Dear Eliana,

4) then the curve in the *.rfd file defines the precipitation time series. ...the time units in the *.rfd file should be the same as defined in the *.tim file.

5) I do not have any experience with overland flow. maybe you should just try it.

Best,

Am Freitag, 11. Mai 2018, 10:38:02 MESZ hat Eliana Medeiros Ferreira da Silva Gallo <eliana...@gmail.com> Folgendes geschrieben:

[elianam...@gmail.com]

Hello again.

My overland flow model for a catchment, with more or less 60km2, located in high altitudes (from 4500m to 6100m) presents really high values for water depth (270m) and a really small flux at the outlet point even when model for a long time. According to the bechmark of V-catchment the CRITICAL_DEPTH should be 1, the OBEN polyline is the boundary line and there is only flux at the outlet where it is assigned as critical depth condition.

Questions:

6) Why is the critical depth assigned as 1? 

7) When and where is the polyline OBEN used in the model? What is the purpose of the OBEN polyline?

Cheers,

Eliana