!!!Fatal Error

[Veronica Escobar]

Hi Steve,

I was running SHETRAN and the model stopped, the legend that appeared in the PRI file was as follow.

!!!FATAL ERROR 1060 at time = 20883.40 hours

OC flow criteria could not be met

I would like to know if this means that I'm using a low stickler number therefore is not possible to reach an adequate flow velocity. What is the meaning of this error?

Hope you can help me. Thanks

Vero

[Steve Birkinshaw]

Hi Veronica,

Occasionally when you do a simulation this error occurs. The problem is generally at the confluence of river channels during a recession period. Normally the timestep is too large to find a solution and the error occurs.

As you probably know in Shetran there is a basic timestep (normally 1 hour) and during heavy rainfall this decreases and then gradually increases back to the basic timestep. The reduced timestep is to produce a stable solution in the overland flow component of Shetran. if there is a lower Strickler coefficient the flow through the catchment is slower and the timestep can increase back to its basic value too quickly.

So the normal solution is to change a parameter (called PALFA) so the timestep returns to its basic value slower.

If you are using the standard version of Shetran with text files this is the second parameter here in the frd file:

:FR20 - BASIC TIMESTEP DATA                                                    
 0.5000 0.0500 9999.0 1.0000      T

so reduce the value to 0.0500 as shown here (or even down to 0.02).

If you are using the new Easy Setup then reduce the value below in the xml file:

<IncreasingTimestep>0.05</IncreasingTimestep> speed of increase in timestep after rainfall back to the standard timestep. The standard value is 0.05. If if there are numerical problems with error 1060 it can be reduced to 0.01 but the simulation will take longer.

If it still does not work then let me know again.

In the long term I need to change the code and automatically reduce the timestep in these circumstances.

I hope this helps,
Steve

[Veronica Escobar]

Hi Steve;

Thanks for the quick reply. The problem do not appears in some simulations but is still there in others, even with an PALFA value of 0.01. What I do not understand is that if I'm using the same catchment (rainfall, evapotranspiration, soil types, etc.) and just changing the types of land use why not all the  simulations that present this problem stopped in the same date (although they are periods of drought).

How can I solve this problem? Is possible to wrote a smallest value of 0.01?  The simulations with 0.01 take a little more time that those with 0.02 so is fine for me to use a less value (if is possible).

 

 Vero

[Steve Birkinshaw]

Hi Veronica,

It is a bit strange. Basically small changes in the flow (such as from a different land-use) can cause the error. This is why it is quite a hard problem to fix properly. If you send me a private email with all the input files I will have a look. I can also send you an executable that completes the simulation even with this problem (it does not solve it - it is just a fix).

Regards,
Steve