Error with Solute Transport File Conversion

[Olawale Quadri Ogunsola]

Hello all,

I have used Python scripts (link) to convert from ATS 1.2 (Docker version) to 1.5, and everything has been behaving as expected. After converting the .xml file from ATS 1.5 to 1.6, I got this notification: 

Found a subcycling MPC with subpk types: ['coupled water', 'surface subsurface transport']

... changing to type "coupled flow and transport"
fixing transport pk
found MD for names: ['salt']
searching for LWC = water_content
fixing transport pk
found MD for names: []
searching for LWC = surface-water_content. 

However, I am encountering the error message below whenever I run the previously working ATS 1.5 .xml file after upgrading to 1.6.

'Errors::Message'

  what():  Evaluator "total_component_concentration" @ "" cannot be created in State. Verify (1) SetEvaluator is called or (2) name exists in state->evaluators.

terminate called after throwing an instance of 'Errors::Message'.

Please see the attached error log and the Python script used for the conversion from ATS 1.5 to 1.6.

I also searched the Python script for "total_component_concentration" and found no changes related to  "total_component_concentration"   in the file.

I would also love to follow up on the depth to groundwater or any closely related variable for the saturated zone.

Thank you for your continued support, and I hope to hear from you soon.

Best regards,

Olawale

[Phong Le]

Hi,

In v1.6, the variable "total_concentration" has been changed to "mole_fraction". Probably, in your observations, you write "total_concentration" and it is not available anymore. That's why you see this error.

To fix this, change "total_concentration" in your XML file to "mole_fraction".

Phong

[Olawale Quadri Ogunsola]

Hi,

Thank you very much. I just did that now and I have having another error "terminate called after throwing an instance of 'Errors::Message'
  what():  Cannot find region of name "computational domain" in the geometric model." Please, do you have any recommendations on this?

Thank you.

Olawale

[Phong Le]

You don't have a region named: computational domain in your regions tag in XML. Probably, through the input converter, it changes to subsurface domain
Solution: change the subsurface domain defined in regions tag to computational domain.

[Olawale Quadri Ogunsola]

Hi,

Thank you for your reply. Please could you shed more light on this?

I hope to hear from you.

Best regards,

Olawale

[Phong Le]

I don't have your xml file, so I can only guess here. If you look at the ParameterList regions in your XML file, there is no "computational domain" defined as below. However, in some of your PKs, you use the name "computational domain". It is undefined, thus you got this error.

I guess the converter changes the "computational domain" to "subsurface domain". So the easiest solution is to change "subsurface domain" defined in regions back to "computational domain".

Thanks.

<ParameterList name="regions" type="ParameterList">

<ParameterList name="computational domain" type="ParameterList">

<ParameterList name="region: all" type="ParameterList">

</ParameterList>

</ParameterList>

[Olawale Quadri Ogunsola]

Hi,

Thank you for your reply. Do you mean something similar to what is attached?

[Sergi Molins Rafa]

Olawale, 

No that is just a point region. Your problem is the with the region that defines the whole subsurface domain.  

What you need to do is to go through the input file carefully and make sure that the region that defines your entire subsurface domain is referred to as consistently (with the same name) throughout. That is if in the Regions block it is named "subsurface domain" you should use the same name elsewhere where the code needs to know you apply something to the entire subsurface domain. You can choose the name, if that is easier for you.

Sergi

--
You received this message because you are subscribed to the Google Groups "Amanzi-ATS Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ats-users+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/ats-users/0c514d39-6a68-478f-82c9-66b3770c683an%40googlegroups.com.

[Olawale Quadri Ogunsola]

Hi,

Thank you for your continued support. Do you mean this?

For the region name, I have checked, and the naming is consistent. Are there any other things I should watch out for?

Best regards,

Olawale

[Bo Gao]

Hi Olawale,

No, the error is caused by inconsistent naming for the region of your subsurface domain. In your input file, go to regions, you should be able to see the region name of your subsurface domain. Throughout your inputfile, wherever you want to define something or use something in or from your subsurface domain, you should use the same name for that subsurface domain to let the code locate the exact region. The error "Cannot find region of name "computational domain" in the geometric model." means you used the name "computational domain" somewhere in your input file in order to define/use something from the region "computational domain". However, in the regions section of your inputfile, it is not named "computational domain" but something else. So check your inputfile to make sure the names of regions are consistent.

Best,

Bo Gao

To view this discussion visit https://groups.google.com/d/msgid/ats-users/a139c7ae-0a5a-4ab2-9a4d-c6121daa7cf4n%40googlegroups.com.

[Olawale Quadri Ogunsola]

Hi Phong, Sergi, and Bo,

Thank you very much. It works now.

Best regards,

Olawale