retrieving time step GOTM-FABM

[Mattias Cape]

Good morning,

A model I'm working to implement (Davis et al. 2014, JGRO) contains a water column denitrification flux term requiring a model time step variable. From the paper:

where SD and LD are small and large detritus concentrations respectively (mmol m-3) and r a remineralization rate (d-1, converted to s-1 using a constant factor). 

I'm wondering how to retrieve the model time step from GOTM. I see a discussion here in fabm development which mentions the way FABM tracks time as provided by the physical model.

Thank you,

Mattias

[Jorn Bruggeman]

Hi Mattias,

 

The first, simple answer is that FABM does not have a mechanism to provide access to the model time step. That’s in part because any time step dependence in the model rates makes the solution dependent on the time discretization, which is not usually a desirable feature. Typically, models with such a time step dependence also implicitly make assumptions on the integration scheme used (almost always, Forward Euler is assumed), and that is something that can also vary between FABM hosts (Euler, leapfrog, and variable time stepping have all been used).

 

Still, that’s a bit of a purist take on things. The Davis et al. formulation seems essentially designed to avoid negative oxygen due to bacterial respiration. In essence, any respiration that would take oxygen into the negative realm with a Forward Euler scheme (delta_t*r(SD+LD) > O2) is attributed to denitrification instead. Such approaches are not uncommon. Since FABM does not currently expose the time step, though, the only solution is to have a parameter for the time step within the biogeochemical model, either configurable in fabm.yaml, or as a hardcoded “maximum time step expected”.

 

Cheers,

 

Jorn

 

 

From: fabm-...@googlegroups.com <fabm-...@googlegroups.com> On Behalf Of Mattias Cape
Sent: 30 November 2021 18:16
To: FABM-users <fabm-...@googlegroups.com>
Subject: [fabm-users: 861] retrieving time step GOTM-FABM

 

Good morning,

 

A model I'm working to implement (Davis et al. 2014, JGRO) contains a water column denitrification flux term requiring a model time step variable. From the paper:

 

where SD and LD are small and large detritus concentrations respectively (mmol m-3) and r a remineralization rate (d-1, converted to s-1 using a constant factor). 

 

I'm wondering how to retrieve the model time step from GOTM. I see a discussion here in fabm development which mentions the way FABM tracks time as provided by the physical model.

 

Thank you,

 

Mattias

--
You received this message because you are subscribed to the Google Groups "FABM-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fabm-users+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/fabm-users/bfa216f9-6ad9-4c5e-9c20-294d7c9f12den%40googlegroups.com.

[Mattias Cape]

Thank you, Jorn, makes perfect sense. I appreciate all the context, makes for a good learning opportunity.

You received this message because you are subscribed to a topic in the Google Groups "FABM-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/fabm-users/3hVS-K4oYRc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to fabm-users+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/fabm-users/015c01d7e61a%24bc5fc490%24351f4db0%24%40bolding-bruggeman.com.