Replies: 1 comment 1 reply
-
|
Thanks for writing @Fu-Weitao. We would recommend that you update to the latest KPP 3.1.1 anyway, since there have been several new features and fixes introduced since KPP 2.2.3. I am not familiar with BOXMOX so I cannot provide any insight. What you might try is to adjust the tolerances ATOL and RTOL for the Rosenbrock solver. You should be able to do this on a per-species basis. |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
I was using BOXMOX (https://mbees.med.uni-augsburg.de/gitlab/mbees/boxmox), which is built upon KPP 2.2.3. When using MOZART-T1 chemical mechanism, the program sometimes aborted, showing "Forced exit from Rosenbrock due to the following error: --> Step size too small: T + 10*H = T or H < Roundoff". I read https://gchp.readthedocs.io/en/stable/geos-chem-shared-docs/supplemental-guides/error-guide.html, and noticed that the error might be related to species concentrations or reaction rates being unusually high or low. So I looked into the output, and found there is indeed one species that reach a ppb value of ~0.38e-167.
I tried to replace the codes (drv, int, src, util) related to KPP 2.33 with KPP 2.6, ./compile, make, make install, but encountered some errors when compiling "MOZART_T1" that I was unable to resolve, because I'm unfamiliar with Fortran.
I wonder if there is any snippet in the newest KPP version that deals with program failure when some chemical species reach very low/high values, so that I can apply it as a patch to BOXMOX-KPP 2.2.3.
Thank you!
Beta Was this translation helpful? Give feedback.
All reactions