GCHP 14.2.2 simulation fails because of KPP error
Closed this issue · 2 comments
Name and Institution (Required)
Name: Hansen Cao
Institution: University of York
Confirm you have reviewed the following documentation
Description of your issue or question
Hello,
My GCHP fullchem simulation (v14.2.2 with debromination on by default) at C48 failed due to KPP error. I tried tightening the convergence criteria, which did not help. It seems that the KPP error is related to halogen explosion, see below the species concentrations in the problem box (4 7 16):
-9.3472878502639567E+019 HOBr
9.3457223154611585E+019 Br2
-9.3475157765068063E+019 BrSALA
The concentrations of these species in that grid box seems normal before they getting into the integrator:
13196.212474068439 BrSALA
6762.2799453671641 Br2
17365924.016521949 HOBr
My GCHP stretched simulation (fullchem, v14.2.2, with debromination on) at C200 (SF2.0) encountered the similar KPP error, wherein the explosion happened to iodine species in grid box (19 16 7):
1.2255442983974835E+018 HOI
1.1386758559108652E+021 I
-4.8607138921225430E+021 IO
Before entering the integrator, the iodine species concentrations also seemed normal:
384693.52015368728 HOI
176312.87417442043 I
225570.16518929612 IO
I wonder if you could provide any suggestions on solving this problem? Also, how do I convert the grid index (e.g. here the problem box in the C48 simulation, 4, 7, 16) in the GCHP into latitude and longitude? Thank you!
Please provide as much detail as possible. Always include the GCHP version number and any relevant configuration and log files.
Hi @hazel008, you may be encountering this known issue in 14.2.0: geoschem/geos-chem#1982.
The fix is in geoschem/geos-chem#2006 which is included in version 14.3.0. We are currently benchmarking 14.3.0 but you can download it by checking out the dev/14.3.0 branch at the top-level of the GCHP source code directory (git fetch; git checkout dev/14.3.0) and then running git submodule update --init --recursive.