floating point exception issue

[SmithSimmon]

Hi, everyone! I ran a simulation from 2015 to 2100 using CMIP6 data. The model crashed in November 11, 2030. I tried to skip this day and continue running the model, but after running for a few years, the model encountered the same error. the error is mainly about surfacEenergyFluxMod and ResistancecanopystomataBallBerryMod.
Here is error message:

year 2030

year 2036
the command line from error message in SurfaceEnergyFluxMod f90 file:
TempTmp = TempUnitConv(TemperatureGrdBare)
I don't know why I am meeting a floating-point issue. I have checked the temperature forcing data and radiation forcing data, and the values seem to be within a reasonable range. Could you please give me some advice?
By the way, is there any method to know the exact grid where the problem occurs？

[cenlinhe]

What model time step are you using? If you are using an hourly timestep (NOAH_TIMESTEP = 3600), then please reduce it to half-hourly (NOAH_TIMESTEP = 1800). The issue might be caused by the inaccurate numerical solution under a large timestep.
If you want to know the exact grid, you could add printout code in the NoahmpDriverMain module for the I, J point.

[SmithSimmon]

thanks for u help! I have solved this issue.