Access of invalid array index

[gold2718]

Brief summary of bug

Running a DEBUG test (see details below) gives the error:

forrtl: severe (408): fort: (3): Subscript #1 of the array IRRIGATED has value -9999 which is less than the lower bound of 0

General bug information

CTSM version you are using: ctsm5.4.002_noresm_v3

Does this bug cause significantly incorrect results in the model's science? Yes

Configurations affected: HIST_CAM70%LT%NORESM%CAMoslo_CLM60%FATES-NOCOMP_CICE_BLOM%HYB%ECO_MOSART_DGLC%NOEVOLVE_SWAV_SESP_BGC%BPRP

Details of bug

Running test, ERS_D.ne30pg3_tn14.NEHIST.betzy_intel.allactive-defaultio.test_emis, where NEHIST is the compset above yields (on Betzy):

1401: forrtl: severe (408): fort: (3): Subscript #1 of the array IRRIGATED has v
alue -9999 which is less than the lower bound of 0
1401:
1401: Image PC Routine Line Source

1401: cesm.exe 000000000E8B161F Unknown Unknown Unknown
1401: cesm.exe 000000000A783231 irrigationmod_mp_ 1627 IrrigationMod.F90
1401: cesm.exe 000000000A77B06D irrigationmod_mp_ 1478 IrrigationMod.F90
1401: cesm.exe 00000000089E2AF8 clm_driver_mp_clm 834 clm_driver.F90
1401: cesm.exe 00000000088A9D28 lnd_comp_nuopc_mp 915 lnd_comp_nuopc.F90

### Important details of your setup / configuration so we can reproduce the bug

This does not appear to crash the model when not running in debug mode, however, adding a `write` statement just before line 1627 generates 41860268 lines of output (I think all with the -9999 index).

[mvdebolskiy]

@gold2718 it's a known bug, irrigate=.false. should be specified.

[mvertens]

All of our NHIST compsets have irrigate=.true. by default. I think it would be important to ensure that any new historical simulations turn this off.
@ tjiputra - could you please restart your simulation and setirrigate = .false.in user_nl_clm
I believe that all historical simulations we have done so far are suspect - if they have irrigate = .true.

[gold2718]

@gold2718 it's a known bug, irrigate=.false. should be specified.

I did not know about it. I looked for an open issue before reporting this. If irrigate has to be .false., can this be made the default?

[maritsandstad]

@gold2718 it's a known bug, irrigate=.false. should be specified.

I did not know about it. I looked for an open issue before reporting this. If irrigate has to be .false., can this be made the default?

This sounds like a reasonable thing to do as I see it. Do we quickly want to make a PR for that and try to have it in along with our other recent fixes @mvdebolskiy ?

[mvdebolskiy]

If we need hist compsets today. Though we will not have propet initial files for them ubtil the 15 pfts are sorted out.

[gold2718]

This NHIST experiment has irrigate = .true.

• nhist.ne16pg3_tn14.nhist_cmip7.constLU.20260118 noresm3_dev_simulations#313

Is it still active?

[mvertens]

@tjiputra - In looking at the experiment

/cluster/projects/nn9560k/tjiputra/NorESM/NorESM3_dev_cases/nHISTesm.ne16pg3_tn14.noresm3_0_beta03b.2025-11-04/
irrigate = .false.

However in the latest experiment

/cluster/projects/nn9560k/tjiputra/NorESM/NorESM3_dev_cases/nHISTesm.ne16pg3_tn14.noresm3_0_beta11_20260225/
irrigate = .true.

So this is definitely problematic. However, I'm not seeing nHISTesm.ne16pg3_tn14.noresm3_0_beta11_20260225 as an entry in noresm_dev_simulations.

[maritsandstad]

If we need hist compsets today. Though we will not have propet initial files for them ubtil the 15 pfts are sorted out.

Is there a reason to not just add this default change asap?