Remove W3_IS2 compile guard from IC5 floe diameter update in w3wavemd

[ezhilsabareesh8]

This PR fixes the missing W3UIC5 update in IC4 builds by removing incorrect #ifdef W3_IS2 guards in w3wavemd.F90.

In ACCESS-OM3 IC4 IS0 configurations, CICE correctly exports floe diameter as Si_floediam, and WW3 imports it into ICEP5, but the W3UIC5 call that copies ICEP5 into ICEF was not compiled because W3_IS2 is not defined. As a result, ICEF stayed at its hard-coded initial value of 1000 m throughout the run.

The fix removes the unnecessary compile guards while keeping the existing runtime check, refer here:

IF ( FLIC5 .AND. DTI50.NE.0. ) THEN
  CALL W3UIC5( FLFRST )
END IF

Validation with access-om3/pr242-1 confirms the fix: ICEF is no longer constant 1000 m, but varies from the CICE FSD floor value of 5.4 m up to ~1255 m, consistent with 2 × fsdrad.

This restores the intended CICE → CMEPS → WW3 floe-diameter coupling path for IC4 wave-ice scattering.

The fixed build produces spatially varying ICEF values instead of the constant 1000 m initialisation value:

Day	Ice cells	ICEF min	ICEF mean	ICEF max
2 Jan 1958	7,907	5.4 m	20.8 m	830.8 m
3 Jan 1958	8,174	5.4 m	41.3 m	839.8 m
4 Jan 1958	8,356	5.4 m	62.0 m	1101.6 m
5 Jan 1958	8,478	5.4 m	75.3 m	1239.9 m
6 Jan 1958	8,534	5.4 m	84.9 m	1255.2 m
Buggy build	17,016	1000.0 m	1000.0 m	1000.0 m

[anton-seaice]

Are there cases where this code shouldn't be included ?

should this say

#if defined(W3_IS2) || defined(W3_IS0) or something ?

[ezhilsabareesh8]

Thanks @anton-seaice, I checked this, and I don’t think this should be changed to:

#if defined(W3_IS2) || defined(W3_IS0)

The reason is that this update is not really specific to either "IS2" or "IS0". "W3UIC5" is the generic updater for the IC5 floe-diameter input field: it copies the imported forcing field "ICEP5" into the internal WW3 fields "ICEF" / "ICEDMAX".

So the correct gate is the existing runtime condition:

IF ( FLIC5 .AND. DTI50.NE.0. ) THEN
CALL W3UIC5( FLFRST )
END IF

"FLIC5" is set from "INFLAGS1(-3)", i.e. whether the IC5/floe-diameter input is actually active. In ACCESS-OM3 this is enabled by "wav_coupling_to_cice = .true." / "ICE_PARAM5 = C". If IC5 forcing is not active, "FLIC5" is false and this block does not run.

So there are cases where the code should not execute, but that is already handled by "FLIC5". I don’t think there are cases where it needs to be compiled out based on the selected sea-ice source/scattering switch.

Using "#if defined(W3_IS2) || defined(W3_IS0)" would fix the current "IC4 IS0" case, but it would still couple a generic forcing-field update to particular ice physics switches. That seems more fragile than using the existing runtime forcing flag.

[anton-seaice]

Ok thanks