[Bug]: 3.30.24 prerelease difference in seasonal model forecasts

[CaitlinAkselrud-NOAA]

Describe the bug

I ran the 3.30.24 prerelease on the 2025 sardine update assessment files. Model runs and converges. The timeseries values are the same until the forecast years, and only for season 2. For bio_all, the season 2 forecast value is about 1/2 of the update assessment value. Other values look similar except mature_num, which is about 1/4 of the update assessment value.

Was there a setting change for this pre-release that I need to update to generate consistent numbers?

Note: I tried using the starter file compatibility marker. Adding the compatibility = 0 still yields the same result. Setting it = 1 results in all the season 2 values being equal to 0.

To Reproduce

Input files are available here: https://drive.google.com/drive/folders/1vpFwAnOJvx7w1x8y4ZaCAptbqqZ_f4UM?usp=drive_link

Expected behavior

I expected forecast values with SS 3.30.22 to be the same as 3.30.24 prerelease when run with the same input files for both seasons.

Screenshots

Which OS are you seeing the problem on?

Windows

Which version of SS3 are you seeing the problem on?

3.30.24

Additional Context

No response

[Rick-Methot-NOAA]

I will take a look immediately.
I fear that sardine may be victim of this newly found and fixed bug: #691

[Rick-Methot-NOAA]

There is an unusual aspect of the sardine setup that has always been a bit of a challenge.

spawning is set for month 7 (July 1) and recruitment is set for month 1 (Jan 1) at age 0.
SS3 expects new recruits to be age 0 only in the year of spawning then to increment to age 1 on Jan 1.
This causes SS3 to throw this note, then to play through with your settings:

Note: Information: settle_month is less than spawn_month, so logical age at settlement calculated to be: 1 for settle_time 1. Does not match read value of 0 are you sure?

Then the first mature age is age 0 and they have a positive fecundity so contribute to the spawning biomass on the spawning date when they are age 0.
It appears that the logic in the forecast is slightly different than the logic in the time series in the revised SS3. I am following that thread.

[Rick-Methot-NOAA]

progression of numbers-at-age in second season of forecast is flawed.

[Rick-Methot-NOAA]

A line of code introduced in #691 caused an error in the unique situation of sardine that has spawning in one year and recruitment at age 0 in the following year. That line of code set natage(t,0) = 0, but close inspection of the code logic showed that the statement was unnecessary.
@CaitlinAkselrud-NOAA