Does CAM6 skip Feb 29 when reading daily emissions in a NO_LEAP run?

[RuiXu]

Hi everyone,

I am running CESM2.1.5 with CAM6, and the case calendar is NO_LEAP.

I created daily, CAM-readable biomass burning emission forcing from CAMS GFAS for 2003-2023. My forcing file retains February 29 in leap years. Its time metadata is:
dimensions:
time = UNLIMITED ; // (7670 currently)
lat = 192 ;
lon = 288 ;

variables:
double time(time) ;
time:calendar = "gregorian" ;
time:units = "days since 1750-01-01 00:00:00" ;
time:long_name = "Time" ;
int date(time) ;
date:units = "YYYYMMDD" ;
date:long_name = "Date" ;
date:_FillValue = -2147483647 ;

I attach the full NetCDF header below.

When comparing the simulated AOD time series with reanalysis data, I noticed what appears to be an approximately one-day offset. When the model advances from February 28 to March 1 in a leap year, will CAM6 automatically skip the February 29 record in the forcing file and read the correct March 1 emissions? Or could retaining February 29 data cause the forcing data read by CAM to be misaligned?

Thank you very much!