How to offset start of year?

I have a dataframe of oceanographic measurements organized by sample date:

The oceanographic year (here) is based on upwelling, so Instead of Jan-Dec, it runs from ~ March of year 1 to Feb of year 2. I’d like to adjust the year start/end dates so that I can organize the data by oceanographic year without altering the calendar year sampling date_time.

Using this code with lubridate almost works:
upw.year ← quarter(Date_time, with_year = TRUE, fiscal_start = 3)

but the oceanographic seasons aren’t in quarters exactly. Does anyone know a function that is slightly more flexible that I could use, or another way to go about it?

Any tips are much appreciated!