As an update I was able to get the new version of the data with the correct coordinates by adding a “nocache“ parameter with a unique numeric string, as described here:
unique_numeric_string = str(int(time.time()))
params = {
"variable": variables,
"year": str(year),
"month": [str(month).zfill(2)],
"day": days,
"time": [f"{str(h).zfill(2)}:00" for h in range(0, 23 + 1)],
"area": [ymax, xmin, ymin, xmax],
"data_format": "netcdf",
"download_format": "unarchived",
“nocache": unique_numeric_string, # <-- add this to invalidate the server cache
}
With this I was able to make sure i have all the new data coordinates regardless of old data cached on the server.