Date incorrect GOES-R test files

[joleenf]

Hi,

I have a GOES-R test file that I am displaying in McIDAS-V. I think the date is being interpreted incorrectly. The file should have a nominal time of 20-Aug-2015 0:15 UTC, instead, the data object has a time of 2015-Feb-05 at 2:08 UTC.

This is the from the same test file series that I provided for the unit="1" question: http://mcidasv.ssec.wisc.edu/forums/vie ... 46479f3c5a

Thanks,
Joleen

[mhiley]

Hi Joleen,

Currently the timestamp in the McV display comes from the "t" variable in the NetCDF file, which for that particular file has a value of "476374086.93498" in units of "seconds since 2000-01-01 12:00:00".

The following Python code seems to confirm that 02:08UTC is the correct timestamp for that value of "t":

Code: Select all

from datetime import datetime, timedelta
print datetime.strptime('2000-01-01 12:00:00', '%Y-%m-%d %H:%M:%S') + timedelta(seconds=476374086.93498)
# result: 2015-02-05 02:08:06.934980

(note: that is not the actual code that determines the timestamp you see in the GUI, so it should be a fairly independent test. The GUI's timestamp is coming from netcdf-java).

So, it seems like the GUI is interpreting the "t" variable correctly. Is there any chance there is a bug with these files and they just have a bad "t" value?

Mike

[joleenf]

Mike,

We are guessing that the HH:mm:ss time is correct, because that is about halfway between the start and end HH:mm:ss, but the month, date, year is not correct in the file, this might the day when the test system produced the data.

Thanks for verifying the code.
Joleen

[wstraka]

As I mentioned to Joleen, we're going to wait until we get updated files to see if this issue was fixed in the GS.