[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
20041019: Bug in ut_calendar?
- Subject: 20041019: Bug in ut_calendar?
- Date: Tue, 19 Oct 2004 12:29:53 -0600
Dennis,
>Date: Tue, 19 Oct 2004 11:47:42 -0600 (MDT)
>From: Dennis Shea <address@hidden>
>Organization: UCAR
>To: address@hidden
>Subject: Re: 20041019: Bug in ut_calendar?
The above message contained the following:
> I was the one who 1st reported this to Mary Haley.
> I recently obtained the dataset from the Climate Research Unit.
> It is one of the datasets used in the IPCC assessments.
>
> http://www.cru.uea.ac.uk/cru/data/temperature/#datdow
>
> specifically: hadcrut2v.nc
>
> netcdf hadcrut2v {
> dimensions:
> lat = 36 ;
> lon = 72 ;
> time = 1616 ;
> variables:
> short temanom(time, lat, lon) ;
> temanom:long_name = "temperature_anomaly" ;
> temanom:units = "celsius" ;
> temanom:scale_factor = 0.01 ;
> temanom:valid_min = -9000 ;
> temanom:missing_value = -9999 ;
> float lat(lat) ;
> lat:units = "degrees_north" ;
> float lon(lon) ;
> lon:units = "degrees_east" ;
> short time(time) ;
> time:units = "months since 1870-1-1" ;
> data:
>
> time = 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
> 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37,
> [snip]
>
> =================
> I am sure they [me, Mary, others I have talked to] assumed that
> time=0 ==> 187001
> 1 187002
> etc
>
> I can see that this is erroneous based upon the above.
>
> I would lay odds that errors have been made as a result.
> Seems "months since ..." is a bad unit to use.
Yup. People (scientists even) have some odd ideas of what constitutes
a unit of a physical quantity. I had one person who thought "flight
level" should be a unit.
> Mary ... maybe NCL's documentation for the udunits
> software should contain some explicit mention of this issue.
Regards,
Steve Emmerson