[Esip-documentation] time_coverage_duration and time_coverage_resolution for climatological data

Fri Mar 16 00:49:05 EDT 2018

Well, there are clearly two time durations related to this data set—one describing the duration of the whole set, and the other describing the duration of the slice that you’ve taken of the data set. Or, to look at it another way, the duration represented by the original data, and the duration represented by the calculated data. So I suppose there’s a solid argument about ambiguity. (I’m pretty sure we didn’t discuss this particular use case, if that helps!)

That said, I’d have to disagree with your science steward on this one. The data that is represented clearly represents a time coverage of 10 years, and if you use start and end times to convey that, it would be too strange for people (and software) to see a duration that is not equal to end-minus-start. Time coverage should mean the same thing for all 3 attributes. (Conversely, if you had used the start and end of the month as a time_coverage_start/end, that would be weird, because which year are we really talking about?)

If you need to reflect that this is not 10 years of data, that clue provided by the fact the time coverage resolution would be 1 month. If your whole data set is one point for each month of the year, that means you have 12 months of data, which indicates climatology pretty clearly to me. If your whole data set is just one point, you could still use 1 month for the resolution, to reflect the fact that if you have a second data point it would be 1 month away. (Which neatly solves the problem of figuring out how big an interval is represented by the one point.)

Note that time_coverage_resolution is the only one of the 4 attributes that refers to the _values_ in the data set, as opposed to the data set as a whole. So it makes sense to me that 1 month fits for that, but 10Y fits for the duration of the set as a whole.

These are metadata specifications not precise data descriptions, so it is fair to say that we’re at the edge of what’s realistic for ACDD to represent. That said, if a user is searching for data sets that span at least X years, they would want you to have used 10Y for your duration.

John

---------------------------------------
John Graybeal
jbgraybeal at mindspring.com

> On Mar 14, 2018, at 20:22, Ajay Krishnan - NOAA Affiliate via Esip-documentation <esip-documentation at lists.esipfed.org> wrote:
> 
> Hi Ed,
> 
> Thanks for your response but that still won't address the issue. We are making use of the time_coverage_start and time_coverage_end attributes (to give the user the idea that it is a decadal climatology). 
> We've omitted the time_coverage_resolution attribute since there is only one time stamp. 
> 
> Is it better to not include the resolution/duration attributes or any of the time_coverage attributes for that matter, for climatological products?
> 
> Thanks,
> Ajay
> 
> On Tue, Mar 13, 2018 at 3:39 PM, Armstrong, Edward M (398G) <Edward.M.Armstrong at jpl.nasa.gov <mailto:Edward.M.Armstrong at jpl.nasa.gov>> wrote:
> Hi Ajay,
> 
>  
> 
> I would suggest you could use this three attributes to described the time series start/stop and time step:
> 
>  
> 
> time_coverage_start
> 
> string
> 
> Describes the time of the first data point in the data set. Use the ISO 8601:2004 date format, preferably the extended format as recommended in the Attributes Content Guidance section.
> 
> time_coverage_start = "2016-09-01T08:12:01" ;
> 
> required
> 
> ACDD 1.3
> 
> time_coverage_end
> 
> string
> 
> Describes the time of the last data point in the data set. Use ISO 8601:2004 date format, preferably the extended format as recommended in the Attributes Content Guidance section.
> 
> time_coverage_end = "2016-09-01T08:17:59" ;
> 
> required
> 
> ACDD 1.3
> 
> time_coverage_resolution
> 
> string
> 
> Describes the targeted time period between each value in the data set. Use ISO 8601:2004 duration format, preferably the extended format as recommended in the Attributes Content Guidance section.
> 
>  time_coverage_resolution = "00:05:58";
> 
> recommended
> 
> ACDD 1.3
> 
>  
> 
> From: 
> 
> https://podaac.jpl.nasa.gov/PO.DAAC_DataManagementPractices <https://podaac.jpl.nasa.gov/PO.DAAC_DataManagementPractices>
> From: Esip-documentation <esip-documentation-bounces at lists.esipfed.org <mailto:esip-documentation-bounces at lists.esipfed.org>> on behalf of Ajay Krishnan - NOAA Affiliate via Esip-documentation <esip-documentation at lists.esipfed.org <mailto:esip-documentation at lists.esipfed.org>>
> Reply-To: Ajay Krishnan - NOAA Affiliate <ajay.krishnan at noaa.gov <mailto:ajay.krishnan at noaa.gov>>
> Date: Tuesday, March 13, 2018 at 10:12 AM
> To: "<esip-documentation at lists.esipfed.org <mailto:esip-documentation at lists.esipfed.org>>" <esip-documentation at lists.esipfed.org <mailto:esip-documentation at lists.esipfed.org>>
> Subject: [Esip-documentation] time_coverage_duration and time_coverage_resolution for climatological data
> 
>  
> 
> Hello All,  <>
> 
> I am looking at populating global attributes for a dataset which contains monthly mean temperature for 1995-2004.
> 
>  
> 
> I had recommended 
> 
> :time_coverage_duration="P10Y" ;
> 
> The science steward is not convinced, and perhaps rightly so, because it is not a 10yr field but a 1 month field which is a mean of 10yrs of that month. How should the global attribute be populated in this case? 
> 
> What about the time_coverage_resolution attribute?
> 
>  
> 
> Are such complicated cases outside the purview of ACDD?
> 
>  
> 
> Thanks!
> 
> -- 
> 
> Ajay Krishnan
> Geospatial Data Developer | Science & Technology Corporation <http://stcnet.com/> - Federal Government Contractor
> NOAA National Centers for Environmental Information (NCEI)
> SSMC3 - 1315 East West Hwy Silver Spring MD 20910 <https://goo.gl/maps/ovow9dP4nhE2>
> ajay.krishnan at noaa.gov <mailto:ajay.krishnan at noaa.gov>
> Phone: 301-713-4864 <tel:(301)%20713-4864>
> Customer Support:  1-301-713-3277 <tel:1-301-713-3277> or ncei.info at noaa.gov <mailto:ncei.info at noaa.gov>
>  
> 
> 
> 
> -- 
> Ajay Krishnan
> Geospatial Data Developer | Science & Technology Corporation <http://stcnet.com/> - Federal Government Contractor
> NOAA National Centers for Environmental Information (NCEI)
> SSMC3 - 1315 East West Hwy Silver Spring MD 20910 <https://goo.gl/maps/ovow9dP4nhE2>
> ajay.krishnan at noaa.gov <mailto:ajay.krishnan at noaa.gov>
> Phone: 301-713-4864 <tel:(301)%20713-4864>
> Customer Support:  1-301-713-3277 <tel:1-301-713-3277> or ncei.info at noaa.gov <mailto:ncei.info at noaa.gov>
> 
> _______________________________________________
> Esip-documentation mailing list
> Esip-documentation at lists.esipfed.org <mailto:Esip-documentation at lists.esipfed.org>
> http://lists.deltaforce.net/mailman/listinfo/esip-documentation <http://lists.deltaforce.net/mailman/listinfo/esip-documentation>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.deltaforce.net/pipermail/esip-documentation/attachments/20180315/471f9b84/attachment-0001.html>