[Go-essp-tech] Time metadata in ESG published files...

[Roland Schweitzer]

Hi,

Bob, I think this is primarily addressed to you.

When looking at the data published at GFDL, I see what I think it a mistake
in the time information that is in the THREDDS catalog.

With this URL:
http://esgdata.gfdl.noaa.gov/thredds/esgcet/1/cmip5.output1.NOAA-GFDL.GFDL-CM3.1pctCO2.mon.atmos.Amon.r1i1p1.v20110601.xml

An aggregation looks like this:

<dataset
name="cmip5.output1.NOAA-GFDL.GFDL-CM3.1pctCO2.mon.atmos.Amon.r1i1p1.ccb.20110601.aggregation"

 ID="cmip5.output1.NOAA-GFDL.GFDL-CM3.1pctCO2.mon.atmos.Amon.r1i1p1.ccb.20110601.aggregation"
restrictAccess="esg-user">
  <property name="aggregation_id"
value="cmip5.output1.NOAA-GFDL.GFDL-CM3.1pctCO2.mon.atmos.Amon.r1i1p1.ccb.20110601.aggregation"/>
  <variables vocabulary="CF-1.0">
    <variable name="ccb"
vocabulary_name="air_pressure_at_convective_cloud_base" units="Pa">Air
Pressure at Convective Cloud Base</variable>
  </variables>
  <metadata inherited="true">
    <geospatialCoverage>
      <northsouth>
        <start>-89.0</start>
        <size>178.0</size>
        <units>degrees_north</units>
      </northsouth>
      <eastwest>
        <start>1.25</start>
        <size>357.5</size>
        <units>degrees_east</units>
      </eastwest>
    </geospatialCoverage>
    <timeCoverage>
      *<start>0001-01-16T12:00:00</start>*
*      <end>0140-12-16T12:00:00</end>*
    </timeCoverage>
  </metadata>
  <access

 urlPath="?catid=55B48299F30BC9828F9E93A18AFAA2FC_ns_cmip5.output1.NOAA-GFDL.GFDL-CM3.1pctCO2.mon.atmos.Amon.r1i1p1.ccb.20110601.aggregation"
     serviceName="LASatGFDL" dataFormat="NetCDF"/>
  <dataset
name="cmip5.output1.NOAA-GFDL.GFDL-CM3.1pctCO2.mon.atmos.Amon.r1i1p1.ccb.20110601.aggregation
- Subset 1"

 ID="cmip5.output1.NOAA-GFDL.GFDL-CM3.1pctCO2.mon.atmos.Amon.r1i1p1.ccb.20110601.aggregation.1"

 urlPath="cmip5.output1.NOAA-GFDL.GFDL-CM3.1pctCO2.mon.atmos.Amon.r1i1p1.ccb.20110601.aggregation.1"
restrictAccess="esg-user">
    <serviceName>gridded</serviceName>
    <property name="aggregation_id"
value="cmip5.output1.NOAA-GFDL.GFDL-CM3.1pctCO2.mon.atmos.Amon.r1i1p1.ccb.20110601.aggregation.1"/>
    <property name="time_delta" value="1 month"/>
    <property name="calendar" value="noleap"/>
*    <property name="start" value="1-1-1 0:0:0.0"/>*
    <property name="time_length" value="1680"/>
    <access
urlPath="?catid=55B48299F30BC9828F9E93A18AFAA2FC_ns_cmip5.output1.NOAA-GFDL.GFDL-CM3.1pctCO2.mon.atmos.Amon.r1i1p1.ccb.20110601.aggregation.1"
            serviceName="LASatGFDL" dataFormat="NetCDF"/>
  </dataset>
</dataset>

The time coverage metadata has the correct start time for the aggregation
(16-Jan-0001 12:00), but the extra property which LAS depends on to
initialize itself the wrong start time (01-Jan-0001).  This is causing many
data sets to have their initial plot fail when LAS is started.  The reason
for the failure are complicated based on trying to find the closest time
point in the data for the requested time (1-jan-0001) which turns out to be
the previous month which does not exist.

We can work on coding around this , but in the end I think fixing the
publisher to be more strict about what it puts as the value of the start
property would be a big help to us.

Roland
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/go-essp-tech/attachments/20120216/b3dc66e5/attachment.html