Roy,
This (hidden/legacy) THREDDS aggregation "glb_analysis" is dated and
not recommended for use anymore. It was an aggregation "of
convenience" joining (crudely) all/multiple GLBa0.08 experiments (90.6
to 91.2). Any time overlap from the experiment that comes next will
cause this time order issue to occur as its time array gets appended
to the existing time array.
The GOFS 3.1 data should be used instead.
https://www.hycom.org/dataserver/gofs-3pt1/analysis
<dataset name="1/12 degree HYCOM + NCODA Global Hindcast Analysis (Combined)"
ID="GLBa0.08-glb_analysis" urlPath="glb_analysis">
<serviceName>all</serviceName>
<netcdf xmlns="
http://www.unidata.ucar.edu/namespaces/netcdf/ncml-2.2">
<aggregation dimName="MT" type="joinExisting">
<netcdf location="dods://<%= @tds_URI %>/thredds/dodsC/GLBa0.08/expt_90.6"/>
<netcdf location="dods://<%= @tds_URI %>/thredds/dodsC/GLBa0.08/expt_90.8"/>
<netcdf location="dods://<%= @tds_URI %>/thredds/dodsC/GLBa0.08/expt_90.9"/>
<netcdf location="dods://<%= @tds_URI %>/thredds/dodsC/GLBa0.08/expt_91.0"/>
<netcdf location="dods://<%= @tds_URI %>/thredds/dodsC/GLBa0.08/expt_91.1"/>
<netcdf location="dods://<%= @tds_URI %>/thredds/dodsC/GLBa0.08/expt_91.2"/>
</aggregation>
</netcdf>
</dataset>
If you explicitly want this GOFS 3.0 data, then I would recommend the
use of the individual experiment aggregations here,
https://www.hycom.org/dataserver/gofs-3pt0/analysis
e.g.,
https://tds.hycom.org/thredds/catalogs/GLBa0.08/expt_91.2.html
...
https://tds.hycom.org/thredds/catalogs/GLBa0.08/expt_90.6.html
> --
> --
> You received this message because you are a member of HYCOM.org
> To ask a question, send an email to
fo...@hycom.org
>
>
--
Michael McDonald
HYCOM.org Administrator