Ferret seems to be reading the data incorrectly; it is running into a problem handling the E dimension. I think I see what is happening; we will be able to get out a fix for this soon. It is not so much in the time aggregation but in reading a timeseries from any multi-file dataset when there is an E or an F dimension, so it would also apply to a classic descriptor file as well.
Ansley
On 1/6/2017 8:57 AM, Patrick Brockmann
wrote:
Hi All,
I am confused with the use of time aggregation from define/agg/t
that is a very cool feature.
I have reduced my initial probem to small files ready to be aggregated.
But discontinuties appear when I trace the members average and then
different specific members.
The problem is quite strange because it depends of the order of the calls.
I suspect a memory problem.
The material to test this is available from
https://vesg.ipsl.upmc.fr/thredds/catalog/IPSLFS/brocksce/ferret_bugs/aggregation/catalog.html
yes? go err700_agg_onT.jnl
Any help and test are welcome on this issue.
Thanks
Patrick
-- Data Analysis and Visualization Engineer LSCE/IPSL, CEA-CNRS-UVSQ laboratory LSCE - Climate and Environment Sciences Laboratory IPSL - Institut Pierre Simon Laplace --
