[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[IDV #GYA-404192]: aggregatin nc files



Hi HP-

> Again thanks for the hints. It works now to my expectation and provides
> even a time stamp. The molec is molecules snd I also entered your
> conversion info into the ncML. Just one last curiosity. I still get the
> warning of unknown units for degrees east and north. Is this also a
> format problem (not that it matters).

It should either be degrees or degrees_east.   Otherwise, it's
parsed as two units multiplied by each other and there is no
unit called "east".  You could use NcML to clean up those also,
just as you did for NO2.

> BTW I got it from the DLR people that nc is not an official format and
> they recommend to use the HD5 files. Unfortunately I still cannot ingest
> them. However, for the moment I can live with nc files wrapped into ncML.

Is that HDF5 or is HD5 a different format.  If the former, do you
have a sample?  John has been working on HDF5 reading and the
latest IDV should support some variations of HDF5 files.
 
> Expecting a thundery night, HP

We got our first significant rain in several weeks last night.
Now that moisture is building into clouds for this afternoon.
I'll probably get wet on the bike ride home. :-(

Don
> Unidata IDV Support wrote:
> > Hi HP-
> >
> >
> >> Thanks for the quick reply. This ncML things is quite nice, though you
> >> have get used to it.I have created the ncML file from your template.
> >> However, there are problems that do not manifest themselves the same way
> >> under XP and 64-bit Linux. I report what happens under Linux.
> >>
> >
> > Maybe I should test my solutions before I send them out. ;-)
> >
> >
> >> When I take you template character by character the last image of the
> >> sequence is displayed (under XP things stop already before I can move on
> >> to the Field Selector) and I get the following message:
> >>
> >> 06-Aug-2008 10:21:14 ucar.nc2.ncml.AggregationNew buildDataset
> >> SEVERE: file:/home/user/Desktop/20080720-31_ma_GOME2_NO2Column/NO2.ncml 
> >> aggNewDimension cant find variable N02
> >> Unknown unit [10^15_molec_cm-2]
> >> (...)
> >>
> >> I then discovered that you put
> >> <variableAgg name="N02"/>
> >> where N02 is supposed to be NO2 (nitrous dioxide).
> >>
> >
> > I must have been brain dead when I responded.   I knew it is NO2, but my
> > fingers typed N02.
> >
> >
> >> Correcting for this and trying to open the ncML file again open a window 
> >> titled No Gridded Data and asks whether I want to it as load another data 
> >> type.
> >>
> >> Looking into the ncML website I could not find a clue on what's going 
> >> wrong.
> >>
> >
> > Well, the docs turned out to be a little misleading.  The default type for
> > a variable is String, to time ends up being string variable.  But for
> > times as strings, they need to be in the ISO standard.
> >
> > So, if you change:
> >
> > <variable name="time">
> >
> > to:
> >
> > <variable name="time" type="int">
> >
> > it should work.  John is going to fix the web page as well.
> >
> >
> >> (BTW "unknown unit" unknown to whom and what could be the consequences?)
> >>
> >
> > Unknown to the units package.  I don't think you could get away with a
> > unit spec of 10^15_molec_cm-2 in a journal publication.  What's a molec
> > anyway? For ISO units, it would be something like "1.66E-9 mole cm-2" since 
> > 1 mole
> > it ~6.02E23 molecules (if that's what a molec is).  The consequence it that
> > you wouldn't be able to use the Change Unit menu since the unit would
> > end up being null.
> >
> > You could use NcML to clean up the units as well.  You could add in:
> >
> > <variable name="NO2">
> > <attribute name="units" value="1.66E-9 mole cm-2"/>
> > </variable>
> >
> > after the time variable block (or after the aggregation block).
> >
> > Let me know if you still have problems with this method.
> >
> > Don
> >
> >> Unidata IDV Support wrote:
> >>
> >>> Hi HP-
> >>>
> >>>
> >>>
> >>>> Full Name: HansPeter Roesli
> >>>> Email Address: address@hidden
> >>>> Organization: EUMETSAT
> >>>> Package Version: 2.6a3 build date:@DATE@
> >>>> Operating System: Linux
> >>>> Hardware: Java: home: /home/user/IDV_2.6a3/jre version: 1.6.0_07 
> >>>> j3d:1.5.2 fcs (build4)
> >>>> Description of problem: HI
> >>>>
> >>>> I tried to aggregate a sequence of daily nc files showing global NO2 
> >>>> columns from GOME-2 on
> >>>> Metop 
> >>>> (http://wdc.dlr.de/data_products/SERVICES/GOME2NRT/archive.php?spec=NO2).
> >>>> However, it does not work and ends this state. I also attach a sample 
> >>>> file.
> >>>>
> >>>>
> >>> These files have no time in them.  Were you trying to use the
> >>> "Make a time sequence from single time grids/images" formula to aggregate 
> >>> these?
> >>> You could create an NcML file to do the aggregation manually.  You can
> >>> see some examples at:
> >>>
> >>> http://www.unidata.ucar.edu/software/netcdf/ncml/v2.2/Aggregation.html
> >>>
> >>> under the JoinNew section.  You would create a new time dimension like:
> >>>
> >>> <netcdf xmlns="http://www.unidata.ucar.edu/namespaces/netcdf/ncml-2.2";>
> >>>
> >>>    <variable name="time">
> >>>      <attribute name="units" value="days since 2008-07-22 00:00"/>
> >>>      <attribute name="_CoordinateAxisType" value="Time" />
> >>>      <values>0 1 2</values>
> >>>    </variable>
> >>>
> >>>    <aggregation dimName="time" type="joinNew">
> >>>     <variableAgg name="N02"/>
> >>>     <netcdf 
> >>> location="file:/home/user/Desktop/20080720-31_ma_GOME2_NO2Column/GOME2.L3.NO2.VCD.NRT.GDP-4.20080722.nc"
> >>>  />
> >>>     <netcdf 
> >>> location="file:/home/user/Desktop/20080720-31_ma_GOME2_NO2Column/GOME2.L3.NO2.VCD.NRT.GDP-4.20080723.nc"
> >>>  />
> >>>     <netcdf 
> >>> location="file:/home/user/Desktop/20080720-31_ma_GOME2_NO2Column/GOME2.L3.NO2.VCD.NRT.GDP-4.20080724.nc"
> >>>  />
> >>>    </aggregation>
> >>>
> >>> </netcdf>
> >>>
> >>>
> >>>
> >>>> BTW marcos
> >>>> field short and
> >>>> long name both deliver the same string (NO2). That's porbably due to the 
> >>>> coding by the provider, isn't it?
> >>>>
> >>>>
> >>> For netCDF files, we use the "long_name" attribute to get a descriptive 
> >>> name.
> >>> However, these files look like:
> >>>
> >>>    float NO2(latitude=720, longitude=1440);
> >>>      :NO2 = "NO2_Total_Column";
> >>>      :units = "[10^15_molec_cm-2]";
> >>>      :value = "Retrieval for given date";
> >>>      :date = "20080722";
> >>>      :version = "GDP-4.2";
> >>>
> >>> if the :N02 attribute was  :long_name, then we'd pick it up.
> >>>
> >>> Also, none of the units are valid specifications.
> >>>
> >>> Don
> >>>
> >>>
> >>>
> >>>> Creating display: Color-Shaded Plan View
> >>>> org.python.core.PyException
> >>>>
> >>>> ******************
> >>>> Stack trace:
> >>>> Traceback (innermost last):
> >>>> File "<string>", line 1, in ?
> >>>> File "<string>", line 230, in makeTimeSequence
> >>>> at visad.Set.<init>(Set.java:148)
> >>>> at visad.SimpleSet.<init>(SimpleSet.java:58)
> >>>> at visad.SampledSet.<init>(SampledSet.java:47)
> >>>> at visad.GriddedSet.<init>(GriddedSet.java:72)
> >>>> at visad.Gridded1DSet.<init>(Gridded1DSet.java:130)
> >>>> at visad.Gridded1DDoubleSet.<init>(Gridded1DDoubleSet.java:107)
> >>>> at visad.Gridded1DDoubleSet.create(Gridded1DDoubleSet.java:174)
> >>>> at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
> >>>> at 
> >>>> sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
> >>>> at 
> >>>> sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
> >>>> at java.lang.reflect.Method.invoke(Method.java:597)
> >>>>
> >>>> visad.UnitException: visad.UnitException: Set: units dimension 1 does 
> >>>> not match Domain dimension 2
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>> Ticket Details
> >>> ===================
> >>> Ticket ID: GYA-404192
> >>> Department: Support IDV
> >>> Priority: Normal
> >>> Status: Open
> >>>
> >>>
> >>>
> >>>
> >>
> >
> >
> > Ticket Details
> > ===================
> > Ticket ID: GYA-404192
> > Department: Support IDV
> > Priority: Normal
> > Status: Open
> >
> >
> >
> 
> 


Ticket Details
===================
Ticket ID: GYA-404192
Department: Support IDV
Priority: Normal
Status: Open