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

Re: 20100513: new NDFD model data for conduit

On Tue, 15 Jun 2010, Timothy Boyer wrote:


Robb,
A few questions in response. I'm not sure where you're getting MDT in your decoding process. We encode all NDFD elements using GMT. 

Timothy,
It's my debugger that converts GMT to MDT, the data is actually in GMT time so it's not a problem. 




Do you have a specific NDFD element you can use as an example?
I can tell you that we've been asked this before. The short answer is that NDFD is not treated like standard NCEP model data, thus we don't really use the reference time in the same manner as NCEP does. For NDFD, reference time is really a product issuance time. One problem we have that NCEP model grids don't is the case where the beginning of the time interval is actually prior to the reference/issuance time. For example, QPF is valid for a 6-hr period. Let's say the issuance time is 14Z today. The beginning of the valid interval for the QPF grid that ends at 18Z today is12Z today. Since this is actually 2 hours prior to the reference/issuance time, there's no WMO-approved way to encode the number of hours from the reference/issuance time to the beginning of the valid interval. WMO does not allow a negative number of hours (in this case, -2). Since NDFD forecasts are issued every hour, you can see that this type of case happens quite often. 

Thanks for the explanation, that was helpful.


I sent a note out our users asking for local Grib2 parameters: ie
Discipline > 191, or Category > 191, or Parameter number > 191.
I looked at the NDFD elements page but some of parameters are missing, could you give me a updated list. 

Note said:
We have already included the local parameters published by
NCEP. If you have been using NCEP local parameters (and you aren't
NCEP, WMO site 7), you will begin to see "Unknown" parameters because the local parameter retrieval is site dependent. 

If you use local parameters, please send us the following information
(or a link to it):
- WMO site # reference http://www.nco.ncep.noaa.gov/pmb/docs/on388/table0.html 
- and a list of parameters including:
  - Discipline
  - Category
  - Parameter number
  - name, description, units

Thanks for your help,
Robb...
Anyway, I hope this helps illustrates some of the differences between NCEP model encoding and NDFD encoding. 

Tim

Robb Kambic wrote:

Timothy,

Thanks for the information in the previous email.
Working with the new NDFD data, I noticed that the given End time intervals were not consistent with other NCEP type models.
I was wondering why the NDFD data uses the start of interval as the given End of Interval using Octets 35-41 in PDS 4.8? It would of made more sense to me to use the End of Interval as it's calculated from the time 
ranges. Here's an example:

productType  8
Reference Time  Wed May 12 15:00:00 MDT 2010
End of Interval from time ranges Mon May 17 06:00:00 MDT 2010
End of Time Interval from octets 35-41 Sun May 16 06:00:00 MDT 2010
Interval Start Stop =87 111

RT = May 12 15:00:00
End of Time Interval from octets 35-41 = RT + 87 hours
End of Interval from time ranges = RT + 111
Almost all other NCEP models use the End of Interval from time ranges as the given End of Interval in Octets 35-41 

Thanks,
RObb...
=============================================================================== Robb Kambic Unidata Program Center 
Software Engineer III               Univ. Corp for Atmospheric Research
address@hidden           WWW: http://www.unidata.ucar.edu/
=============================================================================== 




===============================================================================
Robb Kambic                                Unidata Program Center
Software Engineer III                      Univ. Corp for Atmospheric Research
address@hidden             WWW: http://www.unidata.ucar.edu/
===============================================================================