[Met_help] [rt.rap.ucar.edu #64498] History for Unable to do MET Analysis for my 6-months forecast WRF Run

[John Halley Gotway via RT]

----------------------------------------------------------------
  Initial Request
----------------------------------------------------------------

Hi,

I am Manish Dawar from IBM Research Lab, India. I have been working on
weather modelling for a while which involves running the WRF Model and
doing the analysis using MET Packages.

With respect to MET, I have been majorly dealing with 3 MET commands which
are -

1. pcp_combine
2. Point-stat ( before this, i have converted the WRF output in the MET
required format)
3.  stat-analysis

I have been able to do the anlaysis for my 2-days forecast and 7-days
forecast run using these commands.

But now, I need to do the MET Analysis for my 6-months run which I am not
able to do so.

*Issue:*

pcp_combine takes grib files which gets generated from the UPP. The problem
is as UPP is generating the output in grib1 files only (grib2 format is not
fully functional till now) and *grib1 files have limitation of having the
forecast hour only till 256 hours*.

*By this limitation, I mean, when my data exceeds the 256 hours, grib1
files takes the modulus w.r.t ( value % 256) and then put this value in the
file which is incorrect.*

Let us say, If my file is suppose to have forecast hour of 258 hours, grib
file will have 2 as the forecast hour because of which i can not do
pcp_combine with subtract option.


*With this respect, I have two questions - *

1. Did you able to catch this limitation of grib1 files in your code ?? I
mean, Is there any way out of this..

2. Do pcp_combine works with any other file format except grib format??
Although this might not solve the problem completely but it might provide
some help.


*Please reply as soon as possible. Waiting for your reply.*

Thank you for your valuable time.


Regards
Manish Dawar


----------------------------------------------------------------
  Complete Ticket History
----------------------------------------------------------------

Subject: Re: [rt.rap.ucar.edu #64498] Unable to do MET Analysis for my 6-months forecast WRF Run
From: John Halley Gotway
Time: Mon Dec 09 11:27:48 2013

Manish,

I apologize for the delay in getting back to you.  I see that you're
having problems with the limitations in GRIB1.  Yes, it is a
frustrating limitation.

The current (METv4.1) version of the MET pcp_combine tool should be
able to perform addition and subtraction using any of the gridded file
formats that MET supports (GRIB1, GRIB2, NetCDF output of the
pinterp utility).  If you're able to get GRIB2 output from UPP
working, that would be a good way to go.

I'm surprised by the behavior you describe from UPP - that a 258 hour
forecast would be encoded in GRIB1 as a 2 hour forecast.  I worked on
changes to WPP (WRF PostProcessor), the predecessor to UPP
that handled this problem a bit better.  In WPP, for a 258 hour
forecast, we correctly encoded it as a 258 hour forecast but set the
time range indicator (TRI) to a value of 10
(http://www.nco.ncep.noaa.gov/pmb/docs/on388/table5.html).  For
precipitation, you should really use a TRI of 4 to define the
beginning/ending times of the accumulation interval.  By switching it
to a
10, we treated the precip field as an instantaneous forecast rather
than an accumulation.  But we figured that having the forecast time
correct was more important than having the accumulation interval
correct.

Is that not the behavior you're seeing from the current version of
UPP?

Thanks,
John

On 12/04/2013 02:41 PM, Manish Dawar via RT wrote:
>
> Wed Dec 04 14:41:19 2013: Request 64498 was acted upon.
> Transaction: Ticket created by manishdawar2 at gmail.com
>         Queue: met_help
>       Subject: Unable to do MET Analysis for my 6-months forecast
WRF Run
>         Owner: Nobody
>    Requestors: manishdawar2 at gmail.com
>        Status: new
>   Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=64498 >
>
>
> Hi,
>
> I am Manish Dawar from IBM Research Lab, India. I have been working
on
> weather modelling for a while which involves running the WRF Model
and
> doing the analysis using MET Packages.
>
> With respect to MET, I have been majorly dealing with 3 MET commands
which
> are -
>
> 1. pcp_combine
> 2. Point-stat ( before this, i have converted the WRF output in the
MET
> required format)
> 3.  stat-analysis
>
> I have been able to do the anlaysis for my 2-days forecast and 7-
days
> forecast run using these commands.
>
> But now, I need to do the MET Analysis for my 6-months run which I
am not
> able to do so.
>
> *Issue:*
>
> pcp_combine takes grib files which gets generated from the UPP. The
problem
> is as UPP is generating the output in grib1 files only (grib2 format
is not
> fully functional till now) and *grib1 files have limitation of
having the
> forecast hour only till 256 hours*.
>
> *By this limitation, I mean, when my data exceeds the 256 hours,
grib1
> files takes the modulus w.r.t ( value % 256) and then put this value
in the
> file which is incorrect.*
>
> Let us say, If my file is suppose to have forecast hour of 258
hours, grib
> file will have 2 as the forecast hour because of which i can not do
> pcp_combine with subtract option.
>
>
> *With this respect, I have two questions - *
>
> 1. Did you able to catch this limitation of grib1 files in your code
?? I
> mean, Is there any way out of this..
>
> 2. Do pcp_combine works with any other file format except grib
format??
> Although this might not solve the problem completely but it might
provide
> some help.
>
>
> *Please reply as soon as possible. Waiting for your reply.*
>
> Thank you for your valuable time.
>
>
> Regards
> Manish Dawar
>

------------------------------------------------