[Met_help] Object Definition question

[John Halley Gotway]

Jeff,

I suspect that this may be an issue of units.

The GRIB spec (http://www.nco.ncep.noaa.gov/pmb/docs/on388/table2.html)
indicates that pressure is stored in GRIB as pascals.  So the thresholds
you use in the MODE config file should be in pascals as well.  It is
possible that you're specifying the thresholds in mb as opposed to
pascals?

You can always use the "wgrib" tool
(http://www.cpc.noaa.gov/products/wesley/wgrib.html) with the "-V" option
to look at the range of data values in each GRIB record.  Also, the
colorbar on the first page of the MODE output should give you a good idea
about the range of values.

If you continue to have troubles, feel free to send me a sample fcst, obs,
and config file.  I'd be happy to run it here and try to figure out what's
going on.

If you'd like to do that, you could post that data to our anonymous FTP site:
ftp ftp.rap.ucar.edu
username=anonymous
password="your email address"
cd incoming/irap/met_help
put "your data files"
bye

Thanks,
John

> Hey John,
>
> Yet another MET question! I am trying to use WRF sea level pressure
> forecasts in MODE, specifically trying to track cyclones as objects
> defined below a minimum SLP pressure threshold, and am coming up with
> strange output. The forecasts and observations both appear to run
> through the program, however, the entire theater domain is listed as an
> object. After seeing this I thought that it was likely an object
> definition, in particular threshold, issue, but after a number of trial
> and errors, including trying 1000 mb temperature defined objects, the
> theater-wide object remained. I am not sure if this is a problem with
> the configuration files that I have not identified, or if it is a result
> of the grib files used in the forecast/observations. If you have any
> thoughts/opinions on what I might be missing or overlooking, please let
> me know!
>
> Thanks,
>
> Jeff
>
> -----Original Message-----
> From: John Halley Gotway [mailto:johnhg at ucar.edu]
> Sent: Wednesday, January 06, 2010 9:07 AM
> To: Hamilton, Jeffrey A CIV USAF AFWA 16 WS/WXN; met_help
> Subject: Re: Persistance Calculations in MET
>
> Jeff,
>
> That's interesting that you wrote about persistence today.  I've
> actually been talking to Robert Craig at AFWA over the last couple of
> days about persistence in regards to how you guys compute the Go
> Index.
>
> But the answer to your question is no, MODE does not currently do any
> computations with regards to persistence.  The closest thing to this in
> MET right now is in Point-Stat - you can pass it a gridded
> field containing "climatological" values.  Then it'll dump out
> additional line types, SAL1L2 and VAL1L2, for scalar and vector anomaly
> partial sums.  But we don't have any pre-defined scheme for doing
> anything with persistence.
>
> That being said, we'd be happy to work with you guys to discuss ways to
> enhance the MET tools to better meet your needs.  The MET statistics
> tools, Point-Stat, Grid-Stat, MODE, and Wavelet-Stat,
> operate on a single field at a single point in time.  However, the
> analysis tools, Stat-Analysis and MODE-Analysis, aggregate and summarize
> results through time and/or space.  It may be possible to
> add functionality to these analysis tools to better handle the
> persistence calculations you'd like to see.
>
> Thanks,
> John
>
> Hamilton, Jeffrey A CIV USAF AFWA 16 WS/WXN wrote:
>> John,
>>
>> Have a quick question for you in regards to MET. An issue that keeps
>> coming up from our modelers is the presence of persistence in the MODE
>> calculations. We are wondering if persistence is even calculated in
> MET
>> and, if so, what is the scheme by which it is?
>>
>> Thanks and hope you had a great New Year!
>>
>>
>> // SIGNED //
>> JEFF HAMILTON, Civ SCEP
>> Model Analysis Team
>> 16 WS/ WXN
>> Air Force Weather Agency
>> Offutt AFB, NE 68113
>> (402) 232-7207
>>
>>
>>
>> -----Original Message-----
>> From: John Halley Gotway [mailto:johnhg at ucar.edu]
>> Sent: Friday, November 13, 2009 1:23 PM
>> To: Hamilton, Jeffrey A CIV USAF AFWA 2 WXG/; met_help
>> Subject: Re: masking options
>>
>> Jeff,
>>
>> There is, in fact, a flag in the MODE config file that does exactly
> what
>> you're looking for.  Try turning on the "plot_valid_flag" in the
> config
>> file, as shown below:
>>
>> //
>> // If the plot_valid_flag is set, only the region containing valid
> data
>> // after the masking is applied will be plotted:
>> // (0) Plot the entire domain of the data
>> // (1) Plot only the region containing valid data after masking
>> //
>> plot_valid_flag = 1;
>>
>> Just let us know if any more questions come up.
>>
>> Thanks,
>> John
>>
>> Hamilton, Jeffrey A CIV USAF AFWA 2 WXG/ wrote:
>>> John,
>>>
>>>
>>>
>>> I had a quick question about the masking tool in MET MODE. I have
> been
>>> able to mask an area of the globe using the polygon option, however,
>> the
>>> graphic that appears still shows the worldwide view, making it
>> difficult
>>> to distinguish the objects in the masked area. I am wondering if
> there
>>> was any way to additionally mask the graphic showing the objects,
>> along
>>> with the data itself, in order to visually analyze the objects?
>>>
>>>
>>>
>>> Thanks for any input you might have,
>>>
>>>
>>>
>>> // SIGNED //
>>>
>>> JEFF HAMILTON, Civ SCEP
>>>
>>> Model Verification
>>>
>>> 2 WXG/ WEA
>>>
>>> Air Force Weather Agency
>>>
>>> Offutt AFB, NE 68113
>>>
>>> (402) 232-7207
>>>
>>>
>>>
>>>
>