[Met_help] [rt.rap.ucar.edu #70754] History for Cosine Latitude Weighing
John Halley Gotway via RT
met_help at ucar.edu
Thu Feb 19 11:58:06 MST 2015
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Hi all, we have been tasked to set up verification using the WMO Commission for Basic System standards. Two of these standards are verification on a 1.5 degree grid using area averaging to reduce the model resolution to this standard grid (copygb should not be used) and the calculation of RMSE from verification versus analysis must include a cosine latitude weighting. Does MET have a method to filter the RMSE generation to include the cosine longitude weighting in Grid Stat (is it built in)? Do you have any programs that reduce a grid using area averaging before we end up writing one?
Thanks
Bob Craig
AFWA
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: Cosine Latitude Weighing
From: John Halley Gotway
Time: Wed Feb 18 15:45:35 2015
Hello Bob,
It sounds like you have 2 questions...
(1) How to interpolate your model output to the global 1.5 degree grid
using area averaging?
(2) Does MET support the computation of continuous statistics weighted
by
the cosine of the latitude?
First, copygb does support 5 different types of interpolation options,
including -i3 for "budget" interpolation. I'm not sure what exactly
"area
average" means, but perhaps that'd do it:
http://www.dtcenter.org/met/users/support/online_tutorial/METv4.0/copygb/copygb.txt
Second, we are currently working on an automated regridding feature
for
MET. So you could hand grid-stat two files that aren't on the same
grid
and tell it to regrid one to the other before computing statistics.
One of
our interpolation options is a simple mean, so perhaps that's aerial
averaging. This functionality isn't available yet but will be in the
next
release in the summer of 2015.
Lastly, no, MET doesn't compute statistics weighted by the cosine of
the
latitude. This is the first I've heard of that. So I've asked our
scientists to do some investigation on this. If this is something
that
would be broadly useful, we could consider adding it as an option.
FYI - here's a link Barb Brown found that discusses these standards:
https://software.ecmwf.int/wiki/display/WLD/Updated+standard+verification+procedures
Thanks,
John
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
From: robert.craig.2 at us.af.mil
Time: Thu Feb 19 07:06:19 2015
Thanks for the info. Attached is the document that listed the
standards -
http://www.wmo.int/pages/prog/www/DPFS/documents/485_Vol_I_en.pdf.
Since we are now using the MET Office Unified Model, we have to play
by these rules. Look at page 137-138 for a discussion of the cosine
latitude weighting. The areal averaging method CMC uses is at
http://www.wmo.int/pages/prog/www/BAS/documents/AnnexDoc3-3.pdf. I am
not familiar with the budget option in copygb - is it similar to the
areal averaging mentioned in this article? I haven't been able to
find a good description of the budget method on the web.
Thanks
Bob Craig
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Wednesday, February 18, 2015 4:46 PM.
To: CRAIG, ROBERT J GS-12 USAF AFWA 16 WS/WXN
Cc: SITTEL, MATTHEW C CTR USAF AFWA 16 WS/WXN
Subject: Re: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
Hello Bob,
It sounds like you have 2 questions...
(1) How to interpolate your model output to the global 1.5 degree grid
using area averaging?
(2) Does MET support the computation of continuous statistics weighted
by the cosine of the latitude?
First, copygb does support 5 different types of interpolation options,
including -i3 for "budget" interpolation. I'm not sure what exactly
"area average" means, but perhaps that'd do it:
http://www.dtcenter.org/met/users/support/online_tutorial/METv4.0/copygb/copygb.txt
Second, we are currently working on an automated regridding feature
for MET. So you could hand grid-stat two files that aren't on the
same grid and tell it to regrid one to the other before computing
statistics. One of our interpolation options is a simple mean, so
perhaps that's aerial averaging. This functionality isn't available
yet but will be in the next release in the summer of 2015.
Lastly, no, MET doesn't compute statistics weighted by the cosine of
the latitude. This is the first I've heard of that. So I've asked
our scientists to do some investigation on this. If this is something
that would be broadly useful, we could consider adding it as an
option.
FYI - here's a link Barb Brown found that discusses these standards:
https://software.ecmwf.int/wiki/display/WLD/Updated+standard+verification+procedures
Thanks,
John
------------------------------------------------
Subject: Cosine Latitude Weighing
From: John Halley Gotway
Time: Thu Feb 19 10:04:36 2015
Bob,
Thanks for sending the references. That makes it a lot easier to
understand. I read the description of the aerial average, and it
sounds
like it's a straight-forward weighted average of the source grid
points.
The closest thing currently in MET is the unweighted mean (UW_MEAN),
which
is just an unweighted average of nearby points. But that wouldn't
handle
the partial grid box weighting at the edge, and the number of grid
boxes
going into the average would remain constant over the grid. In the
aerial
average, the number of grid points falling inside would likely change
at
various latitudes as the grid distortion changes.
The budget interpolation option in copygb is the one most similar to
this,
but it isn't exactly right.
As I mentioned, we're working on an automated regridding capability,
and I
added this interpolation option as one to consider. We'll also
consider an
option for cosine latitude weighting when computing continuous
statistics.
We have some proposals in the works where enhancing MET to be
consistent
with these standards would be desirable.
What is your timeline? When do you need to start producing results
following these standards?
Thanks,
John
On Thu, Feb 19, 2015 at 7:06 AM, robert.craig.2 at us.af.mil via RT <
met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70754 >
>
> Thanks for the info. Attached is the document that listed the
standards -
> http://www.wmo.int/pages/prog/www/DPFS/documents/485_Vol_I_en.pdf.
Since
> we are now using the MET Office Unified Model, we have to play by
these
> rules. Look at page 137-138 for a discussion of the cosine latitude
> weighting. The areal averaging method CMC uses is at
> http://www.wmo.int/pages/prog/www/BAS/documents/AnnexDoc3-3.pdf. I
am
> not familiar with the budget option in copygb - is it similar to the
areal
> averaging mentioned in this article? I haven't been able to find a
good
> description of the budget method on the web.
>
> Thanks
>
> Bob Craig
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Wednesday, February 18, 2015 4:46 PM.
> To: CRAIG, ROBERT J GS-12 USAF AFWA 16 WS/WXN
> Cc: SITTEL, MATTHEW C CTR USAF AFWA 16 WS/WXN
> Subject: Re: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
>
> Hello Bob,
>
> It sounds like you have 2 questions...
>
> (1) How to interpolate your model output to the global 1.5 degree
grid
> using area averaging?
> (2) Does MET support the computation of continuous statistics
weighted by
> the cosine of the latitude?
>
> First, copygb does support 5 different types of interpolation
options,
> including -i3 for "budget" interpolation. I'm not sure what exactly
"area
> average" means, but perhaps that'd do it:
>
>
>
http://www.dtcenter.org/met/users/support/online_tutorial/METv4.0/copygb/copygb.txt
>
> Second, we are currently working on an automated regridding feature
for
> MET. So you could hand grid-stat two files that aren't on the same
grid
> and tell it to regrid one to the other before computing statistics.
One of
> our interpolation options is a simple mean, so perhaps that's aerial
> averaging. This functionality isn't available yet but will be in
the next
> release in the summer of 2015.
>
> Lastly, no, MET doesn't compute statistics weighted by the cosine of
the
> latitude. This is the first I've heard of that. So I've asked our
> scientists to do some investigation on this. If this is something
that
> would be broadly useful, we could consider adding it as an option.
>
> FYI - here's a link Barb Brown found that discusses these standards:
>
>
>
https://software.ecmwf.int/wiki/display/WLD/Updated+standard+verification+procedures
>
> Thanks,
> John
>
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
From: robert.craig.2 at us.af.mil
Time: Thu Feb 19 10:12:28 2015
Thanks for looking into it. Our timeline is as soon as possible. My
choices are to modify my existing verification code (written by Matt
Sittel) to add this capability, or wait for MET to do it. If you guys
could deliver the capability within a month, I might just wait. If
not I will modify our existing code to do it and then move to MET once
it has it available.
Bob
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Thursday, February 19, 2015 11:05 AM
To: CRAIG, ROBERT J GS-12 USAF AFWA 16 WS/WXN
Cc: SITTEL, MATTHEW C CTR USAF AFWA 16 WS/WXN
Subject: Re: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
Bob,
Thanks for sending the references. That makes it a lot easier to
understand. I read the description of the aerial average, and it
sounds like it's a straight-forward weighted average of the source
grid points.
The closest thing currently in MET is the unweighted mean (UW_MEAN),
which is just an unweighted average of nearby points. But that
wouldn't handle the partial grid box weighting at the edge, and the
number of grid boxes going into the average would remain constant over
the grid. In the aerial average, the number of grid points falling
inside would likely change at various latitudes as the grid distortion
changes.
The budget interpolation option in copygb is the one most similar to
this, but it isn't exactly right.
As I mentioned, we're working on an automated regridding capability,
and I added this interpolation option as one to consider. We'll also
consider an option for cosine latitude weighting when computing
continuous statistics.
We have some proposals in the works where enhancing MET to be
consistent with these standards would be desirable.
What is your timeline? When do you need to start producing results
following these standards?
Thanks,
John
On Thu, Feb 19, 2015 at 7:06 AM, robert.craig.2 at us.af.mil via RT <
met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70754 >
>
> Thanks for the info. Attached is the document that listed the
> standards -
> http://www.wmo.int/pages/prog/www/DPFS/documents/485_Vol_I_en.pdf.
> Since we are now using the MET Office Unified Model, we have to play
> by these rules. Look at page 137-138 for a discussion of the cosine
> latitude weighting. The areal averaging method CMC uses is at
> http://www.wmo.int/pages/prog/www/BAS/documents/AnnexDoc3-3.pdf. I
am
> not familiar with the budget option in copygb - is it similar to the
> areal averaging mentioned in this article? I haven't been able to
find a good description of the budget method on the web.
>
> Thanks
>
> Bob Craig
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Wednesday, February 18, 2015 4:46 PM.
> To: CRAIG, ROBERT J GS-12 USAF AFWA 16 WS/WXN
> Cc: SITTEL, MATTHEW C CTR USAF AFWA 16 WS/WXN
> Subject: Re: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
>
> Hello Bob,
>
> It sounds like you have 2 questions...
>
> (1) How to interpolate your model output to the global 1.5 degree
grid
> using area averaging?
> (2) Does MET support the computation of continuous statistics
weighted
> by the cosine of the latitude?
>
> First, copygb does support 5 different types of interpolation
options,
> including -i3 for "budget" interpolation. I'm not sure what exactly
> "area average" means, but perhaps that'd do it:
>
>
>
http://www.dtcenter.org/met/users/support/online_tutorial/METv4.0/copy
> gb/copygb.txt
>
> Second, we are currently working on an automated regridding feature
> for MET. So you could hand grid-stat two files that aren't on the
> same grid and tell it to regrid one to the other before computing
> statistics. One of our interpolation options is a simple mean, so
> perhaps that's aerial averaging. This functionality isn't available
> yet but will be in the next release in the summer of 2015.
>
> Lastly, no, MET doesn't compute statistics weighted by the cosine of
> the latitude. This is the first I've heard of that. So I've asked
> our scientists to do some investigation on this. If this is
something
> that would be broadly useful, we could consider adding it as an
option.
>
> FYI - here's a link Barb Brown found that discusses these standards:
>
>
>
https://software.ecmwf.int/wiki/display/WLD/Updated+standard+verificat
> ion+procedures
>
> Thanks,
> John
>
>
>
>
------------------------------------------------
Subject: Cosine Latitude Weighing
From: John Halley Gotway
Time: Thu Feb 19 10:38:14 2015
Bob,
I think modifying your existing code makes the most sense. We
certainly
wouldn't be able to provide it in the next month. Our next release
won't
be out until summer 2015. Even so, our budget is pretty lean this
year and
we'd need to identify a funding source to pay for the new work.
Thanks,
John
On Thu, Feb 19, 2015 at 10:12 AM, robert.craig.2 at us.af.mil via RT <
met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70754 >
>
> Thanks for looking into it. Our timeline is as soon as possible.
My
> choices are to modify my existing verification code (written by Matt
> Sittel) to add this capability, or wait for MET to do it. If you
guys
> could deliver the capability within a month, I might just wait. If
not I
> will modify our existing code to do it and then move to MET once it
has it
> available.
>
> Bob
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Thursday, February 19, 2015 11:05 AM
> To: CRAIG, ROBERT J GS-12 USAF AFWA 16 WS/WXN
> Cc: SITTEL, MATTHEW C CTR USAF AFWA 16 WS/WXN
> Subject: Re: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
>
> Bob,
>
> Thanks for sending the references. That makes it a lot easier to
> understand. I read the description of the aerial average, and it
sounds
> like it's a straight-forward weighted average of the source grid
points.
>
> The closest thing currently in MET is the unweighted mean (UW_MEAN),
which
> is just an unweighted average of nearby points. But that wouldn't
handle
> the partial grid box weighting at the edge, and the number of grid
boxes
> going into the average would remain constant over the grid. In the
aerial
> average, the number of grid points falling inside would likely
change at
> various latitudes as the grid distortion changes.
>
> The budget interpolation option in copygb is the one most similar to
this,
> but it isn't exactly right.
>
> As I mentioned, we're working on an automated regridding capability,
and I
> added this interpolation option as one to consider. We'll also
consider an
> option for cosine latitude weighting when computing continuous
statistics.
> We have some proposals in the works where enhancing MET to be
consistent
> with these standards would be desirable.
>
> What is your timeline? When do you need to start producing results
> following these standards?
>
> Thanks,
> John
>
> On Thu, Feb 19, 2015 at 7:06 AM, robert.craig.2 at us.af.mil via RT <
> met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70754 >
> >
> > Thanks for the info. Attached is the document that listed the
> > standards -
> > http://www.wmo.int/pages/prog/www/DPFS/documents/485_Vol_I_en.pdf.
> > Since we are now using the MET Office Unified Model, we have to
play
> > by these rules. Look at page 137-138 for a discussion of the
cosine
> > latitude weighting. The areal averaging method CMC uses is at
> > http://www.wmo.int/pages/prog/www/BAS/documents/AnnexDoc3-3.pdf.
I am
> > not familiar with the budget option in copygb - is it similar to
the
> > areal averaging mentioned in this article? I haven't been able to
find
> a good description of the budget method on the web.
> >
> > Thanks
> >
> > Bob Craig
> >
> > -----Original Message-----
> > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > Sent: Wednesday, February 18, 2015 4:46 PM.
> > To: CRAIG, ROBERT J GS-12 USAF AFWA 16 WS/WXN
> > Cc: SITTEL, MATTHEW C CTR USAF AFWA 16 WS/WXN
> > Subject: Re: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
> >
> > Hello Bob,
> >
> > It sounds like you have 2 questions...
> >
> > (1) How to interpolate your model output to the global 1.5 degree
grid
> > using area averaging?
> > (2) Does MET support the computation of continuous statistics
weighted
> > by the cosine of the latitude?
> >
> > First, copygb does support 5 different types of interpolation
options,
> > including -i3 for "budget" interpolation. I'm not sure what
exactly
> > "area average" means, but perhaps that'd do it:
> >
> >
> >
http://www.dtcenter.org/met/users/support/online_tutorial/METv4.0/copy
> > gb/copygb.txt
> >
> > Second, we are currently working on an automated regridding
feature
> > for MET. So you could hand grid-stat two files that aren't on the
> > same grid and tell it to regrid one to the other before computing
> > statistics. One of our interpolation options is a simple mean, so
> > perhaps that's aerial averaging. This functionality isn't
available
> > yet but will be in the next release in the summer of 2015.
> >
> > Lastly, no, MET doesn't compute statistics weighted by the cosine
of
> > the latitude. This is the first I've heard of that. So I've
asked
> > our scientists to do some investigation on this. If this is
something
> > that would be broadly useful, we could consider adding it as an
option.
> >
> > FYI - here's a link Barb Brown found that discusses these
standards:
> >
> >
> >
https://software.ecmwf.int/wiki/display/WLD/Updated+standard+verificat
> > ion+procedures
> >
> > Thanks,
> > John
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
From: robert.craig.2 at us.af.mil
Time: Thu Feb 19 11:04:20 2015
John, Steve Rugg wanted me to remind you to not count on us for a
funding source for this (as I am sure you are aware). We hope some of
the rest of the community might fund this as its becoming a standard.
Thanks for your input.
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Thursday, February 19, 2015 11:38 AM
To: CRAIG, ROBERT J GS-12 USAF AFWA 16 WS/WXN
Cc: SITTEL, MATTHEW C CTR USAF AFWA 16 WS/WXN
Subject: Re: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
Bob,
I think modifying your existing code makes the most sense. We
certainly wouldn't be able to provide it in the next month. Our next
release won't be out until summer 2015. Even so, our budget is pretty
lean this year and we'd need to identify a funding source to pay for
the new work.
Thanks,
John
On Thu, Feb 19, 2015 at 10:12 AM, robert.craig.2 at us.af.mil via RT <
met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70754 >
>
> Thanks for looking into it. Our timeline is as soon as possible.
My
> choices are to modify my existing verification code (written by Matt
> Sittel) to add this capability, or wait for MET to do it. If you
guys
> could deliver the capability within a month, I might just wait. If
> not I will modify our existing code to do it and then move to MET
once
> it has it available.
>
> Bob
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Thursday, February 19, 2015 11:05 AM
> To: CRAIG, ROBERT J GS-12 USAF AFWA 16 WS/WXN
> Cc: SITTEL, MATTHEW C CTR USAF AFWA 16 WS/WXN
> Subject: Re: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
>
> Bob,
>
> Thanks for sending the references. That makes it a lot easier to
> understand. I read the description of the aerial average, and it
> sounds like it's a straight-forward weighted average of the source
grid points.
>
> The closest thing currently in MET is the unweighted mean (UW_MEAN),
> which is just an unweighted average of nearby points. But that
> wouldn't handle the partial grid box weighting at the edge, and the
> number of grid boxes going into the average would remain constant
over
> the grid. In the aerial average, the number of grid points falling
> inside would likely change at various latitudes as the grid
distortion changes.
>
> The budget interpolation option in copygb is the one most similar to
> this, but it isn't exactly right.
>
> As I mentioned, we're working on an automated regridding capability,
> and I added this interpolation option as one to consider. We'll
also
> consider an option for cosine latitude weighting when computing
continuous statistics.
> We have some proposals in the works where enhancing MET to be
> consistent with these standards would be desirable.
>
> What is your timeline? When do you need to start producing results
> following these standards?
>
> Thanks,
> John
>
> On Thu, Feb 19, 2015 at 7:06 AM, robert.craig.2 at us.af.mil via RT <
> met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70754 >
> >
> > Thanks for the info. Attached is the document that listed the
> > standards -
> > http://www.wmo.int/pages/prog/www/DPFS/documents/485_Vol_I_en.pdf.
> > Since we are now using the MET Office Unified Model, we have to
play
> > by these rules. Look at page 137-138 for a discussion of the
cosine
> > latitude weighting. The areal averaging method CMC uses is at
> > http://www.wmo.int/pages/prog/www/BAS/documents/AnnexDoc3-3.pdf.
I
> > am not familiar with the budget option in copygb - is it similar
to
> > the areal averaging mentioned in this article? I haven't been
able
> > to find
> a good description of the budget method on the web.
> >
> > Thanks
> >
> > Bob Craig
> >
> > -----Original Message-----
> > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > Sent: Wednesday, February 18, 2015 4:46 PM.
> > To: CRAIG, ROBERT J GS-12 USAF AFWA 16 WS/WXN
> > Cc: SITTEL, MATTHEW C CTR USAF AFWA 16 WS/WXN
> > Subject: Re: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
> >
> > Hello Bob,
> >
> > It sounds like you have 2 questions...
> >
> > (1) How to interpolate your model output to the global 1.5 degree
> > grid using area averaging?
> > (2) Does MET support the computation of continuous statistics
> > weighted by the cosine of the latitude?
> >
> > First, copygb does support 5 different types of interpolation
> > options, including -i3 for "budget" interpolation. I'm not sure
> > what exactly "area average" means, but perhaps that'd do it:
> >
> >
> >
http://www.dtcenter.org/met/users/support/online_tutorial/METv4.0/co
> > py
> > gb/copygb.txt
> >
> > Second, we are currently working on an automated regridding
feature
> > for MET. So you could hand grid-stat two files that aren't on the
> > same grid and tell it to regrid one to the other before computing
> > statistics. One of our interpolation options is a simple mean, so
> > perhaps that's aerial averaging. This functionality isn't
available
> > yet but will be in the next release in the summer of 2015.
> >
> > Lastly, no, MET doesn't compute statistics weighted by the cosine
of
> > the latitude. This is the first I've heard of that. So I've
asked
> > our scientists to do some investigation on this. If this is
> > something that would be broadly useful, we could consider adding
it as an option.
> >
> > FYI - here's a link Barb Brown found that discusses these
standards:
> >
> >
> >
https://software.ecmwf.int/wiki/display/WLD/Updated+standard+verific
> > at
> > ion+procedures
> >
> > Thanks,
> > John
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
Subject: Cosine Latitude Weighing
From: John Halley Gotway
Time: Thu Feb 19 11:15:31 2015
Bob,
Yes, I realize we aren't receiving funding for verification from AFWA.
We
have a couple of proposals in the works with Mexico and Panama. If
either
of those gets funded, we might be able to include this work there. It
would be good to make sure their verification results are consistent
with
WMO standards. But that makes it less likely for us to include it in
the
Summer 2015... perhaps the one after that.
Thanks,
John
On Thu, Feb 19, 2015 at 11:04 AM, robert.craig.2 at us.af.mil via RT <
met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70754 >
>
> John, Steve Rugg wanted me to remind you to not count on us for a
funding
> source for this (as I am sure you are aware). We hope some of the
rest of
> the community might fund this as its becoming a standard.
>
> Thanks for your input.
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Thursday, February 19, 2015 11:38 AM
> To: CRAIG, ROBERT J GS-12 USAF AFWA 16 WS/WXN
> Cc: SITTEL, MATTHEW C CTR USAF AFWA 16 WS/WXN
> Subject: Re: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
>
> Bob,
>
> I think modifying your existing code makes the most sense. We
certainly
> wouldn't be able to provide it in the next month. Our next release
won't
> be out until summer 2015. Even so, our budget is pretty lean this
year and
> we'd need to identify a funding source to pay for the new work.
>
> Thanks,
> John
>
> On Thu, Feb 19, 2015 at 10:12 AM, robert.craig.2 at us.af.mil via RT <
> met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70754 >
> >
> > Thanks for looking into it. Our timeline is as soon as possible.
My
> > choices are to modify my existing verification code (written by
Matt
> > Sittel) to add this capability, or wait for MET to do it. If you
guys
> > could deliver the capability within a month, I might just wait.
If
> > not I will modify our existing code to do it and then move to MET
once
> > it has it available.
> >
> > Bob
> >
> > -----Original Message-----
> > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > Sent: Thursday, February 19, 2015 11:05 AM
> > To: CRAIG, ROBERT J GS-12 USAF AFWA 16 WS/WXN
> > Cc: SITTEL, MATTHEW C CTR USAF AFWA 16 WS/WXN
> > Subject: Re: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
> >
> > Bob,
> >
> > Thanks for sending the references. That makes it a lot easier to
> > understand. I read the description of the aerial average, and it
> > sounds like it's a straight-forward weighted average of the source
grid
> points.
> >
> > The closest thing currently in MET is the unweighted mean
(UW_MEAN),
> > which is just an unweighted average of nearby points. But that
> > wouldn't handle the partial grid box weighting at the edge, and
the
> > number of grid boxes going into the average would remain constant
over
> > the grid. In the aerial average, the number of grid points
falling
> > inside would likely change at various latitudes as the grid
distortion
> changes.
> >
> > The budget interpolation option in copygb is the one most similar
to
> > this, but it isn't exactly right.
> >
> > As I mentioned, we're working on an automated regridding
capability,
> > and I added this interpolation option as one to consider. We'll
also
> > consider an option for cosine latitude weighting when computing
> continuous statistics.
> > We have some proposals in the works where enhancing MET to be
> > consistent with these standards would be desirable.
> >
> > What is your timeline? When do you need to start producing
results
> > following these standards?
> >
> > Thanks,
> > John
> >
> > On Thu, Feb 19, 2015 at 7:06 AM, robert.craig.2 at us.af.mil via RT <
> > met_help at ucar.edu> wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70754 >
> > >
> > > Thanks for the info. Attached is the document that listed the
> > > standards -
> > >
http://www.wmo.int/pages/prog/www/DPFS/documents/485_Vol_I_en.pdf.
> > > Since we are now using the MET Office Unified Model, we have to
play
> > > by these rules. Look at page 137-138 for a discussion of the
cosine
> > > latitude weighting. The areal averaging method CMC uses is at
> > > http://www.wmo.int/pages/prog/www/BAS/documents/AnnexDoc3-3.pdf.
I
> > > am not familiar with the budget option in copygb - is it similar
to
> > > the areal averaging mentioned in this article? I haven't been
able
> > > to find
> > a good description of the budget method on the web.
> > >
> > > Thanks
> > >
> > > Bob Craig
> > >
> > > -----Original Message-----
> > > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > > Sent: Wednesday, February 18, 2015 4:46 PM.
> > > To: CRAIG, ROBERT J GS-12 USAF AFWA 16 WS/WXN
> > > Cc: SITTEL, MATTHEW C CTR USAF AFWA 16 WS/WXN
> > > Subject: Re: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
> > >
> > > Hello Bob,
> > >
> > > It sounds like you have 2 questions...
> > >
> > > (1) How to interpolate your model output to the global 1.5
degree
> > > grid using area averaging?
> > > (2) Does MET support the computation of continuous statistics
> > > weighted by the cosine of the latitude?
> > >
> > > First, copygb does support 5 different types of interpolation
> > > options, including -i3 for "budget" interpolation. I'm not sure
> > > what exactly "area average" means, but perhaps that'd do it:
> > >
> > >
> > >
http://www.dtcenter.org/met/users/support/online_tutorial/METv4.0/co
> > > py
> > > gb/copygb.txt
> > >
> > > Second, we are currently working on an automated regridding
feature
> > > for MET. So you could hand grid-stat two files that aren't on
the
> > > same grid and tell it to regrid one to the other before
computing
> > > statistics. One of our interpolation options is a simple mean,
so
> > > perhaps that's aerial averaging. This functionality isn't
available
> > > yet but will be in the next release in the summer of 2015.
> > >
> > > Lastly, no, MET doesn't compute statistics weighted by the
cosine of
> > > the latitude. This is the first I've heard of that. So I've
asked
> > > our scientists to do some investigation on this. If this is
> > > something that would be broadly useful, we could consider adding
it as
> an option.
> > >
> > > FYI - here's a link Barb Brown found that discusses these
standards:
> > >
> > >
> > >
https://software.ecmwf.int/wiki/display/WLD/Updated+standard+verific
> > > at
> > > ion+procedures
> > >
> > > Thanks,
> > > John
> > >
> > >
> > >
> > >
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
From: robert.craig.2 at us.af.mil
Time: Thu Feb 19 11:36:01 2015
Yes, the realities of government funding :)
Bob
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Thursday, February 19, 2015 12:16 PM
To: CRAIG, ROBERT J GS-12 USAF AFWA 16 WS/WXN
Cc: SITTEL, MATTHEW C CTR USAF AFWA 16 WS/WXN
Subject: Re: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
Bob,
Yes, I realize we aren't receiving funding for verification from AFWA.
We have a couple of proposals in the works with Mexico and Panama. If
either of those gets funded, we might be able to include this work
there. It would be good to make sure their verification results are
consistent with WMO standards. But that makes it less likely for us
to include it in the Summer 2015... perhaps the one after that.
Thanks,
John
On Thu, Feb 19, 2015 at 11:04 AM, robert.craig.2 at us.af.mil via RT <
met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70754 >
>
> John, Steve Rugg wanted me to remind you to not count on us for a
> funding source for this (as I am sure you are aware). We hope some
of
> the rest of the community might fund this as its becoming a
standard.
>
> Thanks for your input.
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Thursday, February 19, 2015 11:38 AM
> To: CRAIG, ROBERT J GS-12 USAF AFWA 16 WS/WXN
> Cc: SITTEL, MATTHEW C CTR USAF AFWA 16 WS/WXN
> Subject: Re: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
>
> Bob,
>
> I think modifying your existing code makes the most sense. We
> certainly wouldn't be able to provide it in the next month. Our
next
> release won't be out until summer 2015. Even so, our budget is
pretty
> lean this year and we'd need to identify a funding source to pay for
the new work.
>
> Thanks,
> John
>
> On Thu, Feb 19, 2015 at 10:12 AM, robert.craig.2 at us.af.mil via RT <
> met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70754 >
> >
> > Thanks for looking into it. Our timeline is as soon as possible.
> > My choices are to modify my existing verification code (written by
> > Matt
> > Sittel) to add this capability, or wait for MET to do it. If you
> > guys could deliver the capability within a month, I might just
wait.
> > If not I will modify our existing code to do it and then move to
MET
> > once it has it available.
> >
> > Bob
> >
> > -----Original Message-----
> > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > Sent: Thursday, February 19, 2015 11:05 AM
> > To: CRAIG, ROBERT J GS-12 USAF AFWA 16 WS/WXN
> > Cc: SITTEL, MATTHEW C CTR USAF AFWA 16 WS/WXN
> > Subject: Re: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
> >
> > Bob,
> >
> > Thanks for sending the references. That makes it a lot easier to
> > understand. I read the description of the aerial average, and it
> > sounds like it's a straight-forward weighted average of the source
> > grid
> points.
> >
> > The closest thing currently in MET is the unweighted mean
(UW_MEAN),
> > which is just an unweighted average of nearby points. But that
> > wouldn't handle the partial grid box weighting at the edge, and
the
> > number of grid boxes going into the average would remain constant
> > over the grid. In the aerial average, the number of grid points
> > falling inside would likely change at various latitudes as the
grid
> > distortion
> changes.
> >
> > The budget interpolation option in copygb is the one most similar
to
> > this, but it isn't exactly right.
> >
> > As I mentioned, we're working on an automated regridding
capability,
> > and I added this interpolation option as one to consider. We'll
> > also consider an option for cosine latitude weighting when
computing
> continuous statistics.
> > We have some proposals in the works where enhancing MET to be
> > consistent with these standards would be desirable.
> >
> > What is your timeline? When do you need to start producing
results
> > following these standards?
> >
> > Thanks,
> > John
> >
> > On Thu, Feb 19, 2015 at 7:06 AM, robert.craig.2 at us.af.mil via RT <
> > met_help at ucar.edu> wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=70754 >
> > >
> > > Thanks for the info. Attached is the document that listed the
> > > standards -
> > >
http://www.wmo.int/pages/prog/www/DPFS/documents/485_Vol_I_en.pdf.
> > > Since we are now using the MET Office Unified Model, we have to
> > > play by these rules. Look at page 137-138 for a discussion of
the
> > > cosine latitude weighting. The areal averaging method CMC uses
is
> > > at
> > > http://www.wmo.int/pages/prog/www/BAS/documents/AnnexDoc3-3.pdf.
> > > I am not familiar with the budget option in copygb - is it
similar
> > > to the areal averaging mentioned in this article? I haven't
been
> > > able to find
> > a good description of the budget method on the web.
> > >
> > > Thanks
> > >
> > > Bob Craig
> > >
> > > -----Original Message-----
> > > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > > Sent: Wednesday, February 18, 2015 4:46 PM.
> > > To: CRAIG, ROBERT J GS-12 USAF AFWA 16 WS/WXN
> > > Cc: SITTEL, MATTHEW C CTR USAF AFWA 16 WS/WXN
> > > Subject: Re: [rt.rap.ucar.edu #70754] Cosine Latitude Weighing
> > >
> > > Hello Bob,
> > >
> > > It sounds like you have 2 questions...
> > >
> > > (1) How to interpolate your model output to the global 1.5
degree
> > > grid using area averaging?
> > > (2) Does MET support the computation of continuous statistics
> > > weighted by the cosine of the latitude?
> > >
> > > First, copygb does support 5 different types of interpolation
> > > options, including -i3 for "budget" interpolation. I'm not sure
> > > what exactly "area average" means, but perhaps that'd do it:
> > >
> > >
> > >
http://www.dtcenter.org/met/users/support/online_tutorial/METv4.0/
> > > co
> > > py
> > > gb/copygb.txt
> > >
> > > Second, we are currently working on an automated regridding
> > > feature for MET. So you could hand grid-stat two files that
> > > aren't on the same grid and tell it to regrid one to the other
> > > before computing statistics. One of our interpolation options
is
> > > a simple mean, so perhaps that's aerial averaging. This
> > > functionality isn't available yet but will be in the next
release in the summer of 2015.
> > >
> > > Lastly, no, MET doesn't compute statistics weighted by the
cosine
> > > of the latitude. This is the first I've heard of that. So I've
> > > asked our scientists to do some investigation on this. If this
is
> > > something that would be broadly useful, we could consider adding
> > > it as
> an option.
> > >
> > > FYI - here's a link Barb Brown found that discusses these
standards:
> > >
> > >
> > >
https://software.ecmwf.int/wiki/display/WLD/Updated+standard+verif
> > > ic
> > > at
> > > ion+procedures
> > >
> > > Thanks,
> > > John
> > >
> > >
> > >
> > >
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
More information about the Met_help
mailing list