[Met_help] [rt.rap.ucar.edu #63182] History for Adjustment of Time
Julie Prestopnik via RT
met_help at ucar.edu
Thu Oct 10 11:19:32 MDT 2013
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
HI,
I would like to ask for some assistance in running MET.
I have a little_r file in local time. I would like to match it against a
WRF output forecast that is in UTC. Is there a way in MET that this can be
set up so that it will match correctly?
Or I really need to recreate my observed little_r file in UTC?
Thank you!
Vladimir A. Malabanan
3F IBM Bldg J, UP Ayala Technohub
Applications Programmer
Advanced Analytics - Data Specialist
Diliman, Quezon City
IBM Solutions Delivery, Inc.
Philippines
Global Delivery Center
Phone:
+632-995-2IBM local 8639
e-mail:
malabava at ph.ibm.com
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #63182] Adjustment of Time
From: Julie Prestopnik
Time: Thu Sep 26 10:17:20 2013
Hi Vladimir.
It sounds like you'll be using PointStat to get verification
statistics.
The discrepancy in valid time can be handled by setting "beg" and
"end" (offset in seconds relative to the valid time of the forecast)
in
the obs_window in the PointStat configuration file:
//
// Point observation time window
//
obs_window = {
beg = -5400;
end = 5400;
}
or by setting the obs_valid_beg time and obs_valid_end time options on
the command line:
"-obs_valid_beg time" in YYYYMMDD[_HH[MMSS]] sets the
beginning of the matching time window (optional).
"-obs_valid_end time" in YYYYMMDD[_HH[MMSS]] sets the
end of the matching time window (optional).
This will allow you to set the matching time window you'd like to use
for your verification.
I hope this helps.
Julie
On 09/26/2013 01:39 AM, Vladimir A Malabanan via RT wrote:
>
> Thu Sep 26 01:39:12 2013: Request 63182 was acted upon.
> Transaction: Ticket created by vlad.malabanan at ph.ibm.com
> Queue: met_help
> Subject: Adjustment of Time
> Owner: Nobody
> Requestors: vlad.malabanan at ph.ibm.com
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=63182 >
>
>
> HI,
>
> I would like to ask for some assistance in running MET.
>
> I have a little_r file in local time. I would like to match it
against a
> WRF output forecast that is in UTC. Is there a way in MET that this
can be
> set up so that it will match correctly?
>
> Or I really need to recreate my observed little_r file in UTC?
>
> Thank you!
>
>
> Vladimir A. Malabanan
> 3F IBM Bldg J, UP Ayala Technohub
>
> Applications Programmer
> Advanced Analytics - Data Specialist
> Diliman, Quezon City
> IBM Solutions Delivery, Inc.
> Philippines
> Global Delivery Center
>
>
> Phone:
> +632-995-2IBM local 8639
>
>
> e-mail:
> malabava at ph.ibm.com
>
>
>
>
>
>
--
Julie Prestopnik
National Center for Atmospheric Research
Research Applications Laboratory
Phone: 303.497.8399
Email: jpresto at ucar.edu
------------------------------------------------
Subject: Adjustment of Time
From: Vladimir A Malabanan
Time: Thu Sep 26 23:13:59 2013
Hello Julie,
Thanks for the quick response.
Yes you are correct. I would like to perform forecast accuracy using
PoinStat tool and my input data are grib forecast files (converted
from
WRF netcdf files) and an NC observation file (converted from little_r
using ASCII2NC).
I am new in weather forecasting and I am just doing some proof-of-
concept
on how to use MET.
Just to make it clear, to be able to match my observations (which is
in
local time, PHT, (+8 UTC)) to my forecast (in UTC), all I need to do
is to
adjust the beg and end time in the config file. Am I correct if I will
set
beg and end to -28800 and 28800 respectively??
On the other hand, if I will add obs_valid_beg and obs_valid_end
options
in the PointStat script, what values will I put? For example I have a
forecast for April 1, 2013 at 00:00:00 UTC. Will I get the correct
matching if I just set obs_valid_beg and obs_valid_end to
20130401_080000
and 20130401_090000 (my observation file is in UTC +8 )?
Thank you!!
Vladimir A. Malabanan
3F IBM Bldg J, UP Ayala Technohub
Applications Programmer
Advanced Analytics - Data Specialist
Diliman, Quezon City
IBM Solutions Delivery, Inc.
Philippines
Global Delivery Center
Phone:
+632-995-2IBM local 8639
e-mail:
malabava at ph.ibm.com
From: "Julie Prestopnik via RT" <met_help at ucar.edu>
To: Vladimir A Malabanan/Philippines/IBM at IBMPH,
Date: 09/27/2013 12:16 AM
Subject: Re: [rt.rap.ucar.edu #63182] Adjustment of Time
Hi Vladimir.
It sounds like you'll be using PointStat to get verification
statistics.
The discrepancy in valid time can be handled by setting "beg" and
"end" (offset in seconds relative to the valid time of the forecast)
in
the obs_window in the PointStat configuration file:
//
// Point observation time window
//
obs_window = {
beg = -5400;
end = 5400;
}
or by setting the obs_valid_beg time and obs_valid_end time options on
the command line:
"-obs_valid_beg time" in YYYYMMDD[_HH[MMSS]] sets the
beginning of the matching time window (optional).
"-obs_valid_end time" in YYYYMMDD[_HH[MMSS]] sets the
end of the matching time window (optional).
This will allow you to set the matching time window you'd like to use
for your verification.
I hope this helps.
Julie
On 09/26/2013 01:39 AM, Vladimir A Malabanan via RT wrote:
>
> Thu Sep 26 01:39:12 2013: Request 63182 was acted upon.
> Transaction: Ticket created by vlad.malabanan at ph.ibm.com
> Queue: met_help
> Subject: Adjustment of Time
> Owner: Nobody
> Requestors: vlad.malabanan at ph.ibm.com
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=63182 >
>
>
> HI,
>
> I would like to ask for some assistance in running MET.
>
> I have a little_r file in local time. I would like to match it
against a
> WRF output forecast that is in UTC. Is there a way in MET that this
can
be
> set up so that it will match correctly?
>
> Or I really need to recreate my observed little_r file in UTC?
>
> Thank you!
>
>
> Vladimir A. Malabanan
> 3F IBM Bldg J, UP Ayala Technohub
>
> Applications Programmer
> Advanced Analytics - Data Specialist
> Diliman, Quezon City
> IBM Solutions Delivery, Inc.
> Philippines
> Global Delivery Center
>
>
> Phone:
> +632-995-2IBM local 8639
>
>
> e-mail:
> malabava at ph.ibm.com
>
>
>
>
>
>
--
Julie Prestopnik
National Center for Atmospheric Research
Research Applications Laboratory
Phone: 303.497.8399
Email: jpresto at ucar.edu
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #63182] Adjustment of Time
From: Julie Prestopnik
Time: Fri Sep 27 10:40:29 2013
Hi Vladimir.
Setting the beg/end obs_window to +/- 28800 seconds is close but not
quite what you should do, because then you'll be comparing 16 hours of
observations to a forecast valid one time. Instead, you should
probably
set beg = 28800 and end = 32400. That's a +/- one hour window with an
8-hour offset applied. You can decide on what matching window you'd
like to use (maybe +/- 10 min, 30 min, or one hour) and then just add
28800 to both settings to account for the local time vs UTC
difference.
Alternatively, you can use obs_valid_beg and obs_valid_end on the
command line, which will override the obs_window settings in the
config
file. The example you provided for setting obs_valid_beg and
obs_valid_end looks correct to me.
Please let us know if you have any other questions.
Thanks,
Julie
On 2013-09-26 23:13, Vladimir A Malabanan via RT wrote:
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=63182 >
>
> Hello Julie,
>
> Thanks for the quick response.
>
> Yes you are correct. I would like to perform forecast accuracy using
> PoinStat tool and my input data are grib forecast files (converted
> from
> WRF netcdf files) and an NC observation file (converted from
little_r
> using ASCII2NC).
>
> I am new in weather forecasting and I am just doing some
> proof-of-concept
> on how to use MET.
>
> Just to make it clear, to be able to match my observations (which is
> in
> local time, PHT, (+8 UTC)) to my forecast (in UTC), all I need to do
> is to
> adjust the beg and end time in the config file. Am I correct if I
> will set
> beg and end to -28800 and 28800 respectively??
>
> On the other hand, if I will add obs_valid_beg and obs_valid_end
> options
> in the PointStat script, what values will I put? For example I have
a
> forecast for April 1, 2013 at 00:00:00 UTC. Will I get the correct
> matching if I just set obs_valid_beg and obs_valid_end to
> 20130401_080000
> and 20130401_090000 (my observation file is in UTC +8 )?
>
> Thank you!!
>
>
>
> Vladimir A. Malabanan
> 3F IBM Bldg J, UP Ayala Technohub
>
> Applications Programmer
> Advanced Analytics - Data Specialist
> Diliman, Quezon City
> IBM Solutions Delivery, Inc.
> Philippines
> Global Delivery Center
>
>
> Phone:
> +632-995-2IBM local 8639
>
>
> e-mail:
> malabava at ph.ibm.com
>
>
>
>
>
>
>
>
>
> From: "Julie Prestopnik via RT" <met_help at ucar.edu>
> To: Vladimir A Malabanan/Philippines/IBM at IBMPH,
> Date: 09/27/2013 12:16 AM
> Subject: Re: [rt.rap.ucar.edu #63182] Adjustment of Time
>
>
>
> Hi Vladimir.
>
> It sounds like you'll be using PointStat to get verification
> statistics.
> The discrepancy in valid time can be handled by setting "beg" and
> "end" (offset in seconds relative to the valid time of the forecast)
> in
> the obs_window in the PointStat configuration file:
>
> //
> // Point observation time window
> //
> obs_window = {
> beg = -5400;
> end = 5400;
> }
>
> or by setting the obs_valid_beg time and obs_valid_end time options
> on
> the command line:
>
> "-obs_valid_beg time" in YYYYMMDD[_HH[MMSS]] sets
> the
> beginning of the matching time window (optional).
> "-obs_valid_end time" in YYYYMMDD[_HH[MMSS]] sets
> the
> end of the matching time window (optional).
>
> This will allow you to set the matching time window you'd like to
use
> for your verification.
>
> I hope this helps.
>
> Julie
>
> On 09/26/2013 01:39 AM, Vladimir A Malabanan via RT wrote:
>>
>> Thu Sep 26 01:39:12 2013: Request 63182 was acted upon.
>> Transaction: Ticket created by vlad.malabanan at ph.ibm.com
>> Queue: met_help
>> Subject: Adjustment of Time
>> Owner: Nobody
>> Requestors: vlad.malabanan at ph.ibm.com
>> Status: new
>> Ticket <URL:
>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=63182 >
>>
>>
>> HI,
>>
>> I would like to ask for some assistance in running MET.
>>
>> I have a little_r file in local time. I would like to match it
>> against a
>> WRF output forecast that is in UTC. Is there a way in MET that this
>> can
> be
>> set up so that it will match correctly?
>>
>> Or I really need to recreate my observed little_r file in UTC?
>>
>> Thank you!
>>
>>
>> Vladimir A. Malabanan
>> 3F IBM Bldg J, UP Ayala Technohub
>>
>> Applications Programmer
>> Advanced Analytics - Data Specialist
>> Diliman, Quezon City
>> IBM Solutions Delivery, Inc.
>> Philippines
>> Global Delivery Center
>>
>>
>> Phone:
>> +632-995-2IBM local 8639
>>
>>
>> e-mail:
>> malabava at ph.ibm.com
>>
>>
>>
>>
>>
>>
------------------------------------------------
Subject: Adjustment of Time
From: Vladimir A Malabanan
Time: Fri Sep 27 19:10:04 2013
Thanks Julie! I will try that and let you know of the results!
Vladimir A. Malabanan
3F IBM Bldg J, UP Ayala Technohub
Applications Programmer
Advanced Analytics - Data Specialist
Diliman, Quezon City
IBM Solutions Delivery, Inc.
Philippines
Global Delivery Center
Phone:
+632-995-2IBM local 8639
e-mail:
malabava at ph.ibm.com
From: "Julie Prestopnik via RT" <met_help at ucar.edu>
To: Vladimir A Malabanan/Philippines/IBM at IBMPH,
Date: 09/28/2013 12:39 AM
Subject: Re: [rt.rap.ucar.edu #63182] Adjustment of Time
Hi Vladimir.
Setting the beg/end obs_window to +/- 28800 seconds is close but not
quite what you should do, because then you'll be comparing 16 hours of
observations to a forecast valid one time. Instead, you should
probably
set beg = 28800 and end = 32400. That's a +/- one hour window with an
8-hour offset applied. You can decide on what matching window you'd
like to use (maybe +/- 10 min, 30 min, or one hour) and then just add
28800 to both settings to account for the local time vs UTC
difference.
Alternatively, you can use obs_valid_beg and obs_valid_end on the
command line, which will override the obs_window settings in the
config
file. The example you provided for setting obs_valid_beg and
obs_valid_end looks correct to me.
Please let us know if you have any other questions.
Thanks,
Julie
On 2013-09-26 23:13, Vladimir A Malabanan via RT wrote:
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=63182 >
>
> Hello Julie,
>
> Thanks for the quick response.
>
> Yes you are correct. I would like to perform forecast accuracy using
> PoinStat tool and my input data are grib forecast files (converted
> from
> WRF netcdf files) and an NC observation file (converted from
little_r
> using ASCII2NC).
>
> I am new in weather forecasting and I am just doing some
> proof-of-concept
> on how to use MET.
>
> Just to make it clear, to be able to match my observations (which is
> in
> local time, PHT, (+8 UTC)) to my forecast (in UTC), all I need to do
> is to
> adjust the beg and end time in the config file. Am I correct if I
> will set
> beg and end to -28800 and 28800 respectively??
>
> On the other hand, if I will add obs_valid_beg and obs_valid_end
> options
> in the PointStat script, what values will I put? For example I have
a
> forecast for April 1, 2013 at 00:00:00 UTC. Will I get the correct
> matching if I just set obs_valid_beg and obs_valid_end to
> 20130401_080000
> and 20130401_090000 (my observation file is in UTC +8 )?
>
> Thank you!!
>
>
>
> Vladimir A. Malabanan
> 3F IBM Bldg J, UP Ayala Technohub
>
> Applications Programmer
> Advanced Analytics - Data Specialist
> Diliman, Quezon City
> IBM Solutions Delivery, Inc.
> Philippines
> Global Delivery Center
>
>
> Phone:
> +632-995-2IBM local 8639
>
>
> e-mail:
> malabava at ph.ibm.com
>
>
>
>
>
>
>
>
>
> From: "Julie Prestopnik via RT" <met_help at ucar.edu>
> To: Vladimir A Malabanan/Philippines/IBM at IBMPH,
> Date: 09/27/2013 12:16 AM
> Subject: Re: [rt.rap.ucar.edu #63182] Adjustment of Time
>
>
>
> Hi Vladimir.
>
> It sounds like you'll be using PointStat to get verification
> statistics.
> The discrepancy in valid time can be handled by setting "beg" and
> "end" (offset in seconds relative to the valid time of the forecast)
> in
> the obs_window in the PointStat configuration file:
>
> //
> // Point observation time window
> //
> obs_window = {
> beg = -5400;
> end = 5400;
> }
>
> or by setting the obs_valid_beg time and obs_valid_end time options
> on
> the command line:
>
> "-obs_valid_beg time" in YYYYMMDD[_HH[MMSS]] sets
> the
> beginning of the matching time window (optional).
> "-obs_valid_end time" in YYYYMMDD[_HH[MMSS]] sets
> the
> end of the matching time window (optional).
>
> This will allow you to set the matching time window you'd like to
use
> for your verification.
>
> I hope this helps.
>
> Julie
>
> On 09/26/2013 01:39 AM, Vladimir A Malabanan via RT wrote:
>>
>> Thu Sep 26 01:39:12 2013: Request 63182 was acted upon.
>> Transaction: Ticket created by vlad.malabanan at ph.ibm.com
>> Queue: met_help
>> Subject: Adjustment of Time
>> Owner: Nobody
>> Requestors: vlad.malabanan at ph.ibm.com
>> Status: new
>> Ticket <URL:
>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=63182 >
>>
>>
>> HI,
>>
>> I would like to ask for some assistance in running MET.
>>
>> I have a little_r file in local time. I would like to match it
>> against a
>> WRF output forecast that is in UTC. Is there a way in MET that this
>> can
> be
>> set up so that it will match correctly?
>>
>> Or I really need to recreate my observed little_r file in UTC?
>>
>> Thank you!
>>
>>
>> Vladimir A. Malabanan
>> 3F IBM Bldg J, UP Ayala Technohub
>>
>> Applications Programmer
>> Advanced Analytics - Data Specialist
>> Diliman, Quezon City
>> IBM Solutions Delivery, Inc.
>> Philippines
>> Global Delivery Center
>>
>>
>> Phone:
>> +632-995-2IBM local 8639
>>
>>
>> e-mail:
>> malabava at ph.ibm.com
>>
>>
>>
>>
>>
>>
------------------------------------------------
Subject: Adjustment of Time
From: Vladimir A Malabanan
Time: Mon Sep 30 04:36:00 2013
HI Julie,
It's already working! I was able to match my forecast file vs my
observed
file!
I have another question though.
Is the measure I take for consideration in the config below the same
as
the field T2 in the WRF forecast?
name = "TMP";
level = [ "Z2" ];
Also, I am trying to compare my matched pairs from my results in MET
against my output from another software. I would just like to know how
MET
computes for its nearest neighbor (W=1). Is it euclidean metric or
city
block metric?
Thanks!
Vladimir A. Malabanan
3F IBM Bldg J, UP Ayala Technohub
Applications Programmer
Advanced Analytics - Data Specialist
Diliman, Quezon City
IBM Solutions Delivery, Inc.
Philippines
Global Delivery Center
Phone:
+632-995-2IBM local 8639
e-mail:
malabava at ph.ibm.com
From: "Julie Prestopnik via RT" <met_help at ucar.edu>
To: Vladimir A Malabanan/Philippines/IBM at IBMPH,
Date: 09/28/2013 12:39 AM
Subject: Re: [rt.rap.ucar.edu #63182] Adjustment of Time
Hi Vladimir.
Setting the beg/end obs_window to +/- 28800 seconds is close but not
quite what you should do, because then you'll be comparing 16 hours of
observations to a forecast valid one time. Instead, you should
probably
set beg = 28800 and end = 32400. That's a +/- one hour window with an
8-hour offset applied. You can decide on what matching window you'd
like to use (maybe +/- 10 min, 30 min, or one hour) and then just add
28800 to both settings to account for the local time vs UTC
difference.
Alternatively, you can use obs_valid_beg and obs_valid_end on the
command line, which will override the obs_window settings in the
config
file. The example you provided for setting obs_valid_beg and
obs_valid_end looks correct to me.
Please let us know if you have any other questions.
Thanks,
Julie
On 2013-09-26 23:13, Vladimir A Malabanan via RT wrote:
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=63182 >
>
> Hello Julie,
>
> Thanks for the quick response.
>
> Yes you are correct. I would like to perform forecast accuracy using
> PoinStat tool and my input data are grib forecast files (converted
> from
> WRF netcdf files) and an NC observation file (converted from
little_r
> using ASCII2NC).
>
> I am new in weather forecasting and I am just doing some
> proof-of-concept
> on how to use MET.
>
> Just to make it clear, to be able to match my observations (which is
> in
> local time, PHT, (+8 UTC)) to my forecast (in UTC), all I need to do
> is to
> adjust the beg and end time in the config file. Am I correct if I
> will set
> beg and end to -28800 and 28800 respectively??
>
> On the other hand, if I will add obs_valid_beg and obs_valid_end
> options
> in the PointStat script, what values will I put? For example I have
a
> forecast for April 1, 2013 at 00:00:00 UTC. Will I get the correct
> matching if I just set obs_valid_beg and obs_valid_end to
> 20130401_080000
> and 20130401_090000 (my observation file is in UTC +8 )?
>
> Thank you!!
>
>
>
> Vladimir A. Malabanan
> 3F IBM Bldg J, UP Ayala Technohub
>
> Applications Programmer
> Advanced Analytics - Data Specialist
> Diliman, Quezon City
> IBM Solutions Delivery, Inc.
> Philippines
> Global Delivery Center
>
>
> Phone:
> +632-995-2IBM local 8639
>
>
> e-mail:
> malabava at ph.ibm.com
>
>
>
>
>
>
>
>
>
> From: "Julie Prestopnik via RT" <met_help at ucar.edu>
> To: Vladimir A Malabanan/Philippines/IBM at IBMPH,
> Date: 09/27/2013 12:16 AM
> Subject: Re: [rt.rap.ucar.edu #63182] Adjustment of Time
>
>
>
> Hi Vladimir.
>
> It sounds like you'll be using PointStat to get verification
> statistics.
> The discrepancy in valid time can be handled by setting "beg" and
> "end" (offset in seconds relative to the valid time of the forecast)
> in
> the obs_window in the PointStat configuration file:
>
> //
> // Point observation time window
> //
> obs_window = {
> beg = -5400;
> end = 5400;
> }
>
> or by setting the obs_valid_beg time and obs_valid_end time options
> on
> the command line:
>
> "-obs_valid_beg time" in YYYYMMDD[_HH[MMSS]] sets
> the
> beginning of the matching time window (optional).
> "-obs_valid_end time" in YYYYMMDD[_HH[MMSS]] sets
> the
> end of the matching time window (optional).
>
> This will allow you to set the matching time window you'd like to
use
> for your verification.
>
> I hope this helps.
>
> Julie
>
> On 09/26/2013 01:39 AM, Vladimir A Malabanan via RT wrote:
>>
>> Thu Sep 26 01:39:12 2013: Request 63182 was acted upon.
>> Transaction: Ticket created by vlad.malabanan at ph.ibm.com
>> Queue: met_help
>> Subject: Adjustment of Time
>> Owner: Nobody
>> Requestors: vlad.malabanan at ph.ibm.com
>> Status: new
>> Ticket <URL:
>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=63182 >
>>
>>
>> HI,
>>
>> I would like to ask for some assistance in running MET.
>>
>> I have a little_r file in local time. I would like to match it
>> against a
>> WRF output forecast that is in UTC. Is there a way in MET that this
>> can
> be
>> set up so that it will match correctly?
>>
>> Or I really need to recreate my observed little_r file in UTC?
>>
>> Thank you!
>>
>>
>> Vladimir A. Malabanan
>> 3F IBM Bldg J, UP Ayala Technohub
>>
>> Applications Programmer
>> Advanced Analytics - Data Specialist
>> Diliman, Quezon City
>> IBM Solutions Delivery, Inc.
>> Philippines
>> Global Delivery Center
>>
>>
>> Phone:
>> +632-995-2IBM local 8639
>>
>>
>> e-mail:
>> malabava at ph.ibm.com
>>
>>
>>
>>
>>
>>
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #63182] Adjustment of Time
From: Julie Prestopnik
Time: Mon Sep 30 11:12:22 2013
That's great news, Vladimir!
Regarding your questions:
#1 Yes, that's 2-m temperature.
#2 I am not completely sure. Perhaps it is the "city block metric".
Here's how it works in the code:
- For each observation, we have a lat/lon value.
- Convert that lat/lon to an x/y location in grid space. Generally,
the x/y values won't fall exactly on a grid point.
- Round the x/values to the nearest integers.
- Use the forecast value at the grid point specified by those rounded
values.
I hope that information will help.
Thanks,
Julie
On 2013-09-30 04:36, Vladimir A Malabanan via RT wrote:
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=63182 >
>
> HI Julie,
>
> It's already working! I was able to match my forecast file vs my
> observed
> file!
>
> I have another question though.
>
> Is the measure I take for consideration in the config below the same
> as
> the field T2 in the WRF forecast?
>
> name = "TMP";
> level = [ "Z2" ];
>
> Also, I am trying to compare my matched pairs from my results in MET
> against my output from another software. I would just like to know
> how MET
> computes for its nearest neighbor (W=1). Is it euclidean metric or
> city
> block metric?
>
> Thanks!
>
>
>
> Vladimir A. Malabanan
> 3F IBM Bldg J, UP Ayala Technohub
>
> Applications Programmer
> Advanced Analytics - Data Specialist
> Diliman, Quezon City
> IBM Solutions Delivery, Inc.
> Philippines
> Global Delivery Center
>
>
> Phone:
> +632-995-2IBM local 8639
>
>
> e-mail:
> malabava at ph.ibm.com
>
>
>
>
>
>
>
>
>
> From: "Julie Prestopnik via RT" <met_help at ucar.edu>
> To: Vladimir A Malabanan/Philippines/IBM at IBMPH,
> Date: 09/28/2013 12:39 AM
> Subject: Re: [rt.rap.ucar.edu #63182] Adjustment of Time
>
>
>
> Hi Vladimir.
>
> Setting the beg/end obs_window to +/- 28800 seconds is close but not
> quite what you should do, because then you'll be comparing 16 hours
> of
> observations to a forecast valid one time. Instead, you should
> probably
> set beg = 28800 and end = 32400. That's a +/- one hour window with
> an
> 8-hour offset applied. You can decide on what matching window you'd
> like to use (maybe +/- 10 min, 30 min, or one hour) and then just
add
> 28800 to both settings to account for the local time vs UTC
> difference.
>
> Alternatively, you can use obs_valid_beg and obs_valid_end on the
> command line, which will override the obs_window settings in the
> config
> file. The example you provided for setting obs_valid_beg and
> obs_valid_end looks correct to me.
>
> Please let us know if you have any other questions.
>
> Thanks,
> Julie
>
> On 2013-09-26 23:13, Vladimir A Malabanan via RT wrote:
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=63182 >
>>
>> Hello Julie,
>>
>> Thanks for the quick response.
>>
>> Yes you are correct. I would like to perform forecast accuracy
using
>> PoinStat tool and my input data are grib forecast files (converted
>> from
>> WRF netcdf files) and an NC observation file (converted from
>> little_r
>> using ASCII2NC).
>>
>> I am new in weather forecasting and I am just doing some
>> proof-of-concept
>> on how to use MET.
>>
>> Just to make it clear, to be able to match my observations (which
is
>> in
>> local time, PHT, (+8 UTC)) to my forecast (in UTC), all I need to
do
>> is to
>> adjust the beg and end time in the config file. Am I correct if I
>> will set
>> beg and end to -28800 and 28800 respectively??
>>
>> On the other hand, if I will add obs_valid_beg and obs_valid_end
>> options
>> in the PointStat script, what values will I put? For example I have
>> a
>> forecast for April 1, 2013 at 00:00:00 UTC. Will I get the correct
>> matching if I just set obs_valid_beg and obs_valid_end to
>> 20130401_080000
>> and 20130401_090000 (my observation file is in UTC +8 )?
>>
>> Thank you!!
>>
>>
>>
>> Vladimir A. Malabanan
>> 3F IBM Bldg J, UP Ayala Technohub
>>
>> Applications Programmer
>> Advanced Analytics - Data Specialist
>> Diliman, Quezon City
>> IBM Solutions Delivery, Inc.
>> Philippines
>> Global Delivery Center
>>
>>
>> Phone:
>> +632-995-2IBM local 8639
>>
>>
>> e-mail:
>> malabava at ph.ibm.com
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> From: "Julie Prestopnik via RT" <met_help at ucar.edu>
>> To: Vladimir A Malabanan/Philippines/IBM at IBMPH,
>> Date: 09/27/2013 12:16 AM
>> Subject: Re: [rt.rap.ucar.edu #63182] Adjustment of Time
>>
>>
>>
>> Hi Vladimir.
>>
>> It sounds like you'll be using PointStat to get verification
>> statistics.
>> The discrepancy in valid time can be handled by setting "beg" and
>> "end" (offset in seconds relative to the valid time of the
forecast)
>> in
>> the obs_window in the PointStat configuration file:
>>
>> //
>> // Point observation time window
>> //
>> obs_window = {
>> beg = -5400;
>> end = 5400;
>> }
>>
>> or by setting the obs_valid_beg time and obs_valid_end time options
>> on
>> the command line:
>>
>> "-obs_valid_beg time" in YYYYMMDD[_HH[MMSS]] sets
>> the
>> beginning of the matching time window (optional).
>> "-obs_valid_end time" in YYYYMMDD[_HH[MMSS]] sets
>> the
>> end of the matching time window (optional).
>>
>> This will allow you to set the matching time window you'd like to
>> use
>> for your verification.
>>
>> I hope this helps.
>>
>> Julie
>>
>> On 09/26/2013 01:39 AM, Vladimir A Malabanan via RT wrote:
>>>
>>> Thu Sep 26 01:39:12 2013: Request 63182 was acted upon.
>>> Transaction: Ticket created by vlad.malabanan at ph.ibm.com
>>> Queue: met_help
>>> Subject: Adjustment of Time
>>> Owner: Nobody
>>> Requestors: vlad.malabanan at ph.ibm.com
>>> Status: new
>>> Ticket <URL:
>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=63182 >
>>>
>>>
>>> HI,
>>>
>>> I would like to ask for some assistance in running MET.
>>>
>>> I have a little_r file in local time. I would like to match it
>>> against a
>>> WRF output forecast that is in UTC. Is there a way in MET that
this
>>> can
>> be
>>> set up so that it will match correctly?
>>>
>>> Or I really need to recreate my observed little_r file in UTC?
>>>
>>> Thank you!
>>>
>>>
>>> Vladimir A. Malabanan
>>> 3F IBM Bldg J, UP Ayala Technohub
>>>
>>> Applications Programmer
>>> Advanced Analytics - Data Specialist
>>> Diliman, Quezon City
>>> IBM Solutions Delivery, Inc.
>>> Philippines
>>> Global Delivery Center
>>>
>>>
>>> Phone:
>>> +632-995-2IBM local 8639
>>>
>>>
>>> e-mail:
>>> malabava at ph.ibm.com
>>>
>>>
>>>
>>>
>>>
>>>
------------------------------------------------
Subject: Adjustment of Time
From: Vladimir A Malabanan
Time: Tue Oct 01 09:13:19 2013
Thanks Julie!
I think you have a different way of getting the nearest neighbor of
the
observed points that's is why my matched pairs from MET is not the
same as
my matched pairs in another software that I used.
Thanks anyway for the assistance! :)
Vladimir A. Malabanan
3F IBM Bldg J, UP Ayala Technohub
Applications Programmer
Advanced Analytics - Data Specialist
Diliman, Quezon City
IBM Solutions Delivery, Inc.
Philippines
Global Delivery Center
Phone:
+632-995-2IBM local 8639
e-mail:
malabava at ph.ibm.com
From: "Julie Prestopnik via RT" <met_help at ucar.edu>
To: Vladimir A Malabanan/Philippines/IBM at IBMPH,
Date: 10/01/2013 01:11 AM
Subject: Re: [rt.rap.ucar.edu #63182] Adjustment of Time
That's great news, Vladimir!
Regarding your questions:
#1 Yes, that's 2-m temperature.
#2 I am not completely sure. Perhaps it is the "city block metric".
Here's how it works in the code:
- For each observation, we have a lat/lon value.
- Convert that lat/lon to an x/y location in grid space. Generally,
the x/y values won't fall exactly on a grid point.
- Round the x/values to the nearest integers.
- Use the forecast value at the grid point specified by those rounded
values.
I hope that information will help.
Thanks,
Julie
On 2013-09-30 04:36, Vladimir A Malabanan via RT wrote:
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=63182 >
>
> HI Julie,
>
> It's already working! I was able to match my forecast file vs my
> observed
> file!
>
> I have another question though.
>
> Is the measure I take for consideration in the config below the same
> as
> the field T2 in the WRF forecast?
>
> name = "TMP";
> level = [ "Z2" ];
>
> Also, I am trying to compare my matched pairs from my results in MET
> against my output from another software. I would just like to know
> how MET
> computes for its nearest neighbor (W=1). Is it euclidean metric or
> city
> block metric?
>
> Thanks!
>
>
>
> Vladimir A. Malabanan
> 3F IBM Bldg J, UP Ayala Technohub
>
> Applications Programmer
> Advanced Analytics - Data Specialist
> Diliman, Quezon City
> IBM Solutions Delivery, Inc.
> Philippines
> Global Delivery Center
>
>
> Phone:
> +632-995-2IBM local 8639
>
>
> e-mail:
> malabava at ph.ibm.com
>
>
>
>
>
>
>
>
>
> From: "Julie Prestopnik via RT" <met_help at ucar.edu>
> To: Vladimir A Malabanan/Philippines/IBM at IBMPH,
> Date: 09/28/2013 12:39 AM
> Subject: Re: [rt.rap.ucar.edu #63182] Adjustment of Time
>
>
>
> Hi Vladimir.
>
> Setting the beg/end obs_window to +/- 28800 seconds is close but not
> quite what you should do, because then you'll be comparing 16 hours
> of
> observations to a forecast valid one time. Instead, you should
> probably
> set beg = 28800 and end = 32400. That's a +/- one hour window with
> an
> 8-hour offset applied. You can decide on what matching window you'd
> like to use (maybe +/- 10 min, 30 min, or one hour) and then just
add
> 28800 to both settings to account for the local time vs UTC
> difference.
>
> Alternatively, you can use obs_valid_beg and obs_valid_end on the
> command line, which will override the obs_window settings in the
> config
> file. The example you provided for setting obs_valid_beg and
> obs_valid_end looks correct to me.
>
> Please let us know if you have any other questions.
>
> Thanks,
> Julie
>
> On 2013-09-26 23:13, Vladimir A Malabanan via RT wrote:
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=63182 >
>>
>> Hello Julie,
>>
>> Thanks for the quick response.
>>
>> Yes you are correct. I would like to perform forecast accuracy
using
>> PoinStat tool and my input data are grib forecast files (converted
>> from
>> WRF netcdf files) and an NC observation file (converted from
>> little_r
>> using ASCII2NC).
>>
>> I am new in weather forecasting and I am just doing some
>> proof-of-concept
>> on how to use MET.
>>
>> Just to make it clear, to be able to match my observations (which
is
>> in
>> local time, PHT, (+8 UTC)) to my forecast (in UTC), all I need to
do
>> is to
>> adjust the beg and end time in the config file. Am I correct if I
>> will set
>> beg and end to -28800 and 28800 respectively??
>>
>> On the other hand, if I will add obs_valid_beg and obs_valid_end
>> options
>> in the PointStat script, what values will I put? For example I have
>> a
>> forecast for April 1, 2013 at 00:00:00 UTC. Will I get the correct
>> matching if I just set obs_valid_beg and obs_valid_end to
>> 20130401_080000
>> and 20130401_090000 (my observation file is in UTC +8 )?
>>
>> Thank you!!
>>
>>
>>
>> Vladimir A. Malabanan
>> 3F IBM Bldg J, UP Ayala Technohub
>>
>> Applications Programmer
>> Advanced Analytics - Data Specialist
>> Diliman, Quezon City
>> IBM Solutions Delivery, Inc.
>> Philippines
>> Global Delivery Center
>>
>>
>> Phone:
>> +632-995-2IBM local 8639
>>
>>
>> e-mail:
>> malabava at ph.ibm.com
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> From: "Julie Prestopnik via RT" <met_help at ucar.edu>
>> To: Vladimir A Malabanan/Philippines/IBM at IBMPH,
>> Date: 09/27/2013 12:16 AM
>> Subject: Re: [rt.rap.ucar.edu #63182] Adjustment of Time
>>
>>
>>
>> Hi Vladimir.
>>
>> It sounds like you'll be using PointStat to get verification
>> statistics.
>> The discrepancy in valid time can be handled by setting "beg" and
>> "end" (offset in seconds relative to the valid time of the
forecast)
>> in
>> the obs_window in the PointStat configuration file:
>>
>> //
>> // Point observation time window
>> //
>> obs_window = {
>> beg = -5400;
>> end = 5400;
>> }
>>
>> or by setting the obs_valid_beg time and obs_valid_end time options
>> on
>> the command line:
>>
>> "-obs_valid_beg time" in YYYYMMDD[_HH[MMSS]] sets
>> the
>> beginning of the matching time window (optional).
>> "-obs_valid_end time" in YYYYMMDD[_HH[MMSS]] sets
>> the
>> end of the matching time window (optional).
>>
>> This will allow you to set the matching time window you'd like to
>> use
>> for your verification.
>>
>> I hope this helps.
>>
>> Julie
>>
>> On 09/26/2013 01:39 AM, Vladimir A Malabanan via RT wrote:
>>>
>>> Thu Sep 26 01:39:12 2013: Request 63182 was acted upon.
>>> Transaction: Ticket created by vlad.malabanan at ph.ibm.com
>>> Queue: met_help
>>> Subject: Adjustment of Time
>>> Owner: Nobody
>>> Requestors: vlad.malabanan at ph.ibm.com
>>> Status: new
>>> Ticket <URL:
>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=63182 >
>>>
>>>
>>> HI,
>>>
>>> I would like to ask for some assistance in running MET.
>>>
>>> I have a little_r file in local time. I would like to match it
>>> against a
>>> WRF output forecast that is in UTC. Is there a way in MET that
this
>>> can
>> be
>>> set up so that it will match correctly?
>>>
>>> Or I really need to recreate my observed little_r file in UTC?
>>>
>>> Thank you!
>>>
>>>
>>> Vladimir A. Malabanan
>>> 3F IBM Bldg J, UP Ayala Technohub
>>>
>>> Applications Programmer
>>> Advanced Analytics - Data Specialist
>>> Diliman, Quezon City
>>> IBM Solutions Delivery, Inc.
>>> Philippines
>>> Global Delivery Center
>>>
>>>
>>> Phone:
>>> +632-995-2IBM local 8639
>>>
>>>
>>> e-mail:
>>> malabava at ph.ibm.com
>>>
>>>
>>>
>>>
>>>
>>>
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #63182] Adjustment of Time
From: Julie Prestopnik
Time: Tue Oct 01 09:18:30 2013
Thanks, Vladimir. Please let us know if you have any other questions.
Have a great week!
Julie
On 10/01/2013 09:13 AM, Vladimir A Malabanan via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=63182 >
>
> Thanks Julie!
>
> I think you have a different way of getting the nearest neighbor of
the
> observed points that's is why my matched pairs from MET is not the
same as
> my matched pairs in another software that I used.
>
> Thanks anyway for the assistance! :)
>
>
>
> Vladimir A. Malabanan
> 3F IBM Bldg J, UP Ayala Technohub
>
> Applications Programmer
> Advanced Analytics - Data Specialist
> Diliman, Quezon City
> IBM Solutions Delivery, Inc.
> Philippines
> Global Delivery Center
>
>
> Phone:
> +632-995-2IBM local 8639
>
>
> e-mail:
> malabava at ph.ibm.com
>
>
>
>
>
>
>
>
>
> From: "Julie Prestopnik via RT" <met_help at ucar.edu>
> To: Vladimir A Malabanan/Philippines/IBM at IBMPH,
> Date: 10/01/2013 01:11 AM
> Subject: Re: [rt.rap.ucar.edu #63182] Adjustment of Time
>
>
>
> That's great news, Vladimir!
>
> Regarding your questions:
>
> #1 Yes, that's 2-m temperature.
>
> #2 I am not completely sure. Perhaps it is the "city block metric".
> Here's how it works in the code:
> - For each observation, we have a lat/lon value.
> - Convert that lat/lon to an x/y location in grid space. Generally,
> the x/y values won't fall exactly on a grid point.
> - Round the x/values to the nearest integers.
> - Use the forecast value at the grid point specified by those
rounded
> values.
>
> I hope that information will help.
>
> Thanks,
> Julie
>
> On 2013-09-30 04:36, Vladimir A Malabanan via RT wrote:
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=63182 >
>>
>> HI Julie,
>>
>> It's already working! I was able to match my forecast file vs my
>> observed
>> file!
>>
>> I have another question though.
>>
>> Is the measure I take for consideration in the config below the
same
>> as
>> the field T2 in the WRF forecast?
>>
>> name = "TMP";
>> level = [ "Z2" ];
>>
>> Also, I am trying to compare my matched pairs from my results in
MET
>> against my output from another software. I would just like to know
>> how MET
>> computes for its nearest neighbor (W=1). Is it euclidean metric or
>> city
>> block metric?
>>
>> Thanks!
>>
>>
>>
>> Vladimir A. Malabanan
>> 3F IBM Bldg J, UP Ayala Technohub
>>
>> Applications Programmer
>> Advanced Analytics - Data Specialist
>> Diliman, Quezon City
>> IBM Solutions Delivery, Inc.
>> Philippines
>> Global Delivery Center
>>
>>
>> Phone:
>> +632-995-2IBM local 8639
>>
>>
>> e-mail:
>> malabava at ph.ibm.com
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> From: "Julie Prestopnik via RT" <met_help at ucar.edu>
>> To: Vladimir A Malabanan/Philippines/IBM at IBMPH,
>> Date: 09/28/2013 12:39 AM
>> Subject: Re: [rt.rap.ucar.edu #63182] Adjustment of Time
>>
>>
>>
>> Hi Vladimir.
>>
>> Setting the beg/end obs_window to +/- 28800 seconds is close but
not
>> quite what you should do, because then you'll be comparing 16 hours
>> of
>> observations to a forecast valid one time. Instead, you should
>> probably
>> set beg = 28800 and end = 32400. That's a +/- one hour window with
>> an
>> 8-hour offset applied. You can decide on what matching window
you'd
>> like to use (maybe +/- 10 min, 30 min, or one hour) and then just
add
>> 28800 to both settings to account for the local time vs UTC
>> difference.
>>
>> Alternatively, you can use obs_valid_beg and obs_valid_end on the
>> command line, which will override the obs_window settings in the
>> config
>> file. The example you provided for setting obs_valid_beg and
>> obs_valid_end looks correct to me.
>>
>> Please let us know if you have any other questions.
>>
>> Thanks,
>> Julie
>>
>> On 2013-09-26 23:13, Vladimir A Malabanan via RT wrote:
>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=63182 >
>>>
>>> Hello Julie,
>>>
>>> Thanks for the quick response.
>>>
>>> Yes you are correct. I would like to perform forecast accuracy
using
>>> PoinStat tool and my input data are grib forecast files (converted
>>> from
>>> WRF netcdf files) and an NC observation file (converted from
>>> little_r
>>> using ASCII2NC).
>>>
>>> I am new in weather forecasting and I am just doing some
>>> proof-of-concept
>>> on how to use MET.
>>>
>>> Just to make it clear, to be able to match my observations (which
is
>>> in
>>> local time, PHT, (+8 UTC)) to my forecast (in UTC), all I need to
do
>>> is to
>>> adjust the beg and end time in the config file. Am I correct if I
>>> will set
>>> beg and end to -28800 and 28800 respectively??
>>>
>>> On the other hand, if I will add obs_valid_beg and obs_valid_end
>>> options
>>> in the PointStat script, what values will I put? For example I
have
>>> a
>>> forecast for April 1, 2013 at 00:00:00 UTC. Will I get the correct
>>> matching if I just set obs_valid_beg and obs_valid_end to
>>> 20130401_080000
>>> and 20130401_090000 (my observation file is in UTC +8 )?
>>>
>>> Thank you!!
>>>
>>>
>>>
>>> Vladimir A. Malabanan
>>> 3F IBM Bldg J, UP Ayala Technohub
>>>
>>> Applications Programmer
>>> Advanced Analytics - Data Specialist
>>> Diliman, Quezon City
>>> IBM Solutions Delivery, Inc.
>>> Philippines
>>> Global Delivery Center
>>>
>>>
>>> Phone:
>>> +632-995-2IBM local 8639
>>>
>>>
>>> e-mail:
>>> malabava at ph.ibm.com
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> From: "Julie Prestopnik via RT" <met_help at ucar.edu>
>>> To: Vladimir A Malabanan/Philippines/IBM at IBMPH,
>>> Date: 09/27/2013 12:16 AM
>>> Subject: Re: [rt.rap.ucar.edu #63182] Adjustment of Time
>>>
>>>
>>>
>>> Hi Vladimir.
>>>
>>> It sounds like you'll be using PointStat to get verification
>>> statistics.
>>> The discrepancy in valid time can be handled by setting "beg"
and
>>> "end" (offset in seconds relative to the valid time of the
forecast)
>>> in
>>> the obs_window in the PointStat configuration file:
>>>
>>> //
>>> // Point observation time window
>>> //
>>> obs_window = {
>>> beg = -5400;
>>> end = 5400;
>>> }
>>>
>>> or by setting the obs_valid_beg time and obs_valid_end time
options
>>> on
>>> the command line:
>>>
>>> "-obs_valid_beg time" in YYYYMMDD[_HH[MMSS]]
sets
>>> the
>>> beginning of the matching time window (optional).
>>> "-obs_valid_end time" in YYYYMMDD[_HH[MMSS]]
sets
>>> the
>>> end of the matching time window (optional).
>>>
>>> This will allow you to set the matching time window you'd like to
>>> use
>>> for your verification.
>>>
>>> I hope this helps.
>>>
>>> Julie
>>>
>>> On 09/26/2013 01:39 AM, Vladimir A Malabanan via RT wrote:
>>>>
>>>> Thu Sep 26 01:39:12 2013: Request 63182 was acted upon.
>>>> Transaction: Ticket created by vlad.malabanan at ph.ibm.com
>>>> Queue: met_help
>>>> Subject: Adjustment of Time
>>>> Owner: Nobody
>>>> Requestors: vlad.malabanan at ph.ibm.com
>>>> Status: new
>>>> Ticket <URL:
>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=63182 >
>>>>
>>>>
>>>> HI,
>>>>
>>>> I would like to ask for some assistance in running MET.
>>>>
>>>> I have a little_r file in local time. I would like to match it
>>>> against a
>>>> WRF output forecast that is in UTC. Is there a way in MET that
this
>>>> can
>>> be
>>>> set up so that it will match correctly?
>>>>
>>>> Or I really need to recreate my observed little_r file in UTC?
>>>>
>>>> Thank you!
>>>>
>>>>
>>>> Vladimir A. Malabanan
>>>> 3F IBM Bldg J, UP Ayala Technohub
>>>>
>>>> Applications Programmer
>>>> Advanced Analytics - Data Specialist
>>>> Diliman, Quezon City
>>>> IBM Solutions Delivery, Inc.
>>>> Philippines
>>>> Global Delivery Center
>>>>
>>>>
>>>> Phone:
>>>> +632-995-2IBM local 8639
>>>>
>>>>
>>>> e-mail:
>>>> malabava at ph.ibm.com
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>
>
>
>
--
Julie Prestopnik
National Center for Atmospheric Research
Research Applications Laboratory
Phone: 303.497.8399
Email: jpresto at ucar.edu
------------------------------------------------
More information about the Met_help
mailing list