Re: [awips2dev] [gfe] Obs Grid QC Has Become Worthless All - Home - RTMA/URMA Discussion Group

There is a VLab page that has instructions for blacklisting a few observations.  If you have entire networks that need removed (like 20+) it is better to contact Jacob Carley and they can tailor this to your needs.  

The SOO in each office is the main POC for this.  

On Wed, Oct 30, 2019 at 4:26 PM Jonathan Lamb - NOAA Federal <jonathan.lamb@noaa.gov> wrote:

Jeff Craven,

What's the procedure for getting obs removed from URMA? (I should probably know this...).

Thanks,

Jonathan

On Wed, Oct 30, 2019 at 3:57 PM Paul Iniguez - NOAA Federal <paul.iniguez@noaa.gov> wrote:

The image below is the box where KPHX (Phoenix Sky Harbor) is located. The red dot is where the ASOS is. Someone tell me why the grid value must match the dot (which is an observation subject to errors/tolerances). Please note the grid contains several terminals, runways, streets, businesses, a few houses, part of an 8-lane freeway, and a river bed.

Paul Iñiguez

Science & Operations Officer, NOAA/NWS Phoenix, AZ

602.275.7002 x224 - Weather.Gov/Phoenix - @NWSPhoenix - Facebook.com/NWSPhoenix

On Wed, Oct 30, 2019 at 10:02 AM Robert Ballard - NOAA Federal <robert.ballard@noaa.gov> wrote:

I'd echo the concerns of others that the RTMA/URMA analysis is very poor (especially in areas of terrain and/or coastline), and falls well short of what's needed to support operations and IDSS at this point.   It's not a philosophical difference, the analysis is simply wrong and unusable in many critical areas.  I have made Pacific Region leadership aware of this on several occasions. 

My 2¢.

- Bob

 --

Robert Ballard
Science & Operations Officer
National Weather Service Honolulu/Central Pacific Hurricane Center

On Sat, Oct 26, 2019 at 4:56 AM Jeffrey Craven - NOAA Federal <jeffrey.craven@noaa.gov> wrote:

Hello Chad.  There is no requirement for RTMA/URMA to exactly match the observations.  

If this is what the field absolutely needs then it has to be elevated through the governance so that EMC is instructed to match the obs.  There is a balance between the background field and overmatching observations.

I agree that meeting halfway between first guess analysis and obs is not how we have done business in the past.  I have suggested to EMC that RTMA URMA always be forced to be within 3F or 3Knots, etc of the METARS which are the gold standard.  However I have no authority to make that call.

The RTMA and URMA METAR inputs  are overwhelmed by the MADIS observations and each and every WFO must own that and use your expertise to remove and blacklist observations to make it the best it can be.   Many of these problem obs can be removed and greatly improve the RTMA.  

  If you need exact matching the METARS, Your SOO and MIC must ensure that the regional offices push up the chain that this is a requirement.

Up to this point it has not been a requirement; there are quite a few folks that also believe overmatching obs is just as scientifically invalid as those that expect nearly exact matching.

On Sat, Oct 26, 2019 at 10:29 AM Chad Gimmestad - NOAA Federal <chad.gimmestad@noaa.gov> wrote:

Amen and thank you from Boulder.  

Jeff is right philosophically in where we want to be going, but we need something that's not embarrassing to use TODAY!  Our forecast process is becoming tied to something that's not working well enough yet.  Mountain winds and humidities are analysis problems here everyday, and that is propagating all the way through to the NBM and our forecasts.

Unfortunately, you have to keep the old car running until you have the new car actually working right.  It's an investment no one wants to make. Quicker delivery of an URMA that meets our needs gets us out of this situation sooner, but in the meantime we need good forecasts.

 

Chad Gimmestad

Senior Forecaster

National Weather Service

Boulder, Colorado

On Fri, Oct 25, 2019 at 2:03 PM Jerry Wiedenfeld - NOAA Federal <jerry.wiedenfeld@noaa.gov> wrote:

Hi All,

I have opened ticket #249109 with NCF.  

Jerry

On Fri, Oct 25, 2019 at 2:30 PM Alexander Tardy - NOAA Federal <alexander.tardy@noaa.gov> wrote:

Us too at San Diego and we have provided examples of locations such as Palm Springs (3F difference) or other legacy climate sites.

URMA/bias correction needs to include climate sites and the exact value that occurred when known.  The media uses this and someone 50 years from now.

I know that we have microclimates and all the complex terrain and that temperatures vary at day or night every 1 mile but

the climate sites need to be reflected. This builds trust and confidence in our ability to forecast basic weather conditions. 

Thanks, Alex

Alex Tardy

Warning Coordination Meteorologist, Manager
Emergency Preparedness and Partner Collaboration

Education and Outreach Coordinator 

Media and Public Information Officer

Cell: 858-442-6016  Office: 858-675-8700

Skywarn Program Manager

NOAA/National Weather Service
11440 W. Bernardo Court, San Diego, CA

weather.gov/sandiego

Register as a Weather-Ready Nation Ambassador (click here)

Facebook page for NWS San Diego

Twitter @NWSSanDiego

http://www.youtube.com/NWSSanDiego YouTube

We need precipitation reports! http://cocorahs.org/

On Fri, Oct 25, 2019 at 12:11 PM Ryan Kittell <ryan.kittell@noaa.gov> wrote:

We (LOX) have a similar situation to the Billings office. We have a number of high profile sites that URMA completely mishandles. We have added mesowest stations and blacklisted sites, which has helped some, but not well enough.  We are still trying to figure out solutions, but until one comes we are reluctant to completely replace MatchObsAll with URMA. 

As such, we still need the ObsQC program and it is crashing and stalling for us after years and years of no issues.

--

Jerry Wiedenfeld 

Information Technology Officer 

National Weather Service, Milwaukee/Sullivan, WI

work phone - 262-392-5526 x 486

--

JPC

Jeff Craven

Chief, Statistical Modeling Division (SMD)

National Weather Service, W/STI-12

Meteorological Development Laboratory (MDL)

Room 10410, SSMC2

Silver Spring, MD 20910

(301) 427-9475 office

(816) 506-9783 cell/text

@jpcstorm

https://www.weather.gov/mdl/nbm_home

--

JPC

Jeff Craven

Chief, Statistical Modeling Division (SMD)

National Weather Service, W/STI-12

Meteorological Development Laboratory (MDL)

Room 10410, SSMC2

Silver Spring, MD 20910

(301) 427-9475 office

(816) 506-9783 cell/text

@jpcstorm

https://www.weather.gov/mdl/nbm_home