Forums

Back

Dewpoint issues around the Great Salt Lake in operational/parallels

RSS (Opens New Window)(Opens New Window)
DV
Darren Van Cleave, modified 7 Years ago.

Dewpoint issues around the Great Salt Lake in operational/parallels

Youngling Posts: 37 Join Date: 1/8/14 Recent Posts

Hi developers,

 

We're using the RTMA/URMA operationally here at Salt Lake. Several of our forecasters have discovered significant issues in the Td fields lately, which extend at least partially to the parallels as well. Within the last few days, we've seen anomalously low artifacts around the Great Salt Lake. In the attached Dewpoint_Problem.png image (6/8/18 01Z), you can see that the issue is not in the first guess field based on the difference plot (> 30° difference!). The issue is also present in the parallel, though not as noticeable and it's more variable from hour to hour (see Dewpoint_Problem_Parallel.png). 

 

Based on the newness of the problem I wonder it is from a bad ob, but if so I'm unable to locate the ob and need some help to find it for blacklisting. The URMA data points option on the Blender Viewer page is only showing temperature obs. I believe there was a way to see all the obs used each hour in Google Earth, is that still the case? If so, can you share that location again? Also, if it is a grossly bad ob, why is the QC process not flagging it?

 

-Darren

 

 

 

Attachments:

Attachment

Attachment

DV
Darren Van Cleave, modified 7 Years ago.

RE: Dewpoint issues around the Great Salt Lake in operational/parallels

Youngling Posts: 37 Join Date: 1/8/14 Recent Posts
Update: I think I found the Google Earth files here: http://ftp.emc.ncep.noaa.gov/mmb/rtma2/for_eval/ob_files/

If it is a bad ob sneaking through, I was unable to find it via the KML files. For now, I've limited our local representation of the URMA to not dip below 0° as a temporary band-aid.

-Darren

On Fri, Jun 8, 2018 at 9:40 AM, VLab Notifications <VLab.Notifications@noaa.gov> wrote:

Hi developers,

 

We're using the RTMA/URMA operationally here at Salt Lake. Several of our forecasters have discovered significant issues in the Td fields lately, which extend at least partially to the parallels as well. Within the last few days, we've seen anomalously low artifacts around the Great Salt Lake. In the attached Dewpoint_Problem.png image (6/8/18 01Z), you can see that the issue is not in the first guess field based on the difference plot (> 30° difference!). The issue is also present in the parallel, though not as noticeable and it's more variable from hour to hour (see Dewpoint_Problem_Parallel.png). 

 

Based on the newness of the problem I wonder it is from a bad ob, but if so I'm unable to locate the ob and need some help to find it for blacklisting. The URMA data points option on the Blender Viewer page is only showing temperature obs. I believe there was a way to see all the obs used each hour in Google Earth, is that still the case? If so, can you share that location again? Also, if it is a grossly bad ob, why is the QC process not flagging it?

 

-Darren

 

 

 


--
Darren Van Cleave RTMA/URMA Discussion Group Virtual Lab Forum https://vlab.noaa.gov/web/715073/home/-/message_boards/view_message/4220269 VLab.Notifications@noaa.gov



--
Darren Van Cleave
Science and Operations Officer
NWS Salt Lake City
(801) 524-5196
weather.gov/slc | @NWSSaltLakeCity
SL
Steven Levine, modified 7 Years ago.

RE: Dewpoint issues around the Great Salt Lake in operational/parallels

Youngling Posts: 174 Join Date: 11/13/14 Recent Posts

Darren,

That is the correct location of the KML ob files.

We haven't figured out exactly what is happening here, but a couple general notes worth making:

-We don't analyze dew point, it is converted from temperature/moisture/pressure analyses.  So any issue there may carry over

-There are no obs (restricted or otherwise) to obviously account for this pattern that I can see.

We will update as we find anything.

Steve

SL
Steven Levine, modified 7 Years ago.

RE: Dewpoint issues around the Great Salt Lake in operational/parallels

Youngling Posts: 174 Join Date: 11/13/14 Recent Posts

Darren,

It looks like there area couple mesonet obs on islands in the lake (Gunnison Island and Hat Island) that are reporting moisture levels much lower than the background.  It also appears that, due to the escarpment relaxation we put in at your request, some of their influence is seeping onto the land, further drying out air that is already quite dry according to the background field. 

The plots are of specific humidity, which is what URMA actually analyzes.  These two obs have innovations of -3 g/kg.  When nearby areas have a background values of less than 3 g/kg, obviously you can't have negative moisture but that can make for some very dry analysis values.  At these levels, even small absolute amounts of moisture can change the dew point significantly.

If you click on the links above, you can see the the dew points at these obs varied wildly at 00-02Z on the 8th (the links are set to that time already).  In each station, I can see a 9-10 F change in dew point in 5 minutes.  I don't know how realistic those values are.  If they are not, we can flag those two stations for moisture and see if that has an impact.

In the current URMA, all those obs will sort of be mushed together.  In the parallel URMA/URMAX we only use the ob closest to analysis time, right on the hour in this case.  I think that may have something to do with the dulling that takes place in the URMAX.

https://docs.google.com/presentation/d/1-Pf1CYSe258nRpjio-pLdX8MaSv3DZ37KjppdDswpv4/edit?usp=sharing

Steve