Nick is
correct here. Patrick, I am assuming the plan view plot is from the
time of your message. This suggests the RTMA matches the actual
observation well. So, the observation trace on the time series
comparison is likely the bogus value from the GLERL adjusted
synthetic observation (which is not a good truth source). Given this
time series is with offshore flow, the GLERL adjustment will be
artificially high. With the HRRRv4 background in place, the
synthetic observations will need to be sunsetted in the next version
of RTMA / URMA.
Greg Mann, Ph.D.
-
Science &
Operations Officer
NOAA /
National Weather Service, Detroit MI
Mobile:
810.423.5188 (currently on Telework)
Hi Patrick,
I think I might know what is going on here....maybe.
I pulled up the
RTMA QC kml and opened it with Google
Earth for 1/30/2021 00 UTC. KDYT is in there twice, once as
"KDYT" with the provider as "METAR" showing
an observed wind speed of 13.02 knots. It is also listed as
"KDYT" with the provider "GLERL ADJUSTED"
with an observed value of 22.04 knots. That value in your plot
looks like it matches the GLERL ADJUSTED. Greg Mann would know
way more about this, but basically, the GLERL Adjusted obs were
created to help with the lake/land mask when the RTMA was
using a course background, RAP I think. Around a year ago this
topic came up and we were able to blacklist the GLERL adjusted
because they were causing issues in the RTMA/URMA analysis and
therefore NBM bias correction, and not really necessary with the
HRRR as the background.
I've cc'd Greg Mann, maybe he can chime in.
Best,
-Nick
All,
I
noticed winds at KDYT Duluth Sky Harbor Airport in the
marine environment (see map below) have been really too
low the past couple of days. All of our sites outside of the
marine environment have been tracking much better:
Thanks!
--
Patrick Ayd
Science
and Operations Officer
NOAA/NWS
WFO Duluth, MN
Office: 218-729-6697
Cell: 651-245-7829
--
Nick Langlieb
Science and Operations Officer
National Weather Service
WFO
Marquette, MI
906-475-5782 ext 766
