Steve,
Really appreciate the follow-up, thanks. In the actual physical
world however, there is no wall of heat there. I'm quite familiar
with both that area and the meteorological conditions then
prevailing, and that simply isn't what one would encounter.
Temperature there on that day would vary appreciably only with
elevation. And I believe that's where the problem is.
What is the actual definition of the temperature/max temp of a
2.5 x 2.5 km NDFD grid pixel? Here's what's stated in Appendix A of
NWSI 10-201:
-------------------------------------------------------------------------------------
"2 Overarching Grid Element Concept and Guidelines
Element values represent conditions of meteorological fields at
the resolution of the grid."
--------------------------------------------------------------------------------------
The mean elevation of the grid pixel containing the peak is 2,818
ft. The one observation site in that pixel, SJS02, is at the summit
elevation of 3,849 ft and reported a MaxT of 72 deg. The 3 closest
stations to the west and southwest have MaxTs of 89 deg at 296 ft,
86 deg at 853 ft, and 89 deg at 817 ft (Fig 1). So around 15 deg in
3000 ft which is close to dry adiabatic (5.38 deg/1000 ft).
Thus 72 deg at the summit elevation would correspond to ~ 77 deg
for the 2,818 ft mean elevation of the pixel. The pixel immediately
to its west (across the "wall of heat") has a mean
elevation of 1,509 ft. Continuing down at 5 deg/1000 ft yields a
pixel max temp of ~ 84 deg.
Now instead of having a "wall of heat" with a 19 deg
difference between adjoining pixels as in the URMA (Fig
2), it's only 7 deg, from 77 to 84 -- perfectly reasonable
given the difference in respective mean pixel elevations.
This in turn is entirely consistent with what our final in-house
Max T obs grid had: 77 deg to 83 deg (Fig 3). It's also consistent
with the short-range HRRR forecast of 79 deg to 84 deg (Fig 4), and
our own final official forecast of 77 to 85 deg (Fig 5).
With apologies, continue to believe there's a fundamental
methodological flaw (or alternatively an approach inconsistent with
the basic NDFD grid definition). And am concerned as to the adverse
consequences it could have on our forecast verification.
Best Regards,
Warren
P.S. Summit Max T obs of 72 appears to be valid (in contrast to
what I wrote in my first email).
