[meteorite-list] How to compute NEXRAD radar hit times

From: Rob Matson <mojave_meteorites_at_meteoritecentral.com>
Date: Sun, 23 Jan 2011 16:15:02 -0800
Message-ID: <GOEDJOCBMMEHLEFDHGMMIELCEHAA.mojave_meteorites_at_cox.net>

Hi Marc,

Thanks for the fast reply, and confirmation that my interpretation
of how NEXRAD operates was reasonably accurate. It's funny -- up
until a few months ago, I always assumed that when you had double
sweeps at a particular cut angle that they were in indeed separate,
physical sweeps. It was only recently that I began to question that
notion, as I occasionally noticed little or no cloud motion between
consecutive scans at the same cut angle, and yet rather noticeable
changes between consecutive full scans at the same scan angle (i.e.
10 minutes apart versus, say, 1 minute 25 seconds). So -- good to
know that every selectable level-II cut angle is its own sweep.

> It turns out that the time hack for each sweep is embedded in the
> data, but the NOAA software doesn't bother to show it (or so I've
> been told).

I figured it might be -- it's a shame that NOAA's tool doesn't
make use of it.

> The GRlevelX software I use reads the time hack for each cut
> and displays it.

Sounds like it might be time for me to ditch NOAA's Weather and
Climate Toolkit ... especially now that it can no longer read
half the Level-II files out there.

> There can be several seconds' discrepancy between what I
> calculate manually and what the hack reports; I tend to believe
> the embedded time hack because there is additional time needed
> in each change the radar's azimuth between scans and possibly
> to let the radar's computer finish crunching numbers.

I wondered whether the radar could actually instantaneously
jump by one degree in elevation angle each time the cut angle
changed. If there is mechanical steering involved, then obviously
the jump can't be instantaneous. Surely they don't stop the
azimuth sweep while making the elevation angle change, so there
must be a brief transitional period surrounding the instants
when the radar points due north where the elevation angle
smoothly sweeps from one cut angle to the next. This transition
must take even longer at the end of the scan, when the elevation
angle drops all the way back down to 0.5 degrees.

--Rob
Received on Sun 23 Jan 2011 07:15:02 PM PST