Radar anomalies

[phaseform]

Any of you Radar aficionados care to explain this?
https://twitter.com/ABOwarrior/status/1243721130109796353
bonus points if you can avoid a conspiratorial answer

[Nerull]

Look at the publicly available and archived level 2 NEXRAD data from the individual radar sites, not a mosaic assembled by who-knows-what software using who-knows-what clutter suppression algorithms. That looks like fairly typical clear air mode (VCP 31/32/35) early morning clutter that always shows up on radar being filtered by by some automated processing system, which for some reason stops removing it for a few frames.

If whatever was assembling the mosaic wasn't actively filtering out data to clean up the image you'd see a fuzzy haze around every single radar site, which constantly varies in size and intensity based on radar scan mode, time of day, where the sun happens to be, humidity, insect activity, etc. The only time NEXRAD radar produces an empty image with no returns is when its broken.

Here's what a fairly typical evening looks like without clutter suppression, the blooming is caused by an atmospheric inversion that sets in as the ground layer cools:


Here's another:

[LaserSteve]

The Big C forming over Detroit is the Cleveland Indians wining against the Tigers.
 
Steve

[phaseform]

Coool. So basically the mosaic has a faulty connection.
So this 'clutter' (=atmospheric inversion) is being poorly filtered on sites KVAX, and KGSP as here

makes sense

[Nerull]

I doubt its a faulty connection. It has some automated algorithm it applies to decide whether or not to suppress the output of a radar site and for whatever reason its toggling state. Without knowing how it works it could be almost anything. Maybe atmospheric ducting steered the beam into a windfarm. (This is actually extremely common).

Electronics can sometimes get into a state where they're not quite stable when given marginal inputs and they start to oscillate back and forth. Algorithms can do the same thing.

[radar_macgyver]

That's unlikely to be wind turbines. Those only show up as point targets. If it's large scale, like that and it was real, it was more likely an inversion layer. However, the regions where it happened were in the coverage of two different radars (KFFC and KJAX) so it was most likely some weirdness in the data ingest into Accuweather's mosaic algorithm, which is "proprietary" and I trust about as far as I can throw it. Notice how the edges are straight lines? There's no datestamp, so I can't go pull the data from NCDC and look at the dual pol variables to see if it was insects/birds.

[Nerull]

I'm not suggesting that the data itself is from wind turbines, but wind turbines could look enough like a storm to trip the algorithm which decides whether or not to include a radar in the mosaic.

The NEXRAD data processing system certainly loves assigning storm attributes to wind farms.

[Qw3rtzuiop]

Those are birds and bats.

https://twitter.com/RadarAndStuff/status/1043575880054988800