Between 5:09 AM and 8:10 AM local time, eight aircraft got TCAS Resolution Advisories, RAs. Several more got (less serious) Traffic Advisories, TAs. The RAs all happened within a 1 mile section of the approach to DCA.
Aircraft reported that TCAS showed a descending intruder (the TCAS term for the other aircraft) at about 2500 feet. Aircraft at 1900 feet or above were instructed to descend, which still leaves an altitude margin but not a big one. And can mess up an approach, which is why 3 jets chose to go around.
A possible explanation is a system being tested on the ground caused the phantom target. @engelsman.bsky.social found this FAA SAFO mentioning an ADS-B system ground test that caused a "false airborne target" leading to a spurious RA. https://www.faa.gov/sites/faa.gov/files/SAFO17002.pdf
I'm not sure the exact scenario of a bogus ADS-B broadcast is what happened in this situation since DCA tower said they didn't see any other traffic on their radar. But maybe their system doesn't show ADS-B targets that don't have any corresponding primary or secondary radar target.
This FAA SAFO seems to imply that a bogus ADS-B broadcast alone is enough to generate a TCAS RA, which is not what I expected given the TCAS docs I've found–I thought "hybrid surveillance" used Mode S to filter out aircraft that didn't need to be actively interrogated. https://www.faa.gov/sites/faa.gov/files/SAFO17002.pdf
You know, I think I read that SAFO as implying the ADS-B Out part of the test is significant, but I think if it had just been a mode C transponder the same thing would have happened.
Comments
The system is insecure, but it's not totally trivial to spoof an aircraft. It can be done, though.
Longo et al managed to trigger false resolutions last year, but it was fairly demanding.
https://www.usenix.org/conference/usenixsecurity24/presentation/longo