On Wednesday, Jan 17 2001, just after 7:34 AM, a magnitude 2.4 earthquake struck the Upper East Side of Manhattan. There were no reports of injuries or damage, despite the location beneath the densest concentration of Ming vases and other tschotkes on the planet. The location of this event, and its relatively large felt area, produced a huge media reaction. Once again, our seismology group found itself in the middle of public attention.

Coincidentally, Won-Young Kim and Jeremiah Armitrage only a few weeks ago finished the installation of a new automated earthquake location system on the Lamont Cooperative Seismic Network. This system intercepts the digital signals telemetered from our outlying stations and analyzes bits of the time series in near real time for anomalous "events". Upon detection of an "event" at one station, it polls the others. If all the stations vote that an event occurred, the program tries to pick arrivals and locate the event. The system then sends an e-mail to interested folks alerting them to the event and its preliminary location. This happens several times a week in our area.

On Wednesday morning, everything worked fine, anomalous arrivals were detected, time-picks were made, and the event got itself located near Newark. This is not too surprising: we have no stations in NYC (too noisy), and automatic P-pickers have a difficult time with distant signals until they've been tuned up. We are in the middle of this fine-tuning process for our network, but one needs events and they are somewhat hard to come by in the metro region. In any case, Won-Young was on top of the event within 20 minutes, and started reviewing the automatic solution. Something didn't look right, so he started P-picking by hand. As it turned out, it wasn't the trigger calibration: something else was wrong.

Meanwhile, we started getting calls from the press about 8:00 AM. The USGS, alerted by a blip on one of its nearby instruments but with no location capability for this region, called around 8:15 AM. Lerner-Lam sent them the computer location with the caveat that we were still reviewing the picks, but that the event was probably within the "southern metro area". The USGS posted the Newark location anyway, and it was this location that hit the national newswires.

This news was a surprise to the good people of Newark, and confusing to the folks on the East Side and in Long Island City who actually felt the quake. Kim worked diligently to refine the location while the rest of us tried to describe error ellipses to the press.

By 10AM, Stacey Vasallo and Bonnie Mayer had organized the incoming press and the callbacks, and had set up a rotating schedule for live and taped interviews. By 10:00 AM, Kim had figured out what was wrong, and Newark was out and Roosevelt Island was in as the preferred location. This was further refined by noon to the Upper East Side. The press usually asks us to show them the Richter Scale. This time, they were asking for the street address of the quake. Scholz, Sykes, Jacob, Armbruster, and Rodriguez all gave interviews, including some appearances on the morning-after news shows.

The moving epicenter caused some confusion in the press, and with the USGS, which was cleared up by late morning. As Kim and Armitrage finally discovered, something, probably a block of ice, was blocking the GPS antenna connected to our seismograph in Basking Ridge, causing that instrument to shift to an internal, less accurate clock and yielding a false arrival time. Once that pick was removed, the location "clicked" into the East Side. But the need to be authoritative and accurate is not easy to reconcile with both the professional and public pressure to be quick. Despite identifying the location with words like "preliminary" and "unreviewed" and "might change", the first public release by the USGS gave the location verbatim to a tenth of a degree. We learn something new with each event; in our post-mortem, we decided to adopt more explicit wording for our location releases, and to isolate the analysts from the press so that they can work. And despite our more than 25-year relationship with the USGS monitoring office, the information handoff can still be fumbled.

Despite its small size, this earthquake is a reminder that New York City is not risk free. A magnitude 5.2 struck in 1884, just off the Rockaways, and there is a one-in-a-few-hundred chance that a similar earthquake could strike NY in the next year. The concentration of "assets" in NY is so high, that NY ranks fourth (behind LA, San Francisco, and Seattle) in earthquake risk normalized by potential impact. And our built environment has not been subjected to the Darwinian removal of poor construction that would occur in a more quake-prone area. Such "low-probability - high-impact" risks contribute substantially to the risk profile we face as individuals, and, in the aggregate, as a society. (Think of the reasons why you wear a seatbelt, or swap your Firestones for Goodyears.) It is an interesting and necessary thought experiment to consider how we might manage such risks.

Herewith the "official, preliminary, reviewed, location":

Origin time: 01/17/2001 07:34:22.6 (12:34:22.6 UTC)
Location: 40.777N, 73.954W, Depth=7 km
Magnitude: Mc=2.4 (coda magnitude)

95% confidence ellipse about 1 km in diameter.

The event was located within Manhattan's Upper East Side.
The shock was felt in Queens and Manhattan.

If you felt the earthquake, please go to :

to report it. The resulting map will help us analyze our
instrumental data, and calibrate historic felt reports in the area.
._______________ END ____________________________.