Brief Tsunami Warning Startles U.S. West Coast, Reveals Strengths and Weaknesses in Tsunami Preparedness
A magnitude 7.2 earthquake off the northern California coast on the evening of June 14 triggered a brief tsunami warning for the entire U.S. west coast, and a flurry of news-media inquiries to earthquake and tsunami experts in the U.S. Geological Survey (USGS) and the National Oceanic and Atmospheric Administration (NOAA). NOAA's National Weather Service issued the tsunami warning about 5 minutes after the earthquake struck and cancelled it about 1 hour later, after data received from shoreline tide gauges and ocean-bottom sensors showed that the earthquake had generated only a small tsunami, about 1 cm high in the open ocean and not readily detectable at nearby tide gauges.
"This was a good test of the [tsunami-warning] system, and the system itself worked very well," said Eric Geist, a USGS geophysicist stationed in Menlo Park, CA. The National Weather Service issued the tsunami warning from its West Coast and Alaska Tsunami Warning Center (WC/ATWC) in Palmer, AK, at 7:56 p.m. PDT, about 5 minutes after the earthquake struck. The warning estimated that the initial wave would take just 38 minutes to travel from the earthquake epicenter to Crescent City, CA, which lies about 150 km (90 mi) northeast of the epicenter and has been hit by tsunamis in the past. One of the National Weather Service's TsunamiReady communities (see URL http://www.stormready.noaa.gov/tsunamiready/), Crescent City sounded its warning sirens shortly after 8:00 p.m., and thousands of residents and visitors were evacuated. In many other areas, communication problems prevented the tsunami warning from reaching the public until after the warning had been canceled. "Once an alert is issued by the National Weather Service, it gets relayed to two large groups: emergency managers and the public. Tuesday's warning was a good wakeup call," said Geist. "It revealed some gaps in communication and planning that local agencies and communities now have an opportunity to fix."
The tsunami warning was based on the earthquake's large magnitudeinitially calculated at 7.4 by the WC/ATWC and later revised to 7.2 by the USGS National Earthquake Information Center (after information about all the long-period seismic waves had been received from stations around the world). Concern was increased by the epicenter's location near the Mendocino Triple Junction, an intersection of three faults that include the Cascadia subduction-zone thrust fault, which is capable of generating large tsunamis. These parametersmagnitude and locationcan be calculated within minutes of the earthquake's occurrence, allowing a warning to be issued fast enough for evacuation of the closest coastal areas.
Information that takes longer to gather is used to determine whether a tsunami warning should be expanded, continued, or canceled. Most important is direct measurement of water levels by NOAA's ocean-bottom sensors and shoreline tide gaugesdata relayed by satellite to the tsunami-warning centers. Well after the predicted arrival times, no tsunami had been clearly detected at the three tide gauges nearest the epicenterin Humboldt Bay near Eureka, CA; at Crescent City, CA; and at Port Orford, OR. (Later analysisafter filtering out "noise" caused by wind waves and other factorsrevealed that the tide-gauge data did record a tsunami; its height at Crescent City was estimated at 10 cm by the WC/ATWC.) An offshore bottom sensor farther from the epicenter recorded a miniscule tsunamiabout 1 cm highin the open ocean. After careful consideration of all available information, the National Weather Service canceled the tsunami warning at 9:09 p.m. PDT.
Analysis of earthquake data helped seismologists explain why only a tiny tsunami had been generated: the earthquake occurred on a strike-slip fault, now believed to be a left-lateral fault within the Gorda plate, a fault-ridden piece of oceanic crust being compressed and deformed by the northward-moving Pacific plate. Earthquakes on strike-slip faults produce much less vertical movement of the sea floor than do earthquakes on thrust faultssuch as the thrust fault whose rupture triggered the devastating December 2004 Indian Ocean tsunamiand so strike-slip earthquakes are unlikely to generate damaging tsunamis. Such earthquakes can, however, trigger landslides capable of generating dangerous tsunamis, as noted by scientist Bruce Turner of the WC/ATWC, who emphasized the importance of using direct measurements of water level, rather than identification of the type of fault the earthquake ruptured, to determine when to call off a tsunami warning.
NOAA uses two types of water-level sensors to detect tsunamis: shoreline tide stations and offshore ocean-bottom sensors called DART (Deep-ocean Assessment and Reporting of Tsunamis) buoys. Currently, NOAA operates several dozen tide stations on Pacific coastlines and six DART buoys in the Pacific Ocean: two offshore the Cascadia subduction zone in northern California and the Pacific Northwest (one of these, station 46405-D130, recorded the 1-cm-high tsunami generated by the recent earthquake), three offshore the Aleutian Island subduction zone in Alaska, and one on the Equator about 5,000 km west of Peru. The Hydrographic and Oceanographic Service of the Chilean Navy (SHOA) operates a seventh sensor off the coast of Chile. (For more information, visit the National Data Buoy Center.)
Prompted by the December 2004 Indian Ocean tsunami, the United States plans to greatly expand its network of DART buoys, not only in the Pacific Ocean but also in the Atlantic Ocean and the Caribbean Sea. NOAA and the USGS held a joint workshop in early July to discuss the selection of sites for the new buoys, with the USGS providing information about which areas are most likely to undergo tsunami-triggering earthquakes, landslides, and volcanic eruptions.
in this issue:
Brief Tsunami Warning Startles U.S. West Coast