- Email: [email protected]
Emergency response and field observation activities of geoscientists in California (USA) during the September 29, 2009, Samoa Tsunami
Earth-Science Reviews 107 (2011) 193–200
Contents lists available at ScienceDirect
Earth-Science Reviews j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e a r s c i r ev
Emergency response and field observation activities of geoscientists in California (USA) during the September 29, 2009, Samoa Tsunami Rick I. Wilson a,⁎,1, Lori A. Dengler b, James D. Goltz c, Mark R. Legg d, Kevin M. Miller e, Andy Ritchie f, Paul M. Whitmore g a
California Geological Survey, 801 K Street MS 12–31, Sacramento, California, USA 95814 Humboldt State University, Geology Department, #1 Harpst Street, Arcata, California, USA 95521 California Emergency Management Agency, 1200 East California Blvd., Mail Code 104–44, Pasadena, California, USA 91125 d Legg Geophysical, 16541 Gothard St # 107, Huntington Beach, California, USA 92647 e California Emergency Management Agency, 1300 Clay Street, Suite 400, Oakland, California, USA 94612 f U.S. Geological Survey, 400 Natural Bridges Drive, Santa Cruz, California, USA 95060 g West Coast/Alaska Tsunami Warning Center, NOAA-NWS, 910 South Felton Street, Palmer, Alaska, USA 99645 b c
a r t i c l e
i n f o
Article history: Received 29 June 2010 Accepted 25 January 2011 Available online 1 February 2011 Keywords: Tsunami emergency management geoscientist field observations warning center local operational areas
a b s t r a c t State geoscientists (geologists, geophysicists, seismologists, and engineers) in California work closely with federal, state and local government emergency managers to help prepare coastal communities for potential impacts from a tsunami before, during, and after an event. For teletsunamis, as scientific information (forecast model wave heights, first-wave arrival times, etc.) from NOAA's West Coast and Alaska Tsunami Warning Center is made available, federal- and state-level emergency managers must help convey this information in a concise, comprehensible and timely manner to local officials who ultimately determine the appropriate response activities for their jurisdictions. During the September 29, 2009 Tsunami Advisory for California, government geoscientists assisted the California Emergency Management Agency by providing technical assistance during teleconference meetings with NOAA and other state and local emergency managers prior to the arrival of the tsunami. This technical assistance included background information on anticipated tidal conditions when the tsunami was set to arrive, wave height estimates from state-modeled scenarios for areas not covered by NOAA's forecast models, and clarifying which regions of the state were at greatest risk. Over the last year, state geoscientists have started to provide additional assistance: 1) working closely with NOAA to simplify their tsunami alert messaging and expand their forecast modeling coverage; 2) creating “playbooks” containing information from existing tsunami scenarios for local emergency managers to reference during an event; and, 3) developing a state-level information “clearinghouse” and pre-tsunami field response team to assist local officials as well as observe and report tsunami effects. Activities of geoscientists were expanded during the more recent Tsunami Advisory on February 27, 2010, including deploying a geologist from the California Geological Survey as a field observer who provided information back to emergency managers. © 2011 Elsevier B.V. All rights reserved.
Contents 1. 2.
3.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . Event response process . . . . . . . . . . . . . . . . . . . . . . 2.1. WC/ATWC tsunami forecast activities . . . . . . . . . . . . 2.2. WC/ATWC tsunami alert criteria . . . . . . . . . . . . . . . 2.3. Communicating tsunami alert information to local jurisdictions The September 29th, 2009 tsunami advisory for California . . . . . . 3.1. Emergency response activities . . . . . . . . . . . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
⁎ Corresponding author. Tel.: +916 327 0981; fax: +916 445 3334. E-mail addresses: [email protected] (R.I. Wilson), [email protected] (L.A. Dengler), [email protected] (J.D. Goltz), [email protected] (M.R. Legg), [email protected] (K.M. Miller), [email protected] (A. Ritchie), [email protected] (P.M. Whitmore). 1 PG 5878, CEG 1881. 0012-8252/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.earscirev.2011.01.010
194 195 195 195 196 196 196
194
R.I. Wilson et al. / Earth-Science Reviews 107 (2011) 193–200
3.2. Tsunami field response activities . . . . . . . . . . . . . 3.3. Evaluation of forecasted information . . . . . . . . . . . 3.4. Conclusions about the Samoa event and future work . . . 4. Addendum: the February 27th, 2010 tsunami advisory in California Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction The State of California's 1700-kilometer-long coastline has over 1 million residents and tens of millions of visitors each year that could be at risk to tsunami hazards (Fig. 1; California Seismic Safety Commission, 2006). State geoscientists (geologists, geophysicists, seismologists, and engineers) work closely with federal, state and local emergency managers to help identify and mitigate tsunami hazards in California. The primary focus of this work in California includes: 1) creating tsunami inundation maps for emergency response planning; 2) providing technical information about tsunami hazards to coastal emergency managers through presentations, workshops, and tsunami response exercises; and, 3) collecting and analyzing tsunami hazard information
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
197 198 198 199 199 200
to improve the overall knowledge base of the scientific community (Patton and Dengler, 2004, 2006; Wilson et al., 2008; Barberopoulou et al., 2009). As one of the original member states in the U.S. National Tsunami Hazard Mitigation Program (NTHMP), California's tsunami program has been very active in helping to establish policy and guidance with regard to tsunami hazard mitigation at a national level. The state's membership on the NTHMP Coordinating Committee includes a geologist from the California Geological Survey (CGS) and an emergency manager from the California Emergency Management Agency (CalEMA), sharing responsibility for California to implement the state tsunami hazard mitigation and preparedness plan. Part of this plan is for geoscientists from CGS and other organizations in the state to provide emergency
Fig. 1. Map of California showing locations of interest during the September 29, 2009 tsunami in California.
R.I. Wilson et al. / Earth-Science Reviews 107 (2011) 193–200
managers with advice and scientific expertise before, during, and after a tsunami. CalEMA is the contact point and liaison between the local jurisdictions and the National Oceanic and Atmospheric Administration (NOAA), West Coast and Alaska Tsunami Warning Center (WC/ATWC), and has the role of coordinating local, state and federal resources during an emergency. The Earthquake and Tsunami Program within CalEMA is responsible for relaying pertinent incident-specific technical information to emergency managers in a practical way. During a “tsunami alert,” the WC/ATWC coordinates with CalEMA and regional emergency managers providing forecast information for a handful of locations in the state. However, this coverage is not sufficient to assist all coastal operational areas within the state (20 coastal counties, 100 cities, and over 120 state and county beaches) that are vulnerable to tsunami hazards. Because the WC/ATWC cannot provide information to each of these local jurisdictions, CalEMA consults with state tsunami experts familiar with tsunami hazards, state-wide, to provide additional technical assistance and information. These experts include geoscientists from CGS, Humboldt State University, the Tsunami Research Center at the University of Southern California, and the private sector. During the September 29, 2009 tsunami alert response, state geoscientists provided support to emergency response decision makers at the federal, state, and local agency levels. In addition, other geoscientists within the state collected tsunami information in the field during and after the tsunami. This paper summarizes the overall activities of geoscientists during this event, and discusses how these scientists can improve and better coordinate their activities during future NOAA tsunami alerts. 2. Event response process 2.1. WC/ATWC tsunami forecast activities In a typical scenario, the WC/ATWC analyzes background information about a potential tsunamigenic event and determines the appropriate tsunami alert level for sections of coastline within its “areas of responsibility,” shown in Fig. 2. In addition to the State of California, the WC/ATWC's areas of responsibility include the Canadian coastal regions, Puerto Rico and the Virgin Islands, and the other ocean coasts of all U.S. States except Hawaii; Hawaii and other coastal areas within the Pacific Ocean are under the responsibility of the Pacific
195
Tsunami Warning Center (PTWC). Earthquakes that activate the WC/ ATWC alarm system initiate an investigation which includes: 1) automatic locating and sizing of the earthquake; 2) earthquake analysis and review; 3) sea-level data analysis to verify the existence of a tsunami and to calibrate models; and, 4) disseminating information to the appropriate emergency management officials (NOAA/NWS/ WCATWC, 2010). Similar protocols and activities are initiated by the PTWC and other tsunami warning centers, worldwide, for their respective areas of responsibility. There are two primary tsunami forecast tools available for the WC/ ATWC to use to estimate tsunami amplitudes and determine the appropriate level of alert along the coast. The Alaska Tsunami Forecast Model (ATFM) uses pre-computed numerical tsunami models for large subduction zone earthquakes around the Pacific Rim (Kowalik and Whitmore, 1991). The pre-computed model amplitudes are obtained using several different earthquake magnitudes for each potential tsunami source region. Based on the location and magnitude of the earthquake, the most appropriate pre-computed model is selected during the event. Adjustments are made to the ATFM forecasts as observed tsunami information is collected in real-time at tide gauge and Deep-ocean Assessment and Reporting of Tsunamis (DART) stations. The resulting amplitude forecasts represent coastal estimates for over a hundred sites in North America and Hawaii. More recently, the NOAA/Pacific Marine Environmental Laboratory (PMEL) has developed a tsunami forecast tool called the Short-term Inundation Forecasting for Tsunamis (SIFT) method (Gica et al., 2008). The SIFT system uses both pre-computed deep-ocean model results and real-time inundation model computations near-shore to forecast both coastal amplitude over time and flooding for approximately 30 communities in the Pacific, including five in California. SIFT uses observed tsunami data from the DART network to constrain the tsunami source. Pre-computed deep-ocean models are scaled by the DART-constrained source and then provide the boundary conditions for coastal inundation computations.
2.2. WC/ATWC tsunami alert criteria Through a combined use of forecast tools and historical information, the WC/ATWC makes a determination regarding what tsunami alert messaging it issues in each event. In an attempt to help clarify the tsunami alert messaging, new criteria were recently developed for
Fig. 2. Map showing West Coast/Alaska Tsunami Warning Center (WC/ATWC) and Pacific Tsunami Warning Center (PTWC) areas-of-responsibility (Whitmore et al., 2008).
196
R.I. Wilson et al. / Earth-Science Reviews 107 (2011) 193–200
notifying the WC/ATWC region about a tsunami or tsunami hazards (summarized from Whitmore et al., 2008): • Tsunami Information Statement - Issued to inform and update emergency managers and the public that an earthquake has occurred, or that a Tsunami Watch, Advisory or Warning has been issued elsewhere in the ocean. • Tsunami Watch - Issued to alert emergency managers and the public of an event which may later impact the Watch area. May be upgraded to an Advisory or Warning - or canceled - based on updated information and analysis. • Tsunami Advisory - Issued due to the threat of a tsunami which may produce strong currents or waves dangerous to those in or near the water; typically called when forecasted tsunami amplitudes of 0.3 to 1 m are expected. Coastal communities are advised that beach and harbor areas could expect rapid, moderate tidal changes and strong currents. • Tsunami Warning - Issued when a tsunami with significant widespread inundation is imminent or expected; typically called when forecasted tsunami amplitudes are greater than 1 m. Coastal communities are advised to evacuate people from low-lying areas identified as vulnerable to tsunamis. 2.3. Communicating tsunami alert information to local jurisdictions Once a Tsunami Advisory or Warning is issued by the WC/ATWC, detailed information about the tsunami and the alert status is broadcast to local jurisdictions via federal and state emergency managers. In addition to information contained in the standard messages, the WC/ATWC holds hourly conference calls with federal and state entities in California, Oregon, Washington, Alaska, and other territories affected by the tsunami alert. The CalEMA event “Duty Officer,” the CGS science advisor, and representatives from the four regional NOAA-National Weather Service (NWS) forecast offices take part in these calls, gathering the most up-to-date tsunami forecast information including expected arrival times and surge amplitudes. This information and additional guidance is then provided to the coastal county operational areas. The NOAA-NWS forecast offices also send out bulletins about the event to coastal jurisdictions within their areas of responsibility. Emergency managers at the county level then communicate this information to the responsible entities along the coast, including cities, port authorities, and harbor masters. Coastal jurisdictions have established emergency response plans that are implemented depending on the severity of the hazard. If the hazard is significant, counties may open their emergency operations centers to better coordinate response activities within their jurisdiction. 3. The September 29th, 2009 tsunami advisory for California At 1048 (PDT) on September 29, 2009, a major earthquake occurred in the Samoa Islands region of the South Pacific (U.S. Geological Survey, 2009). The magnitude (Mw) 8.1 earthquake (upgraded from Mw 8.0) caused strong shaking in the island countries of Samoa, American Samoa, and Tonga, and generated a large local tsunami that caused significant damage and casualties (see other papers in this volume about the local impact from the tsunami). A teletsunami from this event was also generated with moderate wave energies projected towards the California coastline (Fig. 3). With approximately 10 h before the first wave arrival, there was adequate time for the emergency managers in California to implement their tsunami response plans. 3.1. Emergency response activities Fig. 4 shows a timeline of tsunami alert statements issued by the WC/ATWC, and a summary of the actions that federal and state agencies
Fig. 3. Forecasted, Pacific-wide wave energy map produced and provided by the WC/ ATWC for the September 29, 2009 tsunami from the Samoa area. Map shows that initial projections of tsunami amplitudes were relatively high for California.
took during the hours following the earthquake. A total of nine tsunami alert bulletins were released by the WC/ATWC starting at 1102 (PDT), 14 min after the earthquake. In the first several hours following the earthquake, the WC/ATWC reported a number of issues that complicated forecasting the tsunami amplitudes for California: • Although the epicenter was in the region of the subduction zone in the Tonga Trench, the focal mechanism for earthquake indicated that it had occurred on a normal fault on the outer rise of the subduction zone (U.S. Geological Survey, 2009). Because the precomputed forecast models are calculated for subduction zone events and not for earthquake scenarios on normal faults, the accuracy of the initial modeling completed for the ATFM and SIFT systems were called into question. • A critical ATFM output file used by WC/ATWC was corrupted so it could not be used for anything other than generating the initial wave energy forecast map (Fig. 3). This problem made the SIFT forecast data the only one available to the WC/ATWC for making their tsunami alert designation. • Due to the difficulty of matching the observed DART data to forecasted DART results in SIFT (as the Samoa earthquake source was not part of the pre-computed series of sources for SIFT), forecast uncertainty was high. With the information available, three of the five SIFT forecast locations (Los Angeles, Port San Luis, and Crescent City) were projected to have maximum tsunami amplitudes between 0.3 m and 1.0 m. Although there were concerns about the accuracy of the initial forecast models, the WC/ATWC determined there was sufficient information to put all of California in an “Advisory” level of alert at 1502 h (PDT).
R.I. Wilson et al. / Earth-Science Reviews 107 (2011) 193–200
EVENT RESPONSE TIMELINE
September 29, 2009 - Tsunami Alert Messages for California from the NOAA WC/ATWC 10:48 PDT – Earthquake occurs in Samoa Islands region 10:57 PDT – WC/ATWC issues “observatory message” to federal and state tsunami observatories #1 - 11:02 PDT - M 7.9 Information #2 - 11:59 PDT - M 8.3 Information #3 - 15:02 PDT - M 8.0 Advisory • Provided tsunami arrival times to the minute for many locations along California coast (Figure 1). • Shared initial forecast model flow-depth results for 5 locations in California (Figure 3): • Crescent City = 65 cm (“Advisory” conditions) • San Francisco = 20 cm • Port San Luis = 60 cm (“Advisory” conditions) • Los Angeles = 30 cm (“Advisory” conditions) • San Diego = 15 cm #4 - 16:14 PDT - M 8.0 Advisory #5 - 17:21 PDT - M 8.0 Advisory • The peak tsunami amplitudes have passed in Hawaii. Observations in Hawaii support forecasts of 10 cm to 60 cm along the California Coast. #6 - 19:24 PDT - M 8.0 Advisory #7 - 21:28 PDT - M 8.0 Advisory #8 - 23:30 PDT - M 8.0 Advisory • The tsunami is reaching its maximum amplitude at many locations along the west coast. Expectations are for the amplitude to decrease over the next few hours. #9 - 9/30 1:28 PDT - Cancellation
197
September 29, 2009 – Supplemental information provided by CGS (Rick Wilson) and Humboldt State University (HSU; Lori Dengler) to State Emergency Managers (EMs), NOAA WC/ATWC and regional Warning Coordination Meteorologists (WCMs), and Local EMs. 14:39 PDT - CGS to State EMs: Searched NOAA/National Geophysical Data Center Tsunami Event Database for similar past tsunami events from source region and discovered M 8.3 event on June 26, 1917. In California, tsunami amplitudes recorded on tide gauges from San Francisco and San Diego were 10 cm and 4 cm, respectively. Lander and others (1993) reported no damage in California from this event. 15:50 PDT - HSU to NOAA WCMs, State, and Local EMs: During statewide teleconference for emergency managers, provided explanation of NOAA Tsunami Advisory and answered questions about tsunami hazards local communities might expect from potential strong tsunami currents and prolonged tsunami activity that might extend for a period of time. 17:17 PDT - CGS to State EMs and NOAA WC/ATWC and WCMs: Provided information about potential additional communities at risk based on existing numerical models and work related to the statewide mapping project (Figures 1 and 5; Wilson et al., 2008; Barberopoulou et al., 2009). Based on modeled data from the state inundation mapping project (Figure 5), four locations that have conditions comparable to Crescent City and Port San Luis (south facing, crescent-shaped coves) were noted: Bodega Bay, Half Moon Bay, Santa Cruz, and Cayucos. Other information was passed to the EMs by both CGS and the WC/ATWC: • Tidal projections showed area north of San Francisco Bay would be at high tide at the time of first wave arrival. • Local communities were made aware that actual arrival times in California could vary by plus or minus 15 minutes. • NOAA and CGS discussed that potential errors in the forecast modeling could arise from uncertainties/differences in the earthquake source dynamics and location. Crustal complexities exhibited by the tight bend in the source region subduction zone complicate directionality projections. Because the source was a normal fault instead of a thrust fault, reported forecast model results developed for thrust faults might be in error.
Fig. 4. Event response timeline for the September 29, 2009 Tsunami Advisory in California (14:39 PDT - source Lander et al., 1993).
During the time leading up to the first arrival of the tsunami in California, state geoscientists provided technical support to emergency– response decision makers at the federal, state, and local levels during the event: • Supplemental Hazard Information – Two locations, Crescent City and Port San Luis, were projected to receive maximum surge amplitudes of 0.65 and 0.6 m, respectively, values considered capable of causing strong currents and potential damage to harbor facilities such as docks, piers and vessels. Based on tsunami amplitude data from numerical modeling performed for the statewide inundation mapping project (shown in Fig. 5; Wilson et al., 2008; Barberopoulou et al., 2009), CGS recommended that four other south-facing, crescent-shaped coastal areas…Bodega Bay, Half Moon Bay, Santa Cruz, and Cayucos…be included with Crescent City and Port San Luis as areas of potentially higher hazard for the state. In addition, CGS indicated that high-tide conditions would be active north of San Francisco Bay at the time of first tsunami arrival, potentially increasing flow stresses on some port facilities. This information was included in state and federal messaging to local emergency managers. • Technical Support During Conference Calls – CGS and Lori Dengler (Humboldt State University) took part in the state conference calls with the WC/ATWC and the coastal jurisdictions. During the conference calls, geoscientists fielded questions from the coastal jurisdictions. Because of the recent change to the WC/ATWC tsunami alert-level criteria, there were many questions about the meaning of a Tsunami Advisory level of alert. The state emergency managers and geoscientists reinforced the WC/ATWC messaging
that an Advisory-level event indicates potential strong currents can occur in harbors and on beaches. In addition, geoscientists warned that: 1) forecasted arrival times should include a margin of error of up to plus or minus 15 min; 2) projected tsunami amplitudes could also be in error because of forecast model uncertainties; and, 3) the first wave is usually not the largest and that tsunami activity could last for a number of hours. In the days after the tsunami Advisory ended, the state tsunami program sent out questionnaires to the coastal counties to ask what actions they took and observations they made with regard to the tsunami. As a result of the Tsunami Advisory and the information provided by the various experts, most coastal jurisdictions indicated that they closed beaches and limited access within harbors and waterfront areas. The tsunami's first arrival and most of its activity was at night time making it difficult for people to make observations. However, people on beaches reported no unusual wave activity. Strong currents were observed in some harbors and at the mouths of rivers. Other than displacement of navigational buoys in some harbors, there were no reports of significant damage. 3.2. Tsunami field response activities After a significant tsunami hits the coast anywhere in the world, organized field teams of geoscientists typically collect perishable tsunami data. Although there were no organized efforts in California to deploy field response teams prior to the tsunamis arrival, observations were made at two locations by co-authors of this paper, Andy Ritchie in Santa Cruz and Mark Legg in Huntington Beach. These
198
R.I. Wilson et al. / Earth-Science Reviews 107 (2011) 193–200
Maximum Modeled Flow Depth (meters) 0.0
2.0
4.0
6.0
8.0
10.0 12.0
Crescent City Humboldt -Inside Bay Arena Cove Bodega Bay San Francisco Saucilito Richmond Redwood City Half Moon Bay Santa Cruz Monterey Cayucos Port San Luis Pt Arguello Santa Barbara Ventura Oxnard Malibu Santa Monica San Pedro/LA Harbor Huntington Beach Newport Beach Dana Point San Clemente Oceanside Del Mar La Jolla San Diego Bay Coronado Imperial Beach Fig. 5. Offshore maximum flow-depth values, in meters above sea level, from numerical modeling for the worst-case, distant-source tsunami scenarios as part of the state inundation mapping project (Wilson et al., 2008). This information was used to designate Bodega Bay, Half Moon Bay, Santa Cruz, and Cayucos as potential higherhazard areas comparable to areas considered higher hazard by the WC/ATWC (Crescent City and Port San Luis). All cities are listed from north to south along the coastline.
observations were valuable as there were no tide gauge instruments near these locations to record the tsunami. During the first couple hours of the tsunami, Ritchie made observations from the Murray Street bridge which crosses Santa Cruz Harbor about 0.8 km from the mouth of the harbor to the south. Ritchie observed strong tidal currents up to 8 knots and maximum tsunami amplitudes of 0.7 m above normal conditions. This observation was the largest amplitude reported in California from the September 29th tsunami and helps validate CGS's recommendation that adding Santa Cruz as a location that could expect higher amplitude surges during this Advisory. Legg observed the tsunami near the mouth of the Santa Ana River in southern California and noted several dozen small tsunami surges up to 0.2 m in height, riding up and over the outward flow of the river. Legg further noted that the drawdown was so rapid that it caused the outgoing tidal flow to accelerate and sand along the river bed to rapidly erode. In addition to Legg making scientific observations, he arranged to meet a local reporter and was interviewed on video discussing what he observed at the time of the tsunami's arrival. The video was placed on the internet and has since been copied to many other websites and discussion boards. Legg's analysis helped to demonstrate to the public that not all tsunamis are large events, and that even studying small tsunamis can provide useful information. Lack of verified field intelligence is a known gap in the midst of any type of ensuing natural disaster which only serves to aid effective emergency response. The authors believe that field observations like
these can help validate the accuracy of forecast and other numerical modeling done in the state, and are of value to emergency managers in the midst of an event. The field data could potentially be of even greater value if organized, interpreted, and disseminated through a state-level information clearinghouse. Field geoscientists could relay information to the local emergency manager, and to a regional field manager to be shared with the clearinghouse leader. Similar to the tsunami observer programs in places like Hawaii, real-time scientific information could help determine where state response resources could be best directed. 3.3. Evaluation of forecasted information Table 1 shows a comparison between the projected tsunami amplitudes and arrival times to the recorded values in California from the Samoa tsunami. This information was gathered from tide gauge marigrams measuring water levels at one-minute increments. Recorded tsunami amplitudes were approximately one half (48%) of the amplitude initially forecasted at the four sites where comparisons could be made; this error margin is near the outer limit of expected forecast accuracy (according to Whitmore, 2003). Also, the first arrival times were 17 to 48 min later at points recorded along the coast. A post-event evaluation by WC/ATWC determined this difference in wave amplitude and travel time is likely due to several factors: 1) differences in the forecast model fault motion (thrust) and the actual fault motion (normal); 2) complexities in the source region producing uncertainty in the directionality of the tsunami; 3) inaccuracies in matching the true source with the modeled source; and, 4) observed first arrivals being masked by tide gauge noise at some locations in California. A better understanding of the uncertainties in the forecasted amplitudes and arrival times is needed so that it can be explained to non-scientists in future events. Attempts were made to gather information about tsunami amplitudes at the four locations (Bodega Bay, Half Moon Bay, Santa Cruz, and Cayucos) where the tsunami vulnerability was thought comparable to Crescent City and Port San Luis. None of these locations have permanent tide gauges but the field observations in Santa Cruz by Ritchie indicating that maximum tsunami amplitudes reached 0.7 m helped validate adding Santa Cruz as a location to attribute an Advisory-like status for this event. Based on the responses of state and local emergency managers in the post-event surveys, information from state geoscientists helped emergency managers decide what actions to take during the event. Information provided by CGS … areas of potential increased tsunami amplitudes and vulnerability, high-tide conditions north of San Francisco Bay, and uncertainties in forecasted tsunami arrival times and maximum amplitudes…was received by coastal emergency managers through tsunami alert messaging and teleconferences, and incorporated into their decision making process. The participation of state geoscientists during the initial statewide teleconference helped emergency managers better understand the tsunami alert messaging, forecast information, and the expected tsunami hazard. 3.4. Conclusions about the Samoa event and future work As illustrated by California's experience during the Samoa teletsunami event, state geoscientists can provide emergency managers with useful information and assistance during a tsunami alert. They can explain the uncertainties in the NOAA tsunami amplitude forecast and travel time data so that state and local emergency managers can make more informed decisions about actions to take in their jurisdictions. In addition, because forecast modeling for inundation is done for only five of the more than 250 incorporated and unincorporated communities along the California coast, state geoscientists can use existing tsunami modeling data to determine the potential hazard level for comparable communities. As demonstrated
R.I. Wilson et al. / Earth-Science Reviews 107 (2011) 193–200
199
Table 1 Shows comparison between forecasted and recorded tsunami amplitudes and arrival times for the September 29, 2009 tsunami in California. Locations
Crescent City Arena Cove Point Reyes San Francisco Port San Luis Santa Monica Los Angeles Harbor San Diego
Tsunami Amplitudes (zero to peak in cm)
First Arrival Times
Projected
Recorded
% Difference (less)
Projected Time
Recorded Time
Difference (min. later)
65 44 39 20 60 15 30 15
33
51%
10 28
50% 47%
21:20 20:58 21:02 21:31 21:07 21:20 21:15 *
21:44 21:15 21:50 21:48 21:30 21:39 *
24 17 48 17 23 19
13 43% * * Average Difference = 48%
Average Difference = 25 min
*Not measurable.
by Legg, field geoscientists can also provide real-time feedback to state and local emergency managers and the public in general. Although state geoscientists are integrated into the statewide tsunami emergency response network, their role should be expanded during tsunami response activities. CGS and other geoscientists are working more closely with NOAA and state and local emergency managers to formalize a coordinated tsunami response plan to help provide the most valuable and timely information to those in decision making roles for their communities. This response plan will include: 1) participation by primary and backup geoscience contacts in all event-response meetings and teleconferences with NOAA and state and local emergency managers; 2) provision of information about historical tsunamis and expected tidal conditions as they relate to future forecasted events; and, 3) provision of supplemental scenario information of the expected tsunami hazard (flow depths, strong currents, etc.) for areas not covered by NOAA's forecast models. State tsunami geoscientists are continuing to provide additional support in a number of ways. The California tsunami program will continue to work closely with NOAA through the NTHMP to simplify their tsunami alert messaging and expand their forecast modeling coverage. California geoscientists will provide feedback on NOAA's messages that should help clarify the uncertainties in the tsunami forecast data. In addition, California is supporting the expansion of NOAA's development of digital elevation models used for production of SIFT tsunami forecast models within the state. The state tsunami program also plans to create tsunami response “playbooks” containing information from the recent statewide inundation mapping project, for local emergency managers to reference during an event. Information in these playbooks will help identify the relative hazard level for their coastal communities and include estimates on tsunami flow depths, current speeds, and travel times. Consideration is also being given to providing an easily accessed computer interface to provide this pre-run library of tsunami scenarios applicable to coastal California that would assist in interpreting existing data for emergency managers. CGS has proposed developing a pre- and post-tsunami field response team and state-level information “clearinghouse” made up of geoscientists to assist state and local officials. In addition to noting the effects of the tsunami, members of these field response teams will meet regularly and be available before, during, and after a tsunami to provide assistance to local emergency managers and the public. 4. Addendum: the February 27th, 2010 tsunami advisory in California Nearly five months after the Samoa tsunami, another Tsunami Advisory was issued for California when a M8.8 earthquake occurred off the coast of Chile. The earthquake was generated along the plate boundary where the Nazca Plate subducts under the South American Plate, approximately 300 km north of the magnitude (Mw) 9.5 1960
earthquake. A large tsunami was generated locally along the Chilean coast, causing severe damage to coastal communities and port facilities in Chile. Building on the experience during the event from Samoa and other tsunamis, geoscientists at the federal, state, and local level implemented several new response strategies during the event from Chile (Wilson et al., 2010): • The WC/ATWC using ATFM results increased the number of forecasted tsunami arrival time and maximum amplitude values at California sites from five to 30 (and issued forecasts for a total of 77 sites along the U.S. west coast, British Columbia, and Alaska coasts). With more information to use, local jurisdictions were better able to determine the appropriate response activities. Accuracy improved according to a NOAA Tsunami Forecast Assessment (Bernard, et al, in press) which showed the forecast values had an average error of 33% (20 cm rootmean-square error) from those observed at the 40 locations where the values could be compared. Though there was some confusion by local jurisdictions about what to expect during this Advisory, the additional information provided by the WC/ATWC assisted state emergency managers and scientist in their messaging to local authorities. • Because of the 13–1/2 hour delay between the tsunami's generation and its daytime arrival in California, more scientific observers were able to deploy to the coast prior to the tsunami's arrival. Rick Wilson, a co-author on this paper, was deployed to the Port San Luis area, the location of the highest forecasted tsunami amplitude values for this event. Wilson provided real-time information and recommendations to state emergency managers during the event. Some field observations helped confirm what the WC/ATWC noted on tidegauge marigrams during the event: Advisory-level tsunami amplitudes were active for up to 8 h after the first wave arrival. Validated by the field experiences of geoscientists during the two recent Advisories in California from the Samoa and Chile tsunamis, the NTHMP is funding the state's development of pre- and post-tsunami field teams and a centralized information clearinghouse for future Tsunami Advisories and Warnings. The plan includes enlisting the help of 30 to 40 field geoscientists that can be deployed to collect information about the tsunami similar to Legg, Ritchie, and Wilson. The field teams will be developed during 2010–11 and the plan will be ready for operation if needed in 2011. To help coordinate its development, the tsunami clearinghouse may be integrated with the long-established California Natural Hazards Clearinghouse. If successful, the NTHMP could recommend that other states use the California project as a template for development of pre-tsunami field teams and expanding integration of geoscientists into emergency response planning. Acknowledgements The authors would like to thank NOAA/NTHMP for supporting tsunami hazard mitigation and response activities in California. Thank
200
R.I. Wilson et al. / Earth-Science Reviews 107 (2011) 193–200
you also to the NOAA-NWS Warning Coordination Meteorologists and the state and county emergency managers who are part of the California Tsunami Program Steering Committee for their participation in the post-event survey. References Barberopoulou, A., Borrero, J.C., Uslu, B., Kalligeris, N., Goltz, J.D., Wilson, R.I., Synolakis, C.E., 2009. New maps of California to improve tsunami preparedness. Eos, Transactions, American Geophysical Union 90 (16), 137–138. Bernard, E., Whitmore, P., Rhoades, J., McCreery, C., Fox, C., in press. Chile Tsunami Forecast Assessment: February 27-28, 2010. National Oceanic and Atmospheric Administration (internal report). 77 pp. California Seismic Safety Commission, 2006. The tsunami threat to California. Ad Hoc Committee on Tsunami Safety, CSSC 05–03. 24 pp. Gica, E., Spillane, M.C., Titov, V.V., Chamberlin, C.D., Newman, J.C., 2008. Development of the forecast propagation database for NOAA's Short-term Inundation Forecasting for Tsunamis (SIFT). NOAA Technical Memorandum OAR PMEL-139. 89 pp. Kowalik, Z., Whitmore, P.M., 1991. An investigation of two tsunamis recorded at Adak, Alaska. Science of Tsunami Hazards 9, 67–83. Lander, J., Lockridge, P.A., Kozuch, J., 1993. Tsunamis affecting the west coast of the United States 1806–1992. NGDC Key to Geophysical Research Documentation No. 29. USDOC/NOAA/NESDIS/NGDC, Boulder, CO, USA. 242 pp.
NOAA/NWS/WCATWC, 2010. User's guide for the tsunami warning system in the West Coast/Alaska Tsunami Warning Center Area-of-Responsibility. NOAA National Weather Service, West Coast and Alaska Tsunami Warning Center Operations Manual 6.1.4, dated May 20, 2010. 43 pp. Patton, J.R., Dengler, L.A., 2004. GIS-based relative tsunami hazard maps for northern California, Humboldt and Del Norte Counties: (abs.). Eos, Transactions, American Geophysical Union 85 (47) Fall Meeting Supplement, abstract# OS24D-1353. Patton, J.R., Dengler, L.A., 2006. Relative tsunami hazard mapping for Humboldt and el Norte Counties, California. Proceedings of the 8NCEE/EERI Eighth Earthquake Engineering Conference. U.S. Geological Survey, 2009. M8.0 Samoa Islands Region Earthquake of 29 September 2009. U.S. Geological Survey Earthquake Summary Map, published 2 October 2009. Whitmore, P.M., 2003. Tsunami amplitude prediction during events: a test based on previous tsunamis. Science of Tsunami Hazards 21, 135–143. Whitmore, P.M., Benz, H., Bolton, M., Crawford, G., Dengler, L., Fryer, G., Goltz, J., Hanson, R., Kryzanowski, K., Malone, S., Oppenheimer, D., Petty, E., Rogers, G., Wilson, J., 2008. NOAA/West Coast and Alaska Tsunami Warning Center Pacific Ocean Response Criteria. Science of Tsunami Hazards 27, 1–21. Wilson, R.I., Barberopoulou, A., Miller, K.M., Goltz, J.D., Synolakis, C.E., 2008. New maximum tsunami inundation maps for use by local emergency planners in the State of California, USA. Eos, Transactions, American Geophysical Union 89 (53) Fall Meeting Supplement, Abstract OS43D-1343. Wilson, R.I., Dengler, L.A., Legg, M.R., Long, K., Miller, K.M., 2010. The 2010 Chilean Tsunami on the California Coastline. Seismological Research Letters 81 (3), 545–546.
Contents lists available at ScienceDirect
Earth-Science Reviews j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e a r s c i r ev
Emergency response and field observation activities of geoscientists in California (USA) during the September 29, 2009, Samoa Tsunami Rick I. Wilson a,⁎,1, Lori A. Dengler b, James D. Goltz c, Mark R. Legg d, Kevin M. Miller e, Andy Ritchie f, Paul M. Whitmore g a
California Geological Survey, 801 K Street MS 12–31, Sacramento, California, USA 95814 Humboldt State University, Geology Department, #1 Harpst Street, Arcata, California, USA 95521 California Emergency Management Agency, 1200 East California Blvd., Mail Code 104–44, Pasadena, California, USA 91125 d Legg Geophysical, 16541 Gothard St # 107, Huntington Beach, California, USA 92647 e California Emergency Management Agency, 1300 Clay Street, Suite 400, Oakland, California, USA 94612 f U.S. Geological Survey, 400 Natural Bridges Drive, Santa Cruz, California, USA 95060 g West Coast/Alaska Tsunami Warning Center, NOAA-NWS, 910 South Felton Street, Palmer, Alaska, USA 99645 b c
a r t i c l e
i n f o
Article history: Received 29 June 2010 Accepted 25 January 2011 Available online 1 February 2011 Keywords: Tsunami emergency management geoscientist field observations warning center local operational areas
a b s t r a c t State geoscientists (geologists, geophysicists, seismologists, and engineers) in California work closely with federal, state and local government emergency managers to help prepare coastal communities for potential impacts from a tsunami before, during, and after an event. For teletsunamis, as scientific information (forecast model wave heights, first-wave arrival times, etc.) from NOAA's West Coast and Alaska Tsunami Warning Center is made available, federal- and state-level emergency managers must help convey this information in a concise, comprehensible and timely manner to local officials who ultimately determine the appropriate response activities for their jurisdictions. During the September 29, 2009 Tsunami Advisory for California, government geoscientists assisted the California Emergency Management Agency by providing technical assistance during teleconference meetings with NOAA and other state and local emergency managers prior to the arrival of the tsunami. This technical assistance included background information on anticipated tidal conditions when the tsunami was set to arrive, wave height estimates from state-modeled scenarios for areas not covered by NOAA's forecast models, and clarifying which regions of the state were at greatest risk. Over the last year, state geoscientists have started to provide additional assistance: 1) working closely with NOAA to simplify their tsunami alert messaging and expand their forecast modeling coverage; 2) creating “playbooks” containing information from existing tsunami scenarios for local emergency managers to reference during an event; and, 3) developing a state-level information “clearinghouse” and pre-tsunami field response team to assist local officials as well as observe and report tsunami effects. Activities of geoscientists were expanded during the more recent Tsunami Advisory on February 27, 2010, including deploying a geologist from the California Geological Survey as a field observer who provided information back to emergency managers. © 2011 Elsevier B.V. All rights reserved.
Contents 1. 2.
3.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . Event response process . . . . . . . . . . . . . . . . . . . . . . 2.1. WC/ATWC tsunami forecast activities . . . . . . . . . . . . 2.2. WC/ATWC tsunami alert criteria . . . . . . . . . . . . . . . 2.3. Communicating tsunami alert information to local jurisdictions The September 29th, 2009 tsunami advisory for California . . . . . . 3.1. Emergency response activities . . . . . . . . . . . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
⁎ Corresponding author. Tel.: +916 327 0981; fax: +916 445 3334. E-mail addresses: [email protected] (R.I. Wilson), [email protected] (L.A. Dengler), [email protected] (J.D. Goltz), [email protected] (M.R. Legg), [email protected] (K.M. Miller), [email protected] (A. Ritchie), [email protected] (P.M. Whitmore). 1 PG 5878, CEG 1881. 0012-8252/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.earscirev.2011.01.010
194 195 195 195 196 196 196
194
R.I. Wilson et al. / Earth-Science Reviews 107 (2011) 193–200
3.2. Tsunami field response activities . . . . . . . . . . . . . 3.3. Evaluation of forecasted information . . . . . . . . . . . 3.4. Conclusions about the Samoa event and future work . . . 4. Addendum: the February 27th, 2010 tsunami advisory in California Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction The State of California's 1700-kilometer-long coastline has over 1 million residents and tens of millions of visitors each year that could be at risk to tsunami hazards (Fig. 1; California Seismic Safety Commission, 2006). State geoscientists (geologists, geophysicists, seismologists, and engineers) work closely with federal, state and local emergency managers to help identify and mitigate tsunami hazards in California. The primary focus of this work in California includes: 1) creating tsunami inundation maps for emergency response planning; 2) providing technical information about tsunami hazards to coastal emergency managers through presentations, workshops, and tsunami response exercises; and, 3) collecting and analyzing tsunami hazard information
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
197 198 198 199 199 200
to improve the overall knowledge base of the scientific community (Patton and Dengler, 2004, 2006; Wilson et al., 2008; Barberopoulou et al., 2009). As one of the original member states in the U.S. National Tsunami Hazard Mitigation Program (NTHMP), California's tsunami program has been very active in helping to establish policy and guidance with regard to tsunami hazard mitigation at a national level. The state's membership on the NTHMP Coordinating Committee includes a geologist from the California Geological Survey (CGS) and an emergency manager from the California Emergency Management Agency (CalEMA), sharing responsibility for California to implement the state tsunami hazard mitigation and preparedness plan. Part of this plan is for geoscientists from CGS and other organizations in the state to provide emergency
Fig. 1. Map of California showing locations of interest during the September 29, 2009 tsunami in California.
R.I. Wilson et al. / Earth-Science Reviews 107 (2011) 193–200
managers with advice and scientific expertise before, during, and after a tsunami. CalEMA is the contact point and liaison between the local jurisdictions and the National Oceanic and Atmospheric Administration (NOAA), West Coast and Alaska Tsunami Warning Center (WC/ATWC), and has the role of coordinating local, state and federal resources during an emergency. The Earthquake and Tsunami Program within CalEMA is responsible for relaying pertinent incident-specific technical information to emergency managers in a practical way. During a “tsunami alert,” the WC/ATWC coordinates with CalEMA and regional emergency managers providing forecast information for a handful of locations in the state. However, this coverage is not sufficient to assist all coastal operational areas within the state (20 coastal counties, 100 cities, and over 120 state and county beaches) that are vulnerable to tsunami hazards. Because the WC/ATWC cannot provide information to each of these local jurisdictions, CalEMA consults with state tsunami experts familiar with tsunami hazards, state-wide, to provide additional technical assistance and information. These experts include geoscientists from CGS, Humboldt State University, the Tsunami Research Center at the University of Southern California, and the private sector. During the September 29, 2009 tsunami alert response, state geoscientists provided support to emergency response decision makers at the federal, state, and local agency levels. In addition, other geoscientists within the state collected tsunami information in the field during and after the tsunami. This paper summarizes the overall activities of geoscientists during this event, and discusses how these scientists can improve and better coordinate their activities during future NOAA tsunami alerts. 2. Event response process 2.1. WC/ATWC tsunami forecast activities In a typical scenario, the WC/ATWC analyzes background information about a potential tsunamigenic event and determines the appropriate tsunami alert level for sections of coastline within its “areas of responsibility,” shown in Fig. 2. In addition to the State of California, the WC/ATWC's areas of responsibility include the Canadian coastal regions, Puerto Rico and the Virgin Islands, and the other ocean coasts of all U.S. States except Hawaii; Hawaii and other coastal areas within the Pacific Ocean are under the responsibility of the Pacific
195
Tsunami Warning Center (PTWC). Earthquakes that activate the WC/ ATWC alarm system initiate an investigation which includes: 1) automatic locating and sizing of the earthquake; 2) earthquake analysis and review; 3) sea-level data analysis to verify the existence of a tsunami and to calibrate models; and, 4) disseminating information to the appropriate emergency management officials (NOAA/NWS/ WCATWC, 2010). Similar protocols and activities are initiated by the PTWC and other tsunami warning centers, worldwide, for their respective areas of responsibility. There are two primary tsunami forecast tools available for the WC/ ATWC to use to estimate tsunami amplitudes and determine the appropriate level of alert along the coast. The Alaska Tsunami Forecast Model (ATFM) uses pre-computed numerical tsunami models for large subduction zone earthquakes around the Pacific Rim (Kowalik and Whitmore, 1991). The pre-computed model amplitudes are obtained using several different earthquake magnitudes for each potential tsunami source region. Based on the location and magnitude of the earthquake, the most appropriate pre-computed model is selected during the event. Adjustments are made to the ATFM forecasts as observed tsunami information is collected in real-time at tide gauge and Deep-ocean Assessment and Reporting of Tsunamis (DART) stations. The resulting amplitude forecasts represent coastal estimates for over a hundred sites in North America and Hawaii. More recently, the NOAA/Pacific Marine Environmental Laboratory (PMEL) has developed a tsunami forecast tool called the Short-term Inundation Forecasting for Tsunamis (SIFT) method (Gica et al., 2008). The SIFT system uses both pre-computed deep-ocean model results and real-time inundation model computations near-shore to forecast both coastal amplitude over time and flooding for approximately 30 communities in the Pacific, including five in California. SIFT uses observed tsunami data from the DART network to constrain the tsunami source. Pre-computed deep-ocean models are scaled by the DART-constrained source and then provide the boundary conditions for coastal inundation computations.
2.2. WC/ATWC tsunami alert criteria Through a combined use of forecast tools and historical information, the WC/ATWC makes a determination regarding what tsunami alert messaging it issues in each event. In an attempt to help clarify the tsunami alert messaging, new criteria were recently developed for
Fig. 2. Map showing West Coast/Alaska Tsunami Warning Center (WC/ATWC) and Pacific Tsunami Warning Center (PTWC) areas-of-responsibility (Whitmore et al., 2008).
196
R.I. Wilson et al. / Earth-Science Reviews 107 (2011) 193–200
notifying the WC/ATWC region about a tsunami or tsunami hazards (summarized from Whitmore et al., 2008): • Tsunami Information Statement - Issued to inform and update emergency managers and the public that an earthquake has occurred, or that a Tsunami Watch, Advisory or Warning has been issued elsewhere in the ocean. • Tsunami Watch - Issued to alert emergency managers and the public of an event which may later impact the Watch area. May be upgraded to an Advisory or Warning - or canceled - based on updated information and analysis. • Tsunami Advisory - Issued due to the threat of a tsunami which may produce strong currents or waves dangerous to those in or near the water; typically called when forecasted tsunami amplitudes of 0.3 to 1 m are expected. Coastal communities are advised that beach and harbor areas could expect rapid, moderate tidal changes and strong currents. • Tsunami Warning - Issued when a tsunami with significant widespread inundation is imminent or expected; typically called when forecasted tsunami amplitudes are greater than 1 m. Coastal communities are advised to evacuate people from low-lying areas identified as vulnerable to tsunamis. 2.3. Communicating tsunami alert information to local jurisdictions Once a Tsunami Advisory or Warning is issued by the WC/ATWC, detailed information about the tsunami and the alert status is broadcast to local jurisdictions via federal and state emergency managers. In addition to information contained in the standard messages, the WC/ATWC holds hourly conference calls with federal and state entities in California, Oregon, Washington, Alaska, and other territories affected by the tsunami alert. The CalEMA event “Duty Officer,” the CGS science advisor, and representatives from the four regional NOAA-National Weather Service (NWS) forecast offices take part in these calls, gathering the most up-to-date tsunami forecast information including expected arrival times and surge amplitudes. This information and additional guidance is then provided to the coastal county operational areas. The NOAA-NWS forecast offices also send out bulletins about the event to coastal jurisdictions within their areas of responsibility. Emergency managers at the county level then communicate this information to the responsible entities along the coast, including cities, port authorities, and harbor masters. Coastal jurisdictions have established emergency response plans that are implemented depending on the severity of the hazard. If the hazard is significant, counties may open their emergency operations centers to better coordinate response activities within their jurisdiction. 3. The September 29th, 2009 tsunami advisory for California At 1048 (PDT) on September 29, 2009, a major earthquake occurred in the Samoa Islands region of the South Pacific (U.S. Geological Survey, 2009). The magnitude (Mw) 8.1 earthquake (upgraded from Mw 8.0) caused strong shaking in the island countries of Samoa, American Samoa, and Tonga, and generated a large local tsunami that caused significant damage and casualties (see other papers in this volume about the local impact from the tsunami). A teletsunami from this event was also generated with moderate wave energies projected towards the California coastline (Fig. 3). With approximately 10 h before the first wave arrival, there was adequate time for the emergency managers in California to implement their tsunami response plans. 3.1. Emergency response activities Fig. 4 shows a timeline of tsunami alert statements issued by the WC/ATWC, and a summary of the actions that federal and state agencies
Fig. 3. Forecasted, Pacific-wide wave energy map produced and provided by the WC/ ATWC for the September 29, 2009 tsunami from the Samoa area. Map shows that initial projections of tsunami amplitudes were relatively high for California.
took during the hours following the earthquake. A total of nine tsunami alert bulletins were released by the WC/ATWC starting at 1102 (PDT), 14 min after the earthquake. In the first several hours following the earthquake, the WC/ATWC reported a number of issues that complicated forecasting the tsunami amplitudes for California: • Although the epicenter was in the region of the subduction zone in the Tonga Trench, the focal mechanism for earthquake indicated that it had occurred on a normal fault on the outer rise of the subduction zone (U.S. Geological Survey, 2009). Because the precomputed forecast models are calculated for subduction zone events and not for earthquake scenarios on normal faults, the accuracy of the initial modeling completed for the ATFM and SIFT systems were called into question. • A critical ATFM output file used by WC/ATWC was corrupted so it could not be used for anything other than generating the initial wave energy forecast map (Fig. 3). This problem made the SIFT forecast data the only one available to the WC/ATWC for making their tsunami alert designation. • Due to the difficulty of matching the observed DART data to forecasted DART results in SIFT (as the Samoa earthquake source was not part of the pre-computed series of sources for SIFT), forecast uncertainty was high. With the information available, three of the five SIFT forecast locations (Los Angeles, Port San Luis, and Crescent City) were projected to have maximum tsunami amplitudes between 0.3 m and 1.0 m. Although there were concerns about the accuracy of the initial forecast models, the WC/ATWC determined there was sufficient information to put all of California in an “Advisory” level of alert at 1502 h (PDT).
R.I. Wilson et al. / Earth-Science Reviews 107 (2011) 193–200
EVENT RESPONSE TIMELINE
September 29, 2009 - Tsunami Alert Messages for California from the NOAA WC/ATWC 10:48 PDT – Earthquake occurs in Samoa Islands region 10:57 PDT – WC/ATWC issues “observatory message” to federal and state tsunami observatories #1 - 11:02 PDT - M 7.9 Information #2 - 11:59 PDT - M 8.3 Information #3 - 15:02 PDT - M 8.0 Advisory • Provided tsunami arrival times to the minute for many locations along California coast (Figure 1). • Shared initial forecast model flow-depth results for 5 locations in California (Figure 3): • Crescent City = 65 cm (“Advisory” conditions) • San Francisco = 20 cm • Port San Luis = 60 cm (“Advisory” conditions) • Los Angeles = 30 cm (“Advisory” conditions) • San Diego = 15 cm #4 - 16:14 PDT - M 8.0 Advisory #5 - 17:21 PDT - M 8.0 Advisory • The peak tsunami amplitudes have passed in Hawaii. Observations in Hawaii support forecasts of 10 cm to 60 cm along the California Coast. #6 - 19:24 PDT - M 8.0 Advisory #7 - 21:28 PDT - M 8.0 Advisory #8 - 23:30 PDT - M 8.0 Advisory • The tsunami is reaching its maximum amplitude at many locations along the west coast. Expectations are for the amplitude to decrease over the next few hours. #9 - 9/30 1:28 PDT - Cancellation
197
September 29, 2009 – Supplemental information provided by CGS (Rick Wilson) and Humboldt State University (HSU; Lori Dengler) to State Emergency Managers (EMs), NOAA WC/ATWC and regional Warning Coordination Meteorologists (WCMs), and Local EMs. 14:39 PDT - CGS to State EMs: Searched NOAA/National Geophysical Data Center Tsunami Event Database for similar past tsunami events from source region and discovered M 8.3 event on June 26, 1917. In California, tsunami amplitudes recorded on tide gauges from San Francisco and San Diego were 10 cm and 4 cm, respectively. Lander and others (1993) reported no damage in California from this event. 15:50 PDT - HSU to NOAA WCMs, State, and Local EMs: During statewide teleconference for emergency managers, provided explanation of NOAA Tsunami Advisory and answered questions about tsunami hazards local communities might expect from potential strong tsunami currents and prolonged tsunami activity that might extend for a period of time. 17:17 PDT - CGS to State EMs and NOAA WC/ATWC and WCMs: Provided information about potential additional communities at risk based on existing numerical models and work related to the statewide mapping project (Figures 1 and 5; Wilson et al., 2008; Barberopoulou et al., 2009). Based on modeled data from the state inundation mapping project (Figure 5), four locations that have conditions comparable to Crescent City and Port San Luis (south facing, crescent-shaped coves) were noted: Bodega Bay, Half Moon Bay, Santa Cruz, and Cayucos. Other information was passed to the EMs by both CGS and the WC/ATWC: • Tidal projections showed area north of San Francisco Bay would be at high tide at the time of first wave arrival. • Local communities were made aware that actual arrival times in California could vary by plus or minus 15 minutes. • NOAA and CGS discussed that potential errors in the forecast modeling could arise from uncertainties/differences in the earthquake source dynamics and location. Crustal complexities exhibited by the tight bend in the source region subduction zone complicate directionality projections. Because the source was a normal fault instead of a thrust fault, reported forecast model results developed for thrust faults might be in error.
Fig. 4. Event response timeline for the September 29, 2009 Tsunami Advisory in California (14:39 PDT - source Lander et al., 1993).
During the time leading up to the first arrival of the tsunami in California, state geoscientists provided technical support to emergency– response decision makers at the federal, state, and local levels during the event: • Supplemental Hazard Information – Two locations, Crescent City and Port San Luis, were projected to receive maximum surge amplitudes of 0.65 and 0.6 m, respectively, values considered capable of causing strong currents and potential damage to harbor facilities such as docks, piers and vessels. Based on tsunami amplitude data from numerical modeling performed for the statewide inundation mapping project (shown in Fig. 5; Wilson et al., 2008; Barberopoulou et al., 2009), CGS recommended that four other south-facing, crescent-shaped coastal areas…Bodega Bay, Half Moon Bay, Santa Cruz, and Cayucos…be included with Crescent City and Port San Luis as areas of potentially higher hazard for the state. In addition, CGS indicated that high-tide conditions would be active north of San Francisco Bay at the time of first tsunami arrival, potentially increasing flow stresses on some port facilities. This information was included in state and federal messaging to local emergency managers. • Technical Support During Conference Calls – CGS and Lori Dengler (Humboldt State University) took part in the state conference calls with the WC/ATWC and the coastal jurisdictions. During the conference calls, geoscientists fielded questions from the coastal jurisdictions. Because of the recent change to the WC/ATWC tsunami alert-level criteria, there were many questions about the meaning of a Tsunami Advisory level of alert. The state emergency managers and geoscientists reinforced the WC/ATWC messaging
that an Advisory-level event indicates potential strong currents can occur in harbors and on beaches. In addition, geoscientists warned that: 1) forecasted arrival times should include a margin of error of up to plus or minus 15 min; 2) projected tsunami amplitudes could also be in error because of forecast model uncertainties; and, 3) the first wave is usually not the largest and that tsunami activity could last for a number of hours. In the days after the tsunami Advisory ended, the state tsunami program sent out questionnaires to the coastal counties to ask what actions they took and observations they made with regard to the tsunami. As a result of the Tsunami Advisory and the information provided by the various experts, most coastal jurisdictions indicated that they closed beaches and limited access within harbors and waterfront areas. The tsunami's first arrival and most of its activity was at night time making it difficult for people to make observations. However, people on beaches reported no unusual wave activity. Strong currents were observed in some harbors and at the mouths of rivers. Other than displacement of navigational buoys in some harbors, there were no reports of significant damage. 3.2. Tsunami field response activities After a significant tsunami hits the coast anywhere in the world, organized field teams of geoscientists typically collect perishable tsunami data. Although there were no organized efforts in California to deploy field response teams prior to the tsunamis arrival, observations were made at two locations by co-authors of this paper, Andy Ritchie in Santa Cruz and Mark Legg in Huntington Beach. These
198
R.I. Wilson et al. / Earth-Science Reviews 107 (2011) 193–200
Maximum Modeled Flow Depth (meters) 0.0
2.0
4.0
6.0
8.0
10.0 12.0
Crescent City Humboldt -Inside Bay Arena Cove Bodega Bay San Francisco Saucilito Richmond Redwood City Half Moon Bay Santa Cruz Monterey Cayucos Port San Luis Pt Arguello Santa Barbara Ventura Oxnard Malibu Santa Monica San Pedro/LA Harbor Huntington Beach Newport Beach Dana Point San Clemente Oceanside Del Mar La Jolla San Diego Bay Coronado Imperial Beach Fig. 5. Offshore maximum flow-depth values, in meters above sea level, from numerical modeling for the worst-case, distant-source tsunami scenarios as part of the state inundation mapping project (Wilson et al., 2008). This information was used to designate Bodega Bay, Half Moon Bay, Santa Cruz, and Cayucos as potential higherhazard areas comparable to areas considered higher hazard by the WC/ATWC (Crescent City and Port San Luis). All cities are listed from north to south along the coastline.
observations were valuable as there were no tide gauge instruments near these locations to record the tsunami. During the first couple hours of the tsunami, Ritchie made observations from the Murray Street bridge which crosses Santa Cruz Harbor about 0.8 km from the mouth of the harbor to the south. Ritchie observed strong tidal currents up to 8 knots and maximum tsunami amplitudes of 0.7 m above normal conditions. This observation was the largest amplitude reported in California from the September 29th tsunami and helps validate CGS's recommendation that adding Santa Cruz as a location that could expect higher amplitude surges during this Advisory. Legg observed the tsunami near the mouth of the Santa Ana River in southern California and noted several dozen small tsunami surges up to 0.2 m in height, riding up and over the outward flow of the river. Legg further noted that the drawdown was so rapid that it caused the outgoing tidal flow to accelerate and sand along the river bed to rapidly erode. In addition to Legg making scientific observations, he arranged to meet a local reporter and was interviewed on video discussing what he observed at the time of the tsunami's arrival. The video was placed on the internet and has since been copied to many other websites and discussion boards. Legg's analysis helped to demonstrate to the public that not all tsunamis are large events, and that even studying small tsunamis can provide useful information. Lack of verified field intelligence is a known gap in the midst of any type of ensuing natural disaster which only serves to aid effective emergency response. The authors believe that field observations like
these can help validate the accuracy of forecast and other numerical modeling done in the state, and are of value to emergency managers in the midst of an event. The field data could potentially be of even greater value if organized, interpreted, and disseminated through a state-level information clearinghouse. Field geoscientists could relay information to the local emergency manager, and to a regional field manager to be shared with the clearinghouse leader. Similar to the tsunami observer programs in places like Hawaii, real-time scientific information could help determine where state response resources could be best directed. 3.3. Evaluation of forecasted information Table 1 shows a comparison between the projected tsunami amplitudes and arrival times to the recorded values in California from the Samoa tsunami. This information was gathered from tide gauge marigrams measuring water levels at one-minute increments. Recorded tsunami amplitudes were approximately one half (48%) of the amplitude initially forecasted at the four sites where comparisons could be made; this error margin is near the outer limit of expected forecast accuracy (according to Whitmore, 2003). Also, the first arrival times were 17 to 48 min later at points recorded along the coast. A post-event evaluation by WC/ATWC determined this difference in wave amplitude and travel time is likely due to several factors: 1) differences in the forecast model fault motion (thrust) and the actual fault motion (normal); 2) complexities in the source region producing uncertainty in the directionality of the tsunami; 3) inaccuracies in matching the true source with the modeled source; and, 4) observed first arrivals being masked by tide gauge noise at some locations in California. A better understanding of the uncertainties in the forecasted amplitudes and arrival times is needed so that it can be explained to non-scientists in future events. Attempts were made to gather information about tsunami amplitudes at the four locations (Bodega Bay, Half Moon Bay, Santa Cruz, and Cayucos) where the tsunami vulnerability was thought comparable to Crescent City and Port San Luis. None of these locations have permanent tide gauges but the field observations in Santa Cruz by Ritchie indicating that maximum tsunami amplitudes reached 0.7 m helped validate adding Santa Cruz as a location to attribute an Advisory-like status for this event. Based on the responses of state and local emergency managers in the post-event surveys, information from state geoscientists helped emergency managers decide what actions to take during the event. Information provided by CGS … areas of potential increased tsunami amplitudes and vulnerability, high-tide conditions north of San Francisco Bay, and uncertainties in forecasted tsunami arrival times and maximum amplitudes…was received by coastal emergency managers through tsunami alert messaging and teleconferences, and incorporated into their decision making process. The participation of state geoscientists during the initial statewide teleconference helped emergency managers better understand the tsunami alert messaging, forecast information, and the expected tsunami hazard. 3.4. Conclusions about the Samoa event and future work As illustrated by California's experience during the Samoa teletsunami event, state geoscientists can provide emergency managers with useful information and assistance during a tsunami alert. They can explain the uncertainties in the NOAA tsunami amplitude forecast and travel time data so that state and local emergency managers can make more informed decisions about actions to take in their jurisdictions. In addition, because forecast modeling for inundation is done for only five of the more than 250 incorporated and unincorporated communities along the California coast, state geoscientists can use existing tsunami modeling data to determine the potential hazard level for comparable communities. As demonstrated
R.I. Wilson et al. / Earth-Science Reviews 107 (2011) 193–200
199
Table 1 Shows comparison between forecasted and recorded tsunami amplitudes and arrival times for the September 29, 2009 tsunami in California. Locations
Crescent City Arena Cove Point Reyes San Francisco Port San Luis Santa Monica Los Angeles Harbor San Diego
Tsunami Amplitudes (zero to peak in cm)
First Arrival Times
Projected
Recorded
% Difference (less)
Projected Time
Recorded Time
Difference (min. later)
65 44 39 20 60 15 30 15
33
51%
10 28
50% 47%
21:20 20:58 21:02 21:31 21:07 21:20 21:15 *
21:44 21:15 21:50 21:48 21:30 21:39 *
24 17 48 17 23 19
13 43% * * Average Difference = 48%
Average Difference = 25 min
*Not measurable.
by Legg, field geoscientists can also provide real-time feedback to state and local emergency managers and the public in general. Although state geoscientists are integrated into the statewide tsunami emergency response network, their role should be expanded during tsunami response activities. CGS and other geoscientists are working more closely with NOAA and state and local emergency managers to formalize a coordinated tsunami response plan to help provide the most valuable and timely information to those in decision making roles for their communities. This response plan will include: 1) participation by primary and backup geoscience contacts in all event-response meetings and teleconferences with NOAA and state and local emergency managers; 2) provision of information about historical tsunamis and expected tidal conditions as they relate to future forecasted events; and, 3) provision of supplemental scenario information of the expected tsunami hazard (flow depths, strong currents, etc.) for areas not covered by NOAA's forecast models. State tsunami geoscientists are continuing to provide additional support in a number of ways. The California tsunami program will continue to work closely with NOAA through the NTHMP to simplify their tsunami alert messaging and expand their forecast modeling coverage. California geoscientists will provide feedback on NOAA's messages that should help clarify the uncertainties in the tsunami forecast data. In addition, California is supporting the expansion of NOAA's development of digital elevation models used for production of SIFT tsunami forecast models within the state. The state tsunami program also plans to create tsunami response “playbooks” containing information from the recent statewide inundation mapping project, for local emergency managers to reference during an event. Information in these playbooks will help identify the relative hazard level for their coastal communities and include estimates on tsunami flow depths, current speeds, and travel times. Consideration is also being given to providing an easily accessed computer interface to provide this pre-run library of tsunami scenarios applicable to coastal California that would assist in interpreting existing data for emergency managers. CGS has proposed developing a pre- and post-tsunami field response team and state-level information “clearinghouse” made up of geoscientists to assist state and local officials. In addition to noting the effects of the tsunami, members of these field response teams will meet regularly and be available before, during, and after a tsunami to provide assistance to local emergency managers and the public. 4. Addendum: the February 27th, 2010 tsunami advisory in California Nearly five months after the Samoa tsunami, another Tsunami Advisory was issued for California when a M8.8 earthquake occurred off the coast of Chile. The earthquake was generated along the plate boundary where the Nazca Plate subducts under the South American Plate, approximately 300 km north of the magnitude (Mw) 9.5 1960
earthquake. A large tsunami was generated locally along the Chilean coast, causing severe damage to coastal communities and port facilities in Chile. Building on the experience during the event from Samoa and other tsunamis, geoscientists at the federal, state, and local level implemented several new response strategies during the event from Chile (Wilson et al., 2010): • The WC/ATWC using ATFM results increased the number of forecasted tsunami arrival time and maximum amplitude values at California sites from five to 30 (and issued forecasts for a total of 77 sites along the U.S. west coast, British Columbia, and Alaska coasts). With more information to use, local jurisdictions were better able to determine the appropriate response activities. Accuracy improved according to a NOAA Tsunami Forecast Assessment (Bernard, et al, in press) which showed the forecast values had an average error of 33% (20 cm rootmean-square error) from those observed at the 40 locations where the values could be compared. Though there was some confusion by local jurisdictions about what to expect during this Advisory, the additional information provided by the WC/ATWC assisted state emergency managers and scientist in their messaging to local authorities. • Because of the 13–1/2 hour delay between the tsunami's generation and its daytime arrival in California, more scientific observers were able to deploy to the coast prior to the tsunami's arrival. Rick Wilson, a co-author on this paper, was deployed to the Port San Luis area, the location of the highest forecasted tsunami amplitude values for this event. Wilson provided real-time information and recommendations to state emergency managers during the event. Some field observations helped confirm what the WC/ATWC noted on tidegauge marigrams during the event: Advisory-level tsunami amplitudes were active for up to 8 h after the first wave arrival. Validated by the field experiences of geoscientists during the two recent Advisories in California from the Samoa and Chile tsunamis, the NTHMP is funding the state's development of pre- and post-tsunami field teams and a centralized information clearinghouse for future Tsunami Advisories and Warnings. The plan includes enlisting the help of 30 to 40 field geoscientists that can be deployed to collect information about the tsunami similar to Legg, Ritchie, and Wilson. The field teams will be developed during 2010–11 and the plan will be ready for operation if needed in 2011. To help coordinate its development, the tsunami clearinghouse may be integrated with the long-established California Natural Hazards Clearinghouse. If successful, the NTHMP could recommend that other states use the California project as a template for development of pre-tsunami field teams and expanding integration of geoscientists into emergency response planning. Acknowledgements The authors would like to thank NOAA/NTHMP for supporting tsunami hazard mitigation and response activities in California. Thank
200
R.I. Wilson et al. / Earth-Science Reviews 107 (2011) 193–200
you also to the NOAA-NWS Warning Coordination Meteorologists and the state and county emergency managers who are part of the California Tsunami Program Steering Committee for their participation in the post-event survey. References Barberopoulou, A., Borrero, J.C., Uslu, B., Kalligeris, N., Goltz, J.D., Wilson, R.I., Synolakis, C.E., 2009. New maps of California to improve tsunami preparedness. Eos, Transactions, American Geophysical Union 90 (16), 137–138. Bernard, E., Whitmore, P., Rhoades, J., McCreery, C., Fox, C., in press. Chile Tsunami Forecast Assessment: February 27-28, 2010. National Oceanic and Atmospheric Administration (internal report). 77 pp. California Seismic Safety Commission, 2006. The tsunami threat to California. Ad Hoc Committee on Tsunami Safety, CSSC 05–03. 24 pp. Gica, E., Spillane, M.C., Titov, V.V., Chamberlin, C.D., Newman, J.C., 2008. Development of the forecast propagation database for NOAA's Short-term Inundation Forecasting for Tsunamis (SIFT). NOAA Technical Memorandum OAR PMEL-139. 89 pp. Kowalik, Z., Whitmore, P.M., 1991. An investigation of two tsunamis recorded at Adak, Alaska. Science of Tsunami Hazards 9, 67–83. Lander, J., Lockridge, P.A., Kozuch, J., 1993. Tsunamis affecting the west coast of the United States 1806–1992. NGDC Key to Geophysical Research Documentation No. 29. USDOC/NOAA/NESDIS/NGDC, Boulder, CO, USA. 242 pp.
NOAA/NWS/WCATWC, 2010. User's guide for the tsunami warning system in the West Coast/Alaska Tsunami Warning Center Area-of-Responsibility. NOAA National Weather Service, West Coast and Alaska Tsunami Warning Center Operations Manual 6.1.4, dated May 20, 2010. 43 pp. Patton, J.R., Dengler, L.A., 2004. GIS-based relative tsunami hazard maps for northern California, Humboldt and Del Norte Counties: (abs.). Eos, Transactions, American Geophysical Union 85 (47) Fall Meeting Supplement, abstract# OS24D-1353. Patton, J.R., Dengler, L.A., 2006. Relative tsunami hazard mapping for Humboldt and el Norte Counties, California. Proceedings of the 8NCEE/EERI Eighth Earthquake Engineering Conference. U.S. Geological Survey, 2009. M8.0 Samoa Islands Region Earthquake of 29 September 2009. U.S. Geological Survey Earthquake Summary Map, published 2 October 2009. Whitmore, P.M., 2003. Tsunami amplitude prediction during events: a test based on previous tsunamis. Science of Tsunami Hazards 21, 135–143. Whitmore, P.M., Benz, H., Bolton, M., Crawford, G., Dengler, L., Fryer, G., Goltz, J., Hanson, R., Kryzanowski, K., Malone, S., Oppenheimer, D., Petty, E., Rogers, G., Wilson, J., 2008. NOAA/West Coast and Alaska Tsunami Warning Center Pacific Ocean Response Criteria. Science of Tsunami Hazards 27, 1–21. Wilson, R.I., Barberopoulou, A., Miller, K.M., Goltz, J.D., Synolakis, C.E., 2008. New maximum tsunami inundation maps for use by local emergency planners in the State of California, USA. Eos, Transactions, American Geophysical Union 89 (53) Fall Meeting Supplement, Abstract OS43D-1343. Wilson, R.I., Dengler, L.A., Legg, M.R., Long, K., Miller, K.M., 2010. The 2010 Chilean Tsunami on the California Coastline. Seismological Research Letters 81 (3), 545–546.