Abstract
This paper examines communication methods used to inform the vulnerable community of Hilo, Hawaii of the impending tsunamis that struck in 1946 and 1960. These tsunamis caused tragic loss of life and enormous economic damage in Hilo and along the shores of the Hawaiian Island chain. Over 12 h notice of a possible large tsunami was given in 1960 and the siren warning system sounded more than 4 h prior to the event. The government agencies knew there was a tsunami alert and the media were broadcasting warnings. However, the 1960 tsunami took the lives of 61 people in Hilo only 14 years after 96 people were killed during the 1946 event. In order to discover why so many people perished, government agency logs recorded during the 1960 tsunami were examined and personal accounts from survivors of both the 1946 and 1960 tsunamis were analysed. Contributing to the tragic loss of life was a lack of communication between government agencies in addition to media inaccuracies and a public that was not educated in tsunami safety. Effective tsunami mitigation can only be accomplished through continual tsunami awareness education for the public, media and emergency personnel, and with accurate and timely tsunami warnings.
Residents living in tsunami-vulnerable regions have most probably never experienced a tsunami, and therefore lack knowledge of the hazard and understanding of how to behave (Murata et al. 2010). Despite many Hawaiian residents having direct or indirect experience of tsunami disasters in 1946 and 1960, most lack an understanding of tsunami hazard (Dudley & Lee 1998). It is also questionable whether or not people will take appropriate action in response to the next tsunami warning (Gregg et al. 2007).
Research conducted by Nishimoto (2000) and Johnston (2003) uncovered the depth of confusion and misunderstanding present during the 1946 and 1960 tsunamis that impacted Hilo, Hawaii. Comparing interview responses from survivors of both tsunamis also revealed that there was no heightened awareness of tsunami safety. Atwater et al. (2005) illustrated how people survived the 1960 tsunami in Chile, Hawaii and Japan through rational actions that reliably saved lives (e.g. heeding natural warnings, abandoning belongings, evacuating to higher ground until the hazard had passed) and desperate actions of people caught unaware and/or unprepared (e.g. taking refuge in buildings or trees, floating on debris). Survivor stories, such as these, offer invaluable insight into the complexities of human behaviour prior to, during and after a disaster and therefore, provide important lessons for survival (Ripley 2008; Bird et al. 2011). To preserve life and maintain public safety, it is essential that these lessons be incorporated into ongoing public education programs in addition to, promoting increased preparedness among individuals, businesses and communities. Mitigating the effects of tsunami hazards also requires accurate and timely warnings, interagency communication and co-operation, and hazard knowledge and understanding among key emergency management and media personnel.
In this paper, we will demonstrate that continual education is an essential component of risk mitigation, not only with respect to the public but also in regards to emergency management agencies and the media. We achieve this by analysing communication logs obtained from the County of Hawaii Civil Defense Agency in Hilo, Hawaii, and examining interviews with survivors of the devastating 1946 and 1960 tsunamis, which are archived at the Pacific Tsunami Museum. A discussion on the implications of our findings in relation to risk communication and public education is provided. First, we describe the events of the 1946 and 1960 tsunamis, and the warning system, in the following sections.
The 1946 tsunami
On 1 April 1946 a major earthquake centred about 90 miles (145 km) south of Unimak Pass, Alaska (Fig. 1) created a 100 ft (30.5 m) tsunami wave that swept away the lighthouse on Unimak Island in the Aleutian Islands, killing all five US Coast Guard personnel in the facility (Dudley & Lee 1998). The lighthouse, built in 1940, was nearly 100 ft (30.5 m) high and was sited 40 ft (12.2 m) above sea level.
(a) The locations of the 1946 earthquake in the Aleutian Trench and the 1960 earthquake in Concepción. (b) The Hawaiian Islands.
The earthquake occurred at approximately 02:00 Hawaii local time (Hawaii Standard time) on 1 April 1946 in the Aleutian Trench (Shepard et al. 1949). The US Coast and Geodetic Survey at the University of Hawaii at Manoa, Honolulu, Oahu and the Hawaiian Volcano Observatory at Kilauea, Hawaii registered and recorded the Aleutian earthquake shortly after it occurred (Dudley & Lee 1998). A tsunami, however, travelling over 2300 miles (3700 km) from the Aleutian Islands to Hilo at a rate of nearly 500 miles h−1 (805 km h−1) was not expected. The first wave reached Kaua'i at around 06:00, Honolulu at 06:33 and Hilo at approximately 07:00 (Shepard et al. 1949). A total of nine waves, varying from 17 to 25 ft (5.2–7.6 m) in height, inundated Hilo that day (Dudley & Stone 2000).
Dudley & Stone (2000) reported that as the tsunami approached Hilo, it inundated Laupahoehoe Peninsula (Fig. 2) killing 16 school children and eight adults. On all of the islands, almost 500 homes and businesses were totally destroyed, at least 1000 sustained damage and property damage was estimated at approximately $26 million (1946 US$) (Dudley & Stone 2000). The cost in relation to human life, however, was much worse. A total of 159 people lost their lives in the Territory of Hawaii and 96 of those lives were lost in Hilo.
(a) The Big Island, Hawaii showing Hilo, the Laupahoehoe Peninsula and South Point. (b) The township of Hilo showing the HFD headquarters at 28 Shipman Street (St.), located approximately 100 yards (91.4 m) from Hilo Bay and 100 ft (30.5 m) from the Wailuku River. Other prominent geographical points impacted by the 1960 include Coconut Island and Liliuokalani Park.
Almost immediately, Hilo residents started to rebuild their homes and lives. Although some businesses were never rebuilt, most residential areas and businesses in Waiakea were rebuilt with some, once again, located in vulnerable, low-lying areas (Dudley & Stone 2000). However, Shepard et al. (1949, p. 392) noted:If, moreover, a suitable system can be devised for warning of the approach of dangerous tsunamis even a few minutes in advance of their arrival, it should be possible to effect a very considerable saving of life; and warnings an hour or more in advance would allow the saving of much easily movable property.
It is worth noting that much of the vulnerable, low-lying area rebuilt following the 1946 event was eventually set aside for recreational land use following damage caused as a result of the 1960 tsunami.
The 1960 tsunami
Just after midnight, Hawaii local time, on 21 May 1960 an earthquake registering Mw 7.5 struck near Concepción, Chile, 6600 miles (10 622 km) to the SE (Eaton et al. 1961) (see Fig. 1). The earthquake was registered at the US Coast and Geodetic Survey Observatory (warning centre) in Honolulu, Oahu and a tsunami watch was issued at 00:45. The warning centre started to monitor tide gauges throughout the Pacific but the alert was cancelled at 20:49 when only a small wave was recorded in Hilo Bay.
By 11:30 Hawaii time on 22 May 1960, it was known that a second, large earthquake had occurred near Concepción approximately 2.5 h earlier (Eaton et al. 1961). This event, however, was followed 15 min later by a huge Mw 9.5 earthquake (Kanamori 1977). This was the largest instrumentally recorded earthquake of the twentieth century (Johnston et al. 2008).
Once again, a tsunami watch was issued and the expected time of arrival was predicted to be about 15 h later, at around midnight local time. News reports of destructive waves along the Chilean coast were received in Hawaii during the early afternoon. At 18:47 the warning centre issued a Seismic Sea Wave Warning and at 20:30 – the Hawaii Civil Defense Agency (HCDA) communication log states that the first siren sounded at 20:35 – coastal sirens signalled people in the low-lying areas of Hilo to evacuate to higher ground (Eaton et al. 1961). Lachman et al. (1961, p. 1406) reported that the warning sirens sounded ‘for a 20-min period more than four hours prior to the impact of the wave’.
It was not until 22:23 that official notice of unusual wave activity was received from Tahiti, approximately 2800 miles (4506 km) south of Honolulu (Dudley & Lee 1998). This was the first official confirmation received by the warning centre since the first reports that tsunami waves had impacted the Chilean coast.
At 00:07 on 23 May 1960, rising water was noted beneath Wailuku Bridge (see Fig. 2) on the west edge of Hilo Bay (Eaton et al. 1961). The water crested at 00:13 at more than 4 ft (1.2 m) and began to fall slowly until 00:30 to a trough of −3 ft (−0.9 m). At 00:46, the second wave passed under Wailuku Bridge at a level of +9 ft (+2.7 m) (Dudley & Lee 1998). This wave flooded Kamehameha Avenue and the business district of Hilo. As the water withdrew, a trough of 7 ft (2.1 m) below normal was noted.
At 01:04, Eaton et al. (1961, p. 139) reported a ‘20-foot [6.1 m]-high nearly vertical front’ as the tsunami wave flooded past Wailuku Bridge. At 01:05, the power plant at the southern end of Hilo Bay was impacted by the incoming waves and, after a brief moment, the city of Hilo and most of the Island of Hawaii went dark. Just over 1 h later, at 02:15, the height of the waves had diminished sufficiently for people to re-enter the township of Hilo and assess the damage (Eaton et al. 1961) (Fig. 3).
View WSW across Waiakea after the 23 May 1960 tsunami impacted Hilo, Hawaii. (Photograph courtesy of the Pacific Tsunami Museum, photograph number 1960.04, Polhemus Collection.)
The tsunami warning system
According to Dudley & Lee (1998), Hawaii should have approximately 15 h notice of an impending tsunami generated from South America and only 5–10 min for locally generated tsunami. Tsunami disaster communication therefore requires a specialized strategy because of the potentially limited warning time and also due to their rare occurrence. It is imperative that the disaster response community, the media and the public have accurate information as quickly as possible. This is particularly important for locally generated tsunami with respect to preserving life and maintaining public safety.
On 1 April 1946 there was no early warning to the population but by 1960 a tsunami siren warning system was in place in the state of Hawaii. In addition to the sirens, there were warnings broadcast by radio and government agencies. This early warning system provided technically successful warnings of three large tsunamis on 4 November 1952, 9 March 1957 and 23 May 1960. However, a report from the Geophysical Society of Hawaii (1960) found that improvement in the overall system and its acceptance by the public was still needed.
Based on Bonk et al. (1960), Dudley & Lee (1998) explained that many people did not understand the warning system because it was changed only a few months prior to the 1960 tsunami. The previous warning system consisted of three separate alarms: the first siren was to indicate that a tsunami warning was in effect; the second siren meant to evacuate; and the third was supposed to sound just prior to the arrival of the first wave. The new system, however, consisted of only one siren indicating immediate evacuation, although people may not have been aware of this change.
Nevertheless, the warning siren had been activated prior to 22 May 1960. Johnston (2003) reported that sirens signalling the arrival of relatively small tsunamis, such as the one on 21 May, were perceived by the public as ‘false alarms’ and residents were therefore complacent on 22 May.
In the next section we describe the methods used to analyse communication logs and examine interviews with survivors of the devastating 1946 and 1960 tsunamis.
Methods
Agency logs
An in-depth examination of the logs prepared by the three leading governmental agencies responsible for the safety of Hilo's citizens during the 1960 tsunami was performed. The information was collected from documentation located at the County of Hawaii Civil Defense Agency (HCDA) office in Hilo, Hawaii, and it included logs from the HCDA, the Hawaii Police Department – Hilo Station (HPD) and the Hawaii Fire Department (HFD). These logs were taken during the 28 h period from 11:58 on 22 May to 16:03 on 23 May 1960.
Tsunami survivor interviews
Content analysis was conducted on recorded interviews with individuals who survived the 1946 and 1960 tsunamis in Hilo. All interviews were recorded on digital audio-tape (DAT) and video-tape for the Pacific Tsunami Museum, Hilo, Hawaii. The reader is referred to Dudley et al. (2009) for a detailed description of methods used during video interviewing.
The lead author (Johnston) conducted the interviews on a one-on-one basis, with the length of interviews averaging between 15 and 60 min. It is important to note that some of the survivors of these disastrous events found it difficult and emotional to talk about their experiences. For some, it was the first time they had shared their experience since the tsunami. In all cases, the interviewer respected the survivor's pace and content as they shared their stories, and all attempts were made to accurately report the information they relayed.
Copies of the transcriptions and original DATs are archived at the Pacific Tsunami Museum. All interviewees received copies of the transcripts and have released their interviews to the Pacific Tsunami Museum, where they are available to the public.
A total of 67 interview transcripts were reviewed in this study. This included 45 survivor interviews from the 1 April 1946 tsunami and 22 survivor interviews from the 23 May 1960 tsunami. Each transcript was examined in an effort to attain answers to the following three questions:
Where were you when you heard about the tsunami?
Where did you get the information?
What did you do?
In addition to interviews archived at the Pacific Tsunami Museum in Hilo, Hawaii, interviews conducted by Bonk et al. (1960) and Nishimoto (2000) were reviewed and analysed.
Results
Agency logs: 22–23 May 1960
Unofficial notification
The first indication of the impending disaster recorded in the logs was the ‘unofficial’ report from the US Geological Survey's Volcano Observatory, received by the HPD at 11:58 on 22 May 1960. It was 17 min later, at 12:15, when an officer from the HPD notified the HCDA (Table 1). This notification indicated that the arrival time of the tsunami in the Hawaiian Islands would be midnight on 22 May 1960. The log for the HFD in Hilo did not begin until 20:05 on 22 May 1960.
Critical communication events from 11:58 to 21:38 on 22 May 1960 (after Johnston 2003)
Tidal wave alert
The logs of the HCDA disclose that, at 19:05, Hilo's KHBC radio announced that the Coast and Geodetic Survey had issued a tidal wave alert at 18:48. Six minutes later, at 19:11, the logs of the HPD declare that the Honolulu Police Department called and reported the warning. In addition, the account shows that the State Civil Defense Headquarters requested that the HPD notify all civil defence agencies. However, they were instructed to not sound the siren until advised. The HFD logs indicated the announcement of the ‘tidal wave alert’ occurred almost 1 h later at 20:05.
Warning sirens and evacuation
The HPD log indicated that local radio stations were told at 19:59 that no siren would be sounded at 20:00. At this time, the Assistant Chief of the HPD called the HCDA to notify them that he had received a message from the State Civil Defense that sirens were not to be sounded until notified by the State Civil Defense Headquarters.
At 20:15 the HCDA called the State Civil Defense Headquarters and was given the authority to sound the sirens. At 20:35, the HCDA log recorded ‘First Siren Sounded’, the HPD recorded ‘Siren sounded … Evacuation begins in lower areas’ and the HFD recorded ‘Tidal wave sirens sounded’. According to the HCDA, there were five separate sirens that sounded intermittently – at 20:35, 20:40, 20:43, 20:47 and 20:50.
At 21:38, the HPD log indicates that they received a call from the Honolulu Police Department asking them to sound the warning siren. However, no other records could be located to indicate whether this message was passed along to any other agency, nor was there any record of the sirens being sounded at or around 21:38.
The HPD log reports that they were ‘instructed to evacuate lower areas immediately’ at 21:56 (Table 2).
Critical communication events from 21:56 on 22 May to 00:20 on 23 May 1960 (after Johnston 2003)
Tahiti wave report
At 22:31, the HCDA log reports: ‘Wave hits Tahiti at 8:00 pm, 3 ft. high …’. By this time, the HFD had sent one of their employees over to HCDA with a handheld radio. The employee reported the news of the Tahiti wave to the HFD at 22:35 just 4 min after the HCDA received the report. However, the HPD did not log the news until 20 min later, at 22:55, and their log reveals that they received their report from Hilo's KHBC radio. The HPD report stated: ‘… 3 ft 3 in [1 m] waves were recorded in Tahiti. First reports indicate no damage’.
At 23:00, the HCDA log reports that they called the Civil Defense in Honolulu. The Civil Defense confirmed to the HCDA that Tahiti had been impacted by a tsunami and that they should expect a wave to reach Hawaii as previously estimated. Four minutes later, the HPD was notified of the Tahiti wave and at 23:05 there is a second report of the Tahiti wave in the HCDA log.
Tidal wave gauge
The HPD log reported at 23:30 that ‘the tidal wave gauge is normal’. Since there was no record of where the gauge was located in the records, a Hilo police officer at the time of the tsunami was contacted for the purpose of this study. This officer stated that the gauge was located at Pier One in Hilo Harbour ‘on the breakwater [ocean] side of the harbor’. This officer also stated that whenever there was a tsunami warning the police department would send an officer down to watch the gauge. He stated they would wait for the alarm to sound before taking off, and pray that they reached Silva Street and Kalanianaole Avenue before the water came.
Christmas Island and Samoa wave reports
The logs record that Samoa and Christmas Island were impacted by a tsunami at 20:40 and 22:58, respectively. This information was recorded in the HPD log at 23:47 and in the HFD log at 23:53. The HFD log also reports that an official notification had been given by the Governor of Tahiti stating that Tahiti had suffered no damage.
The first wave
At 00:09 on 23 May 1960, the HPD log indicates that the tidal wave gauge (presumably at Pier One) exceeded the normal high tide level. At 00:15, the HCDA log reports that the HPD notified them that the gauge ‘went off’ and that the water had not begun to recede.
Five minutes later, all three agencies became aware that the tsunami had impacted Coconut Island. Also, the HFD log at 00:20 states that the tidal wave alarm at Pier One had sounded and the HCDA log notes that the water had begun to recede at 00:22 (please note: this entry was logged in at 00:20).
At 00:25 the HCDA received a call from the US Coast Guard stating that they should be aware that the drop in wave height is just as important as the rise in water (Table 3). Also at 00:25, the HPD log states that the water was receding at Wailoa and at 00:33 the HCDA log notes: ‘water receding 5′ below normal’. According to the log entries at 00:35, the HFD received a report from the Coast Guard at Pier One that water had receded 5 ft (1.5 m) below normal and that it was continuing to drop.
Critical communication events from 00:25 to 01:05 on 23 May 1960 (after Johnston 2003)
The second wave
The second wave was logged by the HPD at 00:44: ‘Coconut Island covered with water and now flowing into Liliuokalani Park’; at 00:47: ‘Water coming over the pier’; and at 00:52 ‘Water entering Hawaiian village at Wailoa’.
At 00:53, the HCDA log states ‘Secured headquarters at 60 Shipman Street’ and that is the last logged entry until 01:40. The HCDA evacuated their Shipman Street headquarters and relocated to the HFD because at that time their headquarters was located approximately 100 yards (91.4 m) from Hilo Bay and 100 ft (30.5 m) from the Wailuku River.
The HFD log details at 00:53 that the Naniloa Hotel was surrounded by water at 00:48 and, at 00:58, that Kamehameha Avenue was covered with water from the Hilo Theatre to Waiakea. At 01:00, the HPD log relates that ‘water went to intersection of Kalanianaole and Silva Sts. 100 yards [91.4 m]’. Three minutes later at 01:03 the HFD log states that there is water on the road right up to Hilo Armory (28 Shipman Street): ‘Lights on front street went out. Evacuating all cars from Shipman Street. CD Office [HCDA] evacuated from Headquarters at Shipman Street and set up temporary headquarters in the Fire Prevention Bureau Office [HFD]’.
At 01:05 on 23 May 1960, the HPD log attests that the lights went out and at that same time the HFD log states ‘Electricity went out. No power in city. Tidal Wave hit town area’.
After Hilo went dark
At 02:20, the HCDA log states that a captain from the HPD initiated co-ordination between the HPD, HFD and the HCDA. This is the first indication in the logs that there is any attempt to co-ordinate interagency communication.
According to the HCDA log, at 02:30 Governor Quinn announced over the radio, KGMB [Honolulu] and KHBC [Hilo], that all people should stay away from the beach areas. Quinn mentioned the seriousness of the damage in Hilo.
The HPD log indicates that the power was back on at 06:02 and at 06:24 the following announcement was made: ‘Director of State CD [Civil Defense] agency to all islands – NO FURTHER DAMAGES FOR THE PRESENT FROM THIS TIDAL WAVE. ALERT CALLED OFF – 6:10 AM’.
South point, Island of Hawaii
It appears that the HCDA and the HPD believed that the tsunami would impact South Point (see Fig. 2) prior to Hilo. As there was no telephone available at South Point, the HCDA instructed the Kau Police Station to position a police officer as a lookout at South Point and to signal the arrival of the tsunami with a flare. This instruction came from the HCDA at 21:38 on 22 May 1960.
At 22:43, there are two log entries. First, the HPD log states that the HPD contacted the Kau Police Station asking for the police officer stationed at South Point to locate a certain person. The HPD instructed that ‘He is to remain with this man and pass on information on the tidal wave to the Kau Station and in turn to Hilo Station as soon as possible’.
The second entry was in the HCDA log and it revealed that the HCDA had spoken to the HPD about: ‘whether there is any possibility of some flare signal from the South Point area to the phone. Time is the element – to notify Honolulu …’.
At 22:45, the HCDA log reports that a radio amateur operating from South Point is to notify Hilo and Honolulu. HPD notes, at 22:49, that the HCDA requests that any messages received from South Point regarding the tidal wave are transmitted immediately to the HPD in Honolulu via a special use telephone line.
At 22:50, the HCDA called HPD for further information. The HPD reported no change from South Point at that time. Then, at 23:55, the HCDA called the HPD for the latest report by midnight from South Point. The HCDA log notes at midnight ‘No change at South Point’ and, at 00:52, ‘South Point – No Activity’. Similarly, the HFD log states, at 00:53, ‘South Point normal’. The HCDA, HPD and HFD logs do not mention South Point again.
General observations
None of the logs were signed and it is therefore not possible to tell who made the records. The logs of the HCDA were disorganized and it was apparent that there was more than one person journaling in the HCDA as different typewriters were used for the logs as well as different forms. The HCDA logs are time-dated using 24 h time, although the time was out of sequence in some instances. The evacuation of the HCDA headquarters may account for the discrepancies in the time sequence but there is no notation in the logs as to who was transcribing. Overall, the HFD log is the most concise of the three logs.
Tsunami survivor interviews
The Pacific Tsunami Museum, formerly known as the Hilo Tsunami Museum, was founded in 1993. In 1996, the museum began collecting and archiving tsunami survivor interviews. By that time, many of the people who had survived the 1946 tsunami had passed away. The majority of the survivors who were interviewed were therefore under the age of 40 years during the 1946 event (Table 4). The 23 May 1960 Hilo tsunami survivor interviews were also gathered after 1996. In this case, the majority of interviewees were over 31 years of age at the time of the tsunami.
Interviewee demographics
1 April 1946
There was no official tsunami warning issued for the 1946 tsunami. As the first tsunami wave arrived in Hilo shortly before 07:00, most people were either at home or in the process of going to work or school (Table 5). The first inclination of a problem for many people was when they witnessed the actual tsunami. Sixteen of the 45 interviewees stated that they saw the water before they heard the news from another source. Others were informed by word-of-mouth, in most cases from a relative or friend, but some people were only made aware of the tsunami by people yelling in the street.
Interviewee responses in relation to the 1946 tsunami (n=45)
More than half of the survivors left immediately and began running for safety but there were seven individuals who made a decision to watch what was happening. Not realizing the danger, six of the 45 interviewees decided to go back into their homes. Three of the survivors did not have a chance to respond to information because their first knowledge of the tsunami was when they saw the water. One interviewee, who was a policeman during the tsunami, stated that he drove to the police station as soon as he was informed of the tsunami.
23 May 1960
As with the 1946 tsunami, the first warnings of the 1960 tsunami were announced outside of work hours and, as a result, most of the interviewees were at home when they heard about the tsunami (Table 6). Most interviewees received information about the tsunami by word-of-mouth while a few mentioned the warning siren and even less stated a government agency. Upon hearing about the tsunami, most interviewees stated that they stayed at home and a little more than a quarter evacuated.
Interviewee responses in relation to the 1960 tsunami (n=22)
Tsunami survivor interviews – Bonk et al. (1960)
Bonk et al. (1960) administered a questionnaire to 327 people who lived in the impacted region of Hilo. Although there is no information in the report as to where the individuals were when they first heard about the tsunami, their results indicate that 60% were at home when the tsunami impacted Hilo. This compares to 68% in the study presented above.
Bonk et al. reported that 95% of their respondents heard the 20:35 warning siren and, of these respondents, 41% evacuated and 59% did not. In comparison, results of the interviews presented in this paper show that only 14% of interviewees heard the siren.
In their study, Bonk et al. asked ‘Did you have information about the wave before it hit other than the siren?’. They reported that 80% answered ‘yes’ and 20% responded ‘no’. Those respondents who stated yes received information from radio/TV (68%); relatives or friends (17%); police, fireman or civil defence (3%); radio/TV and relatives and friends (8%); radio/TV and government (3%); and other (1%).
When asked directly ‘Did you hear the siren?’ nearly all respondents in the Bonk et al. study stated ‘yes’. Whereas in the interviews presented in this study, the interviewees were asked ‘Where did you get the information?’ and, therefore, probably responded with the method of communication that made the greatest impact regardless of whether or not they had actually heard the siren. This is easily reconcilable since there had been several occasions in previous days when the sirens were used to warn of a tsunami and no destructive tsunami occurred.
Upon hearing the siren, respondents in the Bonk et al. study waited for further information (45%); evacuated (32%); went about normal activity (15%); or other responses (9%). Whereas, those who received information from other sources either evacuated (36%); prepared to evacuate (30%); awaited further instruction (19%); did not respond (10%); went to watch (3%); and continued their normal routine (1%). In the present study, however, only 27% evacuated while 18% went to watch and 36% stayed at home.
Discussion
The research suggests that governmental agencies did not efficiently communicate or co-ordinate during the 1960 tsunami, which led to confusion between the agencies. Exacerbating the situation were media inaccuracies (Eaton et al. 1961) and a public naive of their vulnerability to tsunami hazard (Johnston 2003). These issues are discussed in the following sections.
Procedures and co-ordination
One of the most interesting discoveries was the lack of communication between the agencies and to the public, and a lack of understanding of the phenomena. It took until 02:20 on 23 May 1960, more than 14 h after the HPD first notified the HCDA of the tsunami hazard warning, for the agencies to co-ordinate their efforts. It is noticeable from reviewing the logs that there was virtually no co-ordination between government agencies during the pre-impact period with one exception – the HFD sent one of their employees to the HCDA with a handheld radio so that he could report the latest news to them. Interagency communication and co-operation should be established long before a disaster occurs and much was learnt from this event.
The logs of the HCDA were understandably disorganized as they had to evacuate from their headquarters and move their operations to the HFD. Overall, the HFD log was the most concise and they received information in a timely manner because they sent one of their employees over to the HCDA to act as a liaison.
Although it appears that the HPD was the lead agency for response in Hilo during the tsunami, permission to sound the tsunami warning system had to come from the State Civil Defense on Oahu. It would have been, therefore, more expedient for the State Civil Defense to notify the HCDA directly. Also, the ‘tidal wave alert’ message was received by the HPD 23 min after the message had been issued by the US Coast and Geodetic Survey Observatory. It is apparent that there were no procedures in place and that any action taken was left to the discretion of the individual deputy director.
Another issue was that the authorities did not evacuate homes in the inundation zone. Eaton et al. (1961, p. 140) reported that some time after 02:15 on 23 May 1960, when they ventured back into Hilo that they found ‘Hilo's streets had been evacuated, but its buildings had not’.
Media misinformation
According to Eaton et al. (1961), at around 00:30 when the water from the first wave had begun to recede, a Honolulu radio station reported that the tsunami had not yet impacted Hawaii and the estimated time of arrival had been set back half an hour. It was then realized ‘with growing uneasiness that communications between crucial observation points and public news media had broken down’ (Eaton et al. 1961, p. 139).
However, other residents returned home after the first wave had impacted Hilo after hearing on the radio that the tsunami waves had already impacted the island (Atwater et al. 2005). Adding to the confusion, these residents were met by a police officer who informed them that the hazard had passed, but at around 01:00 they heard the low rumbling sound of the next tsunami wave impacting the coast.
Tsunami awareness
A lack of tsunami awareness was clearly demonstrated by various government agencies. First, the HPD sending an officer down to watch the gauge whenever a tsunami warning was issued shows that the police department lacked understanding and awareness of tsunami hazard and vulnerability. Secondly, following the devastation that occurred in 1946, a more appropriate early warning system should have been installed other than the HPD sending a police officer as a lookout to South Point at the instruction of the HCDA. Thirdly, it should have been apparent that an agency such as the HCDA, whose function it was to respond to a tsunami, should not be located in the evacuation zone.
Furthermore, the US Army Engineer's Report (1960) testified that an employee from the Corps of Engineers was sent to Hilo to observe and report on the tsunami. Sadly, he lost his life in the 1960 disaster. As with the other lives lost, his life could have been spared if the authorities and population were educated on tsunami hazards and how to take appropriate action in response to the warning signal.
Governor Quinn's radio announcement at 02:30 on 23 May 1960, that people should stay away from the beach areas, should have been made in conjunction with the tsunami warning sirens which sounded between 20:35 and 20:50 on 22 May 1960. This time period presented a perfect opportunity to explain the new warning system to the public as it was changed only a few months earlier. The public needed to know that one siren indicated immediate evacuation. In a more recent study, Gregg et al. (2007) concluded that public understanding of the tsunami warning sirens in Hilo still remains low and, therefore, a major change in tsunami education is essential to increase public understanding of, and effective response to, future tsunamis.
Eaton et al. (1961) stated that the failure of nearly two-thirds of the residents to evacuate in spite of notice on 22 May 1960 that a tsunami would reach Hilo ‘points up a fatal flaw in the public's understanding that low-lying seashore areas must be evacuated before the expected tsunami strikes’ (p. 156). Adding to this misunderstanding is the fact that the area in Hilo impacted by the 1960 tsunami was not the same as in 1946. As a result, many people decided to stay home because they perceived that their area was not vulnerable or they did not understand the meaning of the warning sirens (Atwater et al. 2005). This further highlights the need to educate the public on tsunami hazard and warning, and to ensure that they are aware of their vulnerability in all regions.
An effective and compelling approach to tsunami education is through the use of the survivor video interviews. These videos provide immense value to education initiatives because they reveal visual emotive elements that other media sources lack (Dudley et al. 2009). Educational videos are also being used to retell traditional stories that relate to recognizing natural warning signs and suggesting appropriate behavioural responses. For example, the video Run to Higher Ground! uses a mix of traditional stories and contemporary approaches to educate schools and communities about tsunami hazards, warnings and response to warnings (Becker et al. 2008). These educational tools might also prove valuable in increasing awareness among emergency management and media personnel.
In a report written by the Geophysical Society of Hawaii (1960, p. 1) the need for tsunami awareness is emphasized:The poignancy of this tragedy is intensified by the fact that it was obviously technically preventable. An advisory bulletin on a possible tsunami had been made public nearly 12 h before a wave could arrive, and a formal warning was publicly issued four hours in advance. It appears certain that this information reached the people concerned, not only in Hilo but generally in shoreline areas of the state. Yet many failed to take adequate precautionary action and, worse, others, spurred by curiosity, placed themselves in unnecessary danger. This behaviour, and the testimony of many whose lives were spared, indicate a failure on the part of a large segment of the public to understand the nature of magnitude of the risks to which they subjected themselves.
Like other islands in the Pacific Ocean, the Hawaiian Islands are vulnerable to tsunami from both locally (e.g. earthquakes, volcanic eruptions and/or submarine landslides) and distantly generated (e.g. circum-Pacific subduction zones) sources such as the recent 27 February 2010 Chilean Earthquake (Goff et al. 2011). Steps are being taken by government agencies to reduce the risk from the next catastrophic tsunami (e.g. Crawford 2005; Darienzo et al. 2005; Dengler 2005; Eisner 2005; Johnston et al. 2005). These efforts include enhancing collaboration between emergency management and scientific agencies, and providing media training workshops (Jonientz-Trisler et al. 2005). However, these efforts must be continued and improved upon in order to prevent the next tsunami hazard becoming a disaster in Hilo.
Conclusion
No distant destructive tsunamis have impacted Hawaii since 1960 but the 26 December 2004 Indian Ocean tsunami showed that we, as a society, are complacent. This complacency has led to several generations of people who know very little about the phenomena of tsunamis. Tsunami preparation and increased awareness are essential to lessen the impact of future tsunami disasters (Johnston 2003).
Reports following the devastating 1960 tsunami emphasized the lack of public understanding of tsunami hazards and of the risks to which individuals subjected themselves. Fifty years on, events like the 29 September 2009 South Pacific tsunami and the 25 October 2010 Mentawai Islands, Indonesia tsunami, highlight the desperate need for continuing tsunami education for locals, visitors and hospitality staff in regions vulnerable to tsunami hazards.
Hawaii is an island state and, therefore, it is not only those who live and work in the inundation zones who are potential victims, but visitors too. In the case of a locally generated tsunami, anyone in the inundation zone would be at risk. An Australian tourist holidaying in Waikiki is just as vulnerable as a resident to losing their life due to lack of tsunami awareness, understanding and preparedness. Every person must therefore formulate an evacuation plan whether they live, work or occasionally use the beaches and parks for recreation. Continuing tsunami education is the key to creating a culture of awareness, understanding and preparedness.
Education campaigns must also focus on increasing awareness, understanding and preparedness among government officials, the media and hospitality staff. For example, it is the ignorance of the phenomena that leads the media to disseminate inaccurate information prior to the event, and during the watch and warning period – a time when concise and accurate information is essential to save lives. Fischer (1998) explained that disaster researchers have an obligation to the communities and to society at large to lead the educational process so that it will reach the public, the emergency managers and the reporters. In Hawaii, it is imperative that media agencies are included in disaster mitigation meetings and that a continuing programme of media education be instituted.
A catastrophic tsunami will occur in Hawaii in the future and, therefore, there will be a need for timely and accurate communication. An increasingly complex and diverse society with a growing population and commercial environment in the inundation zone increases the need for institutionalizing the dissemination of critical, life-saving information to the population.
- © The Geological Society of London 2012
References
An examination of hazard communication logs and public response during the 1946 and 1960 tsunamis that impacted Hilo, Hawaii
