Cascadia Subduction Zone Residents’ Tsunami Evacuation Expectations
Round 1
Reviewer 1 Report
My primary question to the authors is the role of Wireless Emergency Alerts in disseminating near shore tsunami warning messages to those at risk. WEA is currently enabled for earthquake early warning, delivering messages to drop, cover, and hold on within seconds of earthquake detection. NOAA has designed tsunami WEA to be sent and they are prioritized, along with EEW, by FEMA IPAWS. Furthermore, android operating systems are also adopting earthquake early warning and tsunami warning, which can deliver warnings to persons who are geolocated within the area of impact (an improvement over WEA's cellular broadcasting mechanism). Can the authors address this in their discussion/conclusion? It seems that it might be an overstatement that residents are erroneous in their belief that they might be warned through social sources, including warnings, prior to experiencing a near shore event.
Author Response
Comment: My primary question to the authors is the role of Wireless Emergency Alerts in disseminating near shore tsunami warning messages to those at risk. WEA is currently enabled for earthquake early warning, delivering messages to drop, cover, and hold on within seconds of earthquake detection. NOAA has designed tsunami WEA to be sent and they are prioritized, along with EEW, by FEMA IPAWS. Furthermore, android operating systems are also adopting earthquake early warning and tsunami warning, which can deliver warnings to persons who are geolocated within the area of impact (an improvement over WEA's cellular broadcasting mechanism). Can the authors address this in their discussion/conclusion? It seems that it might be an overstatement that residents are erroneous in their belief that they might be warned through social sources, including warnings, prior to experiencing a near shore event.
Response: We have revised the Method section to note that the survey was conducted during August and September of 2019, so the respondents’ data reflect conditions as they understood them at that time. In addition, the project for which these data were collected was advised and its final report reviewed by the tsunami coordinators for Alaska, Washington, Oregon, and California. Consequently, the data were consistent with conditions as the tsunami coordinators understood them at that time.
We have also revised the Discussion section to call attention to the fact that Table 1 indicates that only 25% of the respondents had subscribed to electronic alerts from the National Tsunami Warning Center, so few of them would have been able to receive an electronic alert. This low level of NTWC subscription clearly indicates a need for electronic alerts that do not require people to opt-in.
We agree that recent developments in WEA represent a significant step toward achieving timely warnings for a CSZ tsunami. However, we have reviewed some of the federal and state agency guidance about WEA (see the URLs listed below) and note that these web pages instruct people to verify that their devices enable emergency alerts and public safety alerts. This instruction clearly indicates that authorities are uncertain whether everyone will be able to receive a tsunami alert through WEA. Moreover, there is also a possibility that a CSZ earthquake could cause some warning transmission failures if, for example, some cell towers lose power or topple (e.g., due to landslides or soil liquefaction). This is similar to the caution that the actual range of tsunami siren signals can be significantly attenuated below their nominal range by the background noise of wind and surf (Lindell & Prater, 2010). Consequently, the possibility of at least partial social warning systems failure means that emergency managers should remind coastal residents of the need to know that earthquake shaking is an important cue to tsunami onset.
https://nws.weather.gov/nthmp/wea/weamessages.html; https://www.weather.gov/media/wrn/WEATsunamiFactSheet.pdf; and https://m.mil.wa.gov/alerts#tsunami
Reference
Lindell, M.K. & Prater, C.S. (2010). Tsunami preparedness on the Oregon and Washington coast: Recommendations for research. Natural Hazards Review, 11, 69-81.
Reviewer 2 Report
The reviewer appreciates the overall quality of the paper, the precision in results reporting and clear-cut indication of study limitations.
However, the reviewer would suggest some little changes to improve clarity and usability of such research. From a methodological point of view the article appears to be flawless, nonetheless it can be improved by providing some general directions to make an overall sense of fine-grained information provided by authors.
In general terms, the paper appears intended to provide several evidence on a number of different, yet very narrow areas of investigation, making a little uneasy to get a comprehensive and conclusive image of the whole results.
Some categories of readers (e.g., disaster risk managers, civil protection officers, decision makers) may find some unanswered questions about how these results may result in effective and viable tsunami mitigation measures to be eventually applied to the Cascadia Subduction Zone and beyond (lesson learned).
The literature review appears to be a little fragmented and might be arranged in a different way to better spot and address critical points.
Analysis is based on a significant number of variables that may (or perhaps should) be aggregated into factors or theoretically relevant indicators: reviewer suggests using multidimensional scaling / factor analysis or other statistical procedures to provide a concise and more effective synthesis of data to be placed at the end of the discussion paragraph.
Author Response
Comment 1: The reviewer appreciates the overall quality of the paper, the precision in results reporting and clear-cut indication of study limitations. However, the reviewer would suggest some little changes to improve clarity and usability of such research. From a methodological point of view the article appears to be flawless, nonetheless it can be improved by providing some general directions to make an overall sense of fine-grained information provided by authors. In general terms, the paper appears intended to provide several evidence on a number of different, yet very narrow areas of investigation, making a little uneasy to get a comprehensive and conclusive image of the whole results.
Comment 1 Response: We agree with the reviewer that this paper addresses a narrow area of investigation. Indeed, the paper’s specific objective is defined in the first two sentences of the Abstract—“The U.S. Pacific Northwest coast must be prepared to evacuate immediately after a Cascadia Subduction Zone earthquake. This requires coastal residents to understand the tsunami threat, have accurate expectations about warning sources, engage in preimpact evacuation preparedness actions, and plan (and practice) their evacuation logistics—including an appropriate transportation mode, evacuation route, and destination.” Thus, we conducted a survey to assess coastal residents’ evacuation preparedness.
We also agree with the reviewer that there are many other aspects of people’s response to tsunami hazard that should be investigated. Indeed, our survey of tsunami zone residents did investigate a broader range of topics but the complete report is 62 pages long. Given journal page limitations, this article only has enough space to adequately examine one topic—preparedness for evacuation from a local tsunami. If the reviewer is interested in the larger report from which this paper was derived, we can provide a copy.
Comment 2: Some categories of readers (e.g., disaster risk managers, civil protection officers, decision makers) may find some unanswered questions about how these results may result in effective and viable tsunami mitigation measures to be eventually applied to the Cascadia Subduction Zone and beyond (lesson learned).
Comment 2 Response: The original submission’s Conclusions section has three paragraphs, each of which provided a specific recommendation for emergency managers to address when working with their communities to improve preparedness for evacuation from a local tsunami. In response to Reviewer 1’s comment, we added another paragraph with a specific recommendation for emergency managers. Nonetheless, we agree that we have not addressed all conceivable recommendations for improving household evacuation preparedness, but we believe that we have identified the four most important recommendations that are direct implications of the data from this survey.
Comment 3: The literature review appears to be a little fragmented and might be arranged in a different way to better spot and address critical points.
Comment 3 Response: The literature review is organized according to our understanding of the issues relevant to preparedness for evacuation from a local tsunami. Since the reviewer did not identify any missing topics or provide any specific guidance on how to revise the literature review, we have made no changes to it.
Comment 4: Analysis is based on a significant number of variables that may (or perhaps should) be aggregated into factors or theoretically relevant indicators: reviewer suggests using multidimensional scaling / factor analysis or other statistical procedures to provide a concise and more effective synthesis of data to be placed at the end of the discussion paragraph.
Comment 4 Response: We would like to begin our response to this comment by noting that multidimensional scaling (MDS) and factor analysis are two substantially different analytic techniques. MDS would be completely inappropriate for these data because it is used to analyze respondents’ judgments of similarity among pairs of stimuli (see Nunnally & Bernstein, 1994, Chapter 14). None of our data are compatible with the MDS model.
By contrast, our data are suitable for factor analysis, which is commonly used to determine which items have high enough intercorrelations that they can reasonably be combined to form a multi-item scale. However, researchers shouldn’t factor analyze data just because they can. Table 1 shows that Variables 16-19 are quite highly correlated, which might be interpreted as a justification for combining them into a four item scale. That would be appropriate in some cases, but not in this one because Table 1 also shows that people received the two brochures with approximately equal frequency and participated in the two types of meetings with approximately equal frequency, but received brochures much more frequently than they participated in meetings. In some studies, the strong intercorrelations among items would be more important than the differences in item means. In this study, the differences in means are more important than the strong intercorrelations. Consequently, Table 1 continues to report the data at the level of individual items.
Reference
Nunnally, J.C. & Bernstein, I.H. (1994). Psychometric Theory, 3rd ed. New York: McGraw-Hill.
