# 2.8: Tsunami Risk and Hazard Mitigation - Geosciences

So far we've talked about the physics of tsunamis and you have investigated tsunami data and made some calculations about tsunami speeds. What's the current state of the art in terms of tsunami risk and hazard mitigation? Where are some other sites in the Atlantic Ocean where tsunamigenic potential lurks?

Readings relevant to the paper assignment

The following readings are either freely available online if they are linked from this page, or if they are not in the public domain then they are linked from Canvas. You should at least skim these articles because they deal with potential Atlantic Ocean tsunamigenic sites.

• Nealon, J. W., & Dillon, W. P. (2001, April). Earthquakes and Tsunamis in Puerto Rico and the U.S. Virgin Islands. U.S. Geological Society. Retrieved June 2, 2008, from http://pubs.usgs.gov/fs/fs141-00/fs141-00.pdf.
• Teeuw, R., Rust, D., Solana, C., & Dewdney, C. (2009). Large Coastal Landslides and Tsunami Hazard in the Caribbean. Eos, 90 (10), 81-82.
• Driscoll, N. W., Weissel, J. K., & Goff, J. A. (2000). Potential for large-scale submarine slope failure and tsunami generation along the U.S. mid-Atlantic Coast. Geology, 28(5), 407-410.
• Gisler, G., Weaver, R., & Gittings, M. (2006). Sage calculations of the tsunami threat from La Palma. Science of Tsunami Hazards, 24(4), 288-301.
• Pérez-Torrado, F. J., Paris, R., Cabrera, M. C., Schneider, J., Wassmer, P., Carracedo, J., et al. Tsunami deposits related to flank collapse in oceanic volcanoes: The Agaete Valley evidence, Gran Canaria, Canary Islands. Marine Geology, 227(1-2), 135-149.
• Wendell, J.(2015). Mysterious boulders suggest ancient 800-foot-tall tsunami.Eos, 96,doi:10.1029/2015EO036845. Published on 2 October 2015.

The following readings will help you become conversant with the way tsunami warning systems work, and what has been done since the 2004 Sumatra-Andaman tsunami:

• Pacific Tsunami Warning Center History
National Weather Service, Pacific Tsunami Warning Center.
Accessed: 2010-09-13.
• Pacific Tsunami Warning Center Responsibilities
National Weather Service, Pacific Tsunami Warning Center.
Accessed: 2010-09-13.
• Tsunami-Forecasting System Tested by Recent Subduction-Zone Earthquakes
Geist, Eric, Titov, Vasily, Kelly, Annabel, and Gibbons, Helen. USGS, Sound Waves.
Accessed: 2008-03-12.
• Gower, J., & Gonzalez, F. U.S. warning system detected the Sumatra tsunami. Eos, 87(10), 105, 108.
• Tracking Tsunamis (2009) AGU video
• Schiermeier, Q. Tsunami Watch. Nature, 462, 968-969.

## Science Application for Risk Reduction

ForewordThe 1906 Great San Francisco earthquake (magnitude 7.8) and the 1989 Loma Prieta earthquake (magnitude 6.9) each motivated residents of the San Francisco Bay region to build countermeasures to earthquakes into the fabric of the region. Since Loma Prieta, bay-region communities, governments, and utilities have invested tens of billions of.

Detweiler, Shane T. Wein, Anne M.

### The HayWired earthquake scenario—Earthquake hazards

The HayWired scenario is a hypothetical earthquake sequence that is being used to better understand hazards for the San Francisco Bay region during and after an earthquake of magnitude 7 on the Hayward Fault. The 2014 Working Group on California Earthquake Probabilities calculated that there is a 33-percent likelihood of a large (magnitude 6.7 or.

Detweiler, Shane T. Wein, Anne M.

Introduction Natural scientists, like many other experts, face challenges when communicating to people outside their fields of expertise. This is especially true when they try to communicate to those whose background, knowledge, and experience are far distant from that field of expertise. At a recent workshop, experts in risk communication offered.

Perry, Suzanne C. Blanpied, Michael L. Burkett, Erin R. Campbell, Nnenia M. Carlson, Anders Cox, Dale A. Driedger, Carolyn L. Eisenman, David P. Fox-Glassman, Katherine T. Hoffman, Sherry Hoffman, Susanna M. Jaiswal, Kishor S. Jones, Lucile M. Luco, Nicolas Marx, Sabine M. McGowan, Sean M. Mileti, Dennis S. Moschetti, Morgan P. Ozman, David Pastor, Elizabeth Petersen, Mark D. Porter, Keith A. Ramsey, David W. Ritchie, Liesel A. Fitzpatrick, Jessica K. Rukstales, Kenneth S. Sellnow, Timothy L. Vaughon, Wendy L. Wald, David J. Wald, Lisa A. Wein, Anne Zarcadoolas, Christina

### Anticipating environmental and environmental-health implications of extreme storms: ARkStorm scenario

The ARkStorm Scenario predicts that a prolonged winter storm event across California would cause extreme precipitation, flooding, winds, physical damages, and economic impacts. This study uses a literature review and geographic information system-based analysis of national and state databases to infer how and where ARkStorm could cause.

Plumlee, Geoffrey S. Alpers, Charles N. Morman, Suzette A. San Juan, Carma A.

### Agricultural damages and losses from ARkStorm scenario flooding in California

Scientists designed the ARkStorm scenario to challenge the preparedness of California communities for widespread flooding with a historical precedence and increased likelihood under climate change. California is an important provider of vegetables, fruits, nuts, and other agricultural products to the nation. This study analyzes the agricultural.

Wein, Anne Mitchell, David Peters, Jeff Rowden, John Tran, Johnny Corsi, Alessandra Dinitz, Laura B.

### Regional analysis of social characteristics for evacuation resource planning: ARkStorm scenario

Local planning is insufficient for regional catastrophes regional exercises are needed to test emergency plans and decision-making structures. The ARkStorm scenario would trigger a mass evacuation that would be complicated by the social characteristics of populations [e.g., vehicle ownership, age, poverty, English language limitation (ELL), and.

Wein, Anne Ratliff, Jamie L. Allan Baez Sleeter, Rachel

### SAFRR tsunami scenario: Impacts on California ecosystems, species, marine natural resources, and fisheries

We evaluate the effects of the SAFRR Tsunami Scenario on California’s ecosystems, species, natural resources, and fisheries. We discuss mitigation and preparedness approaches that can be useful in Tsunami planning. The chapter provides an introduction to the role of ecosystems and natural resources in tsunami events (Section 1). A separate section.

Ross, Stephanie L. Jones, Lucile Brosnan, Deborah Wein, Anne Wilson, Rick

### The search for geologic evidence of distant-source tsunamis using new field data in California: Chapter C in The SAFRR (Science Application for Risk Reduction) Tsunami Scenario

A statewide assessment for geological evidence of tsunamis, primarily from distant-source events, found tsunami deposits at several locations, though evidence was absent at most locations evaluated. Several historical distant-source tsunamis, including the 1946 Aleutian, 1960 Chile, and 1964 Alaska events, caused inundation along portions of the.

Wilson, Rick Hemphill-Haley, Eileen Jaffe, Bruce Richmond, Bruce Peters, Robert Graehl, Nick Kelsey, Harvey Leeper, Robert Watt, Steve McGann, Mary Hoirup, Don F. Chagué-Goff, Catherine Goff, James Caldwell, Dylan Loofbourrow, Casey

### [email protected]: Stakeholder perspectives on vulnerabilities and preparedness for an extreme storm event in the greater Lake Tahoe, Reno, and Carson City region

Atmospheric rivers (ARs) are strongly linked to extreme winter precipitation events in the Western U.S., accounting for 80 percent of extreme floods in the Sierra Nevada and surrounding lowlands. In 2010, the U.S. Geological Survey developed the ARkStorm extreme storm scenario for California to quantify risks from extreme winter storms and to.

Albano, Christine M. Cox, Dale A. Dettinger, Michael D. Shaller, Kevin Welborn, Toby L. McCarthy, Maureen

### Population vulnerability and evacuation challenges in California for the SAFRR tsunami scenario: Chapter I in The SAFRR (Science Application for Risk Reduction) Tsunami Scenario

The SAFRR tsunami scenario models the impacts of a hypothetical yet plausible tsunami associated with a magnitude 9.1 megathrust earthquake east of the Alaska Peninsula. This report summarizes community variations in population vulnerability and potential evacuation challenges to the tsunami. The most significant public-health concern for.

Wood, Nathan Ratliff, Jamie Peters, Jeff Shoaf, Kimberley

### Public-policy issues associated with the SAFRR Tsunami Scenario: Chapter M in The SAFRR (Science Application for Risk Reduction) Tsunami Scenario

The SAFRR (Science Application for Risk Reduction) tsunami scenario simulates a tsunami generated by a hypothetical magnitude 9.1 earthquake that occurs offshore of the Alaska Peninsula (Kirby and others, 2013). In addition to the work performed by the authors on public-policy issues associated with the SAFRR tsunami scenario, this section of the.

## New insights of tsunami hazard from the 2011 Tohoku-oki event

We report initial results from our recent field survey documenting the inundation and resultant deposits of the 2011 Tohoku-oki tsunami from Sendai Plain, Japan. The tsunami inundated up to 4.5 km inland but the > 0.5 cm-thick sand deposit extended only 2.8 km (62% of the inundation distance). The deposit however continued as a mud layer to the inundation limit. The mud deposit contained high concentrations of water-leachable chloride and we conclude that geochemical markers and microfossil data may prove to be useful in identifying the maximum inundation limit of paleotsunamis that could extend well beyond any preserved sand layer. Our newly acquired data on the 2011 event suggest that previous estimates of paleotsunamis (e.g. 869 AD Jōgan earthquake and tsunami) in this area have probably been underestimated. If the 2011 and 869 AD events are indeed comparable, the risk from these natural hazards in Japan is much greater than previously recognized.

### Highlights

► The first results of a geological survey following the 2011 Tohoku-oki tsunami. ► The tsunami inundated and left sediment deposits up to 4.5 km inland. ► 62% of the inundation distance was covered with sand. ► The 869 AD Jōgan earthquake and tsunami have probably been underestimated.

## Risk for an Atlantic Tsunami?

First, I would like you to read an article that appeared in The Boston Globe following the Sumatra-Andaman earthquake. Then, please read a short scientific paper detailing the tsunami threat from a source in the Canary Islands.

Daley, B. (2007, December 28). N.E. is not immune, scientists warn. The Boston Globe. Retrieved April 22, 2008, from http://www.boston.com/news/world/articles/2004/12/28/ne_is_not_immune_scientists_warn/.

Ward, S., & Day, S. (2001). Cumbre Vieja Volcano -- Potential collapse and tsunami at La Palma, Canary Islands. Geophysical Research Letters, 28, 397-400. (See also a press release from The Independent that accompanies the scientific article.)

The first article nicely introduces the topic for this lesson, which is whether or not you think a dedicated tsunami warning system should be developed for the Atlantic Ocean. The main reason I want you to read this article is that it brings up some topics that we will want to delve into further as this lesson goes along.

When you read the first article, keep in mind the following:

• Has the Atlantic coast ever experienced a damaging tsunami?
• If so, what was the source (e.g., earthquake, volcano, landslide)?
• Do scientists have any estimates for the future risk of an Atlantic Ocean tsunami?
• Can you compare this to the risk of other natural disasters?
• The Pacific Ocean does have a tsunami warning system. What's the difference between the Atlantic and Pacific Oceans?
• How are tsunamis produced in the first place?
• How do tsunami warning systems work?

The second article is a scientific paper published in a journal. When you read this paper, keep in mind the following:

• Do you understand all the terminology? Keep a list of terms you don't know so we can discuss them.
• Who is the intended audience for this paper, compared to the first article?
• Can you give a summary of the scenario and tsunami that would result from a volcanic flank collapse such as the one detailed in this paper?

These articles do not answer all these questions, but these are the questions that I would ask if I read them without knowing anything else. What other questions do you have after reading these articles? Please post any questions to the Questions? discussion board in Canvas.

Geoscience Australia has released a new Probabilistic Tsunami Hazard Assessment - the PTHA18. The PTHA18 is freely accessible and is now available to download. The PTHA18 consists of three products:

The Probabilistic Tsunami Hazard Assessment (PTHA) models the frequency with which tsunamis of any given size occur around the entire Australian coast, due to subduction earthquakes in the Indian and Pacific Oceans. The PTHA also provides modelled tsunami data for hundreds of thousands of earthquake-tsunami scenarios around Australia.

Public talks on the PTHA18 from the Geoscience Australia DGAL Series and the HPC-AI Advisory Council 2020 Australia Conference can be viewed online. You can also read journal publications on different aspects of the PTHA18 here and here. A short conference paper giving an overview of the study is available here.

The PTHA provides vital information to emergency managers to plan and reduce the threat of tsunami on the Australian coast, and for the insurance industry to understand the tsunami risk as an input to pricing insurance premiums.

This information provides a nationally consistent basis for understanding tsunami inundation hazards in Australia. It is important to note that the PTHA does not define the onshore tsunami impact, or the effect of tsunamis on communities. However, understanding the frequency of tsunamis offshore from the PTHA is a key input for developing local tsunami inundation models, in conjunction with additional high-resolution bathymetry and elevation data, to derive evidence-based evacuation plans to improve community safety. High risk areas can be identified and prioritised for further analysis or to conduct scenarios to improve risk mitigation and community safety at a local, regional and national level.

Geoscience Australia provides essential evidence based information to government and emergency managers around Australia to improve our communities' ability to prepare for, mitigate against and respond to natural disasters. Contact us at [email protected] if you need further information.

The last PTHA was completed in 2008. The PTHA18 has been significantly updated to include advances in our understanding of earthquakes and the resulting tsunamis and to provide hazard information for all Australian offshore territories.

Compared with the previous iterations of the PTHA, PTHA18 includes more comprehensive treatment of the natural variability of earthquake size and slip. This has an important impact on the predicted tsunami wave heights and hazard.

The new methodologies have been tested using 10 years of deep ocean tsunami observational data from the Pacific Ocean to confirm they give a realistic depiction of tsunami behaviour. This crucial deep-ocean-observational dataset was unavailable for the previous assessments, as most of the tsunamis in our test-set had not even occurred in 2008.

The new PTHA methodologies also reflect advances in our understanding of how frequently large earthquakes occur, and the uncertainties in these frequencies. We also provide outputs at many more sites, which make it easier for other modellers to use these results in local scale hazard studies, including major Australian offshore islands and territories.

Given the changes in the method and the available data, there are changes to the wave heights estimated for a given frequency (or frequency for a given wave height) since the last assessment in 2008. We also provide much more information on the uncertainties in these frequencies.

Currently the PTHA does not include non-earthquake sources that can cause a tsunami such as landslides, volcanic activity or meteorological events. Methods for assessing tsunami hazards for these sources are much less well established than for earthquakes both internationally as well as in Australia. Further research is required to underpin a nationally consistent treatment these tsunami sources.

The PTHA18 data is freely accessible and is available to use. Download now.

National scale annual exceedance maps derived from over 1 million possible scenarios (the 2008 assessment provided

• Hazard curves (stage vs exceedance probability)
• Hazard deaggregation for scenarios and exceedance rates
• Time series for each possible scenario at that site
• Initial conditions of the earthquake source for each possible scenario at that site

Geoscience Australia will plan a future update the PTHA18, as we recognise the importance of incorporating best practice and evidence based science. Science and technology is constantly evolving and improving, and we need to ensure the PTHA reflects these advancements so we can ensure Australian communities are as safe as possible from tsunami events.

## Hazards

Natural hazards such as earthquakes, landslides, hurricanes, floods, and wildfires endanger public health and safety, threaten critical infrastructure, and cost our economy billions of dollars each year. Geoscientists study these hazards to provide information and warnings to populations at risk.

Do you have a question that's not listed here? Search all FAQs

### Explore Related Topics

Since 1980 the United States has experienced more than 24 major droughts, resulting in almost 3,000 deaths and economic impacts exceeding $225 billion. All areas of the U.S. have some drought risk. Since 1900, earthquakes in the United States have resulted in over 1300 deaths and direct damages totaling more than$51 billion. While the West Coast and Alaska have the highest risk, history shows that major earthquakes can also affect the Central and Eastern United States.

Flooding is the most common and costliest natural hazard facing the United States. Each year, flooding causes billions of dollars in damages and dozens of deaths nationwide.

Landslides affect all 50 states and U.S. territories, where they cause 25 to 50 deaths and more than $1 billion in damages each year. Geoscientists study and monitor landslides to identify at-risk areas, prepare populations, and improve our understanding of why, when, and where landslides happen. Sinkholes have both natural and artificial causes. They tend to occur most often in places where water can dissolve the bedrock (especially limestone) below the surface, causing overlying rocks to collapse. Florida, Texas, Alabama, Missouri, Kentucky, Tennessee, and Pennsylvania are most sinkhole-prone. Tsunamis are destructive waves caused by sudden displacement of ocean water. Tsunamis most often appear on shore as a rapidly receding tide or rapidly rising flood. In the United States, the Pacific coastal states – Oregon, Washington, California, Alaska, and Hawaii – are at greatest risk for destructive tsunamis. Volcanoes pose many hazards to their surroundings, from ashfall, mudflows, lava flows, landslides, and associated earthquakes. At least 54 of the United States' 169 active volcanoes pose major threats to public health and safety and to major industries such as agriculture, aviation, and transportation. Although our industrial society produces a variety of solid wastes and waste waters, over the past 50 years we have made progress in disposing of them safely in landfills, by incineration, and in underground injection wells. Many wastes are also increasingly recycled or reused. Weather hazards impact the entire country, with enormous effects on the economy and public safety. Since 1980, weather/climate disasters have cost the U.S. economy more than$1.5 trillion. In an average year, the United States will be affected by six billion-dollar weather/climate disasters, but this number has increased in recent years: from 2013-2017 the average was 11.6 events.

Wildfires are causing more frequent and wider-ranging societal impacts, especially as residential communities continue to expand into wildland areas. Since 2000, there have been twelve wildfires in the United States that have each caused damages exceeding $1 billion cumulatively, these twelve wildfires have caused a total of$44 billion in damages.

## 2.8: Tsunami Risk and Hazard Mitigation - Geosciences

The Alaska Earthquake Information Center (AEIC) conducts tsunami inundation mapping for coastal communities in Alaska. This activity provides local emergency officials with tsunami hazard assessment and mitigation tools. At-risk communities are spread along several segments of the Alaska-Aleutian Subduction Zone, with each segment having a unique seismic history and potential tsunami hazard. As a result, almost every community has a distinct set of potential tsunami sources that need to be considered in order to make a tsunami inundation map. Therefore, an important component of the inundation mapping effort is identification and specification of potential tsunami sources. We are creating tsunami inundation maps for Sitka, Alaska, in the scope of the National Tsunami Hazard Mitigation Program. Tsunami potential from tectonic and submarine landslide sources must be evaluated in this case for comprehensive mapping of areas at risk for inundation. The community of Sitka, the former capital of Russian Alaska, is located in Southeast Alaska, on the west coast of Baranof Island, facing the Pacific Ocean. In this area of southern Alaska, the subduction of the Pacific plate beneath the North America plate becomes a transform boundary that continues down the coast as the Fairweather - Queen Charlotte (FW-QC) transform fault system. The Sitka segment of the FW-QC fault system ruptured in large strike-slip earthquakes in 1927 (Ms7.1) and in 1972 (Ms7.6). We numerically model the extent of inundation in Sitka due to tsunami waves generated from earthquake and landslide sources. Tsunami scenarios include a repeat of the tsunami triggered by the 1964 Great Alaska earthquake, repeat of the tsunami triggered by the 2011 Tohoku earthquake, tsunami waves generated by a hypothetically extended 1964 rupture, a hypothetical Cascadia megathrust earthquake, and hypothetical earthquakes in the FW-QC fault system. Underwater landslide events off the continental shelf along the FW-QC fault zone are also considered as credible tsunamigenic scenarios. We perform simulations for each of the scenarios using AEIC's numerical model of tsunami propagation and runup, which was validated through a set of analytical benchmarks and tested against laboratory and field data. Results of numerical modeling combined with historical observations in the region will be delivered to local emergency management to be used in local tsunami hazard assessment, evacuation planning and public education.

## Conclusions and Directions

In this review, we discuss a large number of research gaps in PTHA and PTRA. It becomes obvious that methods have substantially improved over the past decades, but also that open questions remain in the physical description, conceptualization, modeling, as well as the social and psychological dimensions of the topic.

The physics and geological complexity of tsunamigenic sources are still not captured nor understood adequately, leading to large uncertainties. For SPTHA, neither all earthquake faults nor their exact location, geometry, boundary and initial conditions (e.g., stress, friction) are fully constrained. Statistical models of recurrence constitute the largest uncertainties in large and rare events, including tsunami earthquakes. Uncertainty may become excessive for landslide tsunamis, where statistics on past events often are absent, and our understanding of slope failure probability is limited. The need for covering vast geographical scales, source diversity and related uncertainties, render LPTHA extremely challenging. For VPTHA additional difficulties arise due to the complexity of tsunamigenic volcano sources and triggers, but they are constrained spatially. MPTHA may benefit from a large meteorological data network allowing for (prototypical) forecasting as well as PTHA applications, but sensitivity to source parameters is still unconstrained.

While modeling and parameterization of individual phenomena are possible, they are often excessively computationally expensive or highly uncertain due to missing constraints on input parameters. The multiple scales involved in PTHA from far-field propagation over oceanic distances to the need to resolve small scale inundation features while capturing physics and resolving uncertainties still represent an open challenge. Yet, this solution is needed to convey PTHA information properly into risk analysis.

Even more challenging is the situation in PTRA, where gaps exist in the transformation of physical hazard to risk and quantifying the uncertainties in the assessment of risk and resilience. Key concepts, such as physical vulnerability and mortality and their related uncertainties, are less developed than the main PTHA elements. There are gaps regarding selection of IM, limited observed damage asset- and location-wise, and limited experimental validation.

Furthermore, tsunami science is immature concerning embedding issues with intrinsically multi-hazard and multi-risk aspects, such as the cascading events that are entangled with tsunami hazards. A weakly developed link between quantitative PTRA and the social sciences is a clear gap. At this point, it is worth noting that terms “vulnerability” and “resilience” are multi-dimensional concepts that are used both in the consequence-based–natural sciences inspired𠄺s well as context-based approaches–motivated by social sciences. Therefore, they may have quite different interpretations depending on the analysis context.

The overarching issue is integrating all the above components and developing an overall consistent sensitivity and uncertainty quantification framework, to understand tsunami risk and identify risk drivers, from the probability of the sources causing hazards to the probability of their physical consequences and societal impact. This understanding must be developed and prioritized in future research.

To guide such efforts, we have performed an expert judgment exercise that we discuss in the following subsection. It may help to identify most pressing research needs as well as prioritize research efforts.

### Prioritizing Research Gaps

A scientific sensitivity analysis of the impact of each research gap, as conducted for individual sources in Sepúlveda et al. (2017) or Davies and Griffin (2020), on the overall result of a PTHA or PTRA is certainly out of the scope of a single review paper. However, some guidance on prioritization of efforts is certainly desirable. Since we focused our description on research gaps, we suggest two important metrics for the prioritization: The susceptibility of PTHA and PTRA results on uncertainty due to the research gap (sensitivity) and the difficulty or amount of research effort needed to fill that respective gap (tractability).

In order to assess these two metrics, we conducted a first-pass expert judgment among the more than 50 co-authors of this article𠄺ll experts in one or more of the aspects of our review. A questionnaire was designed that asked three questions for each of the 47 research gap subsections that we have described before. The first two questions addressed the two metrics just mentioned. The third question asked if experts were of the opinion if the research gap existed because of a missing theoretical understanding, a lack of data, or both. While this somewhat ad hoc prioritization is not as solid as a rigorous expert elicitation (e.g., Cooke, 1991 Budnitz et al., 1997 Morgan, 2014 for tsunami hazard see an application in Basili et al., 2021, or the discussion in Grezio et al., 2017) and hence could be somehow biased, we believe it still provides a valuable starting point for future efforts. It is a qualitative broad-brush answer to the question, which research gap may be of highest importance. More details on this exercise are given in the Supplementary Material.

The result of this exercise is visualized in a priority matrix (Figure 2). It may appear natural to respond first to those research gaps that are located in the left upper quadrant of the matrix, since these gaps are considered less difficult to solve, while they are expected to influence the risk considerably. It can be noted that most of the research gaps are judged hard to solve but with a highly sensitive impact on the overall result. This seems natural, since high impact but simple problems would have been solved already.

FIGURE 2. Priority Matrix for all the 47 research gaps identified. Letters indicate seismic source gaps (S), landslide source gaps (L), volcanic source gaps (V), meteorological source gaps (M), hydrodynamical modeling gaps (H), exposure related gaps (E), physical vulnerability related gaps (P), resilience related gaps (R), social vulnerability and risk indicators related gaps (I). The size of each marker relates to the agreement of experts, larger marker size means less spread in the answers. Colors are used to indicate if the gap is caused by missing theoretical understanding (blue), a lack of data (red), or both (cyan).

Based on our qualitative assessment, we can therefore identify some overall trends. First, we see some common challenges related to establishing annualized source probability of occurrence, which tend to cluster in the upper right corner of Figure 2. This means that they are considered relatively most important, yet hardest to solve. Of these, obtaining landslide related annual source probabilities (L1) is considered both the largest yet most important obstacle, while a just slightly lower similar prioritization is evident for earthquake and volcano sources (S1 and V2). Another aspect that is considered important (and challenging) is the multi-hazard and cascading hazard aspect (R5). On the other hand, the research gaps that appear to be least sensitive and also easy to be filled are related to the numerical modeling of wave propagation (H3), as well as lack of joint intensity measures (I3) and gaps related to earthquake scaling relations (S4). Finally, we also note Figure 2 allows us to analyze several instances of components with similar sensitivity but with clearly different tractability. For instance, the lack of tsunami exposure data (E2) is considered as important as modeling complicated aspects of inundation (H6), but the former is assumed by the authors of this paper to be more easily achieved. Several other similar examples can be analyzed from Figure 2.

It is noteworthy that most of the research gaps that most experts find consensus on are highly sensitive in their impact (all located at the upper margin of the point cloud). It is also worth noting that most research gaps are considered to relate to data and theory gaps and that those that relate to only a missing theoretical understanding are considered of relatively low sensitivity. This may be related to the fact that when we don’t understand a phenomenon, we cannot really judge whether it affects our results or not. In other words, this may be an “unknown”. Whereas a data related research gap may already have proved to be sensitively influential by a specific example, but due to a lack of data cannot be involved concisely into the workflow.

This priority matrix is just a very first approach. Since tsunami research eventually aims at protecting life from natural hazard, one could also prioritize those research gaps with direct impact on this goal. These would be in particular those topics mentioned in sections “Gaps in Physical Vulnerability,” “Gaps in Risk and Resilience Metrics,” and “Gaps in Social Vulnerability, Multi-Dimensional Vulnerability and Risk Indicators” (marked with P, R, and I respectively).

### Final Considerations

We have described and prioritized a comprehensive list of research gaps in PTHA and PTRA. While our approach to prioritization and the metric used to do so are to some extent subjective, it remains for the scientific community and further investigation as well as future incentives to decide, which directions to choose from. Nevertheless, our priority matrix will serve as a first impression on the weight of each of the identified research gaps.

An important part of the future puzzle will be exploring how uncertainties propagate to risk across disciplines. While uncertainties are more extensively explored in earthquake-related hazard analysis, non-seismic hazard, vulnerability, exposure and risk are lagging behind. On the other hand, different levels of maturity of methods and understanding will always exist. Hence, it is imperative to develop PTRA standards and guidelines to appropriately merge all risk analysis components considering their different uncertainty exploration and maturity level.

While validation of individual components has been addressed in several of the sections in our text, validating the PTHA and PTRA workflow as a whole is still ongoing research. Marzocchi and Jordan (2014) propose a methodology for a meaningful test of general probabilistic hazard models and an example of a successful application can be found in Meletti et al. (2021).

Certainly, research gaps exist also outside of the scope of PTHA and PTRA. New computational methods, like fuzzy methods, machine learning techniques and even advances in classical computational methods have to be considered. Rigorous, information theory inspired approaches to validation may also be explored.

Considering the goals of the Sendai Framework for Disaster Risk Reduction and acknowledging the vast number of challenges outlined in the sections before, a concerted interdisciplinary effort to close the most pressing gaps is required. Attempts to gather expertize, facilitate exchange and development, and coordinate community efforts are represented by the Global Tsunami Model (GTM, 2020) and the COST Action AGITHAR. A thorough consolidation of available sources of information in openly accessible databases, documentation of standard workflows, unification of terminology and metrics, as well as information hubs need to be established.

## 18 - Hazard assessment for risk analysis and risk management

The focus in this chapter is on the client – what is it that hazard and risk managers want from geomorphologists and what do geomorphologists believe that their science can constructively offer hazard and risk management? However, communicating skills and requirements can be difficult because scientists and practitioners come from different backgrounds and work within different constraints. On the one hand, the geomorphologist primarily needs to satisfy the research community, while the manager, on the other hand, has to deal with their client base and the public in general, often within a strict statutory, regulatory, policy and financial framework. Clearly, the basic information demands of hazard assessment, of where (location), what (type of event), when (how often) are fundamental to reducing risk but the manager might also legitimately ask ‘which areas are free from hazard?’, ‘what type of mitigation might be appropriate?’, ‘what sort of monitoring should be undertaken?’, ‘what changes can we expect in the future?’ and ‘what is the cost effectiveness of different management options?’.

In post-event situations, geomorphologists may also be required for forensic investigation. In many cases this will be to establish the cause, apportion weight to the causative factors, and to determine the relative importance of human versus natural factors in creating both cause and consequences.

By understanding the geomorphic system, not only in space but also through time, the geomorphologist should be capable of predicting or at least indicating the hazardous characteristics of processes and places within the system, at a range of spatiotemporal scales.

## Tsunamis and Tsunami Hazards

You don't hear about tsunamis very often, but when they do strike, they can be huge newsmakers and can have drastic and devastating effects. The occurrence and potential for tsunamis on the coasts of the United States is not out of the question. Read on to learn about tsunamis.

### Tsunami Facts

• Tsunamis are triggered by earthquakes, volcanic eruptions, submarine landslides, and by onshore landslides in which large volumes of debris fall into the water. All of these triggers can occur in the United States.
• If a tsunami-causing disturbance occurs close to the coastline, a resulting tsunami can reach coastal communities within minutes.
• Although many people think of a tsunami as a single, breaking wave, it typically consists of multiple waves that rush ashore like a fast-rising tide with powerful currents. Tsunamis can travel much farther inland than normal waves.

### A Real Risk for the United States

The west coast of the U.S. has experienced tsunami impacts in the past

In December 2004, when a tsunami killed more than 200,000 people in 11 countries around the Indian Ocean, the United States was reminded of its own tsunami risks.

In fact, devastating tsunamis have struck North America before and are sure to strike again.

Especially vulnerable are the five Pacific States — Hawaii, Alaska, Washington, Oregon, and California — and the U.S. Caribbean islands.

In the wake of the Indian Ocean disaster, the United States is redoubling its efforts to assess the Nation's tsunami hazards, provide tsunami education, and improve its system for tsunami warning.

The U.S. Geological Survey (USGS) is helping to meet these needs, in partnership with the National Oceanic and Atmospheric Administration (NOAA) and with coastal States and counties.

This map shows seven earthquake-generated tsunami events in the United States from the years 900 to 1964. The earthquakes that caused these tsunamis are: Prince William Sound, Alaska, 1964, magnitude 9.2 Chile, 1960, magnitude 9.5 Alaska, 1946, magnitude 7.3 Puerto Rico/Mona Rift, 1918, magnitude 7.3 to 7.5 Virgin Islands, 1867, magnitude undetermined Cascadia, 1700, magnitude 9 and Puget Sound, 900, magnitude 7.5. Map not to scale. Sources: National Geophysical Data Center, NOAA, USGS

### Tsunami Impacts

• The 2004 Indian Ocean tsunami reached heights of 65 to 100 feet in Sumatra, caused more than 200,000 deaths from Indonesia to East Africa, and registered on tide gauges throughout the world.
• The 1964 Alaska tsunami led to 110 deaths, some as far away as Crescent City, Calif.
• In 1918, an earthquake and tsunami killed 118 people in Puerto Rico. Several such events have struck this region in historic times
• A tsunami that originated along the Washington, Oregon, and California coasts in 1700 overran Native American fishing camps and caused damage in Japan.

### Life of a Tsunami

1. INITIATION

Earthquakes are commonly associated with ground shaking that is a result of elastic waves traveling through the solid earth.

Note: In this figure, the waves are greatly exaggerated compared to water depth. In the open ocean, the waves are at most several meters high spread over many tens to hundreds of kilometers in length.

However, near the source of submarine earthquakes, the seafloor is "permanently" uplifted and down-dropped, pushing the entire water column up and down. The potential energy that results from pushing water above mean sea level is then transferred to horizontal propagation of the tsunami wave (kinetic energy). For the case shown above, the earthquake rupture occurred at the base of the continental slope in relatively deep water. Situations can also arise where the earthquake rupture occurs beneath the continental shelf in much shallower water.

Within several minutes of the earthquake, the initial tsunami (Panel 1) is split into a tsunami that travels out to the deep ocean (distant tsunami) and another tsunami that travels towards the nearby coast (local tsunami). The height above mean sea level of the two oppositely traveling tsunamis is approximately half that of the original tsunami (Panel 1). (This is somewhat modified in three dimensions, but the same idea holds.) The speed at which both tsunamis travel varies as the square root of the water depth. Therefore, the deep-ocean tsunami travels faster than the local tsunami near shore.

3. AMPLIFICATION

Several things happen as the local tsunami travels over the continental slope. Most obvious is that the amplitude increases. In addition, the wavelength decreases. This results in steepening of the leading wave — an important control of wave runup at the coast (next panel). Note that the first part of the wave reaching the local shore is a trough, which will appear as the sea recedes far from shore. This is a common natural warning sign for tsunamis. Note also that the deep ocean tsunami has traveled much farther than the local tsunami because of the higher propagation speed. As the deep ocean tsunami approaches a distant shore, amplification and shortening of the wave will occur, just as with the local tsunami shown.

Tsunami runup occurs when a peak in the tsunami wave travels from the near-shore region onto shore. Runup is a measurement of the height of the water onshore observed above a reference sea level.

Except for the largest tsunamis, such as the 2004 Indian Ocean event, most tsunamis do not result in giant breaking waves (like normal surf waves at the beach that curl over as they approach shore). Rather, they come in much like very strong and fast-moving tides (i.e., strong surges and rapid changes in sea level). Much of the damage inflicted by tsunamis is caused by strong currents and floating debris. The small number of tsunamis that do break often form vertical walls of turbulent water called bores. Tsunamis will often travel much farther inland than normal waves.

Do tsunamis stop once on land? No! After runup, part of the tsunami energy is reflected back to the open ocean and scattered by sharp variations in the coastline. In addition, a tsunami can generate a particular type of coastal trapped wave called edge waves that travel back-and forth, parallel to shore. These effects result in many arrivals of the tsunami at a particular point on the coast rather than a single wave as suggested by Panel 3. Because of the complicated behavior of tsunami waves near the coast, the first runup of a tsunami is often not the largest, emphasizing the importance of not returning to a beach many hours after a tsunami first hits.