Encyclopedia of Complexity and Systems Science

Living Edition
| Editors: Robert A. Meyers

Volcanic Hazards Warnings: Effective Communications of

  • C. J. FearnleyEmail author
Living reference work entry
DOI: https://doi.org/10.1007/978-3-642-27737-5_634-1

Glossary

CAS

Complex Adaptive System

EWS

Early Warning System

IDNDR

International Decade for Natural Disaster Reduction

ISDR

International Strategy for Disaster Reduction

NIMS

National Incident Management System

NOTAM

Notice to Airmen

NVEWS

National Volcano Early Warning System

USGS

US Geological Survey

VAA

Volcanic Ash Advisory

VAAC

Volcanic Ash Advisory Center

VALS

Volcano Alert Level System

VAN

Volcanic Activity Notice

VEWS

Volcano Early Warning System

VHP

Volcano Hazard Program

VONA

Volcano Observatory Notices for Aviation

Introduction: Challenges to Volcanic Crisis Communication

There have been numerous volcanic crises and disasters over the last 100 years, despite significant advances in scientific knowledge, monitoring technologies, and the development of statistical and GIS-based models. Often these disasters have resulted from a breakdown in communication that has often been the result of no, or poor, early warning systems in place, or the failure of society to respond effectively. One such tragic example is the 1985 eruption of Nevado del Ruiz in Colombia which, despite having only a volcanic explosivity index (VEI) of 3, killed over 23,000 people in the city of Armero, when a fatal lahar that was forecast and expected by scientists traveled over 70 km to strike the city (Hall 1990). There was no formal early warning system in place; however, hazard maps had been drawn by the Colombian Geological and Mine Institute (INGEOMINAS) and United States Geological Survey (USGS), with scientists identifying numerous prior lahars that had traveled through Armero. The resulting high death toll was because local authorities and communities did not act on warnings, as Voight eloquently states “the catastrophe was not caused by technological ineffectiveness or defectiveness, nor by an overwhelming eruption, or by an improbable run of bad luck, but rather by cumulative human error – by misjudgment, indecision and bureaucratic shortsightedness. Armero could have produced no victims, and therein dwells its immense tragedy” (1990, p. 349). This tragedy, among many others, highlights the important role of effective communications in reducing risks posed by volcanic hazards.

The Diversity of Volcanic Hazards

Volcanic hazards are complex because of the diversity in the types of hazards that a volcano can produce, both primary and secondary, and the differing geographical locations that these hazards impact, and the differing time scales (for an extensive overview see Sigurdsson et al. 2015). For example, volcanic ash generally only occurs during an explosive eruption, and can affect a large area downwind of the volcano as the ash travels through the atmosphere and is deposited. In contrast, lava flows only impact areas within relatively close proximity of an erupting volcano, but these flows can occur over decades during a prolonged eruption. A nonerupting volcano can still generate lahars (particularly if there is a storm, heavy rainfall, or glacial melt), landslides (due to unstable slopes), and subsequently tsunamis (displacement of water in or near to the volcano). Volcanoes produce the most diverse range of hazards of any phenomenon, which can lead to extensive secondary hazards such as crop failure, famine, disease, contamination, and climate change that can result in more fatalities and socioeconomic impact than the primary hazards of a volcano (Oppenheimer 2011). It is this diversity and complexity of the hazard that creates problems when trying to communicate potential hazards and risks. In addition, volcanoes are complex phenomenon, a nondeterministic emergent system, making then exceedingly challenging to predict, but there is however scope to forecast potential activity (Kilburn 2003; Sparks and Aspinall 2004). Forecasting and prediction are often considered synonymous; however, it is recommended to adopt the following definitions. A forecast is a comparatively imprecise statement of the time, place, and nature of expected activity. Prediction is a comparatively precise statement of the time, place, and ideally, the nature and size of impending activity (Swanson et al. 1985). With complex phenomenon that is not easily forecast, a wide range of generated hazards, there is also the additional risk that society generates by building and living near or even distal to volcanoes.

The communication and negotiation of potential hazards and their impact on society is typically classified as Disaster Risk Reduction measures (Kelman and Kelman 2017; Wisner et al. 2012). There are numerous preparedness and mitigation strategies that can be put into place. Mitigation strategies are policies or procedures that lead to preplanned actions that operate before or during a hazard event to reduce its impact on vulnerable populations. Common examples include land-use and development planning; engineering strategies such as tsunami barriers, river or tidal flood defenses, and seismically resilient buildings; and warning systems that foster education, evacuation plans, and communication to enable mitigation actions at the time of hazard events or in anticipation of them. Different mitigation strategies require differing timescales and methods of implementation according to both the nature of the hazards and the vulnerabilities of the exposed societies. Mitigation systems consist of all the mitigation strategies implemented together to ensure “the lessening or limitation of the adverse impacts of hazards and related disasters” (UNISDR 2007). Day and Fearnley (2015) divide mitigation strategies into three classes according to the timing of the actions that they prescribe. Permanent mitigation strategies prescribe actions such as construction of tsunami barriers or land-use restrictions: they are frequently both costly and “brittle” in that the actions work up to a design limit of hazard intensity or magnitude and then fail. Responsive mitigation strategies prescribe actions after a hazard source event has occurred, such as evacuations, that rely on capacities to detect and quantify hazard events and to transmit warnings fast enough to enable at risk populations to decide and act effectively. Anticipatory mitigation strategies prescribe use of the interpretation of precursors to hazard source events as a basis for precautionary actions, but challenges arise from uncertainties in hazard behavior. Day and Fearnley’s classification provides an insight into the challenges of the temporal dynamics of natural hazards that constrain the ability to provide warnings and key parameters surrounding potential events and hazards. It is these timing constraints that shape the actions prescribed by a strategy, and the adaptability of these actions to individual hazard events, both of which shape the types of warnings and communication that are possible for volcanic hazards.

Volcano Mitigation and Communication Strategies

For volcanoes, most mitigation strategies are anticipatory. Even with permanent structures in place such as SABO dams (commonly found in Japan), should an eruption be imminent, populations are still likely to be evacuated. Anticipatory mitigation strategies prescribe use of the interpretation of precursors to hazard source events as a basis for precautionary actions, but challenges arise from uncertainties in hazard behavior. Societies also struggle to manage precautionary approaches as the cost of evacuations is high, many people do not want to leave their home or their businesses or farms as the cost is great, particularly for extended times (Tobin and Whiteford 2002; Stirling 2007). Therefore, either greater certainty is needed or greater support from the government to rehouse and support populations for the time as required. To try and negotiate these differing needs given the inherent uncertainties, Early Warning Systems (EWS) are usually implemented as a cost-beneficial tool to manage the precautionary and anticipatory issues that emerge from volcanic crises.

EWS are a key component of mitigation strategies, whereby communication and decision-making based upon information provided by monitoring and warning technologies leads to actions in response to that information. Despite an abundance of EWS-related research, there is little consensus about what they are or how they are defined (Glantz 2004). EWS are seen as “a means of getting information about an impending emergency, communicating that information to those that need it, and facilitating good decisions and timely response by people in danger” (Mileti and Sorenson 1990, pp. 2–1). While this is a simple definition, the operation of an EWS is far more complex, partly due to variations spatially (global, national, regional, local), temporally (rapid onset, slow onset, frequent, infrequent), in function (safety, property, environment), and in hazard (weather, climate, geo-hazard). EWS also operate in different economic, political, and social circumstances; use different communicative tools (from technology to word of mouth); and link many different organizations (or actors) such as science (government and private), engineering, technology, government, news/media, and the public. This leads to different perspectives of what EWS are, and what they should do. For governments, EWS are an important tool for disaster-risk reduction (DRR) measures; consequently, EWS tend to be highly centralized. Decisions have to be made about the benefits of EWS relating to: cost-benefit, timeliness (what constitutes a warning; are they a forecast, projection, or trend; and how early is early), establishing different levels of warning, and lastly accountability.

As part of the warning process, there are a number of key tools used by volcano observatories and institutions globally that communicate during times of escalated unrest. These include various written statements, alert level systems, notification systems, and the use of social media. These are explored further in this chapter, but it is important to note that there are a diverse and wide range of communication tools that can be used during volcanic crises. These range from (and can all be found in Fearnley et al. 2017b): understanding historical texts and knowledge to help prepare for future events; developing community knowledge and DRR practices that embrace oral and indigenous knowledges; stakeholder engagement and communication that recognizes cultural, disciplinary, and economic divides; decision-making tools; the use of statistic (particularly Bayesian) models to forecast volcanic unrest and activity; the role of insurance to mitigate against losses; the use of hazard and risk maps; the use of satellite imagery and geo-spatial technologies inform real-time analysis, communication, and decision-making; the role of art and performance in communicating volcanic hazards and risks, alongside folklore; the role of education for all ages in various teaching environments; and the role of social media. While these approaches are all vital to modern day volcanic crisis communication, only some aspects of these approaches contribute to the development of volcano hazard warnings. This chapter subsequently aims to examine the role of early warning systems, various types of volcano hazard warnings, the challenges of standardization focusing on a case study of the United States Geological Survey’s Volcano Alert Level Systems that were standardized in 2006, and the value of complexity approaches when considering what is effective communication for volcano hazard warnings.

Early Warning Systems

Early warning systems (EWS) have existed for a number of hazards, but the most significant and oldest system is the Pacific Tsunami Warning System, established in 1949 following the 1946 Aleutian Island earthquake that generated a tsunami that killed 165 people in both Alaska and Hawaii. Despite growing populations near volcanoes, little research has been devoted to establishing best practice for volcano EWS to minimize loss of life and socioeconomic damage prior to and during volcanic crises.

Generating effective warnings is particularly challenging for several key reasons. First, volcanologists and related scientists are still developing theories to understand the origin, processes, and eruptive behavior of volcanoes and their numerous associated hazards. Second, volcanic hazards occur within different social contexts involving different cultures, and economic and political circumstances. In addition, volcanic activity tends to occur over long time frames relative to human time-scales and, in particular, periods of political office, and therefore are not normally a political priority. This commonly results in limited funding and resources for research and volcano observatory upkeep, leading to limited volcanic hazard awareness. Finally, institutional influences can lead to increasing levels of bureaucracy so that decisions become complex and take a long time to make. Managing a volcanic crisis can involve numerous institutions from emergency managers/civil protection, to weather services, to land managers and owners, to the media, to businesses and key infrastructure, to the public, making it difficult to maintain effective communications, both internally and externally.

To devise effective EWS, there is a need to consider what these systems are and how they have evolved over time, what they do, how they operate, and how their success can be measured. These aspects are addressed below.

An Evolution from the Linear to the Complex

The diversity in EWS and of the agencies involved results in different ways of conceptualizing EWS. This section explores theoretical approaches to EWS by reviewing the concept and its evolution from a linear to a complex system.

EWS form a relatively new area of inquiry within the context of disaster research, obtaining growing recognition in 1960s. Prior to this, studies typically viewed early warning as a linear process (Gillespie and Perry 1976), where there is a clear relationship between a hazard occurring and generating a warning, forming a cause and effect relationship. Linear processes are characteristically embodied in the Newtonian paradigm, often referred to as reductionist, but in 1969 a study by Barton changed this to view early warnings as a “system.” Barton’s work (1969) on disaster classification generated a paradigm shift from the descriptive to the analytical, by developing four classifying variables in his typology of disasters: scope of impact, speed of onset, duration of impact, and social preparedness. This work influenced Gillespie and Perry who said that “by adopting a systems perspective, the disaster researcher can not only describe and classify disasters more effectively, but can also move towards a more analytic approach” (Gillespie and Perry 1976, p. 305). A systemic approach enabled the development of models for the prediction of individual, group, and organizational behaviors, going beyond the simplistic cause and effect relationships within an early warning.

Although the idea of “systems” influence on disasters had first been identified in 1958 by Form and Nostow, it took decades to take hold. General Systems Theory (GST) emerged following the Second World War as an interdisciplinary approach to the field of science and the study of the complex systems in nature and society (Bertalanffy 1975). The term “systems” has many definitions, although the one adopted here is of a group of interacting, interrelated, or interdependent elements forming a complex whole, which is nearly always defined with respect to a specific purpose (Kim 1994). Originating in biological studies in the 1920s, GST recognized that systems are greater than the sum of their parts (Bertalanffy and Woodger 1933), providing a holistic approach to disaster studies by demonstrating that the processes involved are interrelated. In 1975, models of idealized EWS were developed such as in Fig. 1, which rather than showing an EWS as a linear progression through the different stages of disasters in chronological order, indicated that an EWS comprised of subsystems (in this case evaluation-dissemination and response) that have inputs, outputs, and feedback between them.
Fig. 1

Systems model of a warning system (White and Haas 1975, p. 185)

By the 1980s, Foster (1980) identified that decision-making and communication processes between different actors in EWS were nonlinear and could be understood better within the context of systems theory as a dynamic system. Foster also developed an idealized EWS to represent the different stages, using a system style layout as seen in Fig. 2. Although this model recognized the role of organizations and policy, it maintains an element of linearity rather than presenting a series of feedback loops that are multidirectional enabling a systems approach, as Foster states “every warning system should be designed to facilitate a two-way flow of information” (Foster 1980, p. 203) (author’s emphasis).
Fig. 2

Idealized warning systems (Foster 1980, p. 172)

The EWS models developed by White and Haas (1975) and Foster (1980) (in Figs. 1 and 2 respectively) divide EWS into component parts and consider each part separately to ensure their proper function (White 1995). While these models are idealized and are not descriptive of what actually happens in an EWS, the models also struggle to view EWS as a system because they fail to “identify emergent properties arising from interacting elements and because it does not consider that the behaviour of systems is due as much to their external environment as to their internal mechanisms” (White 1995, p. 41). White argues that disaster studies tools that provide a holistic approach, by considering how human behavior and context can affect the management of risk, should be used.

By the late 1990s, there was growing recognition that interactions between natural environments, human perception, actions, and organizations are part of a genuinely “complex” system (Mileti 1999). The term “complex” has become a popular and often misused term both in the physical and social sciences. The complex systemic approach “focuses on interaction among the elements of a system and on the effects of its interactions; it examines a variety of factors at one time; it integrates time, feedback, and uncertainty” (Mileti 1999, p. 107). It is the reciprocal interactions or feedback among variables or subsystems, as well as time delays in seeing the results, that create complexity, making the system difficult to understand (Senge 1990). As a result, complexity highlights serious limitations to our scientific knowledge because it breaks traditional reductionist Newtonian thinking that regards science as infinitely divisible and measurable (Capra 1996); “complexity argues against reductionism, against reducing the whole to the parts” (Urry 2005a, p. 401). The concept of complexity and chaos has questioned the naivety that science depends on patterns by establishing a link between determinism and predictability (Nowotny et al. 2001; Sardar and Ravetz 1994). Therefore, it is important that analyses of complex systems are not left solely to scientists, since these systems are transdisciplinary, involving human agents, science, and society (Nowotny 2005).

Mileti (1999) was not alone in recognizing that systems are firmly entrenched in thinking and research on hazards and disasters. According to Gillespie et al. (2004), knowing how to mitigate the negative consequences of natural disasters and respond effectively requires three steps: “understanding the physical and social systems involved in disasters, communicating that understanding clearly to decision-makers, and knowing what interventions may be effective” (Gillespie et al. 2004, p. 82). Complex systems theory provides a holistic approach to integrate these three steps and understand how complex interactions generate certain behavior, although it is difficult to monitor these complex interactions. Theoretically, the framing of EWS has evolved through systems thinking throughout the last 50 years with growing recognition of the social systems involved in an EWS.

Early Warning Systems within Disaster Management

Individuals, who developed theories on how disasters and EWS operate, as outlined above, were not alone in recognizing that social systems have a significant role in disaster management. In the last century, human geographers have influenced disaster management thinking by challenging top-down expert-driven approaches, by instead suggesting bottom-up locally integrated ones. In the 1930s and 1940s, the “dominant approach” was widely accepted (Wisner 2004), stating that factors such as “material wealth, experience of hazardous events, systems of belief, and psychological considerations are all important in controlling how individuals, social groups, and indeed, whole societies respond to disasters” (Chester et al. 2005, p. 416). This approach implied there are adjustments that individuals and societies can make to deal with natural hazards. In the 1980s, Kenneth Hewitt (1983) discussed the inherent complexities in natural-disaster planning in “Interpretations of Calamity” disputing the dominant approach. This new approach adopted the view that most disasters in developing countries are the result of poverty and deprivation rather than extreme natural hazard events, and so those economically or geographically marginalized suffer the most (Susman et al. 1983). Radical alternatives changed the way natural hazards are studied by scientists, social scientists, and policy-makers to emphasize the uniqueness of the location and the socioeconomic and cultural conditions of the society. To be successful, adjustments to hazards should be sensitive to the local environment and be intercultural. The common acceptance of this alternative way of thinking did not occur until the World Conference on Natural Disaster Reduction in 1994, where the published Yokohama Strategy reviewed, in part, how we could transform society to reduce disasters using radical alternatives (UN ISDR 2004).

By the 1990s, EWS became an area of focused research within disaster-management studies. EWS are difficult to understand because they encompass the physical hazard and the context of the “society” affected. Mileti and Sorenson (1990) provide one of the first detailed reviews of EWS from a social science perspective. Based on 200 studies in the USA, they established three key findings. First, variation in the nature and content of warnings has a large impact on whether the public responds. Second, the characteristics of the population receiving the warning affect the response (i.e., gender, ethnicity, and age, and other social, psychological, and knowledge characteristics). Third, many current myths about public response to emergency warnings are at odds with field investigation results, for example, “cry-wolf” syndrome, public panic, and hysteria. These results indicate there is a difference between “ideal” models and those in practice. Drawing on case studies, the authors outline guidance for what information warning messages should contain: the hazard, location, guidance, time, and sources. For many hazards, including volcanoes, this is extremely challenging to achieve since hazards have different levels of predictability, detectability, certainty, lead time, duration of impact, and visibility as scientific capabilities remain limited, making it difficult to generate “specificity, consistency, accuracy, certainty and clarity” (Mileti and Sorenson 1990, pp. 3–11) in warnings. Mileti and Sorenson (1990) state it is not possible to review EWS in a comprehensive manner by just isolating the social and physical elements, because there is a need to establish organizational effectiveness, work with other organizations, and maintain flexibility during warnings. The report presents a model of EWS (see Fig. 3) with a detection component (monitoring and detection, data assessment and analysis, prediction and informing), emergency management component (interpretation, decision to warn, method and content of warning, and monitoring of response), and response component (interpretation and response). This builds on the White and Haas (1975) model (Fig. 1) by emphasizing the different subsystems and their relationships and institutional roles, rather than only focusing on the relationship between the hazard, warning, and response.
Fig. 3

The general components of an integrated warning system (Mileti and Sorenson 1990, pp. 2–4)

Every aspect of an EWS involves a decision, from interpreting monitoring information, to issuing a warning, to evacuating a town, to the vulnerable individual deciding what to do. Decision-making is still considered under a systemic approach, using a logical sequence between the definitions of the problem, the risk assessment, and its solution, rather than considering the complexities involved (UNDRO 1990). A wide range of institutions have to make decisions about appropriate actions, and the conventional view is these move along a linear chain as shown in Fig. 4, taking a top-down approach. Some countries, such as the USA, adopt this top-down approach to decision-making wherein populations turn to their local civil authorities for information and advice to make informed decisions. Countries that adopt a bottom-up approach place greater responsibility on the individual or community.
Fig. 4

The decision-makers within EWS adopting a top-down approach (Fearnley 2011, p. 55)

In 2004, the Humanitarian Practice Network developed a model of EWS (see Fig. 5) (Twigg 2004) that shows EWS as a more complex system, with feedback loops and variables, but also identifies the need for risk assessment, understanding vulnerability, and public education. Unlike the linear models shown in Figs. 2, 3, and 4, this model illustrates that decision-making is a core component of EWS and is not a linear process, but the result of feedback from different actors involved in the EWS.
Fig. 5

Generic model of forecasting/warning systems developed by Schlosser, C (Twigg 2004, p. 301)

Following the Indian Ocean tsunami of 2004, Hurricane Katrina in the USA in 2005, and the Tohoku tsunami in 2011, recent publications highlight that EWS are becoming an increasingly topical and important area within disaster risk reduction methods (Glantz 2009; Hall 2007; IFRC 2009). Despite the importance of EWS, research about their application and effectiveness is fragmented, unconsolidated, and patchy. More holistic perspectives of EWS are emerging, although still not widely implemented, viewing them as a system that attempts to interact with a number of complex systems (such as the physical phenomenon, hazard, and society) to provide sufficient warnings for appropriate action to take place. Historically EWS research focused on two key areas: forecasting techniques for natural hazards within the scientific community and exploring strategies to disseminate warnings effectively and credibly to vulnerable populations, often referred to as the “last mile” within disaster studies. Studies on the last mile relate to large literatures on risk perception (Gaillard and Dibben 2008; Slovic 2000), vulnerability (Bankoff et al. 2004; Birkmann 2006; Wisner 2004), resilience (Bankoff 2007; Kelman and Mather 2008), and capacity and communication (Tierney and Dynes 1994). There is a need to also consider the “first mile,” an often overlooked but key component in EWS in an increasingly globalized yet patchily standardized world. This “first mile” relates to the design and operation of EWS and raises questions about how effective they are in communicating warnings and information to all the users of the system. These user groups are growing in diversity as trade and travel becomes increasingly international. Understanding the “first mile” requires investigation into how scientists understand volcanoes, how they manage the associated uncertainties and risks, and how they attempt to manage them both theoretically and practically.

Historically high levels of uncertainty in natural hazard science have resulted in scientists becoming core stakeholders in EWS, due to their expertise and responsibilities, so that EWS became “hazard-focused, linear, top-down, expert/driven systems, with little or no engagement of end-users or their representatives” (Basher 2006, p. 2712). From this, mistrust of expert and local authorities can develop based on criticism that implementing an EWS is a long-term process where local populations can sustain themselves and thus benefit for generations to come. Twigg (2004, p. 306) highlights that:

The bulk of effort and expense is put into transmitting detailed clearly presented information to decision-makers and government emergency management services. Far less effort and funding go into disseminating this information right down to individual communities or households through accessible messages that will warn them and help them to make sensible decisions about how to respond.

To date, there has been little evaluation of the influence of institutional organization and the flow of information between different actors in an EWS on making decisions. Typically, government institutions that manage potential disasters use simple policy, often prescriptive in manner; however, with the recognition that decision-making is more complex, local practitioners and vulnerable populations are increasingly managing disasters relevant to them using community-based EWS. These EWS are based upon local capabilities and technologies where communities can have ownership, generating an EWS that adopts a bottom-up approach. The idea of community-based EWS has gained momentum (Maskrey 2011), in line with the radical approach developed by Hewitt (1983), and is suggested as an approach to develop people-centric EWS by the UN ISDR PPEW (2006).

Institutional Approaches to EWS

In recent decades, global institutions that provide guidelines and best practices for EWS have increasingly recognized the role of EWS in disaster management, largely the United Nations (UN). The UN General Assembly designated the 1990s as the International Decade for Natural Disaster Reduction (IDNDR), which in 2000 the International Strategy for Disaster Reduction (ISDR) replaced. Throughout the 1990s and 2000s, the UN held a number of EWS conferences resulting in a number of publications (Kuppers and Zschau 2002; UN ISDR 2006a, b). In 2005, the UN established the Hyogo Framework, a global blueprint for disaster-risk reduction (DRR) efforts during the next decade with the goal to substantially reducing disaster losses by 2015. One of its five key priorities for actions is to “identify, assess and monitor disaster risks and enhance early warning” (UN ISDR 2005, p. 6), highlighting growing awareness of the role EWS has within institutional governance.

Following the catastrophic Indian Ocean tsunami of 2004, the Secretary-General of the United Nations called for the development of a global EWS for all natural hazards and communities. It was felt that if an EWS were in place when the tsunami struck the Indian Ocean region, many thousands of lives could have been saved (230,000 are estimated to have been killed in 11 countries (Thieren 2005)). In March 2005, the UN ISDR Platform for the Promotion of Early Warning (PPEW) undertook a global survey to identify existing capacities and gaps in EWS, intended as a wake-up call for governments and other agencies about the value of EWSs in reducing human and economic loss from natural hazards. Published in 2006 the “Global Survey of Early Warning Systems” was the culmination of this research and advocated that EWSs should be “people-centered” (i.e., community based) and encompass spanning four key elements: risk knowledge, monitoring and warning service, dissemination and communication, and response capability (see Fig. 6) (UN ISDR PPEW 2006).
Fig. 6

The elements of a people-centered early warning system (UN ISDR PPEW 2006, p. 2)

According to the UN, an EWS “can only be effective if the element and the linkages are well-understood, well-designed and well-operated” (Basher 2006, p. 2176). Yet, the model presented in Fig. 6 does not indicate what these linkages are. The survey concludes that the world is far from having the global system for all hazards and communities called for by the UN Secretary-General, but it does make five key recommendations that illustrate the difficulties and contradictions involved in achieving this goal (UN ISDR PPEW 2006, p. vi):
  1. 1.

    Develop a globally comprehensive EWS, rooted in existing EWS and capacities

     
  2. 2.

    Build national people-centered EWS (i.e., community based)

     
  3. 3.

    Strengthen the scientific and data foundation for early warnings

     
  4. 4.

    Fill the main gaps in global early warning capacities

     
  5. 5.

    Develop the institutional foundations for a global EWS

     

First, the recommendations raise questions about the viability of uniformity, with different hazards, countries, varying levels of scientific capabilities and communication technologies available, and different local decision-making structures, and institutions involved. Second, it appears contradictory to develop a system that can be globally comprehensive, yet built by the local community. Third, the role of “scientific and data foundation” in preventing EWS failure is questionable, given that this chapter has already reviewed the case study of the Nevado del Ruiz tragedy that shows that frequently it is not scientific or technological deficiencies that cause failure, but social and institutional elements. Fourth, identifying gaps may be difficult given that what may be a gap in resources and capabilities for one country may not pose a problem in another, due to differing social and institutional contexts such as available funding. Last, there may be issues with developing institutional foundations for a global EWS when the requirements for emergency response vary in different nations.

Hall (2007) has commented that despite the efforts by the UN events focused on EWSs, there still lacks “coordinated, collaborative international action” (p. 32) to make the move from debate to tangible results. Additionally, Hall outlines that the emphasis within EWS has consequently been more to do with funding of current capabilities and development in science and technology, which has “distracted us from the central issue of address the real needs of the communities and people at risk” (Hall 2007, p. 32). Some scientists agree, suggesting that they must step outside their “ivory tower” and try to anticipate the consequences of developing warning tools and to make sure they will actually lead to hazard reduction (Malone 2008).

To improve EWS, the UN has called for more effective procedures via standardization and the application of new technologies and enhanced scientific understanding (UN ISDR PPEW 2006). Such a strategy poses two potential weaknesses. First, standardization, by definition, tends to exclude the importance of incorporating local factors into a global procedure (discussed below in section “General Information Statements”). Second, the focus on science and technology implicitly assumes that social and cultural variations are secondary factors, when the materials presented in this chapter clearly illustrate the importance of social context in making EWS effective. The UN has not developed approaches to EWS that consider the complexities involved; instead they focus on the need for developing global platforms and standardizing; as standardization is reductive, this counteracts systematic approaches to managing crisis. Despite this, standardized methods are frequently used to manage hazards or complex situations.

Garcia and Fearnley (2012) highlight that, while an EWS may have four key components as outlined by the UNISDR, it is often the links between these categories that are the focus of systemic failure. Analyzing several case studies conducted over the last 40 years, the authors discovered common emerging factors that improve links between the different components of EWS. They identified four key factors (see also Fig. 7): (1) establishing effective communication networks to integrate scientific research into practice; (2) developing effective decision-making processes that incorporate local contexts by defining accountability and responsibility; (3) acknowledging the importance of risk perception and trust for an effective reaction; and (4) consideration of the differences among technocratic and participatory approaches in EWS, when applied in diverse contexts. In the context of volcano warnings, these vital processes include (1) clear communication (Solana et al. 2008), (2) effective decision-making processes (Leonard et al. 2008), (3) trust building and participatory activities (Haynes et al. 2008a), and (4) defining accountability and responsibility so people know what to do clearly (Glantz 2004) as per the points above. All these factors show the importance of flexibility and the consideration of local context in making EWSs effective, whereas increasing levels of standardization within EWSs nationally and globally might challenge the ability to incorporate the required local expertise and circumstances.
Fig. 7

Diagram of Early Warning System (EWS) with factors to improve the linking of subsystems (Garcia and Fearnley 2012, p. 133)

Volcano Warnings

There are numerous volcano early warning systems in place globally that have been designed to cope with a wide range of hazards. In 2005, the United States Geological Survey (USGS) devised a ranking system called the US National Volcano Early Warning System (NVEWS) (Ewert et al. 2005) to enable recommendations for varying levels of monitoring (Moran et al. 2008). They ranked 169 volcanoes in the USA in a combined assessment of 15 hazard and 9 exposure factors to generate a threat score. Well-established volcano EWSs have evolved over the last centuries, bringing together different practices, technologies, and procedures; these are discussed extensively with supporting case studies in Fearnley et al. (2017a) and Fearnley and Beaven (2018). Each warning system adopts its own approach unique to the nature of the volcano and the social context in which it operates. Some are technically driven, others are community-based EWS adopting a bottom-up approach. However, most volcano EWS adopt varying types of procedures and protocols to assist in the management of a volcanic crisis. A volcano EWS can be seen of comprising of four key elements (adapted from Potter et al. 2017, p. 4):
  1. 1.

    Technical: This relates to the equipment type, deployment (distribution/location/density), telemetry (radio, wire, internet, etc.), visualization (software packages), and analysis; all receive some level of standardization through manufacturing standards, detection limits, and international scientific best practice.

     
  2. 2.

    Analytical tools: Analysis of monitoring data may be further structured through statistical approaches such as expert elicitation, Bayesian event trees, or Bayesian belief networks.

     
  3. 3.

    Warning tools: Notification may be standardized through message content (e.g., standard messages, terminology, alert level criteria), packaging (e.g., bulletins, alert levels, maps), and delivery channels (e.g., phone, internet, siren). Some standards lend interoperability, such as a Common Alerting Protocol.

     
  4. 4.

    Response: Decision-making and action by the end-user can be standardized to some extent through communication and education approaches and message content.

     

Typically, volcano warnings are issued using a diverse range of tools, or products, but of significant note is the integration of the volcano alert level system.

Volcano Alert Level Systems

A Volcano Alert Level System (VALS) is the part of a volcano EWS that relates to the processes occurring before and during the issuance of a volcano warning. The USGS defines a volcano alert level system as a “series of levels that correspond generally to increasing levels of volcanic activity” (Gardner and Guffanti 2006, p. 2). As a volcano becomes increasingly active toward eruption, a higher alert level is issued that offers the public and civil authorities a framework they can use to gauge and coordinate their response to a developing volcanic emergency. VALS are based on a linear design, where the alert level assigned is directly proportional to the volcanic activity (see Fig. 8 for an example). In addition, alert levels carry information from the observatory to those who use it in a univalent (one directional) manner. Globally, many VALS (also referred to as status levels, condition levels, or color codes) are used providing volcanic warnings and emergency information in relation to volcanic unrest and eruptive activity based on data analysis or forecasts (Potter et al. 2017).
Fig. 8

Aviation color code, now internationally adopted by the International Civil Aviation Organization (Gardner and Guffanti 2006, p. 3)

There is a growing body of knowledge that discusses and challenges the role of VALS including: a review of VALS and the role of communication during volcanic crisis (Fearnley and Beaven 2018), assigning an alert level (Fearnley 2013; Winson et al. 2014), standardization of VALS (Fearnley et al. 2012; Potter et al. 2014), and the use and value of alert levels (Papale 2017).

Other Volcano Warning Tools

Supplementing the VALS is a range of other warning tools that can be classified as the following:

Event-Driven (Urgent) Messages

These messages are designed specifically to fulfill users’ requirements, for example, by using a number of standardized templates such as products specifically aimed at the aviation sector, such as a “Volcano Observatory Notice for Aviation” (VONA), A Notice to Airmen (NOTAM), A Volcanic Ash Advisory (VAA), and for ground hazard focused users a “Volcano Alert Notification” (VAN). Providing information in specified formats, these are bespoke tools that enable fast and quick communication of the facts at the time of issuance to specific user groups.

Time-Driven (Scheduled) Status Messages

A majority of communication that occurs during a crisis is multivalent, involving communications with several people, usually formalized via a number of protocols such as telephone call-down lists, and meetings between the relevant actors usually as part of a coordination plan, media talking points, and personal communication between the decision-makers. These usually follow prearranged schedules and protocols.

General Information Statements

These are more traditional communication tools that typically consist of information statements, status reports, updates, and other longer text documents that aim to disseminate the scientific data, interpretation, forecasts, and in some cases guidance. Often, hazard maps and longer reports are also used to release useful information, alongside links to various online resources. For the scientists, information statements provide a greater level of flexibility in communicating information than just issuing an alert level, although they follow a univalent format of information. Scientists tailor these messages to be relevant and of interest to the local users; however, these messages are still limited to text so there is no opportunity for dialogue or for users to add context.

With increasing use of social media and online website and networks, warning information is traveling quicker and to a wider audience than ever, placing pressures on the need for information to be credible, accurate, and of relevance. Most volcano observatories now have their own Twitter or Facebook accounts that support their websites to quickly disseminate messages, alongside the use of Short Message Service (SMS) (Sennert et al. 2015).

While VALS are intended to be linear, in practice the process of issuing a volcanic warning is complex since there is multiplicity of legitimate perspectives, nonlinearity, self-organization, multiplicity of scales, and areas of continuing uncertainty. Therefore, the self-organizing and adaptability of the communication networks provide the flexibility to accommodate the user’s needs, requirements, and capabilities in making decisions, and communicating such information to their users. This is not achieved solely through the VALS, which acts more as a “heads up” about the current status of a volcano (although this itself depends on the design of the system that varies by country). During a crisis, the communication that occurs becomes a complex network, or system, that generates feedback loops and enables the communication between the different actors to adapt and evolve as per the requirements of each of the actors involved (see Fig. 9). For every crisis, this system will be different, even if it is the same volcano, because the actors involved and the circumstances are constantly changing. By having a flexible and adaptive communications network, it is possible to accommodate the needs of the diverse range of users and varying hazards over time.
Fig. 9

Communication tools employed between volcano observatories and key user groups during volcanic crises. (Adapted from Fearnley 2013, p. 1896)

The Challenges for Effective Volcano Warning

Warning effectiveness is not just a function of good hazard knowledge and the generation of a warning message, but needs to be complemented by accurate knowledge of risk and risk management actions (Leonard et al. 2008) (see Fig. 10). However, the effectiveness of an integrated response can be compromised by communication, coordination, training, and organizational constraints (Paton et al. 1998). Once a decision to warn has been made, communication of it in an understandable format to decision-makers and the public is fundamental. It is imperative that all warning communication must be one consistent message, with no contradiction to generate confusion, to help establish trust between the public and other users that the information is correct and useful (Mileti and Sorenson 1990). This creates a problem because often there is scientific controversy.
Fig. 10

Effective early warning systems model from Leonard et al. (2008, p. 204)

A number of volcanic crises have highlighted the importance of effective communications between different actors of a volcano EWS that are discussed extensively in Fearnley et al. 2017a but can be summarized as: (1) the need to communicate danger to the public and decision-makers as successfully illustrated by the 1991 eruption of Mt. Pinatubo in the Philippines and the use of the Krafts’ educational video (Tayag et al. 1996; Newhall and Solidum 2017); (2) communicating with stakeholders (e.g., civil agencies, land owners, the public) to prepare for volcanic crises as seen at Vesuvius volcano observatory, Italy (Solana et al. 2008); (3) miscommunication between scientists and the media as seen at Soufriere Hills at Guadeloupe during 1976 (Fiske 1984); (4) scaremongering as seen at Galeras volcano in Colombia when it reawakened in 1989 resulting in the public loss of confidence in the scientists (Cardona 1997; Velasco 2000); and (5) within a volcano observatory, culture can shape the ability to communicate and discuss contentious views at a time of crisis as seen at Long Valley Caldera in the USA (Hill 2002; Hill et al. 2017).

Local context is very important to the success of volcano EWS, in particular there are four key local contingencies. First, the political context, as seen in the 1902 eruption of Mt. Pelee that destroyed Saint-Pierre, Martinique, in part the result of politicians who, in the middle of an election “obliging them [the inhabitants] to stay in the city and vote,” effectively, resulting in the death of approximately 30,000 people (Scarth 2002, p. 43). Second, issues of expertise, trust, and credibility as seen in Montserrat (1995–ongoing to present). The decision to evacuate nearly two thirds of the island took much longer than expected, because the government remained uncertain as to the status of the volcano due to challenges over the “experts” view for scientific advice (Aspinall et al. 2003; Haynes 2008).The possibility of lawsuits and false alarms leading to a loss in credibility and potential inappropriate decisions in the future are consistent concerns (Denis 1995), as also seen during the L’Aquila trial (Alexander 2014; Bretton et al. 2015). Insights from volcano EWS case studies demonstrate there is no one formula for transmitting scientific knowledge, so that the credibility of experts is in a sense always being negotiated and evaluated; therefore, trust cannot be routinized (Wynne 1996). The relationship between scientific expertise and the public is therefore far more complex than typically recognized in calls for “public understanding” that emanate from the scientific establishment. A study in Montserrat discovered the most trusted source for volcanic information is “friends and relatives” (Haynes et al. 2008a, b), thus highlighting the need for volcanologists to negotiate acceptable levels of risk and trade-offs with the public. Third, the resources available to operate the volcano EWS such as adequate resources and scientists/staff maintain full communications with stakeholders, develop policy, and manage crises. However, this optimal response on an active volcano rarely coincides with an actual crisis (Peterson and Tilling 1993). There are too few observatories, many with limited staff, funding, and equipment for monitoring, resulting in poor communication with local civil officials; moreover, scientists sometimes are so engrossed in their work that they regard interactions with the press and public as annoyances and distractions. The final key local contingency is deterined by the type of volcanic activity, both past and future. Vulnerable populations vary around a volcano, from those that live on the volcano and nearby, to those that live 10’s of kilometers away in the river valleys formed by the volcano, to those that live 100’s of kilometers away that can be affected by ash. These vulnerable groups have different needs relating to the different hazards and their knowledge of them, and the ability to communicate warning information. Peterson and Tilling (1993, p. 340) identified key factors that lead to some of the complexities involved in operating a volcano EWS. For example, “small, frequent eruptions induce good communications and promote good relations between scientists and the public,” as “uncertainty about the outcome of volcanic unrest, especially if major violence is among the possibilities, seems to induce poor inter-relations,”. In addition, there is recognition that “the public often has unrealistic expectations of scientists’ forecasting ability” (p. 348). Therefore, the volcano’s eruptive style, activity, and hazards are integral to making a warning relevant to the affected community.

Many lessons have been learnt from volcano EWS including members of the volcanological community who have subsequently reviewed their professional conduct during volcano crises (Newhall et al. 1999). The few case studies addressed demonstrate the value that a more comprehensive understanding of decision-making, communication, and the relevance of local contexts can make toward designing more effective volcano EWS (Ronan et al. 2000). It is clear volcano EWS need to be flexible in their design to accommodate variation in both the physical hazard and the social context, both of which are locally dependent. This raises questions about the ability for a standardized volcano EWS to achieve its objective.

The Emergence and Challenge of Standardization

Over the last 40 years, volcanic crises have supported the argument that EWS are not linear, but have to negotiate numerous complex systems. This requires a bottom-up approach that considers local context (contingency) that needs to respond to changes over time, and is socially constructed and adapted by the relevant society’s requirements. This contradicts increasing levels of standardization in EWS that do not facilitate local flexibility or recognize the complexities involved, and how to best govern them. However, examples outside of disaster management have shown that although standardization is reductive, it helps establish responsibilities and cooperation between the different groups involved. This section draws on examples to review issues that standardization raises that may be relevant to standardizing EWS, addressing the key question as to how a warning can be standardized to consider local context and yet appeal to a diverse range of users?

Why Standardize?

Globally, the levels of standardization in protocols and procedures for disasters and emergency management have risen, including the development of an Indian Ocean Tsunami EWS following the 2004 Boxing Day tsunami. Within the USA, the 9/11 terrorist attacks led to significant changes in government policy resulting in the standardized National Incident Management System (NIMS), the Homeland Security Alert Level, and other alerting and warning protocols for electronic technological warning capabilities. Standardizing warnings is not a new concept, but as disaster practitioners learn more about the complexity of natural disasters, concerns are being raised that it is increasingly difficult to use “nonlinear” methods of communication and that “faced with the nature and complexity of challenges involved in societal responses to hurricanes [or other hazards], interdisciplinary work that, for example, integrates appropriate meteorological and social science research will be critical” (Gladwin et al. 2009, p. 4). In addition, there appears to be insufficient literature on the effectiveness of standardization as a tool to manage complex disaster-related issues; subsequently, there is minimal understanding of what benefits or limitations standardization can bring. Within disaster studies, guidelines and models for applying standards have been developed, such as consistency and quality control, for developing and using emergency plans (Alexander 2005). Alexander argued that while viewing standards as unnecessarily restrictive and overly prescriptive, they could also help guarantee the quality, content, and relevance of these plans. Given the lack of other data around the standardization of EWS, reviewing other standardized processes such as medical procedures or technological processes can demonstrate issues that standardization raises as a method of managing complexity.

While some regard standardization as a constraint, a number of features also make standardization attractive. First, it improves the “doability” of work to enable scientists to “constrain work practices and define, describe, and contain representations of nature and reality,” and enables a “dynamic interface to translate interest between social worlds” (Fujimura 1987, p. 205). Second, it permits simpler procedures for people to learn from and carry out. Third, in a number of spheres, particularly medical and ethical, it provides answers to concerns relating to the processes or procedures by the public (Hogle 1995). Medical practices regard standardization as necessary to control processes and make outcomes more effective and reproducible. Fourth, standardization provides political ordering and control. In summary, standardization offers a tool to communicate in compatible ways (via language or protocols), ensure minimum quality, and provide a reference point (David and Greenstein 1990).

Standardizing a process is difficult, predominantly because it fixes the process in an ever-changing and dynamic world. In addition, there is no guarantee that researchers or users in different locations will use them in the same way. Scientists tend to “tinker” with standard procedures, often making assumptions of the standard application, so although standardization can increase “doability” it does not guarantee reproducibility (Fujimura 1987). In fact, it can create problems that begin to work against the benefit of standardization, creating tension between the efforts to rationalize work, and changes in the local conditions, which affect the work (Fujimura 1987). Often local practice can render a process less standard, rather than more predictable and uniform (Hogle 1995). Since the cultural, organizational, and institutional relations that characterize a process change, it seems difficult to remove contingency and national variation; for example, medics acting within a standard process bring their own experience and technical contingencies that mean local cultural meanings and categories remain. It is difficult for standardized technologies to be flexible, unless black boxed like a computer, because once a standardized system is in place, it already has a number of users geographically and organizationally that are difficult to change (Hanseth et al. 1996).

Standardization requires establishing boundaries, often in complex scenarios, making it difficult to decide what to leave outside of standardization, and what to include. Most studies on standardization across different practices have demonstrated it is not possible to factor in uncertainty or ignorance when designing a standard, and that knowledge, practices, and technologies of the present shape the standardization. Clearly, these aspects are not static, but to reflect this within the tool of standardization is not easy. Standardizing an EWS that is diverse and pluralistic is particularly challenging.

A Case of Standardization at the United States Geological Survey (USGS) VALS

The emergence and implications of standardization for managing the scientific complexities and diverse agencies involved in volcano crises have been charted for the USGS, which manages five volcano observatories in Alaska (AVO), Cascades (CVO), Hawaii (HVO), Long Valley (LVO)(now California), and Yellowstone (YVO) (Fearnley 2011). With a wide diversity of volcanic types, many active volcanoes, excellent monitoring resources, and international experience via their Volcano Disaster Assistance Program (VDAP), the USGS provided an excellent case study to review the impact of standardizing their VALS. In 2006, the USGS replaced all previous, locally developed systems at four of its observatories with a common standard. Their VALS comprises of two systems, one for ground hazards, and the other for ash hazards (see Fig. 8), which this case study focuses on. The rationale for standardizing the VALS for ash hazards stemmed from demand for the aviation sector for a standard warning for ash throughout the USA to prevent confusion. Additionally, following the 9/11 terrorist attacks in 2001, increasing levels of standardization for warning procedures and protocols were enforced. There was also a desire to develop uniform warnings that would be more consistent across the USGS.

Research at all five observatories revealed numerous different interpretations in the meaning of an alert level. One such example recalled at both AVO and at HVO concerned a commercial Alaskan pilot flying from Alaska to Hawaii. The pilot, used to flying in Alaska and dealing with the aviation color code frequently in place there, was concerned that the Kilauea volcano on the island of Hawaii was assigned an Orange alert level. Based on his experience with volcanoes in Alaska, he anticipated that the volcano would be exhibiting unrest with increased potential for eruption with ash. When the pilot arrived in Hawaiian airspace, he expected some form of diversion or information (such as a Volcanic Ash Advisory) regarding Kilauea but received nothing and landed with no problems. He later discovered that Kilauea is erupting, but only emitting a small ash plume that prohibits low-level flying within close proximity of the volcano. What he expected was based on his experience with volcanoes assigned alert level Orange in Alaska. Although alert level terms are standardized throughout the USA, they mean different things to users, in different locations, demonstrating both flexibility and inconsistency in the meaning and interpretation of VALS by users. Often these interpretations build on an individual’s local experiences and interactions with a VALS. In addition, the meanings of alert levels change between government agencies. An Orange alert level does not affect the local Volcanic Ash Advisory Centers (VAAC) or the National Weather Service (NWS) in Hawaii as it does in Alaska.

Fearnley (2011) discovered there are tensions between using local VALS, and those that are standardized as summarized in Fig. 11, illustrating that there are benefits associated with both local and national systems. Using a local system provides greater flexibility to adapt to the local needs and integrate the VALS into the management processes of the crisis. However, local systems are becoming increasingly limited by nationally standardized disaster protocols such as the USA National Incident Management System (NIMS). Dependence on common terminology for each alert level may help streamline communications but equally can be misleading as a standardized VALS cannot provide specific information that a locally developed VALS can. Limitations in the ability to provide diversity and pluralism suggest that there may not be enough flexibility in the design. The principle of “one size fits all” does not apply to VALS; they need to adapt to reflect changes in volcanic behavior and their impact on people, and this is better done when they are viewed from a holistic perspective to incorporate all the variables involved, many of which will be unknown prior to the crisis.
Fig. 11

This graphic compares pros and cons of local (left) and standardized (right) VALS (Fearnley 2011 p. 247)

The case study concludes that it is difficult for a VALS to be standardized, yet maintain the benefits of a local system and, in addition, to be understood by users both local and global (e.g., aviators). This creates a problem as the more flexible a system becomes, the less standardized it is. Although consistency is frequently identified as a key element of standardization, in practice it does not seem to work. Currently, the USGS standardized VALS works around limitations in flexibility through the many communication and information products and networks developed between the scientists and the users as seen in Fig. 9 (Fearnley and Beaven 2018).

VALS attempt to manage the many complex systems within a volcanic crisis and standardization provides a “one-size-fits-all” approach that is reductive in nature and potentially unable to accommodate local flexibility required to effectively prevent loss of life and minimize economic impact. Instead, standardization appears to be a tool that helps simplify the organizational elements of VALS for policy-makers and large-scale decision-makers, such as government institutions. There is a growing literature reviewing the impacts of the standardization of volcano EWS or VALS from either a national or a local perspective (Potter et al. 2014), although it remains unknown as to how standardization can compromise the need to effectively manage complexity. This is an important issue that warrants further research, and if left answered could generate a number of problems that could lead to further disasters, rather than reducing them.

Future Directions

There is a growing recognition that volcano EWS interface with complex systems and have to negotiate many issues, in addition to globalization, pluralization, and an erosion of expertise. Linear models of EWS are unable to represent the relationships and feedback within these complex systems because of their constraints in design. If society wants to be prepared for volcano crises then it needs a “truly complex-systemic approach to both the practice and method of science” (Gallopin et al. 2001, p. 223). There is a need to understand the connectedness, relationships, and contexts of volcanic warnings and their dynamics in order to investigate “how the different components and processes interact functionally to generate system responses and emergent properties, how the system adapts and transforms itself” (Gallopin et al. 2001, p. 223). A complex system can be defined as “a system in which large networks of components with no central control and simple rules of operation give rise to complex collective behaviour, sophisticated information processing and adaptation via learning or evolution” (Mitchell 2009, p. 13). In addition, complex systems “exhibit nontrivial emergent and self-organising behaviours” (Mitchell 2009, p. 13).

Complexity theory has been applied to core topics of sociology, spurring a diverse range of studies (Nowotny 2005; Urry 2005b), but this application has not been without contention (Hemaspaandra and Ogihara 2001). Pragmatists suggest that complexity provides a lens that helps us to look at our world and shape our actions, but it should not be seen as the only way to look and do things. Many, however, still have concerns with the reliance of complexity thinking as a method of solving all apparent woes, including volcanic hazards. The most robust critique is that key concepts of complexity are often poorly understood, with issues of their relevance and applicability often ignored or glossed over (Piepers 2006). Regardless, complexity is becoming a popular method to view an ever increasingly interconnected and uncertain world that is unable to be viewed or understood using reductionist methods. There will, however, always be concern that the theory that tries to explain everything may in fact explain nothing at all.

A core element of complexity is that traditional “boundaries” no longer exist, although they may be in place institutionally. Drawing boundaries is an eminently social process and boundaries are routinely drawn between science and nonscience, experts and lay persons, science and politics, and the social and the natural (Gieryn 1983), with consequences for what is taken into account when understanding and managing risk. Complexity appeals most to those who feel that top-down and reductionist approaches are inappropriate in real-world situations because complexity approaches “use rules which promote and permit complex, diverse, and locally fitting behaviour; decentralise, minimise controls and enable local appraisal, analysis, planning and adaption for local fit in different ways” (Chambers 1997, p. 221).

Reconceptualizing Volcano Warnings

Volcano EWS need to manage and interact with a number of complex systems, including the volcano, society, and environment. But how are these complexities negotiated? Historically cost-benefit models and producing policy that tends to revert to a “reductionist” nature have been adopted. The complexity literature offers little resolution as to how to model and use complexity to help manage knowledge and make decisions. There is, however, one such model, Cyneform (Kurtz and Snowden 2003; Snowden 2005), that addresses differing levels of complexity, not by narrowing opportunities through compartmentalizing them into frameworks, but by moving from different stages of known, knowable, complex, and chaotic systems (Fig. 12). Cyneform is a model or approach to policy formation and operational decision-making that recognizes the value of uncertainties and risk, by reducing pattern entrainment. It is a sense-making framework and its value is not in the logical arguments of empirical verification, but in decision-making and facilitating shared understandings to emerge through the many discourses of the decision-makers (Ravetz 1999). This model enables people to make sense of complexity by relaxing three basic assumptions prevalent in organizational decision-making: assumptions of order, rational choice, and of intent.
Fig. 12

Cyneform model (Kurtz and Snowden 2003, p. 468)

The four different domains in the model represent the dynamics of situations, decisions, perspectives, and conflicts when making a decision under uncertain conditions. The boundaries shown are more like phase changes than physical boundaries, so it is possible to consider the problem as it moves between different phases, such as “knowable” to “complex.” This model helps understanding and the interpretation of problems by indicating they are not always static, which is highly applicable to volcanic crises where complexity and chaos are all involved at different stages and within different systems.

By the 2000s, numerous papers highlighted the need for holistic systemic approaches that accommodate the complexities involved, and provide integrated approaches to disaster management (Geis 2000; McEntire and Fuller 2002; Paraskevas 2006). Yet, to date no single overarching theory has been ascribed that captures every variable and issue associated with disasters (McEntire 2004). For this reason, complexity and chaos theories have gained recognition with the growing understanding that disaster responses should be flexible and adaptive (Koehler 1995; Mileti 1999). Likewise, climate change debates have generated a wealth of literature relating to uncertainty, risk, and the plural values of society built on the theory of systems through the concept of complexity and chaos.

Studies of organizational crises that adopt a complex science approach demonstrate that a complexity-informed framework can aid the design of response to a crisis by developing a co-evolving system that essentially self-organizes, learns, and adapts to their dynamically changing environment, a complex adaptive system (CAS) (Paraskevas 2006; Zhong and Low 2009). A CAS is defined as “a number of components of agents, that interact with each other according to sets of rules that require them to examine and respond to each other’s behaviour in order to improve their behaviour" (Stacey 1996, p. 10), and evolve (Axelrod and Cohen 1999). Interactions within the system can produce unexpected patterns or behaviors that can have unexpected effects on other parts of the system creating nonlinear feedback networks. During a crisis, feedback is required to monitor the progress of the crisis response, and this feedback enables a system to self-correct or modify behavior, learning from experience. Crisis response communication systems, such as volcano EWSs, can be viewed as a CAS where agents self-organize and restructure at a local scale.

In 2008, the Overseas Development Institute (ODI) reviewed the applicability of complexity approaches within real-world crises (Ramalingam et al. 2008). The report highlights that complexity science can generate useful insights into managing complex problems, with a more realistic and holistic approach, supporting useful intuitions, actions, and policy. The idea of self-organization indicates that “actors at all levels of a given system need to be empowered to find solutions to problems, challenging the existing dichotomies of ‘top-down’ versus ‘bottom-up’ so often discussed in disaster practice and international aid agencies” (Ramalingam et al. 2008, p. 62). The concepts of complexity challenge the very method in which current governance conducts its work, as outlined in the following quote (Telford et al. 2006, p. 119):

International agencies need to pay as much attention to how they do things, and their capacities to do them, as they do to the content of their policies and programmes […] sensitivity to context and the flexibility to adapt to evolving realities are essential, instead of applying predetermined strategies and one-size-fits-all solutions.

Complexity theory and models can provide a tool for practitioners, policy-makers, managers, and researchers to reflect collectively on how they are trying to solve problems, by providing better awareness of why disaster or development and humanitarian work is so problematic.

The Need for an Integrated Volcano Early Warning System

Social contexts affect the use and success of VALS far more than previously acknowledged. There are still many scientific uncertainties within volcanology; scientists are continuously developing theories to better understand the origin, processes, and eruptive behavior of volcanoes and the numerous associated hazards. However, it is not just scientific constraints involved in determining warnings but also constraints from social and institutional contexts. Volcanic hazards pose a significant problem in society because they generally occur on a longer time frame than political terms or human generations and therefore are not normally a priority. This generally results in limited funding and resources for monitoring volcanoes and conducting research on their past behaviors, and limited volcanic hazard awareness. From an institutional perspective, the wide-ranging impact of volcanic hazards tends to result in the involvement of numerous institutions and agencies, and it is often difficult to maintain communication both within and external to each body involved. Increasing levels of bureaucracy and contending stakeholders mean that decisions can be complex and take a long time to make and implement. Decision-making is a highly pressured process, particularly for the scientists in charge and federal agency users who have a legal obligation to respond. To reduce this pressure, emergency response plans are established prior to crisis to aid and generate communication and understanding; but this is not enough.

Managing volcanic crises requires careful consideration and understanding of how to take action in the context of extreme uncertainty and complexity, both scientifically and socially. To do this successfully a volcano EWS should be fully integrated to cover everything from monitoring and detection, to analysis and interpretation of the data, understanding risk, to communicating information to stakeholders, and generating an effective response. This requires planning, cooperation, the execution of drills, education, and discussion, to name a few processes, between all actors so that during a crisis effective decisions can be made quickly. In summary, this is reconceptualizing a volcano EWS as a complex adaptive system of communication networks.

Bibliography

  1. Alexander D (2005) Towards the development of a standard in emergency planning. Disaster Prev Manag 14:158–175CrossRefGoogle Scholar
  2. Alexander DE (2014) Communicating earthquake risk to the public: the trial of the “L’Aquila Seven”. Nat Hazards 72:1159–1173CrossRefGoogle Scholar
  3. Aspinall WP, Woo G, Voight B, Baxter PJ (2003) Evidence-based volcanology: application to eruption crises. J Volcanol Geotherm Res 128:273–285ADSCrossRefGoogle Scholar
  4. Axelrod R, Cohen MD (1999) Harnessing complexity: organizational implications of a scientific frontier. Free Press, New York/LondonGoogle Scholar
  5. Bankoff G (2007) Dangers to going it alone: social capital and the origins of community resilience in the Philippines. Contin Chang 22:327–355CrossRefGoogle Scholar
  6. Bankoff G, Frerks G, Hilhorst T (2004) Mapping vulnerability: disasters, development, and people. Earthscan Publications, LondonGoogle Scholar
  7. Barton AH (1969) Communities in disaster: a sociological analysis of collective stress situations, 1st edn. Doubleday, New YorkGoogle Scholar
  8. Basher R (2006) Global early warning systems for natural hazards: systematic and people-centred. Phil Trans R Soc 364:2167–2182ADSCrossRefGoogle Scholar
  9. Bertalanffy LV (1975) Perspectives on general system theory: scientific-philosophical studies. G. Braziller, New YorkGoogle Scholar
  10. Bertalanffy LV, Woodger JH (1933) [Kritische Theorie der Formbildung.] Modern theories of development. An introduction to theoretical biology ... Translated and adapted by JH Woodger. p x. 204. Oxford University Press: LondonGoogle Scholar
  11. Birkmann J (2006) Measuring vulnerability to natural hazards: towards disaster resilient societies. United Nations University, TokyoGoogle Scholar
  12. Bretton RJ, Gottsmann J, Aspinall WP, Christie R (2015) Implications of legal scrutiny processes (including the L’Aquila trial and other recent court cases) for future volcanic risk governance. J Appl Volcanol 4:18CrossRefGoogle Scholar
  13. Capra F (1996) The web of life: a new synthesis of mind and matter. Flamingo, LondonGoogle Scholar
  14. Cardona OD (1997) Management of the volcanic crises of Galeras volcano: social, economic and institutional aspects. J Volcanol Geotherm Res 77:313–324ADSCrossRefGoogle Scholar
  15. Chambers R (1997) Whose reality counts? Putting the first last. Intermediate Technology, LondonCrossRefGoogle Scholar
  16. Chester DK, Marti J, Ernst GG (2005) Volcanoes, society, and culture. In: Volcanoes and the environment. Cambridge University Press, Cambridge, pp 404–439CrossRefGoogle Scholar
  17. David PA, Greenstein S (1990) The economics of compatibility standards: An introduction to recent research. Economics of innovation and new technology 1(1-2):3–41CrossRefGoogle Scholar
  18. Day S, Fearnley C (2015) A classification of mitigation strategies for natural hazards: implications for the understanding of interactions between mitigation strategies. Nat Hazards 79:1219–1238CrossRefGoogle Scholar
  19. Denis H (1995) Scientists and disaster management. Disaster Prev Manag 4:14–19ADSCrossRefGoogle Scholar
  20. Ewert JW, Guffanti M, Murray TL (2005) An assessment of volcanic threat and monitoring capabilities in the United States: framework for a National Volcano Early Warning System. U.S. Geological Survey, RestonGoogle Scholar
  21. Fearnley CJ (2011) Standardising the USGS volcano alert level system: acting in the context of risk, uncertainty and complexity. UCL (University College London), Great BritainGoogle Scholar
  22. Fearnley CJ (2013) Assigning a volcano alert level: negotiating uncertainty, risk, and complexity in decision-making processes. Environ Plan A 45:1891–1911CrossRefGoogle Scholar
  23. Fearnley CJ, Beaven S (2018) Volcano alert level systems: managing the challenges of effective volcanic crisis communication. Bull Volcanol 80(5):46ADSCrossRefGoogle Scholar
  24. Fearnley C, McGuire W, Davies G, Twigg J (2012) Standardisation of the USGS Volcano Alert Level System (VALS): analysis and ramifications. Bull Volcanol 74:2023–2036ADSCrossRefGoogle Scholar
  25. Fearnley C, AEG W, Pallister J, Tilling RI (2017a) Volcano crisis communication: challenges and solutions in the 21st century. In: Observing the volcano world: volcanic crisis communication. Springer, ChamGoogle Scholar
  26. Fearnley CJ, Bird DK, Haynes K, McGuire W, Jolly G (eds) (2017b) Observing the volcano world: volcanic crisis communication advances in volcanology. Springer.  https://doi.org/10.1007/978-3-319-44097-2 Google Scholar
  27. Fiske RS (1984) Volcanolgists, journalists, and the concerned local public: a take of two Citites in the Eastern Caribbean. In: Studies in geophysics: explosive volcanism: inception, evolution, and hazards. National Academy Press, Washington, DC, pp 170–1976Google Scholar
  28. Form W, Nostow S (1958) Community in disaster. Harper & Row, New YorkGoogle Scholar
  29. Foster HD (1980) Disaster planning: the preservation of life and property. Springer, New YorkCrossRefGoogle Scholar
  30. Fujimura JH (1987) Constructing ‘Do-Able’ problems in cancer research: articulating alignment. Soc Stud Sci 17:257–293CrossRefGoogle Scholar
  31. Gaillard JC, Dibben CJL (2008) Volcanic risk perception and beyond. J Volcanol Geotherm Res 172:163–169ADSCrossRefGoogle Scholar
  32. Gallopin GC, Funtowicz S, O’Connor M, Ravetz J (2001) Science for the twenty-first century: From social contract to the scientific core. Int Soc Sci J 53:219–229CrossRefGoogle Scholar
  33. Garcia C, Fearnley CJ (2012) Evaluating critical links in early warning systems for natural hazards. Environ Hazards 11:123–137CrossRefGoogle Scholar
  34. Gardner CA, Guffanti MC (2006) U.S. geological survey’s alert notification system for volcanic activity. U.S. Geological Survey, RestonCrossRefGoogle Scholar
  35. Geis D (2000) By design: the disaster resistant and quality-of-life community. Nat Hazard Rev 1:151–160CrossRefGoogle Scholar
  36. Gieryn TF (1983) Boundary-work and the demarcation of science from non-science-strains and interests in professional ideologies of scientists. Am Sociol Rev 48:781–795CrossRefGoogle Scholar
  37. Gillespie DF, Perry RW (1976) An integrated systems and emergent norm approach to mass emergencies. Mass Emergencies 1(4):303–312Google Scholar
  38. Gillespie DF, Robards KJ, Cho S (2004) Designing safe systems: Using system dynamics to understand complexity. Nat Hazards Rev 5:82–88CrossRefGoogle Scholar
  39. Gladwin H, Lazo JK, Morrow BH, Peacock WG, Willoughby HE (2009) Social science research needs for the hurricane forecast and warning system. Bull Am Meteorol Soc 90:25–29CrossRefGoogle Scholar
  40. Glantz MH (2004) Usable science 8: early warning systems: do’s and don’ts, report of workshop 20–23 Oct 2003, ShanghaiGoogle Scholar
  41. Glantz MH (2009) Heads up! Early warning systems for climate, water and weather-related hazards. United Nations University Press, New YorkGoogle Scholar
  42. Hall ML (1990) Chronology of the principal scientific and governmental actions leading up to the November 13, 1985 eruption of Nevado-Del-Ruiz. Colombia J Volcanol Geothermal Res 42:101–115ADSCrossRefGoogle Scholar
  43. Hall PH (2007) Early warning systems: reframing the discussion. Aust J Emerg Manag 22:32–36Google Scholar
  44. Hanseth O, Monteiro E, Hatling M (1996) Developing information infrastructure: the tension between standardization and flexibility science. Technol Hum Values 21:407–426CrossRefGoogle Scholar
  45. Haynes K (2008) Volcanic isnad in crises: investigating environmental uncertainity and the complexities it brings. Aust J Emerg Manag 21:21–28Google Scholar
  46. Haynes K, Barclay J, Pidgeon N (2008a) The issue of trust and its influence on risk communication during a volcanic crisis. Bull Volcanol 70:605–621ADSCrossRefGoogle Scholar
  47. Haynes K, Barclay J, Pidgeon NF (2008b) Whose reality counts? Factors affecting the perception of volcanic risk. J Volcanol Geothermal Res 172:259–272ADSCrossRefGoogle Scholar
  48. Hemaspaandra LA, Ogihara M (2001) The complexity theory companion. Springer, New YorkzbMATHGoogle Scholar
  49. Hewitt K (1983) Interpretations of calamity from the viewpoint of human ecology. Allen & Unwin, BostonGoogle Scholar
  50. Hill DP, California. Division of Mines and Geology., Geological Survey (U.S.) (2002) Response plan for volcano hazards in the Long Valley Caldera and Mono Craters region, California U.S. Geological Survey bulletin 2185Google Scholar
  51. Hill DP, Mangan MT, McNutt SR (2017) Volcanic unrest and hazard communication in long valley volcanic region, CaliforniaGoogle Scholar
  52. Hogle LF (1995) Standardization across non-standard domains: the case of organ procurement science. Technol Hum Values 20:482–500CrossRefGoogle Scholar
  53. IFRC (2009) World disasters report 2009- focus on early warning, early action. International Federation of Red Cross and Red Crescent Societies, GenevaGoogle Scholar
  54. Kelman I, Kelman I (2017) Linking disaster risk reduction, climate change, and the sustainable development goals. Disaster Prev Manag Int J 26:254–258CrossRefGoogle Scholar
  55. Kelman I, Mather TA (2008) Living with volcanoes: the sustainable livelihoods approach for volcano-related opportunities. J Volcanol Geotherm Res 172:189–198ADSCrossRefGoogle Scholar
  56. Keys H, Green P (2010) Mitigation of volcanic risks at Mt Ruapehu, New Zealand. In: Proceedings of the mountain risks international conference. Firenze, pp 485–490Google Scholar
  57. Kilburn CRJ (2003) Multiscale fracturing as a key to forecasting volcanic eruptions. J Volcanol Geotherm Res 125:271–289ADSCrossRefGoogle Scholar
  58. Kim DH (1994) Systems thinking tools: a user’s reference guide. Toolbox reprint series. Pegasus Communications, CambridgeGoogle Scholar
  59. Koehler GA (1995) What disaster response management can learn from chaos theory: conference proceedings. In: Koehler GA (ed) What disaster response management can learn from chaos theory, 2001. California Research Bureau, California State Library, Sacramento, pp 293–308Google Scholar
  60. Kuppers AN, Zschau J (2002) Early warning systems for natural disaster reduction. In: Early warning systems for natural disaster reduction. Springer, Springer-Verlag Heidelberg, pp 817–821Google Scholar
  61. Kurtz CF, Snowden DJ (2003) The new dynamics of strategy: sense-making in a complex and complicated world. IBM Syst J 42:462–483CrossRefGoogle Scholar
  62. Leonard GS, Johnston DM, Paton D, Christianson A, Becker J, Keys H (2008) Developing effective warning systems: ongoing research at Ruapehu volcano. N Z J Volcanol Geotherm Res 172:199–215ADSCrossRefGoogle Scholar
  63. Malone S (2008) A warning about early warning. Seismol Res Lett 79:603–604.  https://doi.org/10.1785/gssrl.79.5.603 CrossRefGoogle Scholar
  64. Maskrey A (2011) Revisiting community-based disaster risk management. Environ Hazards 10:42–52.  https://doi.org/10.3763/ehaz.2011.0005 CrossRefGoogle Scholar
  65. McEntire DA (2004) The status of emergency management theory: issues, barriers, and recommendations for improved scholarship. In: Paper presented at the FEMA higher education conference, Emmitsburg, 8th JuneGoogle Scholar
  66. McEntire DA, Fuller C (2002) The need for a holistic theoretical approach: an examination from the El Nino disasters in Peru. Disaster Prev Manag 11:128–140CrossRefGoogle Scholar
  67. Mileti D (1999) Disasters by design: a reassessment of natural hazards in the United States. Joseph Henry Press, Washington, DCGoogle Scholar
  68. Mileti DS, Sorenson JH (1990) Communication of emergency public warnings: a social science perspective and state-of-the -art assessment. Oak Ridge National LaboratoryCrossRefGoogle Scholar
  69. Mitchell M (2009) Complexity: a guided tour. Oxford University Press, Oxford/New YorkzbMATHGoogle Scholar
  70. Moran SC et al (2008) Instrumentation recommendations for volcano monitoring at US volcanoes under the national volcano early warning system. Geological Survey (US)Google Scholar
  71. Newhall C, Solidum RU (2017) Volcanic hazard communication at Pinatubo from 1991 to 2015Google Scholar
  72. Newhall C et al (1999) Professional conduct of scientists during volcanic crises. Bull Volcanol 60:323–334ADSCrossRefGoogle Scholar
  73. Nowotny H (2005) The increase of complexity and its reduction: emergent interfaces between the natural sciences, humanities and social sciences. Theory Cult Soc 22:15–31CrossRefGoogle Scholar
  74. Nowotny H, Scott P, Gibbons M (2001) Re-thinking science: knowledge and the public in an age of uncertainty. Polity Press, CambridgeGoogle Scholar
  75. Oppenheimer C (2011) Eruptions that shook the world. Cambridge University Press, CambridgeCrossRefGoogle Scholar
  76. Papale P (2017) Rational volcanic hazard forecasts and the use of volcanic alert levels. J Appl Volcanol 6:13CrossRefGoogle Scholar
  77. Paraskevas A (2006) Crisis management or crisis response system? A complexity science approach to organizational crises. Manag Decis 44:892–907CrossRefGoogle Scholar
  78. Paton D, Johnston D, Houghton BF (1998) Organisational response to a volcanic eruption. Disaster Prev Manag 7:5–13CrossRefGoogle Scholar
  79. Peterson DW, Tilling RI (1993) Interactions between sceintists, civil authorities and the public at hazardous volcanoes. In: Kilburn CRJ, Luongo G (eds) Active lavas. UCL Press, London, pp 339–365Google Scholar
  80. Piepers I (2006) The International System: “At the Edge of Chaos”. arXiv preprint nlin/0611062.Google Scholar
  81. Potter SH, Jolly GE, Neall VE, Johnston DM, Scott BJ (2014) Communicating the status of volcanic activity: revising New Zealand’s volcanic alert level system. J Appl Volcanol 3:13CrossRefGoogle Scholar
  82. Potter SH, Scott BJ, Fearnley CJ, Leonard GS, Gregg CE (2017) Challenges and benefits of standardising early warning systems: a case study of New Zealand’s volcanic alert level systemGoogle Scholar
  83. Ramalingam B, Jones H, Reba T, Young J (2008) Exploring the science of complexity: ideas and implications for development and humanitarian efforts. In: Working paper 285. Overseas Development Institute (ODI), LondonGoogle Scholar
  84. Ravetz IR (1999) What is post-normal science. Futures-the Journal of Forecasting Planning and Policy, 31(7):647–654Google Scholar
  85. Ronan KR, Paton D, Johnston DM, Houghton BF (2000) Managing societal uncertainty in volcanic hazards: a multidisciplinary approach. Disaster Prev Manag 9:339–349CrossRefGoogle Scholar
  86. Sardar Z, Ravetz JR (1994) Complexity: fad or future? Futures 26:563–567CrossRefGoogle Scholar
  87. Scarth A (2002) La catastrophe Mount Pel’e and the destruction of Saint-Pierre. Terra Publishing, HarpendenGoogle Scholar
  88. Senge PM (1990) The fifth discipline: the art and practice of the learning organization, 1st edn. Doubleday/Currency, New YorkGoogle Scholar
  89. Sennert SS, Klemetti EW, Bird DK (2015) Role of social media and networking in volcanic crises and communicationGoogle Scholar
  90. Sigurdsson H, Houghton B, McNutt S, Rymer H, Stix J (2015) The encyclopedia of volcanoes. Elsevier, San DiegoGoogle Scholar
  91. Slovic P (2000) The perception of risk. Risk, society, and policy series. Earthscan Pub, LondonGoogle Scholar
  92. Snowden D (2005) Strategy in the context of uncertainty. Handb Bus Strateg 6:47–54CrossRefGoogle Scholar
  93. Solana MC, Kilburn CRJ, Rolandi G (2008) Communicating eruption and hazard forecasts on Vesuvius. South Ital J Volcanol Geotherm Res 172:308–314ADSCrossRefGoogle Scholar
  94. Sparks RSJ, Aspinall W (2004) Volcanic activity; frontiers and challenges in forecasting, prediction and risk assessment. In: Stephen R, Sparks J (eds) The state of the planet; frontiers and challenges in geophysics. American Geophysical Union, Washington, DCCrossRefGoogle Scholar
  95. Stacey RD (1996) Complexity and creativity in organizations. Berrett-Koehler, San FranciscoGoogle Scholar
  96. Stirling A (2007) Risk, precaution and science: towards a more constructive policy debate: talking point on the precautionary principle. EMBO Rep 8(4):309–315MathSciNetCrossRefGoogle Scholar
  97. Susman P, O’Keefe P, Wisner B (1983) Global disasters, a radical interpretation. In: Hewitt K (ed) Interpretations of calamity: from the viewpoint of human ecology. Allen & Unwin, London, pp 263–283Google Scholar
  98. Swanson DA, Casadevall TJ, Dzurisin D, Holcomb RT, Newhall CG, Malone SD, Weaver CS (1985) Forecasts and predictions of eruptive activity at Mount St. Helens, USA: 1975–1984. J Geodyn 3:397–423CrossRefGoogle Scholar
  99. Tayag J, Insauriga S, Ringor A, Belo M, Newhall CG, Punongbayan RS (1996) People’s reposnse to eruption warning: the Pinatubo expereince, 1991–92. In: Fire and mud: eruptions and lahars of mount Pinatubo, Phillippines. University of Washington Press, Seattle/London, pp 87–99Google Scholar
  100. Telford, J, J Cosgrave and R Houghton (2006) Joint Evaluation of the international response to the Indian Ocean tsunami: Synthesis Report. London: Tsunami Evaluation CoalitionGoogle Scholar
  101. Thieren M (2005) Asian tsunami: death-toll addiction and its downside. Bull World Health Org 83:82Google Scholar
  102. Tierney KJ, Dynes RR (1994) In: Dynes RR, Tierney KJ (eds) Disasters, collective behavior, and social organization. University of Delaware Press/Associated University Presses, Newark, London/CranburyGoogle Scholar
  103. Tobin GA, Whiteford LM (2002) Community resilience and volcano hazard: the eruption of Tungurahua and evacuation of the Faldas in Ecuador. Disasters 26:28–48CrossRefGoogle Scholar
  104. Twigg J (2004) Disaster risk reduction: mitigation and prepardness in development and emergency programming. HPNGoogle Scholar
  105. UN ISDR (2004) Yokohama strategy and plan of action for a safer world: guidelines for natural disaster prevention, preparedness and mitigation. In: World conference on disaster reduction, Yokohama, 23–27 May 1994Google Scholar
  106. UN ISDR (2005) Hyogo framework for action 2005–2015: building the resilience of nations and communities to disastersGoogle Scholar
  107. UN ISDR (2006a) Early warning-from concept to action-the conclusions. In: Third international conference on early warning: from concept to action, Bonn, 27–29 Mar 2006Google Scholar
  108. UN ISDR (2006b) Early warning – from concept to action: the conclusions of the third international conference on early warning (EWC III). In Bonn, 27–29 Mar 2006, p 32Google Scholar
  109. UN ISDR PPEW (2006) Global survey of early warning systems: an assessment of capacities, gaps and opportunities toward building a comprehensive global early warning system for all natural hazardsGoogle Scholar
  110. UNDRO (1990) Mitigating natural disasters: phenomena, effects and options. United Nations GenevaGoogle Scholar
  111. UNISDR (2007) Terminology. http://www.unisdr.org/we/inform/terminology-letter-m. Accessed 27 Feb 2014
  112. Urry J (2005a) The complexities of the global. Theory Cult Soc 22:235–254CrossRefGoogle Scholar
  113. Urry J (2005b) The complexity turn theory. Cult Soc 22:1–14CrossRefGoogle Scholar
  114. Velasco MLC (2000) Mt Galeras: activities and lessons to be learned philosophical transactions of the Royal Society of London series. Math Phys Eng Sci 358:1607–1617CrossRefGoogle Scholar
  115. Voight B (1990) The 1985 Nevado-Del-Ruiz volcano catastrophe-anatomy and retrospection. J Volcanol Geotherm Res 42:151–188ADSCrossRefGoogle Scholar
  116. White D (1995) Application of systems thinking to risk management: a review of the literature. Manag Decis 33:35–45CrossRefGoogle Scholar
  117. White GF, Haas JE (1975) Assessment of research on natural hazards. MIT Press, CambridgeGoogle Scholar
  118. Winson AE, Costa F, Newhall CG, Woo G (2014) An analysis of the issuance of volcanic alert levels during volcanic crises. J Appl Volcanol 3:14CrossRefGoogle Scholar
  119. Wisner B (2004) At risk: natural hazards, people’s vulnerability, and disasters, 2nd edn. Routledge, London/New YorkGoogle Scholar
  120. Wisner B, Gaillard JC, Kelman I (2012) Handbook of hazards and disaster risk reduction and management. Routledge, New YorkCrossRefGoogle Scholar
  121. Wynne B (1996) May the sheep safely graze? A reflexive view of the expert-lay knowledge divide. In: Lash S, Szerszynski B, Wynne B (eds) Risk, environment and modernity: towards a new ecology. Sage, London, pp 44–83Google Scholar
  122. Zhong Y, Low SP (2009) Managing crisis response communication in construction projects – from a complexity perspective. Disaster Prev Manag 18:270–282CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media LLC 2019

Authors and Affiliations

  1. 1.Department of Science and Technology StudiesUniversity College LondonLondonUK