Introduction:

Natural disasters in the light of management cybernetics

Natural disasters seem to be on the rise worldwide and their increasing frequency and dimension [Munich Re Group, 2004] make them more and more the focus of society's concern [Annan, 1999]. But do natural disasters really occur more often than before? Are they more disastrous because of their physical manner or because they are socially constructed, with society increasingly 'getting in nature's way'? The latter appears to be the case. For example, globalization has led to more direct linkages of distant places than existed in the past. The rising interconnectedness and dependency of elements within human systems increase this complexity while the nature of those connections gets more complicated and the number of system elements increases. Factors such as population growth, agglomeration of population and capital value in metropolitan areas, rising living standards, settlement and industrialization of very exposed areas, vulnerability of certain elements and groups in modern society, and the increasing number of high-risk technologies, all play a role [Munich Re Group, 2004]. Further, an increasing complexity of infrastructure, especially communication systems, makes human society more vulnerable to natural hazards. Trust and dependency on information technology, in particular, enhances vulnerability even more. Environmental degradation, such as surface sealing, global warming and climate change, are other dynamic pressures on the stability of human systems. Consequently, the effects of events like natural disasters are felt more quickly. A more effective response is needed in order to address these many negative impacts.

In the Pacific, small island states are especially vulnerable to hurricanes ('typhoons', 'cyclones') due to their small size, isolation, fragile ecological systems, poorly developed infrastructure, limited fresh-water and other natural resources, fragile economies, limited financial and human resources and low elevation above sea level. Among these islands, the Hawai'ian Islands have the highest population density of them all [Pacific Regional Environment Program, 2003]. Urbanization has increased the concentration of people and capital in Hawai'i's coastal areas, especially on O'ahu. A direct hit on O'ahu by a hurricane would put in jeopardy a significant portion of its population and economic wealth. Further, Hawai'i's isolation makes outside assistance very difficult to provide - the neighbor islands, the west coast of the United States, and Guam are the nearest responders. Transportation of resources by air or sea takes on average five hours or several days, respectively, coming from the west coast of the United States or Guam. Infrastructure damage to the islands will limit the functionality of those life-sustaining transportation corridors.

In response to these vulnerabilities, an effective disaster-management system must be based on the fact that the nearest responders are the State's least damaged islands. Except for the limited assistance that the least damaged islands might provide, Hawai'i State Civil Defense estimates that the population of the Hawai'ian Islands will be without outside assistance for at least one week after a major hurricane event [Teixeira, 2007b]. Focusing on O'ahu, Ed Teixeira, Deputy Director of Hawai'i State Civil Defense, has observed that planning for a high-impact-low-probability event like Hurricane Katrina on O'ahu has not much evolved: 'because it was unthinkable and too hard to think about' [Teixeira, 2007b].

Society's concerns are often underrepresented in disaster management decisions since long-term mitigation measures often cannot compete with short-term politically motivated measures and funding sources are polarized, which results in a continuation of the present situation regarding the disaster vulnerability in Hawai'i. Often, disasters are exacerbated by policy problems and thereby caused vulnerabilities due to failure to address root causes, for example by certain land use or settlement policies, population distribution or degrading habitats [Comfort, 1999]. These are initiated under the legal framework of development policy. The complex problem of disasters consists of four factors: first, the rate of social and environmental change exceeds organizational capacity to manage it effectively. Secondly, the understanding of the components and consequences of that change is inadequate. Third, interactions among individuals, organizations and governments are uninformed and fourth, change in public policy and practice is needed. 'If the complexity of interacting scientific, social, political, and economic conditions exceeds the existing capacity for organizational control, decisions taken by local actors govern the direction of the evolving process' [Comfort, 1999, p. 42]. Yet, an integrated process of hazard reduction requires coordinated action across jurisdictional and disciplinary boundaries. Building resilient communities as a policy is at the core of the participatory approach and is one of the main solutions academia offers. Management Cybernetics offers such participatory and non-hierarchical approaches that are capable to reduce vulnerabilities on all scales.

The assumptions of the Federal Emergency Management Agency (FEMA), expressed in the Concept of Operations (CONOP) for a catastrophic hurricane impacting the State of Hawai'i, offer little consolation [Federal Emergency Management Agency, 2007]. A massive federal effort will be needed, it states, because O'ahu's support infrastructure faces potentially catastrophic inundation and damage: the main power production facilities (both electric generation and liquid fuels), Honolulu International airport, and the cargo-handling facilities at Honolulu harbor and Pearl Harbor. The problematique is further exacerbated because 80% of Hawai'i's population lives on O'ahu and is especially vulnerable due to lack of resources. Making the situation worse, the other islands are dependent on O'ahu for energy, food and other commodities, which will greatly limit their ability to assist O'ahu. No transportation for evacuation will be available or feasible; the evacuation of tourist population faces similar issues of capability and feasibility [Rosenberg, 2007b]. Air evacuation of visitors leaving the State of Hawai'i would require 400-500 aircrafts, Boeing 747 equivalent [FEMA, 2007]. Even if it would be considered, prioritizing evacuees would become a problem. The most pressing problem will arise in the aftermath when in a competitive situation mass control becomes a major issue. In case of a false warning, the governor would possibly face 180,000 angry tourists, an upset airline and hotel industry and a low chance of getting elected again. Overall, the governor can make only recommendations to the Tourist Board, which is an independent commercial board. The Board could start canceling people coming in and making people leave [Rosenberg, 2007a]. Reality, as seen the night before Hurricane Katrina hit New Orleans 2005, looks very different: hurricane parties are abound.

Overall, hurricanes are by far the most costly disasters in Hawai'i (see Table 1).

From 1860 to 1962, floods from tsunamis, hurricanes and rainstorm caused more than 350 deaths and over $82 million in property damage in Hawai'i. Damage from floods from 1963 to 1982 total about $395 million [Department of Land and Natural Resources, 1996]. Hence, within 20 years the damage developed fivefold compared to the 100 year period. For a period of 20 years in comparison, it would be a 25 fold increase. Numbers cannot be directly related in this way, but the trend for the tremendous increase in damage potential is obvious. It results mostly from increase in population density and increase in value and agglomeration of value in the Hawai'ian Islands. Therefore, a massive federal effort is assumed for a major hurricane impacting the whole State of Hawai'i.

Since the risks and damage potential of natural events cannot be changed or managed, it is crucial that human-caused vulnerabilities be kept to a minimum. One way to achieve this is through an effective and efficient disaster management system, which this dissertation aims to explicate by critiquing and offering suggestions to improve the effectiveness of disaster management on O'ahu. The VSM aims at enhancing effectiveness of the elements already in place, rather than proposing new disaster management elements. Lessons learned from this case could be applicable to the management of disasters of any type and size in Hawai'i.

Vulnerability leads to destabilized social systems - what does management cybernetics have to offer? Vulnerability can be defined in a Variety of ways. For the purpose of disaster management, it is 'the characteristics of an individual or social group or a situation to anticipate, cope with, resist and recover from the adverse effects of a natural hazard' [Blaikie, et al., 2004]. In cybernetic terms, vulnerability is the potential for a system to become unstable. When unstable for a certain time and not returning to a stable system, the system becomes non-viable. The timeframe for a system to become then viable again is called 'Relaxation Time'. Vulnerability can be caused by a Variety of sources internal and external to the system. For example, disasters can be socially produced displaying internal disturbances: instabilities persist in the daily routine of people's lives such as a incorrect flood-plain mapping system leaving a family ignorant to the fact that evacuation is necessary at certain times or simply a non-working fire-extinguisher. Those facts by itself are not a threat but in connection to other events, such as a flood or fire, they become key to a circumstance developing into a disaster or not. Ultimately, the term lends itself to many aspects of interest - e.g., physical, natural, environmental, social, economic or cultural vulnerability - and therefore underlines the need for a holistic approach. This also means that the vulnerability of the whole system depends on the vulnerability of its units. This gives rise to the argument that the government should be responsible for or give leadership in making society a safe and secure place in form of integrating anticipatory disaster mitigation into development strategies. This systemic viewpoint underlines the statement that the whole is more than the sum of its parts, which argues against reductionism and advocates a holistic approach, such as the VSM. It is also helpful to elucidate the challenge of complex systems through relating viability and stability. A system is viable if it remains stable in the face of an unexpected event [Beer, 1994a]. In terms of the disaster management system, viable means capable of responding effectively, or remaining invulnerable. Even though small instabilities can endanger the viability of the system as a whole a certain minimal degree of instability within the system does not yet risk its viability because those systems are capable to survive by absorbing certain perturbations [Holling, 1977]. Setting those limits on a small scale will help to alarm the system at large to avert a viability-threatening concatenation of events, even if every single circumstance seems only as a small disregardable risk. But, risk and vulnerability are two sides of one coin. Increasing vulnerability leads to increased disaster risk. Hence, reducing vulnerability by building disaster-resilient communities is key to disaster risk reduction. In a systems thinking view, vulnerability and the risk of incurring instabilities within the system at large has great potential to be stabilized through revealing the driving mechanisms and deep structures. In another stance, the social sciences emphasize that risk cannot only be defined in terms of physical damage, but have to include risk perception or acceptance. Behavior is greatly affected by whether or not risk is taken voluntarily, since people's risk posture (i.e., if they are risk averse, risk neutral, or risk-seeking) has been shown to vary along this dimension.

Overall, a disaster is the destabilization and disruption of the social system, its units - communities, social groups, individuals - and its connectivity. The environmental, economic and social reverberations caused by this destabilization lead to positive feedback loops until countermeasures balance the system as a whole and it returns to a stable state again. This systems approach is valuable, since it is multi-dimensional, includes all temporal and spatial scales, and emphasizes the disturbance of the collective routine. The Environment of the system affects the system with natural and human-induced disturbances - for example, a natural hazard such as a hurricane. My work incorporates all natural disturbances, also termed hazards, that are the trigger of 'natural' disasters as well as the human-induced aspects. Since a disaster can be caused by a Variety of hazards, e.g. by hurricanes, earthquakes or tsunamis this systemic approach is applicable to all hazard types.

Management cybernetics was chosen because much of hazard management research only describes what goes wrong during hazards and why things do not work. Some approaches offer explanations for damages, such as people's poor perception of the phenomena and poor choices of response, the nature of the geophysical phenomena themselves or the nature of institutions that create or exacerbate risk and vulnerability for society or particular groups within it. Ways to improve hazard management in general and in a proactive stance are rarely addressed. As a fresh breeze, management cybernetics offers solutions. This work is original and unique because the VSM was never applied to disaster management systems, a loosly coupled network of systems that are hibernating and not always in place.

The VSM lends itself perfectly for this analysis because it deals with messes - not defined problems - and can illuminate why things go wrong. Its theory says that the sum of the elements is greater than the sum of its parts. Instead of examining the cause and effect in a linear manner, the VSM specifically looks at the links that hold the system together in a holistic fashion and therefore takes a system's full complexity into account. Since the VSM can integrate quantitative and qualitative measures, it can provide a common language and framework to discuss the management support and coordination needed by the groups working in the very complex field of disaster management - private and governmental agencies, non-governmental and volunteer organizations. Improvements to communication channels within and between disaster management organization can save valuable time and can promote high levels of effectiveness and efficiency. The VSM can support research incorporating different disciplines without having its basic structure and dynamics obfuscated. Highly valuable is its ability to diagnose hibernating and temporary systems that jump in and out of existence.

Viability is most commonly understood in terms of longevity and persistence, where success is measured in terms of survival, but the VSM does not neglect organizations that are less long-lived and based on goal-oriented action. Beer's main message in The Heart of Enterprise [Beer, 1994a] specifically states that the aim is not single-goal oriented, e.g. maximizing profits, but viability and survival, which is ensured through effective organization. The reader needs to understand that in essence the means for survival, the 'how', are most important. Hence, effectiveness of the Hawai'ian disaster management system is the measure of viability per se, independent of how long this organization exists. Moreover, two aspects should be highlighted to show the VSM's applicability to disaster management: time - a disaster needs a quick response - and smooth coordination - not confusion. One advantage of using Beer's VSM in a disaster management context is its convenient framework for experiencing and examining interactions among several groups responding to a disaster. This also includes the democratic management style within volunteer organizations and the command and control approach used by the military. The VSM enables one to distinguish among the different operations and management units along with their communication channels and this secures its flexibility in terms of different management styles. Specifically temporary disaster management systems, which jump in and out of existence depending on when a disaster strikes, do not necessarily have a base of commonly understood conventions and relationships on which to build. The methodology is explicitly linked to this purpose and involves 'qualitative measures of cohesion, identity and ethos as well as the more usual quantitative measures' [Leonard, 1993, p. 79].

From a practical standpoint, the federal disaster management system is constituted by the Federal Emergency Management Agency's (FEMA) Incident Command System (ICS), which is a systems approach. So it is both inviting and highly useful to diagnose this system with a similar approach coming from the same field of thinking. Overall, the VSM seemed to have great potential to improve the effectiveness of hurricane hazard management.