Decision Making and Communication Issues
Floodplains constitute about 7 percent of the U.S. land area (Kusler and Larson, 1993) and represent a valuable national resource. Floodplains store flood waters during high flows (helping recharge groundwater supplies in the process); are a source of biological productivity and diversity; and are used for many human activities, including agriculture, grazing, parks and recreation, transportation, housing, and commercial development. However, because most of these activities preclude water storage during high flows, they need to be properly managed.
Individuals, local authorities, state governments, and several federal agencies make decisions about floodplain management. Even though it is in the national interest to do so, coordinating these decisions is exceedingly difficult, as different decision-making authorities have different interests and mandates. Furthermore, the overall goals of U.S. floodplain management are neither clearly specified nor well organized. Floodplain management decisions thus tend to be fragmented, as pointed out in the Galloway Report and elsewhere.
The Corps's risk analysis techniques and flood damage reduction studies will produce their greatest benefits if these techniques and studies are executed within a comprehensive planning paradigm and framework designed to make the best social, economic, and environmental uses of the nation's floodplain resources. Even the best analytical techniques will fall short of their potential contributions if flood damage reduction project goals are not consistent with public values, which can often be better determined through public participation and communication. This chapter reviews goals, multiple project objectives and trade-offs, decision making, and communication in floodplain management.
THE GOAL OF FLOODPLAIN MANAGEMENT
Neither the U.S. Congress nor the Corps of Engineers has identified an explicit goal for management of the nation's floodplains. Perhaps the closest that any federal water planning document comes to identifying a goal for floodplain management is the Principles and Guidelines (USWRC, 1983), which states that flood damage reduction projects (and other federal water projects) are to “contribute to the national economic development, consistent with protecting the Nation's environment. ”
The committee believes that the goal for management of the nation 's floodplains should be broader: to use the land for the greatest social benefit, accounting for the risks of flooding and steps that can be taken to reduce those risks. In contrast, a goal such as minimizing damages from floods necessitates removal of people and activities from the floodplain. Removing people and activities from the floodplain, however, forfeits the many benefits of floodplain use and may thus be economically and socially undesirable.
The goal of maximizing social benefit in floodplain management leads to a strategy that recognizes the availability of land not in danger of flooding, the probabilities and magnitudes of potential floods, the availability of insurance, and the costs of flood damage reduction structures to reduce damages should a flood occur. As geographer Gilbert White stated: “It is striking that in a century of evolving public policy the prevailing aim has been to minimize losses from floods and not to optimize the net social benefits from using floodplain resources . . . . In simplest terms, it is the contrast between ‘loss reduction' and ‘wise use' (White, 2000).
It is unclear whether this approach promotes the wisest use of the nation's floodplains. The issue is important not only in its historical context, but also because of the damages exacted by floods: in most years, floods cause more deaths and damages than any other natural phenomenon, and the damages from floods in the U.S. are increasing over time (Richards, 1999). The distribution, frequency, and intensity of extreme weather events are changing in ways that are difficult to understand and predict (Karl et al., 1996), and an increasing Gross Domestic Product and population make it natural to expect greater demand for land in the floodplain. This in turn means greater damages in future extreme floods. Increases in wealth and increases in population in the nation's floodplains put more property at risk from floods. The trend of increasing damages thus does not necessarily imply a failure of the nation's approach to flood management (cf. Lave et al., 1990; Pielke, 1999).
Floodplain structures should not be considered fixed and immutable. In some areas the probabilities of flooding are sufficiently large, and the ability to mitigate flooding so expensive, that structures in the floodplain should be removed and activities relocated. Such a floodplain could be devoted to the highest-value uses consistent with periodic flooding (e.g., parks and recreation areas). In contrast, some currently underused floodplains are becoming increasingly valuable because of increasing population and economic activity. If after paying the costs of the protection structures and accounting for the costs of periodic flooding, net social benefits remain, these floodplains should be developed.
Land use controls, zoning, and planning are other important factors that complicate floodplain management. As much of the land in the nation's floodplains is privately owned, decisions about uses of those lands lies beyond the direct responsibilities of the Corps and most water management agencies. On these lands, there are often few incentives that encourage proper flood planning and preparedness (e.g., devoting lands to outdoor recreation activities, or elevating buildings above the 100-year flood stage). There are often also inadequate regulations that limit or prohibit development in flood-prone areas.
In the United States, strong tensions often exist between regulation and zoning on the one hand, and individual property rights on the other. But some U.S. communities have enacted programs for the purchase of floodplain properties that have sustained repetitive flood damages. These buyout programs aim to move susceptible property and its inhabitants out of high-hazard areas, while zoning these areas for land uses such as golf courses and hiking trails (which also serve as stormwater retention basins). For example, results from a comprehensive flood hazard mitigation and zoning program in Tulsa, Oklahoma are impressive: Tulsa's flood insurance rates have dropped by 25 percent and are now the lowest in the nation. In 1992, Tulsa received the nation 's highest rating in the National Flood Insurance Program's Community Rating System (National Wildlife Federation, 1998). The Tulsa experience demonstrates the importance of local-level planning and decisions in effective floodplain management.
Although the Principles and Guidelines focus on economic benefits, issues other than economic damages should figure in the design and function of flood damage reduction projects. These issues include water
quality, recreation, ecological protection, biodiversity, the quality of life, and life itself. In addition to the planning requirements of the Principles and Guidelines, the Corps is required by the National Environmental Policy Act to identify environmental implications of its projects. The Endangered Species Act requires the Corps to give special attention to select species whose survival might be compromised by a project. Many aspects of flood damage reduction projects transcend strict National Economic Development concerns.
Making decisions within this multiobjective framework is challenging. In some cases a single alternative dominates across all flood damage reduction objectives, making for an easy decision. In practice, however, the situation is seldom so agreeable. If a dominant alternative is quickly identified, it is likely that not enough thought was given to identify alternatives that are either less expensive but provide less flood protection, or are more expensive but provide more flood protection. The flood damage reduction project that provides the most benefits on one of the objectives seldom provides most benefits on all of them. In most cases, there are a range of project alternatives, all of which contain a complex mix of benefits and costs that must be weighed against each other.
If all people agreed on the values to be assigned to all project consequences, and if all these values allowed translation into a single dimension (e.g., dollars), multiobjective decisionmaking would be easy. For example, if a dollar value could be unequivocally assigned to extinction of a species, another dollar value to increases in water pollution, and so forth, the decision would come down to choosing the proposal with the greatest monetary net benefit, where each dimension is measured in dollars. However, there generally is little agreement about, for example, the social and economic consequences of species extinction, of reduced salmon migration, or of frequent flooding, and less agreement about how to express such outcomes in monetary units. At its worst, the inability to reduce complex issues to a small number of variables makes the situation akin to an environmental impact statement in which hundreds or thousands of impacts are identified, but there is no way to compare the impacts with one another.
The committee does not wish the Corps to become mired in such a morass. However, to the extent possible, the Corps should account for important social consequences of each project alternative, such as lives at risk in the event of flooding, and important environmental consequences, such as loss of wetlands or biodiversity. The Principles and Guidelines mandate the Corps to adopt the National Economic Development (NED)
alternative in its water resources project planning studies (the alternative with the largest net economic benefits to whomever they accrue). However, the range of benefits to be counted in flood damage reduction studies, as specified by the Principles and Guidelines, may be narrowly construed (see Box 1.1). For instance, a previous NRC committee charged to review the Corps 's water project planning procedures concluded, “Today, ecological and social considerations are often of great importance in project planning and should not necessarily be considered secondary to the maximization of economic benefits. Strict adherence to the NED account may discourage consideration of innovative and nonstructural approaches to water resources planning . . . . The notion of NED as formulated in 1983 may not fit contemporary planning and social realities” (NRC, 1999a, p. 4). That committee went on to recommend a comprehensive review and modification of the Principles and Guidelines.
To enhance social benefits of floodplain management, this committee recommends that the Corps account for (or, more properly, be guided by the Principles and Guidelines to account for) a broad range of social and environmental considerations in its flood damage reduction studies and projects. Environmental, health, safety, and other social considerations of flood damage reduction projects should be quantified to the extent possible and included in floodplain management decisions.
COMPARING PROJECT ALTERNATIVES
The specific purposes of using risk analysis for flood damage reduction studies are to define project objectives, create desirable alternatives, evaluate those alternatives, guide analytical efforts, and facilitate communication. The ultimate intent should be good decisions that maximize net social benefits.
Complex decisions like those involved in flood damage reduction studies can be analyzed with a two-part decision model. The first part of the model relates decision alternatives to possible consequences. The second part assesses the relative desirability of the possible consequences. The two parts of the model are combined to establish relative desirability among a set of alternatives. This ranking derives from (1) the likelihood that particular consequences will result from an alternative, and (2) the relative desirability of those consequences. The likelihoods of consequences are estimated using scientific reasoning from data, while the desirabilities are based on value judgments. Clearly, the
approaches to quantifying likelihoods and desirabilities are different, as are the people who should be making the quantifications.
Sound and comprehensive floodplain management is not solely a technocratic process. The values that society places on its rivers, floodplains, wetlands, and water resources should be central to comprehensive flood damage reduction studies. The foundation for specifying public values is a logical set of specific concerns. As these concerns define the scope of public values relevant to a decision, it is important to directly involve citizens or representatives of citizen groups.
A value model can be constructed in four steps (Keeney, 1992): (1) identify values appropriate for the problem being addressed, (2) define and structure specific objectives related to those values, (3) specify attributes (i.e., metrics) with which to measure each objective, and (4) specify trade-offs among objectives. A specific objective might be stated as, “minimize economic damages,” or “avoid loss of natural habitat.” The attributes associated with these objectives might be, “damage measured in dollars,” and “destruction of habitat measured in acres.”
The objectives can typically be categorized into four types (Keeney et al., 1996): (1) fundamental objectives, the ends used to describe consequences that are of concern to the public; (2) means objectives, the objectives that affect eventual consequences but which are themselves important only for their influence on the fundamental objectives; (3) process objectives, those concerned with how a decision is made rather than what decision is made; and (4) organizational objectives, those influenced by the complete set of decisions made over time by the organization with responsibility for acting in the public interest. type="internal">Table 2.1 gives examples of these different types of objectives in the context of flood damage reduction.
The final step is to articulate value trade-offs among the objectives. Value trade-offs indicate willingness to forgo the achievement of one objective in order to increase achievement on another objective. Value trade-offs can be determined directly from the public or its representatives, and an experienced analyst can facilitate the assessments.
Schemes for making value trade-offs have been developed by, for example, Fishburn (1970), Keeney and Raiffa (1976), von Winterfeldt and Edwards (1986). Multiattribute decisionmaking is used to indicate the superiority of one alternative versus another, even when there are multiple attributes that cannot be compared quantitatively. Valuing non-market attributes, such as injury, disease, or polluted air, is done by looking at related decisions that involve market values. Examples in-
clude willingness to take more dangerous jobs in return for greater compensation, the increased value of property that is less polluted, the greater safety of some cars, and the willingness to pay for private campgrounds that are less crowded or less polluted.
A previous NRC committee charged to review the strategic plan of the U.S. Department of Interior's Grand Canyon Monitoring and Research Center noted the value of trade-off analysis in water resources management decisions: “It should be recognized that adaptive management for the Grand Canyon ecosystem will require trade-offs among management objectives favored by different stakeholder groups. The committee recommends that the Adaptive Management Work Group begin to consider mechanisms for equitable weighting of competing interests . . . . The Center's revised Strategic Plan should include a strategy for scientific evaluation of management alternatives, both in terms of ecological outcomes and satisfaction of stakeholder groups” (NRC, 1999b, p. 9).
TABLE 2.1 Representative Objectives and Their Relationships for Setting Public Policy for Levee Design
Maximize net economic benefit
Maximize public health and safety
Minimize construction costs
Minimize environmental impacts
Minimize social disruption
Promote equity and fairness
Ensure quality control
Promote conservation of resources
Minimize construction impacts
Communicate with all stakeholders
Coordinate with other decisions
Involve the public
Use reliable and accurate information
Corps organizational objectives
Contribute to public trust
Ensure public acceptance
Reasonable value trade-offs can help decisionmakers make good public policy decisions. For other problems, they allow for the justifiable elimination of some inferior alternatives, leaving a smaller set of better alternatives.
FLOODPLAIN MANAGEMENT ALTERNATIVES
The Corps's flood-related programs have traditionally focused on structures intended to modify flood flows through storage or by changing channel and floodplain hydraulics. These include dams and reservoirs, levees, walls, diversion channels, bridge modifications, channel alterations, pumping, and land treatment. The alternatives considered are typically combinations of possible structures that will reduce flood frequency and magnitude. However, water resources project planning studies may benefit by considering, both alone and in combination with structural alternatives, alternatives that manage possible consequences of floods just prior to or during a potentially serious flood. These alternatives include warning systems, evacuation plans, and flood triage.
One way to lessen potential loss of life and flood damages is to warn people in sufficient time so that they, and even some of their possessions, can be removed from harm's way. Even a short warning time can be sufficient to avert loss of life from floods (Brown and Graham, 1988; Paté-Cornell, 1984). For example, the National Weather Service reports the formation of tropical depressions that could develop into hurricanes. It gives periodic warnings as the storm develops and indicates whether it appears likely that the storm will strike land. As time passes the quality of information concerning the storm's intensity and path increases. People can interpret this information to take steps to protect their lives and property, even when it is not certain that a damaging storm will strike them. Periodic updates help individuals make decisions about when the risk is high enough to take actions such as evacuation, boarding up windows, and buying emergency food and water supplies. This warning system has developed to the point where even a major hurricane may cause few deaths and where property damages are significantly reduced.
When a large flow threatens the integrity of a flood damage reduction system, a variety of trade-offs must be faced. For example, levees protecting farmland could be breached to lower the chance that levees protecting an urban area will fail. Even levees protecting one urban area could be breached to prevent levee failure in another urban area, where failure could result in much more catastrophic damage or loss of life.
Such triage decisions are difficult, because they may involve subjecting one area to flooding in order to lessen the chance that another will be flooded. These difficult decisions are often easier to make if they are discussed before flood conditions are imminent.
Identifying sound, credible, and effective risk reduction priorities and solutions depends greatly on a well-informed public. The public should be knowledgeable about risk issues and should be given opportunities to express opinions and become involved in risk assessment and risk management activities. This involves risk communication: the effective understanding of risks and the transfer of risk information to the public, and the transfer of information from the public to decisionmakers.
Risk communication covers a range of activities directed at increasing the public's knowledge of risk issues and its participation in risk management. It includes, for example, public education about hazards and public hearings on risk management. Much risk communication research has been conducted since the early 1980s, when risk communication emerged as a distinct element of risk analysis (Fischhoff et al., 1981; Morgan et al., 1992).
The intent is to explain how the public perceives and interprets risks and to identify ways to improve the transfer of information to the public. A large fraction of the public is unfamiliar with the nature of the risks to which they are exposed. As the Galloway Report concluded: “As the Midwest Flood of 1993 has shown, people and property remain at risk, not only in the floodplains of the upper Mississippi River Basin but also throughout the nation. Many of those at risk neither fully understand the nature and potential consequences of that risk nor share fully in the fiscal implications of bearing that risk” (IFMRC, 1994, p. xxi).
Risk management decisions should not simply be made by technical experts and public officials and then imposed on, and justified to, the public after the fact. Risk communication involves a dialogue among interested parties—risk experts, policy makers, and affected citizens. It also involves the news media, as citizens often receive their information from the media. If the media do not report knowledgeably and accurately, constructive public involvement becomes more difficult.
The public's response to risk issues is complex because “the public”
contains groups with different values and stakes. The risk assessment process should be opened to participation and scrutiny by affected stakeholders. This will increase the need to facilitate the public 's ability to understand risk information and the ability of policy makers to understand public perception of risk. It is worth noting that public involvement including large numbers of stakeholder groups requires a significant commitment of time and effort from members of the public, as well as staff members from the agencies involved in flood damage reduction. With larger and more expensive projects, years of commitment may be required to help facilitate communications.
Decisions about appropriate floodplain management strategies differ from decisions about proper communication of those decisions. The goal of floodplain management should be to use the floodplain for the greatest social benefit. The goal of the communication decisions is to involve and inform the public and to have them understand what floodplain management decisions were made and why. These decisions are ideally addressed simultaneously, as it is important to establish two-way communication with interested parties in the course of developing and analyzing floodplain management plans.
Ideal communication decisions involve several steps:
identify the audiences that should be communicated with and involved in the decision process about floodplain management,
specify the objectives of communication, including the information people should provide and should receive, and what requires action and what those actions are,
create alternatives for communication that include oral and written communication (and possibly internet options) for experimenting with models, and
select the best combination of alternatives after appraisal.
The quality of communication is greatly enhanced when trust exists among the parties involved in the communication. This trust is built through an open process in floodplain management decision making and by involving stakeholders early in the process (which the Corps often does in many of its water resources project planning studies). If individuals feel they are involved in analyzing floodplain alternatives that affect them, they are more likely to understand and accept the implications of “their study.”
The methods for analyzing the complexities of floodplain manage-
ment are not simple to understand. This makes it difficult to communicate with citizens who are unfamiliar with scientific principles (e.g., hydrology, structural design) necessary to design floodplain management facilities. Indeed, few of the individuals involved in floodplain management understand all these principles well. It is thus a challenge to have individuals understand the full details of a flood damage reduction planning study.
It is important to use simple models to describe methodological ideas and the results of analysis rather than, for instance, models that focus on the mathematical and scientific concepts used in the analysis. Indeed, most of the public is more concerned about a specific application of a method than about the method itself. It is thus often easier to illustrate both the ideas of a method and the specific application together.
Once an analysis is completed, the critical factors that influenced the selection of recommendations can usually be identified. Simple models that illustrate these key ideas in simple situations that can be more easily understood might be especially useful. For instance, imagine a complex computer model involving more than 20 stages that analyzes alternative plans. The recommended plan depends strongly on the interaction of upstream and downstream management strategies. In this case it should be possible to build a simple two-or three-stage model with hypothetical, but realistic, information that reproduces the key interactions that are critical to the recommendations for the real problem. A simple situation model may also be very helpful in communicating key insights from a more complex analysis. This should enhance the likelihood that individuals will understand the interaction and hence be able to see how it is relevant to them. In illustrating these simple models, it is important to explicitly include all important assumptions and judgments, about both facts and values, that are relevant to the results.
Documentation of floodplain management studies is another critical aspect of communication. The standard for documentation is that an interested party should be able to understand everything that was done, why it was done, how it was done, and the range of implications. All assumptions and summaries of the value judgments and data used should be provided for anyone to examine. In the end, while they do not have to agree with all agency (the Corps and others) planning decisions, stakeholders should fully understand all the steps involved in the flood damage reduction study.