On The Line

Meeting the Promise of Flood Forecasting


Decision makers in the private sector and at all levels of government rely on flood forecasts to decide how to respond to the threat of flooding. The U.S. is currently spending more than $4 billion to "modernize" its National Weather Service (NWS), including improvements to its river and flood program (Fread et al., 1995). Such improvements carry with them a promise to "greatly improve the capability of water facility and emergency managers to take . . . effective actions that will significantly mitigate the impact of major floods." (Braatz et al., 1997). At the same time, our country has experienced considerable economic and other losses; during the 1990s, flood-related damage has averaged almost $5 billion a year.

"Improved" flood forecasts are themselves insufficient to reduce flood losses. As Gilbert White noted in 1939, "a forecast is of no value unless those who receive it are prepared to act promptly and efficiently." Further, a forecast that is "inadequate," meaning either mistaken, misleading, or misused, "may cause more loss than if there had been no forecast at all." Thus, improved flood forecasting requires success on two fronts: the technical aspects, where information is generated, and the choices that are made by those who use this information.

The Red River of the North

Mounting evidence suggests that many decision makers inadequately use existing flood forecast products. Consider the case of recent flooding in the Red River of the North Basin in the spring of 1997. In April, the river, which flows north along the North Dakota-Minnesota border, experienced extreme flooding (see the Observer, Vol. XXI, No. 6, p. 1). Damage is estimated at $1-2 billion, with most occurring in Grand Forks, North Dakota, and East Grand Forks, Minnesota. Considerable attention has been focused on flood stage outlooks and forecasts and their role in decisions leading up to the flooding.

The North Central River Forecast Center issues two types of forecast products for this basin. A numerical outlook is issued one to two months prior to the expected peak flooding, typically in early-March. Such a long lead time is possible largely because most large-scale floods in the basin are the result of snowmelt, not rainfall. Operational forecasts are issued periodically in the weeks prior to and following peak flooding and are the product of a hydrologic modeling system.

When the flood outlooks are issued, two numbers are presented for the expected river stage for each forecast location. One is based on a scenario of average temperature and no subsequent precipitation, the other on average temperature and average precipitation (i.e., "normal" climate). For East Grand Forks, the outlooks issued in mid-February 1997 were for 47.5 and 49 feet.

In interviews conducted in May 1997 with various decision makers in the flooded region, it was clear that different people interpreted the flood stage outlooks in different ways. Some viewed the two numbers as a range, i.e., that the maximum flood stage would be between 47.5 and 49 feet. Others viewed the higher number as a maximum; for example, on April 8, 1997, the Grand Forks Herald reported that NWS "experts are still forecasting a maximum 49-foot crest at East Grand Forks." Others viewed the flood outlook as exact, i.e., "the crest will be 49 feet." Still others viewed it as somewhat uncertain; believing it could vary one to six feet above or below the outlook level.

Which perception might have been correct is not known because NWS flood outlooks do not include any quantitative information regarding uncertainty. A qualita-tive disclaimer is included with the outlook that warns of uncertainty. Looking back at East Grand Forks crest data, the actual crest equaled or exceeded the "normal" level outlook in six of the 12 years outlooks were issued, and, at East Grand Forks, in four of 12 years the "normal" crest outlook was off by more than 10% of the outlook value.

When decision makers ascribe uncertainty to the forecasts, this can potentially introduce a number of pathologies to the decision process, including: misplaced responsibility for flood fight decision making, misunderstandings of the uncertainty associated with forecasts, potential over-confidence in forecasts, inability of the forecasting community to develop appropriate justifications for improvements to (or maintenance of) forecast products, and obstacles to effective preparation for and response to the flood event. Final determination of the extent to which these occurred in, or resulted from, the flooding in the Red River of the North Basin in 1997 awaits further study. Nevertheless, the way decision makers in the region interpreted the flood outlooks affected the actions they took in preparing for the flood.

Little attention has been paid to the manner in which flood forecasts are interpreted by decision makers, and subsequently, how this information is incorporated into decisions. In short, the use and value of existing flood forecast products is not well understood, much less the potential increased usefulness and value that might be attained through "improving" the products and manner in which they are delivered. Other recent experience suggests that this circumstance may be fairly common beyond the Red River of the North Basin (see Changnon, 1996).

Effective use of flood forecasts cannot be attained by simply providing "more information," such as confidence intervals or exceedance probabilities. If decision makers have difficulty using existing products, these difficulties will not go away simply by providing more or "better" information. More attention must be paid to how forecasts are issued, who actually receives what information, and with what effect. Several recent studies suggest that as the amount of information available for decision increases, the judgment process may actually become less reliable, especially in contexts of high uncertainty such as flood forecasting. As a result, decision making skill can decrease as the amount of information increases (see Stewart et al., 1992).

The Advanced Hydrological Prediction System (AHPS), initially being implemented by the NWS in the Des Moines River Basin, provides an opportunity to learn more about how decision makers perceive and use flood forecast information. Decision makers in the basin were sent a questionnaire about their information needs. The results of this first AHPS implementation need to be closely evaluated regarding use of information. If successful, the AHPS may provide a model that can be used in other basins.


We find ourselves at a fork in the road. In the past, it has been enough to focus on developing and refining technical capabilities, because most decision makers had little or no reliable information about flood forecasts at their disposal. Today, we may have reached a point where our capability to produce sophisticated flood forecasts has outstripped our ability to use those forecasts effectively. As we move forward, we must decide whether to continue to focus scarce resources solely on improving our technical capabilities or to apply some resources to better understand--and ultimately better use--the sophisticated information that is now (or soon will be) available. Meeting the promise of flood forecasting lies in the balance.

Roger Pielke, Jr., Environmental and Societal Impacts Group, National Center for Atmospheric Research

The author can be contacted at PO Box 3000, Boulder, CO 80307-3000; e-mail: rogerp@ucar.edu


Braatz, D.T., J.B. Halquist, R.J. Warvin, J. Ingram, J.J. Felt, and M.S. Longnecker. 1997. "NWS Hydrologic Products and Services: Moving from the Traditional to the Technically Advanced." Presented at the AMS 13th International Conference on Interactive Information and Processing Systems for Meteorology, Oceanography, and Hydrology, Long Beach, California.

Changnon, S.A., ed.. 1996. The Great Flood of 1993: Causes, Impacts, and Responses. Westview Press: Boulder, Colorado.

Fread, D.L., R.C. Shedd, G.F. Smith, R. Farnsworth, C.N. Hoffeditz, L.A. Wenzel, S.M. Wiele, J.A. Smith, and G.N. Day. 1995. "Modernization in the National Weather Service River and Flood Program," Weather and Forecasting 10:477-484.

Stewart, T., W.R. Moninger, K.F. Heideman, and P. Reagan-Cirincione. 1992. "Effects of Improved Information on the Components of Skill in Weather Forecasting," Organization Behavior and Human Decision Processes, 53:107-134.

White, G.F. 1939. "Economic Aspects of Flood-Forecasting," American Geophysical Union Transactions, 20:218-233.

UND Researchers to Aid Community and Examine Effects of Red River Floods

The purpose of the Community Redevelopment and Research Network, University of North Dakota (UND), is to provide a focus point for community redevelopment and research efforts in the Grand Forks region following the severe Red River floods that occurred this spring (see the Observer, Vol. XXI, No. 6, p. 1). The hallmark of the network's approach, as exemplified in the on-going Grand Forks Business Emergency Census, is to combine the strengths of traditional university-based research with outreach and intervention to address social problems and community needs following the disaster. Network members will use the expertise and resources of the university to aid the community and in the process develop research and learning opportunities for faculty and students.

The network is seeking working relationships and partnerships with organizations and groups in and out of the Red River Valley to generate resources and identify research needed to address social problems arising both from the disaster and from the economics of recovery.

If you or your organization would like to explore partnership or collaboration with this group, or if you would like to learn more about the Community Redevelopment and Research Network, contact Clifford L. Staples, Department of Sociology, University of North Dakota, Grand Forks, ND 58202; (701) 777-4417; fax: (701) 777-2468; e-mail: staples@badlands.nodak.edu.

Water and Disasters . . .

IDNDR Plans New Internet Conference for 1997 World Disaster Reduction Campaign

The United Nations International Decade for Natural Disaster Reduction (IDNDR) Secretariat invites all interested persons to join its 1997 Internet Conference on "The Socio-Economic Impact of Water-Related Disasters." The conference is part of the 1997 World Disaster Reduction Campaign: "Water: Too Much . . . Too Little . . . Leading Cause of Natural Disasters" (see the Observer, Vol. XXI, No. 6, p. 8).

What is the Conference About?

The conference will focus on floods and drought and will feature reports that outline the impact of water-related disasters on specific locations around the world. From there, participants will work toward building a culture of prevention, focusing on recent actions that have been successful in mitigating disaster impacts.

The conference will take place from mid-September to mid-October 1997 on the Internet using electronic mail and the World Wide Web.

Who Can Participate?

The 1997 IDNDR Internet conference is for anyone concerned about water-related disasters. Persons interested in participating are invited to provide brief (one- to three-page) reports on floods and drought, based on their organizational and geographic perspective. Other individuals can comment or introduce new cases during the debate.

Key policy-making institutions and eminent specialists will contribute. Thus the conference will help to link people and organizations with similar concerns, since it will bring together professionals from local, national, regional, and international bodies in developed and developing countries--all with different perspectives on water-related disasters. It will also allow participants to post announcements regarding related events, projects, or initiatives and thereby stimulate bilateral networking among conference participants.

Conference participation is free. Most presentations will be in English, with some in Spanish.

How Can You Participate?

To participate in this conference, send an e-mail message to: listserv@thecity.sfsu.edu, and in your message write, "subscribe risk [your first name] [your last name]." You will then receive information about the conference and a registration form. You can also register via the conference Web site: http://www.quipu.net/.

For More Information

The conference is jointly organized by the IDNDR Secretariat, part of the United Nations Department of Humanitarian Affairs, and by Quipunet, a nonprofit organization whose aim is to promote education via the Internet. For more information, contact the IDNDR Secretariat, United Nations, Palais des Nations, CH-1211 Geneva 10, Switzerland; tel: (41-22) 798 6894; fax: (41-22) 733 8695; e-mail: idndr@dha.unicc.org. Persons in Latin America and the Caribbean should contact the IDNDR Regional Office, P.O. Box 3745-1000, San José, Costa Rica; tel: (506) 257 2139; fax: (506) 257 2139; e-mail: hmolin@undpcos.nu.or.cr.

Insurance Industry to Work with UNEP on Environmental Disaster Issues

Over 60 leading international insurers from 23 countries have joined together to form the UNEP (United Nations Environment Programme) Insurance Industry Initiative. The group will address risks caused by environmental change. The organization is concerned because:

The UNEP initiative gives the industry its own voice on climate change and other important global environmental issues, and insurance companies of all kinds (life, health, property, etc.) are invited to participate. For more information, contact UNEP, Environment, Trade, and Investment Unit, Geneva Executive Center, C.P. 356, CH-1219 Geneva, Switzerland; fax: (41-22) 796-9240.

[Adapted from the Network Newsletter - a publication of the Environmental and Societal Impacts Group, National Center for Atmospheric Research]

International Hurricane Center Offers Certificate Program and Courses in Hazards Management

The International Hurricane Center, Florida International University, now offers an Emergency Management and Hazard Mitigation Certificate Program--continuing education to meet the needs and interests of all professionals who could benefit from the flexible format and extended schedule of this program.

The first seminar of the program series, "Our Communities and Hazards," was offered in May. Upcoming seminars include (dates tentative):

In addition, a new graduate-level course "Vulnerability Analysis" is being offered this fall through the Department of Construction Management at Florida International University. This course will complement "Topics in Hazard Mitigation," which was introduced in 1995 and which is also being taught this fall. Both of these courses are offered to students pursuing a Master's degree in Construction Management, and both are offered by videotape through the FEEDS (Florida Engineering Education Delivery System) program to locations anywhere in the U.S.

For details about the International Hurricane Center's education programs, contact Ricardo A. Alvarez, International Hurricane Center, Florida International University, University Park Campus, Miami, FL 33199; (305) 438-1607; fax: (305) 348-1605; e-mail: ricardoalfonso@msn.com; or, malvar04@solif.fiu.edu; WWW: http://www.fiu.edu/~hurrican/. Persons interested in the videotape program should contact Mercy Rueda; (305) 348-2801.

SNDR Publishes National Plan for Disaster Reduction

In December 1996, the Subcommittee on Natural Disaster Reduction (see the article above), published Natural Disaster Reduction: A Plan for the Nation, a document that proposes an interagency approach for the coordination and adva ncement of programs, strategies, and research to reduce the social, environmental, and economic costs of natural hazards. The plan focuses on resilience, rather than resistance, to natural hazards through the creation of anticipatory practices with regard to risk assessment, mitigation, and warning.

Natural Disaster Reduction summarizes existing federal research programs dealing with hazards and identifies promising approaches that federal agencies might undertake to reduce disasters' toll. It calls for three major policy shifts in the nation's approach to disasters: 1) anticipating and assessing risk; 2) focusing on comprehensive mitigation that builds resilience; and 3) implementing warning and dissemination systems that permit resilience. To support these goals, the plan identifies four areas of needed research: 1) better understanding of the character of the hazards themselves; 2) improved risk assessment; 3) holistic understanding of the socioeconomic factors driving societal vulnerability and the full range of strategies available to improve mitigation; and 4) improved use of new information technologies to disseminate warnings and provide information on disaster reduction. Finally, in coordination with existing programs, such as the National Mitigation Strategy, the National Earthquake Hazards Reduction Program, and the modernization of the National Weather Service, the plan calls for new efforts, including a National Risk Assessment and an Integrated Natural Disaster Mitigation Information Network.

Natural Disaster Reduction is now available via the World Wide Web at http://www.usgs.gov/sndr/report. Individual copes of this 50-page report are also available from William Hooke, National Weather Research Program, National Oceanic and Atmospheric Administration, SSMCIII, Room 11360, 1315 East-West Highway, Silver Spring, MD 20910; (301) 713-0460, ext. 218; fax: (301) 713-0666; e-mail: bhooke@rdc.noaa.gov.

Presenting PPP 2000

Public Private Partnership 2000 (PPP 2000) is a cooperative endeavor of the federal agencies comprising the Subcommittee on Natural Disaster Reduction (SNDR--a subcommittee of the Committee on Environment and Natural Resources of the president's National Science and Technology Council), the Institute for Business and Home Safety (IBHS--formerly, the Insurance Institute for Property Loss Reduction), and other private-sector organizations. The partnership is committed to reducing deaths, injuries, property damage, economic loss, human suffering, and detrimental environmental impacts caused by natural disasters.

The goal of PPP 2000 is to seek new and innovative opportunities for government and nongovernment partners to work together to reduce losses from and vulnerability to natural hazards. One of the group's principal means of doing this will be a series of forums on public policy issues affecting natural disaster reduction. In each forum the partnership will seek a wide range of ideas and opinions in order to identify common agendas and determine ways to resolve scientific, technical, economic, and policy issues to reduce disasters. The aim is to improve risk assessment, risk management, and emergency response.

The first forum, hosted by IBHS, will introduce concepts and a framework that will be common to all subsequent forums. All PPP 2000 forums will be held in Washington, D.C.; the first two are: Natural Disaster Reduction Initiatives of the Insurance Sector--September 10, 1997 (host: IBHS); and The Uncertainty of Managing Catastrophic Risks--December 11, 1997 (host: Catastrophic Risk Management Project, Wharton School, University of Pennsylvania).

Additional forums will address such topics as cities at risk, a domestic and international broadcast media partnership to support natural disaster reduction, reduction of earthquake vulnerability in California: 1998-2003, assisting communities to deal with vulnerability to natural hazards, global perspectives on natural disasters, a disaster recovery business alliance, real-time monitoring and warning for natural hazards, natural hazards safety and reliability of utility and transportation systems, grass roots mitigation awareness and education, a national risk assessment, and demonstration studies for natural disaster reduction.

For more information about the PPP 2000 project, contact the Institute for Business and Home Safety, 73 Tremont Street, Suite 510, Boston, MA 02108-3910; (617) 722-0200; fax: (617) 722-0202; WWW: http://www.iiplr.org; or Walt Hays, U.S. Geological Survey, 955 National Center, Reston, VA 20192; (703) 648-6711; fax: (703) 648-6747; e-mail: whays@usgs.gov.

Some Communication/Information Initiatives . . .

SNDR Establishes Working Group on Natural Disaster Information Systems

The Subcommittee on Natural Disaster Reduction (SNDR) (see the articles on the previous page) has recently created a new Working Group on Natural Disaster Information Systems (NDIS). The group's goal is to "evaluate and foster ways to integrate public and private resources and infrastructure to ensure that the most accurate and timely technical information regarding natural disasters is available instantly to everyone who can take action to save lives, reduce damage, and speed response and recovery." The group is made up of 17 representatives from the spectrum of federal agencies dealing with natural disasters. For more information, or to contribute information to the group, contact Peter Ward, U.S. Geological Survey, 345 Middlefield Road, MS 977, Menlo Park, CA 94025; (415) 329-4736; fax: (415) 329-5163; e-mail: ward@andreas.wr.usgs.gov.

NRC/CSTB Seeking Information on Disaster Communication Needs

The information needs of disaster managers are extraordinary, especially in times of acute crisis, and information technology developed for ordinary business may well be inadequate under such demanding conditions. Inspired by the positive reception by the disaster management community of its report Computing and Communications in the Extreme (see the Observer, Vol. XXI, No. 1, p. 23; the report is now available on the World Wide Web at http://www2.nas.edu/cstbweb), the Computer Science and Telecommunications Board (CSTB) of the National Research Council (NRC) is now working to understand this issue in greater depth and to identify possible solutions.

In the broadest sense, the board is seeking to build intellectual bridges between disaster managers and the information technology research community. Hence, anyone--particularly emergency management professionals--wanting to be informed about CSTB plans in this area should send his or her name, organizational affiliation, address, phone and fax numbers, and e-mail address to the board: fax: (202) 334-2318; e-mail: cstb@nas.edu.

Moreover, anyone who can identify operational needs that are not being met by today's information technology is asked to contact the board and let the NRC know what those needs are. Stated differently, the board is asking, "What is your wish list? What information technology would you want to have during a crisis?"

ERLink Created to Ease Electronic Information Exchange in Disasters

The U.S. Federal Response Plan outlines the coordinated roles of federal agencies when responding to significant disasters. The plan consists of 12 "Emergency Support Functions," and Support Function Two--"Communications"--is headed by the National Communications System (NCS). To improve that function, and to support emergency communication generally, NCS is hosting a pilot program called Emergency Response Link (ERLink).

ERLink offers key organizations a way to share information electronically, simplifying disaster response planning and operations. It employs a collection of communications protocols, information structures, and software that use World Wide Web technology for data sharing. Thus, ERLink provides a simple "point and click" method of retrieving all sorts of information, from text, to graphics, to databases, to modeling tools.

ERLink is a controlled access Web site that allows Federal Response Plan partners to both obtain and post information. The system is currently being tested by several federal agencies to determine if ERLink can truly support the community's information requirements. To obtain more information about ERLink, contact program manager John O'Connor; (703) 607-6130; e-mail: oconnorj@ncs.gov.

[Adapted from Aware Report, Spring 1997, a newsletter of the National Oceanic and Atmospheric Administration]

Introducing the UCLA Center for Public Health and Disaster Relief

Recently, the Los Angeles County Board of Supervisors awarded funding to the University of California at Los Angeles (UCLA) School of Public Health for a two-year project, beginning July 1, 1997, to develop a curriculum on the public health aspects of disasters. This curriculum will address the interdisciplinary roles of public health professionals in preparing communities prior to a disaster and assisting them following a mass population emergency. This program, combined with an existing research project on community response to the Northridge earthquake, will form the core of the new multidisciplinary UCLA Center for Public Health and Disaster Relief.

The director of the center, Steven J. Rottman, has stated that the center's objectives will include:

This effort is the first significant collaboration between a major educational institution--the UCLA School of Public Health--and a major local health agency--the Los Angeles County Department of Health Services--to address the effects of natural and human-generated disasters on public health.

For further information, contact Loc H. Nguyen, Program Coordinator, UCLA Center for Public Health and Disaster Relief, P.O. Box 951772, Los Angeles, CA 90095-1772; (310) 794-6646; fax: (310) 794-1805; e-mail: locn@ucla.edu.

Introducing the Greig Fester Centre for Hazard Research

The Greig Fester Centre for Hazard Research, sponsored by the major reinsurance broker, Greig Fester International, is the first multidisciplinary natural hazards research group in the United Kingdom. The center was officially launched June 1, 1997, and is housed in the Department of Geological Sciences, University College London. The center incorporates staff from 10 departments within the university, who are engaged in a broad range of natural hazards research, including studies of volcanic eruptions, earthquakes, floods, windstorms, and landslides. Center researchers have recently attracted over two million pounds in research funding from the European Commission, the U.K. Research Councils, and other sources. Current projects include: analysis of Atlantic sea-surface temperatures for hurricane forecasting; formulation of digital elevation models to identify active, potentially seismogenic, faults; determination of mechanisms of formation and transport of long run-out landslides; development of a PC-based hazard simulator for training civil authorities; and mitigation of volcanic risk among vulnerable island communities. For more information about this new center, contact Bill McGuire, Director, Greig Fester Centre for Hazard Research, Department of Geological Sciences, University College London, Gower Street, London WC1E 6BT, U.K.; tel: +44 (171) 419 3449; fax: +44 (171) 388-7614; e-mail: ucfbkwg@ucl.ac.uk.

Green Cross UK Takes on Environmental Disasters

When news of a disaster breaks, everyone's first concern is naturally humanitarian. It may be some time before attention and funds turn to the effect on the environment, even though this impact could eventually cause more suffering than the original event.

Recognizing this problem, Green Cross UK, established in 1994 and based at Kingston University, is focusing on the prevention and mitigation of environmental damage due to disasters, and the group is establishing the Green Cross Disaster Response Network as one of its principal projects.

Funded by the European Commission and undertaken in collaboration with Kingston University and others, including the Oracle Corporation, the Disaster Response Network is intended to prevent or mitigate environmental damage following any catastrophe by providing accredited environmental experts to organizations responding to a disaster. The experts will go to the scene in the earliest stages, assess impacts, point out the environmental risks and associated humanitarian risks, evaluate needs, and recommend actions to mitigate or prevent further damage. The network will provide people who can also organize and execute those actions.

For practical and economic reasons, this project will probably use experts from the U.K. (or Western Europe) in its development and trials, although the aim is to provide a model for adoption globally. The network experts will need to have knowledge and expertise in several areas, including:

Besides experts who can assess and mitigate the environmental consequences of disasters, the network will also acquire, manage, and distribute information (including appropriate environmental impact assessment procedures, action plans, material lists, and equipment databases) that experts can draw upon during an emergency. Again, the overall aim is to avert unnecessary long-term damage to the environment.

Because the Environmental Response Network is just starting, the organizers are seeking information on previous work in this area--especially any assessment tools or plans that have already been developed. They are also still identifying the qualities, expertise, and experience, as well as the supporting information and tools, that network experts will need to succeed in particular environmental disaster situations. Finally, they are interested in learning about any earlier work to create a logical and manageable framework for managing and storing the large body of information concerning environmental disasters that this undertaking will require.

Any information, comments, or suggestions are welcome and should be addressed to Nicholas Fickling, Green Cross UK, Kingston University-Reg Bailey Building, Penrhyn Road, Kingston-upon-Thames, Surrey KT1 2EE, U.K.; tel: +44 181 547 8274; fax: +44 181 547 7980, e-mail: greencross@kingston.ac.uk, or n.fickling@kingston.ac.uk; WWW: http//www.kingston.ac.uk/~ad_s611/gc.h tm.

The Green Cross currently includes 16 national organizations worldwide, with many additional countries now preparing to join. All are linked to Green Cross International, founded by Mikhail Gorbachev in the wake of the Rio Summit and based in Geneva. Information about Green Cross International is available from the World Wide Web: http://www.gci.ch.

Next Page

Return to Index of the Natural Hazards Observer

Return to Hazards Center Home Page