A Unified Approach To Probabilistic Risk Assessments for Earthquakes, Floods, Landslides, and Volcanoes

Proceedings of a Multidisciplinary Workshop held in Golden, Colorado
November 16-17, 1999
Sponsored by the U.S. Geological Survey Urban Hazards Initiative

U.S. Geological Survey Open-File Report 01-324
2001

Compiled by Aldo V. Vecchia

Report Options

View or Download Entire Report as a pdf file

View Introduction

View Contents

View or Download

The full report is available in pdf format and is 2.2 MB in size.  Click HERE to view report.

If you have Adobe® Acrobat® or Adobe® Acrobat® Reader® installed on your computer, you may view and print the PDF version of this report. Acrobat Reader, is a free download it from Adobe Systems, Inc.

Return to Top

Introduction

In November 1999, the U.S. Geological Survey (USGS) Urban Hazards Initiative sponsored a workshop to foster communication among experts in probabilistic risk assessment for earthquakes, floods, landslides, and volcanoes and to determine if a more unified framework for assessing risk from the different hazards can be developed. The workshop was held November 16-17, 1999, at the USGS Building on the Colorado School of Mines Campus in Golden, Colo. The following scientists participated:

1Workshop coordinator.

2Representative of Urban Hazards Executive Committee.

The first day of the workshop consisted of a series of presentations aimed at describing the current tools and techniques used for hazard investigations and risk assessment. The purpose of these presentations was to increase awareness of the different approaches that can be used for assessing hazards and associated risk and to stimulate discussion of the similarities and differences that exist among the different disciplines (earthquakes, floods, landslides, and volcanoes). The consensus of participants after the first day was that, although the language of probability theory is a common thread among hazard investigations, numerous differences exist among the specific approaches used for assessing risk. These differences exist among the different disciplines as well as among different applications within the same discipline. Imposing a rigid recipe for risk assessment in every application clearly is not advantageous. However, a great deal of progress can and should be made in unifying the products, such as hazards maps, the USGS provides for risk assessment and in improving the way hazards are communicated to our customers.

During the second day of the workshop, participants were divided into three panels with representatives from each discipline on each panel. Each panel was asked to discuss the similarities and differences in techniques among the disciplines and suggest areas in which the USGS can make its hazards investigations more unified and consistent among the disciplines. The panels also discussed emerging areas of research that may have significant consequences on the way in which risk assessment is done in the future.

A synopsis of the presentations given on the first day of the workshop is provided in the next section, and general results of the panel discussions on the second day of the workshop are summarized in a later section. A summary of the more technical aspects of hazard assessments is given in appendix A, abstracts of the presentations describing the current tools and techniques used for hazard investigations and risk assessment are given in appendix B, and selected papers are given in appendix C. 

As indicated in the title, this workshop dealt with the general subject of probabilistic risk assessment for natural hazards. Risk assessment quantifies the potential economic and social impacts (loss of property, life, etc.) resulting from natural hazards. Hazard assessment, which is a necessary precursor to risk assessment, deals with evaluating the likelihood of occurrence of events, such as earthquakes or floods, that may have serious economic and social impacts.

Most USGS products currently being produced relate to hazard assessment. However, the customers of USGS hazard assessment products are individuals, government entities, and cooperators who perform risk assessment. Therefore, the ultimate goal of this workshop is to make USGS hazard-assessment products more useful to risk assessors.

The term “hazard” is used in this report in two separate contexts. In the first context, hazard is a generic term for earthquakes, floods, landslides, and volcanoes. In the second context, which is standard in the technical literature on probabilistic risk assessment, hazard is synonymous with the probability of occurrence. Thus, for example, landslide hazard refers to the probability of occurrence of landslides, which are a particular type of hazard. Clarifying this terminology up front will hopefully avoid confusion on the part of the reader.

Much was accomplished from the workshop, and participants should be commended for an impressive effort.  However, as expected, 2 days was not nearly enough for the group to explore all the issues involving risk assessment for earthquakes, floods, landslides, and volcanoes, much less recommend a master plan for unifying the disciplines. Future efforts should be made to build on this workshop and further enhance cooperation between the disciplines.

Return to Top

Contents

Introduction

Synopsis of presentations and related discussion

Flood-risk assessment

Landslide-risk assessment

Earthquake-risk assessment

Volcano-risk assessment

Overview of panel discussions and recommendations

Discussion topic 1--What are the similarities and differences among earthquake-, flood-, landslide-, and volcano-hazard assessments?

Temporal and spatial distribution of source events

Hazard assessments

Discussion topic 2--Which U.S. Geological Survey products are beneficial for risk assessment, and how can products be improved or new products developed to support risk assessment?

Discussion topic 3--Which research topics are most important for the future of earthquake-, flood-, landslide-, and volcano-hazard assessments?

Modeling temporal and spatial dependence of source events

Reducing spatial uncertainty in predicted consequences of source events

Development of improved short-term hazard prediction capability

Discussion topic 4--How can hazard and risk assessments be made more unified among earthquakes, floods, landslides, and volcanoes?

Establish a national flood hazard map

Develop a “location-free” measure of flood magnitude

Develop a common data base and base map for all hazards

Contour probabilities, not magnitudes

Quantify regional versus point-wise risk

References

Appendix A--Technical aspects of hazard assessments

Appendix B--Abstracts of presentations

Appendix C--Papers

Characteristics of floods as a spatial process (Brent M. Troutman and Michael R. Karlinger)

Landslide risk assessment (William Roberds)

Probabilistic landslide hazard assessment (Randall W. Jibson)

Fuzzy sources, maximum likelihood, and the new methodology (David M. Perkins)

Estimation of volcanic hazards related to tephra fallout (Charles B. Connor and others)

Probabilities of volcanic eruptions and application to the recent history of Medicine Lake volcano (Manuel Nathenson)

Hazards mapping: A need for guidelines (Ute J. Dymon) 

Return to Top

U.S. Department of the Interior || U.S. Geological Survey Maintainer: Webmaster Last update: Wednesday, 23-Feb-2005 09:04:57 EST Privacy || Disclaimer|| Accessibility || FOIA URL: http://nd.water.usgs.gov/pubs/ofr/ofr01324/index.html