Edited Version of March 25, 1998 Transcript
EIIP Tech Arena Online Presentation
Director of NIBS Multihazard Loss Estimation Program
EIIP Tech Arena Moderator: Amy Sebring
The original transcript of March 25, 1998 online Tech Arena discussion is available in EIIP Virtual Forum Archives (http://www.emforum.org). The following version of the transcript has been edited for easier reading and comprehension. Typos were corrected, date/time/names attributed by the software to each input were deleted but the content of questions and responses are as stated by each participant. Answers from the participants to questions by the audience are grouped beneath the appropriate question to facilitate meaning.
Amy Sebring: On behalf of the EIIP, I am pleased to welcome you to a special event in our Tech Arena. Before I introduce our special guest, I would like to review how to use links to display Web pages in another browser window for the benefit of our newcomers.
When a full URL is typed in the message area, it becomes a hot link, so you can just click on it, and a webpage will display in another browser window. Your browser may display behind the chat window, so you may have to look for it. Make sure you don't accidentally close the EIIP chat login window, or you will be disconnected from the conversation. If that happens, please log back in using a slight variation on your name.
I will put up an opening screen URL so you may take a moment to size and arrange your windows so that you can swap easily between windows.
Amy Sebring: This is a moderated session and our guest will take questions or comments about 30 minutes into the hour. When you type in your question, it does not appear on everyone else's screen until submitted by the moderator. One more reminder --- please do not send direct messages to the speaker or the moderator. We are very busy trying to keep things running smoothly. If your question does not get answered during the session, you will have a chance in the Virtual Forum afterward.
Amy Sebring: And now, it is my pleasure to introduce Philip Schneider, Director of the Multihazard Loss Estimation Program for the National Institute of Building Sciences (NIBS) who will tell us about their multi-hazard loss estimation methodology, and HAZUS. Mr. Schneider, welcome and thank you for being here today.
Phil Schneider: My presentation today is about HAZUS, a standardized nationally applicable earthquake loss estimation methodology implemented through PC-based geographic information system software.
HAZUS is an essential element of FEMA's National Mitigation Strategy to promote sustained action to reduce long-term risk to people and property from earthquakes. HAZUS also will assist local governments in facilitating short-term recovery through emergency preparedness in response to earthquakes. Reducing earthquake losses begins before the earthquake. Loss estimates provide land use and development agencies with a basis for planning, zoning, building codes and development regulations, and policy that would reduce the risk posed by violent ground shaking and ground failure. Loss estimates can also be used to evaluate the cost effectiveness of alternative approaches to strengthening hazardous buildings. Understanding the scope and complexity of earthquake damage is essential to effective preparedness.
HAZUS can forecast damage to buildings, casualties, and disruption of utilities. These estimates can be the basis for developing emergency response plans and for organizing tests and exercises of response capability. A rapid response to a damaging earthquake will reduce life loss, complications to injuries and secondary damage and loss, and will expedite relief to victims. Response and relief agencies can use HAZUS to project damage, loss, and the number of homeless and to estimate what financial and material resources will be necessary to assist victims. Regional, state, and federal officials can use HAZUS estimates of casualties and injuries to project demand on medical resources. HAZUS is being expanded into a multihazard methodology with new models for estimating potential losses from wind (hurricanes, thunderstorms, tornadoes, extra tropical cyclones and hail) and flood (riverine and coastal) hazards.
FEMA's National Mitigation Strategy recognizes that mounting dollar losses cannot be adequately addressed by a fragmented approach to natural hazards. Instead, estimated losses for other hazards are needed to support FEMA's risk-based approach to mitigation and emergency preparedness, and comprehensive mitigation programs by local communities.
The Multihazard Loss Estimation Program under which HAZUS is being developed is sponsored by the Federal Emergency Management Agency (FEMA) through a cooperative agreement with the National Institute of Building Sciences (NIBS). The Cooperative Agreement began in September 1992 and is ongoing. The National Institute of Building Sciences is a Congressionally-authorized 501(c) 3 nonprofit. Our mission is to facilitate the introduction of new technology into the building process and disseminate nationally recognized technical information.
The project is overseen by technical committees for earthquake, wind and flood. Committees are responsible for the detailed technical oversight of project work for a particular hazard, assist in the selection of contractors for methodology development, calibration and pilot testing and oversee the formulation of project findings and conclusions.
User workshops will be scheduled as needed to assist in the development of user requirements and objectives for the wind and flood methodologies and test HAZUS software products. The first flood workshop will be held in Phoenix, AZ on April 13 - 15. The first wind workshop will be held in Miami, FL on May 11 - 13. Participation is by invitation only.
Private contractors are used to accomplish the technical development of the methodology. Risk Management Solutions of Menlo Park, CA is the developer of the HAZUS earthquake methodology and software. Dames & Moore of Seattle, WA conducted the first pilot study in Portland, OR, and EQE of Irvine, CA, the conducted the second in Boston, MA.
Applied Research Associates of Raleigh, NC has just been selected to develop the wind methodology. A contractor to develop the flood methodology will be selected in late May.
Now, I will discuss HAZUS. Earthquake loss estimates are forecasts of damage and human and economic impacts that may result from future earthquakes. They are not precise predictions, but rather estimates based on current scientific and engineering knowledge.
The FEMA HAZUS loss estimation methodology is a software program that uses mathematical formulas and information about building stock, local geology and the location and size of potential earthquakes, economic data, and other information to estimate losses from potential earthquakes. HAZUS uses a geographic information system to map and display ground shaking, the pattern of building damage, and demographic information about your community.
HAZUS is comprised of two major parts: an inventory data collection module and a methodology that estimates losses from potential earthquakes. The inventory data collection module provides guidance to emergency managers, planners and local officials in collecting inventory data required for regional multi-hazard (earthquake, wind and flood) loss estimation. Its major software features are data fields for building and lifeline attributes and tools for importing data.
The major features of the earthquake loss estimation methodology are a default data base, census tract level of resolution, classification systems for information on buildings and lifelines, mathematical formulas for calculating earthquake damage and losses, and displays of earthquake loss data and results including maps, tables & summary results for use in publications.
Estimating earthquake losses with HAZUS is performed with the following steps:
Identify the study region. Any combination of census tracts, counties or states (up to eight) can be used in a study site.
Identify the size and location of the earthquake. Three types of earthquakes may be selected for a scenario: a historic earthquake, a probabilistic earthquake (500, 1,000 or 2,500 year recurrences), or an arbitrarily chosen earthquake. Localities are expected to consult with local geologists when selecting an earthquake magnitude and epicenter.
Once you identify the study site and location and size of a hypothetical earthquake, HAZUS will estimate earthquake losses to provide vital tools to government decision makers including: A map-based analysis of the potential intensity of ground shaking from a postulated earthquake that identifies those parts of your community that will experience the most violent shaking and the structures at greatest risk of damage. Such a map would be a useful tool in land use planning and facility siting decisions.
An estimate of building damage that provides the basis for establishing programs to mitigate or strengthen buildings that may collapse in earthquakes. Estimates of damage and life loss will help in setting priorities for retrofit or abatement programs. An estimate of damage to buildings and utilities and of casualties, which provides the basis for emergency response and contingency planning at the local, regional, and state level. An estimate of casualties and homelessness, providing a basis for medical and relief agency preparedness and response. An estimate of dollar losses that will help the state and federal governments plan for assistance to jurisdictions and disaster victims.
HAZUS also estimates many other types of loss including: loss of functionality of hospitals, and fire and police stations, the amount of damage to transportation systems, debris generation, the number of people displaced from their homes, and losses to the regional economy.
I have provided three maps to illustrate the types of results HAZUS produces. The maps were generated based on a hypothetical Magnitude 7.0 earthquake situated under Indianapolis, IN. (A Magnitude. 7.0 earthquake is theoretically possible anywhere in the Eastern U.S.)
The first map shows the distribution of peak ground acceleration (PGA) by census tract. The smaller census tracts indicate proximity to the downtown. The lower PGA bound of .22 is roughly equivalent to MMI VII. The upper PGA bound of 1.18 exceeds MMI XII. Thus, this would be a very catastrophic event.
In HAZUS, one can overlay inventory features such as hospitals or airports to see where they lie in relation to ground shaking intensity. Please display the first map.
Phil Schneider: The second map shows the levels of moderate damage to the residential building stock. There are four levels of building damage described by HAZUS: slight, moderate, extensive and complete. Generally, a study site will have greater percentages of slight damage and less of extensive. Not surprisingly the pattern of damage follows the PGA map. Forty percent moderate damage in the central area of the map fairly severe. Please display the second map.
Phil Schneider: The third map shows the dollar loss to the residential building stock which is very high. If the entire building stock were included, the losses would be on the order of $20 billion. Losses for the entire event exceeded $27.5 billion. In HAZUS, these maps may be customized by changing colors, fonts, legends and other features. Please display the third map.
Phil Schneider: The loss estimation methodology is available through several products: HAZUS97, the earthquake loss estimation methodology CD-ROM based GIS software including a MapInfo version (east & west) and an ArcView version (east & west).
HAZUS stands for hazards U.S. MapInfo and ArcView are the two GIS platforms HAZUS uses. The east edition covers the U.S. east of the Rocky Mountains, the west version, west of the Rocky Mountains. InCast, inventory collection and survey tool on CD, the HAZUS tutorial on CD.
The user's manual that describes how to perform loss estimates with HAZUS. The technical manual (three volumes) that describes the theory behind the loss estimation methodology. It is written principally for scientists and engineers. The validity of results HAZUS produces has been tested through calibration with data from the Northridge, Loma Prieta and Whittier earthquakes and pilot testing in Portland, OR and Boston, MA.
HAZUS can be used at three levels of complexity. Level 1 uses HAZUS default data to create rapid impressions of the type of damage that an earthquake scenario may produce. All the information you will need to produce this rough estimate is included in the HAZUS software. Default data from national databases, describes the geology of your region, and the building inventory and economic structure of your community.
A Level 1 study can be performed with relatively low level computer skills. Although results are crude, they may be used to convince a local government council to appropriate funds for further study.
Level 2 utilizes user modified default data and user supplied data to achieve more refined results. To produce a Level 2 estimate of losses, you will have to provide detailed information about local geology, an inventory of buildings in your community, and data about utilities and transportation systems.
Level 2 implies the user is conversant with computers and GIS. Assistance from geotechnical and structural engineers may be required for this analysis. These results may be used for to develop definitive plans for mitigation, emergency planning and response and recovery.
Level 3 uses expert supplied techniques to study special conditions of the study site, e.g. a potential dam break, exposure to tsunami, or a network analysis of the electrical lifelines system.
Level 3 implies the use of geotechnical and engineering experts with the ability to input specialized software routines in HAZUS. The results may be used to supplement the information provided in Level 2.
Hardware and software requirements for running HAZUS are as follows: minimum Pentium IBM compatible computer (the faster the better), gigabyte or greater hard drive, CD-ROM drive, color screen, color printer, Windows 95 or greater, and Mapinfo 4.1.2 or ArcView 3.0 or greater.
The HAZUS earthquake model will be updated tentatively in July when HAZUS98 is issued. Some of the expected technical enhancements HAZUS98 will include are: a revised nonstructural loss estimation, a revised transportation lifelines bridge damage functions, and optimization (acceleration) of HAZUS for use in response and recovery. HAZUS98 will operate with the latest versions of MapInfo and ArcView.
In conclusion, HAZUS has been introduced to emergency managers in most of the 50 states and U.S. territories in four workshops held last year. Workshops for FEMA personnel and others are ongoing. A HAZUS home page will be available shortly. Technical support is available to HAZUS users. I am now ready for any questions.
Amy Sebring: We have quite a few questions lined up for you Phil.
Vernon Adler: Did Risk Management Solutions of Menlo Park, CA develop its earthquake methodology and software independent of or in concert with the Consequences Assessment Tool Set (CATS) module for Earthquakes and will other CATS modules be utilized as a 'touchstone' for the future HAZUS developmental activities for other than earthquake events? Please explain the relationships and different uses of these various modules.
Phil Schneider: I am not that familiar with CATS. You need to ask FEMA what its plans for HAZUS and CATS.
Sam Winningham: It seems to me that the system is only a useful tool if it has the necessary detailed data. If we have not previously collected the data in the type of format that is useful for HAZUS, then someone is going to be required to make an investment in the data collection. Who will that be?
Phil Schneider: The answer to this question depends on what you intend to use HAZUS results for. If you want results for use in a hazard awareness program, Level 1 is sufficient and requires no data collection. To develop mitigation plans, Level 2 results should be utilized. However, data for Level 2 will have to come from local agencies.
James Cohen: Regarding collection of building inventory data, current work in New Jersey and New York City is seeking to use high schools, summer interns and other similar sources to involve the public. Additionally, there is the possibility of using non-profit organizations such as AmeriCorp and the American Society of Civil Engineers membership for data collection.
Phil Schneider: The state of Vermont is effectively using Americorp for data collection.
Chip Hines: How responsive to actual events is the system? If you've done the ground work with engineers, can you quickly provide the strength and epicenter based on an actual event and have the system respond to the new data?
Phil Schneider: Absolutely yes to Chris Hines question.
Ron Bolyard: I am very interested in the Wind and Flood Modules when they are completed. Cost?
Phil Schneider: It takes roughly five years to development a hazard methodology. FEMA and NIBS are negotiating how to speed up the process. As for the cost, it is in the multimillions.
Amy Sebring: Can you tell us User Cost for HAZUS CD's?
Phil Schneider: HAZUS products are free to all federal, state and local users. We will sell HAZUS98 at cost to private users.
Sudha M: Is the ArcView version available already? If so where can we get it from? We already have MapInfo but would rather use Arcview.
Phil Scheider: An ArcView version of HAZUS is available, also at no cost. Just contact NIBS at 202-289-7800 to request a copy.
Keith Bea: What challenges do you expect in the transition from an earthquake to a flood and wind model? Are you moving toward an all hazards model?
Phil Schneider: One of the challenges is utilizing as the technology developed for the earthquake model in the wind and flood models to save costs. Our goal for the next three to five years is to have integrated models for earthquake, wind and flood since these represent the greatest dollar loss. In the years ahead, we will consider urban wildfire.
Sudha M: To create a study region, you can only extract data at the census tract level, or county or state level. What about at the level of a city? Does one need to know each census tract that comprises the city and then extract?
Phil Schneider: Generally yes, but if the city boundaries are the same as the county boundaries, you have no problems. Census tracts do not cross county boundaries. We faced this problem in Portland, site of the first pilot study. Their urban growth boundary conformed to one of the county boundaries. However, on the edge of the urban growth boundary, population and building stock in census tracts was sparse, so it was easy to average across the boundary.
Amy Sebring: Phil, would I be correct in summarizing that the wind and flood modules will be worked on simultaneously by different contractors and that the process will take at least 3 years?
Phil Schneider: That is correct.
Amy Sebring: OK, one more question.
James Cohen: What is the methodology used for developing the building damage functions? How easy is it to modify these for local variations in design and construction?
Phil Schneider: Building damage is defined through a pushover/fragility curve relationship to the spectral parameters of the scenario earthquake. To modify damage relationship, you have to modify the fragility curves in HAZUS - which can be done.
James Cohen: True, but this will not be useful when considering wind and flood.
Phil Schneider: For wind and flood we will have different sets of damage functions. For example, most wind damage occurs to a building's roof and cladding. In the earthquake model, damage is principally to the building structure. Nonstructural damage for wind and earthquake are also very different.
Amy Sebring: Thank you very much Philip for a most informative session. We will have the transcript of today's session posted in a few days with the background material. Thank you audience, and since our time is up, we will close down the Tech Arena for today, but we will be in the Virtual Forum room for a few minutes longer, and you are welcome to join us there for open discussion.
Phil Schneider: I appreciate the opportunity and enjoyed it very much.
BREAK - SECOND HOUR - INFORMAL -'BROWN BAG' SESSION:
Further Discussions and Questions in the Virtual Room - After additional expressions of appreciation to Phil Schneider, a few participants stayed for a few minutes with more comments and discussion related to the Tech Arena presentation; the following are excerpts from the Brown Bag session that convey additional information about "HAZUS".
Amy Sebring: Anyone out there using HAZUS?
Joe Fletcher: Amy, Utah is using HAZUS. The person to talk to is Bob Carey at 801/538-3784.
Amy Sebring: James, you seem somewhat knowledgeable. Are you using HAZUS?
James Cohen: I am involved with HAZUS through NJOEM and at FEMA DC through a fellowship with ASCE and IBHS.
Neil Blais: Phil, excellent discussion. I was curious about the BIT tool. Initially the tool would not allow a user to carry the LAT/LONG from their data set if they had one. Has that been modified?
Phil Schneider: Neil, the BIT Tool is scheduled to be overhauled for HAZUS98.
James Cohen: Ann Karemijian did some work with cost benefit analysis using HAZUS. I am interested in trying to expand HAZUS to allow for this. Is there any work being done in this area?
Phil Schneider: James, I spoke to Stephanie King just yesterday about using this work in HAZUS. The cost-benefit model is accessible from a set of FEMA publications.
James Cohen: For information, there is a HAZUS initiative through NJOEM, NYCEMO, NYSOEM for the New York City metropolitan area. Rutgers University, New Jersey Geological Survey and Lamont Doherty Observatory are all involved.
Phil Schneider: Evansville is the site of a disaster resistant community project. HAZUS is an integral part of their work.
James Cohen: I understand from FEMA that work with HAZUS is also being done in Evansville, Indiana and in Kentucky Emergency Management.
Cindy Rice: KY Disaster And Emergency Services is currently with an university for data collection to use with the HAZUS program; they have been collecting for about 4 years I think.
James Cohen: Do you have contact information for Stephanie King?
Phil Schneider: Yes, 650-725-0360.
Neil Blais: Phil, is the B/C tool you mentioning the model developed by Goettle?
Phil Schneider: Yes
Neil Blais: Are there plans to utilize HAZUS' regional capabilities, and expand the B/C tool to a more regional tool rather than site specific?
Phil Schneider: Neil, I just started looking into the B/C tool. The important thing now is to assess its usability in HAZUS.
Amy Sebring: Phil, were you aware of the Canadian effort, and how is that similar or different?
Phil Schneider: Chris Tucker demonstrated his model to me and Claire Drury of FEMA about 1 1/2 years ago. At that time the Canadian model addressed hazards in four or five major cities only. And, it only generated results for canned scenarios. I am not sure whether it has evolved beyond that.
Avagene Moore: This has been an excellent session --- remember tomorrow night at 8 PM EST --- Claire Rubin and ICMA reps meet at the Round Table for an informal discussion. Thanks to everyone. Please join us when you can!