The Australian Journal of Emergency Management Marvborouah- Bushfires: On Mondav 14 Januarv 1985.. a dav, of total fire ban throuohout- the State of Victoria, a large number of serious fires occurred in country areas, continuing in some cases into Tuesdav, Armv,. oersonnel were emoloved. . in fuel reduction activities (to contain the fires and limit losses). In the Maryborough Region alone the fire took one life, caused 100 other casualties, destroyed 101 homes and other inhabited dwellings, devastated hundreds of farm and holiday propenies and killed approximately 40.000 sheep.

Cover: Winning entry of the 'Emergency Volunteers in Action' phorqraphic competition, 'Devastation' by Angela Trapani, Senior Photographer with The Shepparton News. The shot features Victorian Country Fire Officerz Peter Martini, cooling-off after battling a house fire in Shepparron East in DKember 2001. Contents Vol 17 1 No 3 1 November ZOO2

Please note that contributions to the Australian loumal of Emergency Managemenr are reviewed. Academic papers (denoted bv- GB- ) are peer reviewed to appropriate academic standards bv The Australian independent; qualified expea. Journal of Emergency Training can be a recruitment and retention 4 Management tool for emergency service volunteers Vul 17. No 3. November 2W2 I55N 1324 1540 This article profiles a research study of Tasmania's Volunteer Ambulance Officers indicating training may be PUBUSHER The Aurrrol~anJournalo~Emprgcncy Manngemm a strateeic- recruitment and retention solution hel~ine. - to s the officul journal of Emergency Management stahilise Tasmania's emergency rural health workforce. Australra and s the nauon3 mart htehlv rated

management industry in Auaralta. It provides The implementation of the incident control 8 acces to information and knowledge for system in NSW. Span of Control and an active Emergency Management research community and practitionen of Management by Objectives Emergency Managcmcnt. Jennifer Bean examines the implementation of EDKOR-IN-CHIEF management by objectives (MBO) and span of control David Templeman. Director General. (SO0 concepts in the context of the Incident Command Emergency Management Australla. System in NSW AlEM ADVISORY COMMITTEE Louisa Graham. Oke of Emergent). Sernces Commissioner. Melbourne Flood warning reviews in North America 17 Christine Jenkinson. Emergency and Europe: Statements and Silence Management Australia John Handmer examines the main themes, issues and Roger Jones,TEM Consulunts. MI Macedon gaps in the many recent reviews of flood warning Char Keys. NSW State Emergency Service systems in North America and Europe are presented Robert Lee, Emergency Management Australla Graeme Nicholas. Queensland Emergency in this paper. Services David Parsom. Sydney Water Corporation. Sydney Perceptions by Counter Disaster personnel of 25 Michael Tarrant. Erncrgcncy Management Australia the risks of bulk transport of dangerous goods along the Brisbane-Gladstone tmponcorridor PUBUSHER & EDITORIAL TEAM Grey Worldwide Canberra This article presents the results of research conducted in Manager, Mark Gadlrey Queensland with Counter Disaster personnel stationed Editorial Coordinator, Anita Cleaver along the Brishane to Gladstone road-rail conidor. Researcher. Margaret Wallis Design and Typesetting by Whlubang An CIRCULATION Perceived change in risk of narural disasters 34 Published on the law day of August. November. Februan, and Mav each war Over 5.000 coo~u caused by global warming Chris De Freitas examines the extent to which the expectation of the increasing frequency and severity COPYRIGHT of extreme climate events is justified. Matonal in The AustralianJnrml olEmprgmry Manaprnr is protected under the Commonwalrh Copyright Act 1968. No material may be reproduced in pan or in whole without witten consent Imm the copyright holders The views in this journal are Post Disaster Surveys: experience and methodology 39 not necessarily the vlews al Emergency David King questions whether there is a right or Management Auaralu. standard research methodology for rapid response SUBMISSIONS post disaster studies. The Autral~anJournnlojEm~rgmcy Monngema welcomes submlsslons Please contact us on ajcm@ema gavau for a copy ol our Contnbuton' Guldellnes Antcles for pubhcauon may be fonvarded to The Australian Journal of The Politics of Emergency Response and 48 I Recovery: Preliminary Observations on Taiwan's 921 Earthquake This paper presents preliminary results of a study on the organisational and institutional response to the 921 or Chi Chi earthquake. FOREWORD Lessons from violent tragedy Rob Cameron discusses the impact of the Bali terrorist attack on Australia's emergency management arrangements

The observations after the domestic arena. Any likelihood of the 11 September attacks that prompted need for a multi-agency, multi- Rod McKinnon's argument for the jurisdictional and possibly greater integration of Crisis and international response indicates our Consequence Management have been arrangements must be effective and reinforced by experience and national well understood by all those likely to ellort following the Bali bombings. be involved. The requirement for With the overriding concern of all ongoing inter-agency contacts and involved being the rapid provision of partnerships entails an appreciation of appropriate care to those injured and the interdependencies of respective the welfare of the families and friends agencies' roles. of all victims and casualdes, effective To ensure the experiences and coordination, cooperadon and observations of the Bali response to communication between all the date are captured and made available agencies involved was a premium to the crisis and consequence requirement. As noted earlier, the management communities, EMA has location of the bombings outside commenced a series of debriefs. The In the Summer 2001-2002 edition of Australia, together with the numbers Candle lighting first was held at EMMt Macedon ceremonyat the this journal. Rod McKinnon, the of casualties from each Australian campus on 8 November and focussed Director of Planning and Operations jurisdiction meant that agencies who largely on opportunities for Service 'Or at EMA discussed the implications for may not have had such a prominent vicrims of the Bali improvement in inter-agency linkages. atrack Australia's emergency managemem role in an event that occurred inside Preliminary feedback indicates a need arrangements of the terroria attacks our national borders were now very the G~~~ Hall of to develop, implement and exercise a parliament ~ouse, on the World Trade Cemre in New important contributors to, if not national plan for the management of Canberra, Australia York and the Pentagon in Washington managers of, key phases of the incidents involving evacuation and on 24 October on I I September 2001. He high- response and recovery processes. 2002. lighted the opportunity for Australia's repatriation to Australia of large In each phase of the Australian numbers of injured and deceased. emergency managers lo review our response to the 12 October bombings The next acdvity in the debrief process arrangements and the level of there was and still are a significant will be in Danvin in December and preparedness for events of a similar number of public and private agencies will focus more closely on learning nature and foreshadowed a wries of with responsibilities to discharge and from the experiences and observations lessons learned seminars to be hosted contributions to make. The agencies of those more closely associated with by EMA with guen speakers arranged and companies involved accomplished the initial response. The reports from from the United States. some remarkable achievements the EMA-convened debriefs will The violent terrorist bombings in Bali For example, by early Wednesday contribute to Federal Cabinet5 decision on 12 October tested Australia's following the Saturday night bombing. for a whole of government review of capability for managing disaster all Australians requiring hospitalisation the response to the bombings to be events impacting on Australians had been evacuated and were completed by early 2003. travelling overseas. At the time of receiving care in Australian hospitals - writing, the explosions and subse- a noteworthy ellon. As the response Learning from the Bali bombings quent fire ou~sidePaddy's Bar and the condnues and the recovery work needs to be reflected in our plans and Sari Nightclub in the popular tourist gathers momentum, it is becoming arrangements while the experience is destination of Kuta in Bali killed increasingly clear that this event has current and our observations fresh. 67 Australians with the Department of far reaching consequences for families We have an opportunity now to use Foreign Affairs and Trade holding and communities, and that a sustained the experiences of those involved in very serious concerns for another 20 multi-agency recovery effon will the response to and recovery from who remain unaccounted for be required. both the 11 September attacks and Indonesian authorities continue to the Bali bombings to ensure that advise that the total death toll is So what can we learn where required, our arrangements expected to be approximately 180. from this tragedy? are improved. We need to learn from This is the first time Australian Since the attacks in the United States violent tragedy citizens in such numbers have been on 11 September. Australia has victims of deliberate terrorist acts remained on a heightened security Rob Cameron is Ading Assistant Diredor. outside our national borders, and the alert. The bombings in Bali confirm Owelopment (Pmjects). Emergency extra-temtorial dimension has added the assessment that the alert is valid Management Australia a layer of complexity and uncertainty and is not limited to the Australian Training can be a recruitment and retention tool for emergency service volunteers

University of Tasmania researchers argue that training may give incentives for recruitment and retention of Tasmania's emergency rural health workforce.

VAO are imponant not only for the provision of By Christine Fahe)! Professor ludi Walker emergency senices, but also health senices in most and Associate Professor Adrian Sleigh rural and remote Australian towns, where there is total - University Department of Rural Health, reliance on VAO for first-line response to medical University of Tasmania emergencies. Those living in rural and remote areas have poorer health indicators than those living in This paper reports the training findings metropolitan areas, but of particular note is the rate of avoidable deaths (particularly from injury, asthma, and of a larger study of Tasmania's Volunteer suicide), which is 40% greater than in urban areas I Ambulance Officers, the first-line response (Bryant and Strasser 1999). VAO form an important first link for rural areas in Tasmania, Western Australia. to medical emergencies in rural and remote South Australia. Northern Tenitory, and Queensland. areas. They are a dwindling resource in an Unfortunately, lack of specific information about factors isolated state with a great need for such that boost recruitment and retention of rural ambulance oKicers makes it dillicult to plan strategic solutions to services due to its large rural population and the looming crisis (Hudgings 1988, Swan 1991, Federal numerous tourists. The project surveyed all Emergency Management Agency 1995, Halpin 1998, Tasmanian Volunteer Ambulance Officers and Reinholdt and Smith 1998. Aitken 1999). Tasmanian Ambulance Senices, like other emergency senices, have then conducted 10 focus groups. We found experienced growing dificulties with VAO recruitment that training is important to VAO. It is not a and retention. It is important that Tasmania. with a high disincentive. If done well, training will be a proportion of rural residents (41%). develop and maintain a skilled rural ambulance workforce. strategic recruitment and retention tool and will help to stabilise Tasmania's emergency A compounding factor is that Australian emergency senices have recently initiated national competency rural health workforce. This research has a standards for training volunteers, and the Tasmanian wider application for emergency services as Ambulance Se~ceshave been quick to use this approach. The aim is to increase the standards of they undergo similar changes and pressures training to ensure organisations meet 'a duty of care' related to training volunteers. (Howard 1999), but the accompanymg increase in training and accreditation requirements could be a deterrent to potential volunteers (Federal Emergency Over the past two decades Australia has dismantled its Management Agency 1998, Aitken 1999). trade barriers and successfully joined a globalised economy But over the same period rural areas have To address this rural health workforce problem in entered a period of economic decline and rising Tasmania we gathered statewide data from volunteer unemployment with high social costs. Thus it is not ambulance officers regarding recruitment, retention, surprising that rural Australia now finds it increasingly training and support. This study was the research difTicult to recruit and retain volunteer ambulance component of a larger project (More than a Band-Aid) officers (VAO). In both the USA and Australia there is a financed by Emergency Management Australia to devise worrisome fall in such emergency senice workers strateges to improve the situation (Fahey and Walker (Morisey 1993. Fitch 1994, Reinholdt and Smith 1998) 2001). We surveyed all Tasmanian VAO, and then and the 1994 Senate Standing Committee Report on conducted ten focus groups yielding both quantitative Disaster Management in Australia (EMA 1999) stressed and qualitative data. Here we report our findings on the that volunteers need to be encouraged and supported if substantial potential of training as a strategic recruitment response capability is to be maintained. and retention tool. Training connectr self-interested needs with altruistic urges to assist the communify

Then we chose 10 focus groups on the basis of a A questionnaire was developed with advice from an representative mix of VAO from different unit types advisory committee of major stakeholders, and minor (such as remote or semi-urban), including three groups changes were made aher piloting. We included items on formed among VAO geographic subsets with low VAO length of senice, age groups, sex, motivations. response rates. Overall, representatives from nineteen recruitment promprs and training attitudes. The ambulance units were consulted through the focus questionnaire was mailed to the entire population of group component of the study and all were asked to VAO in Tasmania. estimated at 380 at the time of survey, define issues affecting VAO and to explore strategies to with a 55% (206) response rate. improve recruitment, retention, training and support.

After coding the open-ended items on the question- naire we processed the data using SPPS software. Responses were analysed for frequency and compared according to age group, sex, geographical location and length of se~ce.Ordinal categories with five levels were collapsed to three levels before testing relation- ships. Each focus group resulted in extensive field notes and the texts were analysed to identify common issues and themes.

Results Tasmanian VAO closely resemble other Tasmanian and Australian volunteers (Australian Bureau of Statistices 2001) in age (most are 3145 years old) and sex (54% female); however more Tasmanian VAO are employed (84% versus 44-58%). and fewer have completed tertiary study Most respondents had been VAO for five years or less (Figure 1). Assisting the community was rated as an imponant respondents (76.6%) wished to upgrade to the next skill motivation by 94% of questionnaire respondents; 94% level. Increasing age correlated with less interest in also acknowledged that they receive benefits by gaining upgrading qualifications: 87.5% of 18-30 year olds new skills and over 50% reported various social benefits wished to upgrade but only 46% of those over 60 years (Table I). Many VAO activities were reported as old had similar desires enjoyable, especially training and skills maintenance, helping people and group friendships (Table 2). The majority of respondents were satisfied with current The major factors that made volunteering diflicult training delivery methods. Ninety-two per cent found were lack of time (33%) and inadequate provision of training sessions interesting and 83% agreed that resources (29%). training sessions usually have all the necessary equipment. Delivery of training by electronic methods was accepted. with 61% being aware of the benefits of video and teleconferencing and 54% having access to the Internet. But only 18% of respondents used the training information available on the Internet. Important (%) i Focus group findings revealed a strong consensus about / Asssting the Community 94 what is important to VAO. They felt a of lack adequate t I Learning New Skills 94 support, particularly those located in remote regions. t Specifically, the VAO identified that one of their priority A Sense of Achievement 90 1 needs was training, universally complaining of Being Part of a Group 58 inconsistent provision. This problem was described as Gaining Self-esteem 57 most serious in remote groups, and irritating in units f Meeting New People 55 with a paramedic attached. The remote island groups had the least training over the last two years with only Improving Employment Prospects 33 ; one professional training session delivered in the last Other 10 twelve months, and many last minute cancellations due to 'stafl shortages'.

Units with a paramedic attached had more consistent training, but these groups also suffered from last-minute cancellations if the paramedic was called out to an emergency This might happen for several training I Count* % t . . sessions in a row, causing difficulties for the paramedic I Training &skills maintenance 206 2€ to cover the required training topics and sign ofl all the 1 Helping people, patients 175 2i logbooks. Last-minute training cancellations were felt to [ ~riendship& being member of a group 131 11 lessen enthusiasm for attending training. j Call-outs, on road 68 9 The irregular availability of the training modules ! I Social events necessary to improve qualifications was another identified problem. There were many VAO who wished [ Meeting members of the public as a VAO 39 5 to complete Level 1, 2, and 3 qualifications. The 1 Job satisfaction & sense of achievement 24 3 I availability of these modules was felt to be very irregular Case discussion & medical interest 23 3 in two of the regions, creating frustrations for VAO ' Other 19 3 wishing to gain national accreditation. Significantly, i a new recruit could only act as an observer until they 18 2 / Adventure & challenge had completed Level 1, and yet they had difficulty t Driving, work around the station 18 2 accessing the relevant modules. One recruit said "I've i Coordinating operations & leadership 1 7~ 2 been waiting twelve months and I still can't do anything because I'm only an observer". Other VAO identified IGe=in;qualifici~ons that some new members had left because they were tired improving employment prospects 2 0 1- -. of waiting for Level 1 training. In addition some groups 1 Total 790 101 admitted that they did not want to recruit new members / 'Count = no. of timer activity was cited as enjoyable if there was no training to offer them.

Overall, focus group participants indicated that poor Most respondents (94.5%) felt that training did not take training implementation made them lose confidence in too much time and was usually run at a convenient hour their ability to respond appropriately to emergencies. (84.5%). Seventy-eight percent of respondents spent less Maintaining confidence emerged as a major reason for than thirty minutes to reach the training location. Most training with one respondent stating that "lack ojtraining paradigm for at least half of the 20th Century (Marwell and Oliver 1993). However. more recent opinion embraces the altruistic and civic engagement motivations for volunteering that are often cited by volunteers themselves (Bell 1999; KO 1999). It is probable that most volunteers are motivated by both self-interest and altruism, acting to reinforce each other. This fits well with the evident desire for education: training connects self-interested 'learning new skills' with altruistic urges to 'assist the community'. The new skills have little purpose or use in ambulance service unless they are used to 'assist the community', and VAO ax unable to provide adequate emergency senices to the community without new skills. These two motivations reinforce each other; our focus group data strongly support this view as participants feared being inadequate in emergency situations and desired training to ensure competency

When individuals choose to volunteer they have many opportunities within the community to provide services. Most of our respondents identified an interest in medical care and first-aid as a major reason for joining an ambulance service instead of another volunteer service. Volunteer Ambulance Officers are unable to provide adequate emergency semces to So ambulance senices now know that highlighting the community without new skills. provision of appropriate training will be a recruitment leads to lack of interest, lack qfskills and competence, loss of tool, offering a window into the medical world for a sense ojbelonging and personal confidence, which all interested community members. reduces volunteer retention rates." Once volunteers are recruited. ambulance services must retain them. Retention of current volunteers benefits Discussion organisations by retaining expertise. Many groups in the Our use of both quantitative and qualitative data USA are realising that retention is a major part of the collection methods provided information that was both solution to falling volunteer numbers (Federal wide and deep, an 'enlivened' account of volunteer Emergency Management Agency 1995. 1998, Halpin concerns. Other advantages include triangulation of 1998). Our data reveal that retention is a problem in measurements for increased internal validity, and Tasmania as most VAO remain for less than five years. additional flexibility in sampling (Fahey and Walker Emergency services understand that retention means 2001). The sampling method and geo-demographic ensuring volunteers are happy with their role, and use attributes of our respondents suggest the data represent incentives, recognition and reimbursement as strategies well the Tasmanian Volunteer Ambulance Officers. The to ensure this. (Hudgings 1988, Swan 1988, 1991, similarity of our results on motivation and volunteer Federal Emergency Management Agency 1995). profiles to those noted in other studies (Australian But training is not mentioned in this context. Bureau of Statistics 1995. Aitken 1999) mean that these data enrich and support that body of knowledge, with However, we found training should be considered not implications for other volunteer emergency services. only as a recruitment tool, but also as a strong retention Overall we learned that adequate training is very tool. Training and related activities were by far the most important to VAO. It is unfortunate that many frequently stated activities enjoyed by respondents. Poor emergency senice planners do not consider training a provision of training also formed a strong theme from recruitment and retention tool, but more a means for the focus group data and the final open-ended services to ensure 'duty of care'. When emergency comments section of the survey If 'keeping volunteers services are looking to attract and retain volunteers happy' means providing them with activities they enjoy training may be their greatest weapon. and avoiding things that upset them, then ambulance senices can deploy a powerful retention tool by Volunteer recruitment literature stresses the importance providing quality training. of understanding volunteer motivations and successful recruitment methods. Volunteer motivations are the Appropriate training should be high quality, flexible, focus of much debate, with the belief that volunteers timely, and meet set standards. Despite time commit- operated largely from 'self-interest' being the dominant ments being listed as the major difficulty in involvement in VAO activities, VAO were clearly willing to volunteer References their time to attend training, as many respondents Aitken A,. (1999) ldentifylng key issues affecting the retention identified that training did not take too much of their of emergency senice volunteers AutralianJoumal of time. When volunteers commit time to attend training Emergency Management, 14: 16-24, they don't deserve inconsistency and irregularity. Australian Bureau or Statistics, (2001) Voluntary Work, Australia, Australian Bureau of Statistics. Canberra. Providing appropriate training for VAO requires a Auaralian Bureau or StatLstics. (1995) Voluntary Worh: substantial financial investment, and ambulance services Australia. Australian Bureau of Statistics: Canberra. may be wary of overtraining volunteers who may not Bryant D Land Strasser P R., (1999). The delivery of stay within the ambulance service. Training, however. sustninnble rural and remote heolth services. Regional Australia is a key incentive to emergency service volunteers as it Summit. 27-29 October 1999. Canberra. ensures they are competent, and assists them to feel Emergency Management Australia.. (19991, Record ofthe confident. Services should acknowledge that VAO are Dennis Mileti Workshop 25-26 August 1999, Emergency Management Australia. Mt Macedon. not a free service. The cost required to sustain the Fahey C and WalkerJ.. (2001). Asking Volunteers - Volunteer VAO workforce is timely, consistent, quality training. Ambulance Oflcers have their soy! The More than a Band-Aid Such training has benefits that far outweigh the actual Project Repon, University of Tasmania: Hobart. cost of not providing it -high staff turnovers, limited Fahey C and WalkerJ.. (2001). 113 more than numbers! The skills and dwindling emergency capacity The question advantages of mixed-method inquiry in volunteer now becomes 'Can ambulance services afford not to organisations AurtrolianJournal on Volunteering. 6: 49-53. provide adequate training for volunteers? Our study Federal Emergency Management Agency.. (1995). Emergency Medical Services Recmitment and Retention Manual, Federal indicates that the answer is 'no' - any savings arising Emergency Management Agency: Virginia. from cutbacks or cancellations of VAO training would Federal Emergency Management Agency. (1998). Recruitment be illusoly and Retention in the Volunteer Fire Service, Pmblems and Solutions, Federal Emergency Management Agency: Virginia. Conclusion Fitch 1..(1994). Endangered Species: Volunteers. Emergency The role of VAO in health and emergency services is Medical Services, 23: 3E45. important because of the numbers reliant on this Halpin T. (1998). Recruitment and retention of volunteers: Find 'em, keep 'em. Obstacle or opportunity?Joumal of service, and the health profile of the rural population. Emergency Medical Services, 25: 5&72. The utility of training to VAO is an important finding as Howard 8.. (1999). Managing Volunteers. AustralianJournal 01 ambulance services are experiencing difficulties with Emergency Management, 14: 37-38. recruiting and retaining volunteers. Tasmanian Hudgings E.. (1988). Volunteer incentives; Solving Ambulance Service appears to have developed a recruitment and attrition prablems.Journal ojEmergency satisfactory training formula, with national standards Medical Services. 13: 58-61 and training plans, but is failing to deliver it regularly. Manvell G and Oliver P. (1993). The Critical Mass in Collective Action: A Micm-Social Theory, Cambridge University Press: Cambridge. Emergency services must not be afraid that training MoriseyJ.. (1993). 'Here to stay!': Recruiting EMS volunteers. standards are a deterrent to potential volunteers. It is Journal of Emergency Medical Services, 18: 53-58. clear that appropriate training will act as a powerful Reinholdt S and Smith P. (1998). Directions in Volunteer recruitment and retention tool for VAO if provided in a Development in Australian Emergency Services, Country Fire timely and suitable manner. Poorly delivered or Authority and Emergency Management Australia: Mt. constantly re-scheduled training is a disincentive to Waverly, Victoria VAO and the same principles are likely to apply for Swan T. (19881, Keeping Volunteers in Senrice. Journal of Emergency M~ediialServicer. 13: 55. other emergency service volunteers. All emergency services aiming to recruit and retain volunteers should Swan T.. (1991). Great Expectations: What do EMS volunteers want from management?Journal of Emergency Medical Services. investigate the flexibility, quality and timeliness of 16: 47. the training they deliver to ensure it reinforces the Swan T.. (1991). No-cost benefits and incentives. Emergency, motivations of their workforce. Barriers to provision 12: 21. of such training need to be identified and removed...... Christine Fahey is a nune with a Master of Public Administration and Acknowledgements an interest tn wdl.atnon and rommm ty parlmpaLon As a rseaaher at toe Unwenq Depanmenl of R~ralrlealtn (UDRH) The More than a Band-Aid Project was undertaken with Talman a me 0 d tnls vrork on vol~nleerAmo~ ance On

- --- -~ - all types and complexities' (Irwin 1989). NSW fire By lennifer Bean, University of New England. agencies are currently using ICS, having adopted the I ~. . . .. ~ ~ - 7 system in 1987. Interwoven with ICS are the concepts of Jennifer Bean conducted research using Management by Objectives (MBO) and Span of Control (SoC) in as much that they are the principles that the interviews and questionnaires on the imple- Australian version of ICS is based upon. mentation of 'management by objectives' and 'span of control' in the context of the lncident Management by Objectives is 'a process of consultative Control System (ICS) in NSW. The research found management where the management team determines the desired outcomes of the incident. These outcomes or that objectives and strategies are being objectives are then communicated to those' (AIIMS, developed for evely fire and communicated 1994). MBO is important therefore, so that people know effectively to the lncident Management Team where they are, why they are there and what direction is (IMT) but not to the field. In addition, the being taken during the operation. objectives being developed are not useful or Span of Control is a concept that relates to the number meaningful and the process of reassessing of groups or individuals which one person can objectives and strategies is not well understood successfully supenise (AIIMS. 1994). The inference here by the IMT. The lncident Action Plan (IAP) is being that if the SoC is exceeded supenision will be compromised and therefore the safety and effectiveness often not perceived as being relevant or useful of subordinates is compromised as well. to the incident, and preplans are not used extensively at incidents. Logistics are being Investigation of these two concepts can provide useful insights on how ICS is implemented in NSW This paper duplicated by agencies rather than having one look, at their adoption in NSW and suggests function for an incident, and duplication of tasks recommendations for the future. is occurring between the planning and operations sections. Span of Control is well Methodology understood and adhered to but not PerIormance indicators were derived from the principles proactively reassessed. of ICS. Each performance indicator was given a 'measure' or 'measures', that were used to provide a Many of these issues can be resolved with focus or benchmark for research and analysis. Data was gathered on the measures, and conclusions were made interagency traininghefresher exercises and on whether the performance indicator was achieved or standardised procedures (such as briefings, not based on this information (Table 1). development of the IAP and planning meetings) to encourage agencies to implement ICS in the Various techniques were used to collect data. The two questionnaires were used to obtain quantitative data to same manner. identify trends in implementation. They were sent to incident management team participants and crew The lncident Command (Control) System (ICS) is 'a set leadedcrew members, Interviews were used to follow of personnel, policies, procedures, facilities and equip- up the trends from IMT members and obtain more ment, integrated into a common organisational structure qualitative, detailed information on specific subjects. designed to improve emergency response operations of Operations Plans were analysed for inter-agency PERFORMANCE INDICATORS MEASURES METHOD Management by Objectives Joint planning of incident IAP is written and used IMT Questionnaire .Objective is communicated Field Quenionnaire .Preplanning IMT Interview Operations Plans Unified command .One organisational structure. Field Questionnaire one lncident Control Centre. one planning process, one logistics centre and one communications framework. -- . .------. . - - - -- . .- ~. Span of Control Unity of Command Everyone has one boss. IMT Questionnaire Field Questionnaire Duplication is avoided No duplication in tasks. IMT lntelviews arrangements and Rural Fires Act (1997) Section 44 control lines by 2100 in order to protect Brown's village'. repons were used to analyse the management of Cowardin (1984) refers to this as 'action centred individual fires. Management by Objectives (MBOY, where an lncident Controller takes available information, and based on The responses were dominated by the Rural Fire Service their experience, sets a single overall strategy including and National Parks and Wildlife Service. NSW Fire an objective for the emergency Brigades participated in the research to a lesser extent because they are primarily an urban fire fighting Often a person will not remember a piece of information organisation. State Forests declined to participate in because it is not directly relevant to them personally or the research. their situation, or is not 'action centred'. 927 situation repordincident action plans for the period 199711998 1. Management by Objectives -199912000 were analysed. The objectives that occurred Joint planning of the incident: Communication in this period can be summarised into categories, these of the objective are shown in Table 2. For management by objectives to be effective, the Table 2 shows that not one objective was set with time incident objective needs to be communicated to all and space parameters and an intent. The majority of agencies and personnel involved in the incident. Crew objectives (34%) were generic statements that do not leaderskrew members were asked if they knew the convey the incident specific goal. These generic incident objective. 59% responded yes, 41% responded statements can be considered as the overarching mission no. The results were identical when asked if they knew statements for all fire fighting, and are included in the the strategies. It is evident from these results that the BFMC Operations Plans as 'aims'. 27% of the objectives incident objective is communicated to the field only provided some details and guidance in that they half the time. As Management by Objectives requires contained what was to be done within a space that all participants on an incident should know the parameter. 26% of the objectives however, only outlined objective, this principle is not being implemented as what was to be done without any details on the well as it should. parameters or intent, and 11% of the objectives weren't The lncident Management Team (IMT) results contrast objectives at all, but strategies. The lack of detail and with the field results. 100% of respondents said that an direction in these objectives makes them irrelevant and objective was set for the incident. 93% of respondents therefore difficult for subordinates to remember. Perhaps knew the objective and 7% didn't. This shows that the the communication of objectives to the field is incident management team is given the objective and occurring, but the objective is not meaningful enough strategies. However the method of communicating the for the subordinates to use or recall. objective was not researched. The objective and corresponding strategies should be An objective should define what the lncident Controller reassessed regularly for relevance and achievability wants to achieve in a particular timeframe. This with the changing incident context. This process should objective should be realistic and achievable. To do this. be driven primarily by the lncident Controller and the an objective should have time and space parameters and planning section with consultation with the other an intent that is meaningful. An example of an objective sections. IMT members were asked whether the may read. 'to contain the fire within the established objectives and strategies were reassessed. 92% said Said they were reassessed following liaison and briefings Generic TO protect catchment values 34% from the field To proted biodiversity values To protect life and property ldentified IMT meetings To ensure the safety of crews reassessed the objective and strategies Strategy Patrol and mop-up the fire 11% Inspect the fire perimeter ldentified that after weather Establish an accurate plot of the fire reports, aerial inspections and field inspections the What To minimise the area burnt 26% strategiedobjective were To prevent re-ignition reassessed To contain and blackout the fire To prevent escape from the control lines Identified that planning To blackout the control lines meetings reassessed the To secure the control lines strategieslobjective To mop-up the hotspots Said that they were reassessed To monitor the fire as the shift progressed Whatispace To contain the fire within control lines 27% Said that they were assessed To contain the fire to the current area for the next IAPIshift change WhatJtime To contain the fire by 1800 hours 2% yes, 6% said no and 2% didn't know Table 3 outlines asked if an IAP was developed for each shift. 83% said how the objectives and strategies were reported as yes, 14% said no and 3% didn't know The 93% positive being reassessed. response is not surprising, as an IAP is a reporting mechanism required by most agencies and the Rural Table 3 demonstrates that the process of reassessing Fire Senice Commissioner (for s44 incidents). objectives and strategies is not well understood. The lncident Controller is responsible for determining the IMT members were asked if they used the IAP to carry objective, which sets the scene for the incident, and out their role. The results are shown in Table 4. provides everyone on the incident with a 'direction'. Not one person responding to the questionnaire identified the lncident Controller carrying out this role or having this responsibility. 13% Said never The primary forum for reassessing strategies (and the objective) Is via the planning meetings. All functional 4% Said rarely oficers should be present for this meeting, and should 23% Said sometimes analyse the intelligence from the field and other sources 25% Said frequently to assess the success of activities and the relevance of the strategies. The responses reflect the dillerent components 34% Said always and triggers of assessment, but only 12% specifically identified the importance of the planning meetings. Table 4 shows that 40% of the respondents said that they used the IAP sometimes, rarely or never. As the IAP The objective of an incident therefore, is being is the basis for Management by Objec~ives,this response communicated well to the IMT, but less so to members is surprising. It shows that either the role and usefulness in the field. The process of reassessing objectives and of the IAP is not well understood, or that the lAPs being strategies is not well understood. written are not useful and are therefore not being used.

Joint planning of the incident: lncident action The purpose of an LAP is to communicate the objective. plan development and use strategies and support mechanisms provided for a fire. An lncident Action Plan (IAP) is the primary tool used If the plan is not useful and relevant then the purpose of to communicate ihe objective of the incident to all staff the plan is negated. InteMew participants were asked if involved. IMT members were asked whether there was the IAP was useful. 39% responded no. 61% responded an IAP for a nominated fire on which they bad yes. It was reported that the three main reasons for the participated. 93% said yes. 7% said no. They were then usefulness of the plan was that it was a form of In summary there are two primary problems associated with the objective setting process. These are that although the IAP is being written, the communication of the objectives to the field and IMT is not widespread or 56% Felt the IAP didn't reflect the field operations. equitable, nor is monitoring of the achievement of the 33% Felt that the field (or operations section) does objectives and subsequent feedback occumng. what they want regardless of an IAP. Joint planning of the incident: Pre-planning 11 % Felt that the IAP was too prescriptive and outlined tactics Multi-agency pre-planning is an important tool in assisting the establishment of objectives and the development of the LAP (Riley, 1988). Each agency should have a good understanding of their neighbourk concerns and ideals, available qualified personnel and available resources before a major incident. Pre-planning 34% ldentified the ODerations Plan can take many forms, IMT members were asked what 19% Said that there were agency databases and pre-planning was evident at the incident they management plans participated in. The results are shown in Table 6. 17% Said that there was rapport between agencies and interagency contact through the BFMC In NSW, the main sources of information should be the forum, inter-agency training or previous BFMC Operations Plan and agency management plans incidents (e.g. NPWS Plans of Management, Resen.e Fire Management Plans). 13% Said that agency staff were trained and emerienced The Operations Plan is a standard document 9% Referred to internal agreements (such as for each bush fire district in the State, outlining professional officer awards) and standard procedures and protocols, and identifies resources. operating procedures communications and assets at risk. It should be .- 6% ldentified control lines such as trail maintenance routine for the IMT to access the Operations Plan and hazard reductions to identify any issues. In Table 5 only 34% referred to Operations Plans. 3% ldentified the DISPLAN Agency documents oulline area specific information including assets, procedures and relevant legislation. Only 19% referred to these sources of information. These plans should be consulted during incidents in those areas to assist in the rapid identification and prioritisation of tasks. 24% It saved time The BFMC is a forum that can establish rapport 18% The preplanning identified positions and people and understanding between agencies and performs 18% There were non-locals involved who benefited a valuable role in facililating pre-planning. Only from the preplanning 17% reported that these forums fulfilled a pre-planning role. 12% It summarired local information Compiling information into documents for an incident 12% Preplanning assisted team cooperation and lessens conflict by outlining the ground rules is only a valuable exercise if the information and format can be used and accessed rapidly. IMT members were 6% It improved response asked if the pre-plans they used were useful and why 6% People knew their roles 65% said yes, the usefulness of pre-planning is outlined in Table 7. documentation, it gave clear direction and it was a measure of progress and performance. Table 5 outlines The responses in Table 7 are generally referring to the reasons why the 39% that felt the LAP wasn't useful. the Operations Plans. The Operations plans are now required to identify people and roles they can fill during The results in Table 5 demonstrate that when the IAP an incident, and outlines the 'ground rules' for command, is not used it is generally because the information in control and cooperation. the plan wasn't useful (56%). If the IAP is useful, Management by Objectives (using the IAP as a tool) Of the other responses, 10% said they weren't sure if should be implemented by everyone. 33% of the reasons the pre-planning was useful. 10% said pre-planning were that the field 'does what they want' because they was not useful and 15% said there was no pre-planning do not participate in process to set the objectives. evident at the incident. 20% of the respondents This can be remedied with training and use in ICS therefore were ambivalent or negative about the use roles and responsibilities. 01 pre-planning. This result could stem from the in its pre-planning policy for coordination. For the purposes of this research, the American five ICS characteristics (one organisational structure, one Incident Control Centre, one planning procw, one logistics centre and one communications framework) are used as measures of unified command.

It is evident that the ICS roles and structure are being filled. The IMT questionnaire results showed that the five characteristics of unified command were generally occurring at every fire. The particular jobs undertaken varied between responses, however the core tasks were reported as being carried out, such as 'management of the section' for IMT Officers. The survey responses were quite simplistic (often identifying one task). This may be a function of the amount of space provided for respondents (4 lines).

Generally the reporting structure was theoretically National Emergency Management Coordination Centre, EMA. correct. All respondents who were IMT Officers knew that they reported to the Incident Controller. The positions reporting to the IMT Officers varied pre-planning not being in a useful formathot considerably. Some responses included: containing relevant information or from the participant Division Commander reports directly to not understand how pre-planning can benefit incident Deputy Appointee. management. All BFMO have Operations Plans Crew Leaders report to Division Commander. approved by the Coordinating Committee, these plans Crew Leader reports to Operations Officer. should therefore be used as the minimum form of . pre-planning. There is no justifiable reason why these Planning OlTicer repons to Appointee. plans were not present at incidents (15%). Planning Officer reports to Operations Officer. Logistics Officer repons to Planning Olficer. The information provided by respondents shows that Logisdcs Officer repons to Operations Officer. the use of pre-planning is not occurring consistently Although the value of pre-planning is recognised, the Operations OlTicer reports to Planning Officer. pre-planning tools identified were not varied or many 21% of the respondents outlined reporting structures in number. that did not cormpond with the theoretical ICS Unlfled command structure. There was no agency-based trend in these results. This shows that there is either a considerable lack lrwin (1989) defines unified command as a method for of understanding of the reporting structure and roles of all agencies or individuals who have jurisdictional the positions, or that the structure is being adapted to responsibility, or in some cases who have functional local conditions. Either response will have the responsibilities at the incident, to contribute to the: didvanlage of inconsistencies and non-standardisation. determination of overall objectives for the incident, Correspondingly, there could be advantages if the and variations were agreed to by the agencies and were due selection of strategies to achieve the objecdves. to adaptation to local circumstances. It is interesting to note the range of responses from the Irwin (1989) suggests that by all agencies using ICS Incident Controllers. When asked who they reported to. during a major incident (and therefore having the same the responses included: organisational structure, same terminology, same management procedures), unified command occurs. The local Council General Manager; This means that instead of preparing several sets of RFS Regional Manager; plans (with no guarantee of coordination among them). NPWS (District. Region) Manager; only one set need be prepared to inform all participants. The s44 Appointee; In place of several logistical and communications processes, only one system of collective and integrated RosehilVRFS State Operations; procedures is used. The Australian ICS dmnot A 'Higher Authority'; specifically include the concept of 'unified command', RFS Commissioner; and NSW addresses the components of unified command NPWS Park Operations. to have received good accommodation and feeding. while another agency received different arrangements that were perceived to be worse, and therefore not equitable. This is an inherent problem associated with 30% They needed more staff different agencies organising their own logistics. 1 - i 30% They had too many staff to do the jib All changeovers were reported as being scheduled by the 1 20% They accept everything. - initially and then IMT, but organised by each agency by swapping their i reairess the span of control 1 I own resourres with like resources. The IMT in every ; 20% They reassess span of control alwayr instance changed over together regardless of agency and implicitly L~_. _ - . -- J A number of respondents (20%) raised the issue that there are often different shift times for different agencies because of the volunteer-paid firefighter concept. Volunteers work flexible and often different hours compared to a paid workforce with an award. These 33% Reported that if the job was being done. hours and conditions were not researched in this study. the span of control was correct The implications of different hours and conditions of volunteer fire fighters to paid fire fighters (with an 26% Said that the span of control depended on the objectives, strategies and tactics (e.g. mop up award with set working conditions) could be span of control could be 1:8, direct attack span considerable, not only in terms of coordination and of control could be 1:3) logistics of the actual incident, but for occupational 26% Said they used the 1:5 proportion as a basis health and safety requirements. for decision making but twk into account other factors Unified command is therefore being carried out well as a result of the ICS hierarchy and the existing policy 11 % Reported that they kept strictly to the 1:s proportion framework. Establishing one control centre and one planning process is a routine and normal practice. 4% Reported that if they couldn't manage their Areas where there is some variation are related to the staff properly the span of control was likely to be wrong ICS structure, lines of reporting, and logistid changeovers, In addition, logistics and changeovers need to be proactively incorporated into the IMT and This shows that there is some confusion about the organised across agencies. reporting structure above the Incident Controller. Ultimately, an Incident Controller appointed under 544 2. Span of Control of the Rural Fires Act primarily reports to the Assessment of Span of Control Commissioner of the Rural Fire Service, and maintains liaison with all other concerned parties. During interviews. IMT members were asked if they reassessed the Span of Control in their section as a One incident control centre is required by Coordinating manager. 65% responded yes, 35% responded no. Committee policy via the Operations Plans for all Table 9 outlines the reasons that people said that they Class 2 and 3 fires. The control centre is agreed on by reassessed the Span of Control. the local Bush Fire Management Committee in the Operations Plan and is generally the Rural Fire Service Table 9 shows that 60% of the participants reactively Control Centre. All section 44 reports identified that one managed Span of Control, that is, when the situation control centre was used for the major incidents. There was too dilficult or too easy, the Span of Control and mav be some over la^ and confusion of control centres structure was reassessed. 20% reported a philosophy for Class 1IClass 2 fires on public managed land (such of accepting all resources initially and assessing the as National Parks and Wildlife. State Forests) as they are required Span of Control and structure. This approach escalating. This was not researched. may have serious implications for cooperative fire fighting on a statewide basis. Often when there is 53% of interviewees said that logistics were organised on severe fire weather, fires will be occurring across the an agency basis. 47% said that the incident management State, resources will be shared and allocated on an as teamk logistics section organised the logistics for the needs basis. As resources are finite, accepting every- incident. When asked why the agency organised their thing initially will freeze the resources from attending own logistics, the ovenvhelming response was that another fire and contribute to the potential loss of different agencies have different needs and therefore assets. Span of control should proactively assess each need to be organised separately One issue that was shift both on current and predictive basis, because raised a number of times was the issue of 'rich cousin - reactive or opportunistic management may result in poor cousin'. This means that one agency was perceived delays and confusion. Of the people who responded in the negative to the question 'as manager, did you reassess the Span of Control in your section?, 66% said the structure and Span of Control was acceptable and didn't need reassessing. This corresponds to the results that show that the majority of people are reactive in reassessing Outlined that the IMT discusses it, the lncident Controller approves the decision and the logistics Span of Control. 17% said the structure was standard officer does it. and non-negotiable, and 17% said it was someone elseh job to reassess the Span of Control. Reported that they would have contacted their agency to organise it To determine how Span of Control is interpreted and Reported that they would report the proposal to implemented in an incident. IMT members were asked the Operations Officer 'How did you know if the Span of Control was enough? Felt that the planning section organises the Table 10 outlines the results. reductionkxpansion of resources through the IAP At emergency incidents, the environment in which Felt that the field carried out the supervision is required can rapidly change - possibly reductiotVexpansion as required with dangerous consequences. Five reporting groups or individuals are considered to be the optimum, as this maintains a supervisor5 ability to effectively task. monitor and evaluate performance. The results in Table 10 reflect this theoty the majority of people (63% of people interviewed) look at the task and the Responded that the operations section and planning section exhibited duplication of tasks environment and make a judgement on how well subordinates can do the job, how safe the environment Said that it was human nature for there to be is and how well they can supenise them. Managing staff duplication based on a designated ratio is limiting and inefficient, Reported that the deputy Incident Controller the ratio should be used as a guide only. also acted as an operations officer for their own agency To gain an understanding how Span of Control is being Reported that there were too many resources for implemented within the ICS structure, participants were the job asked 'How would you have expandedlreduced the Reported blurring of roles such as agency liaison Span of Control in your section?. The results are and planning officer outlined in Table 11 Reported that there were two operations officers (one for NPWS, one for RFS) Table 11 shows that less than half of the people interviewed understand the chain of command and Said that the field set their own strategies process to expandkduce the Span of Control. The correct method is to discuss the resources required in an 100% of the crew leaderdcrew members knew who IMT forum, the IC to approve the decisions and the they reponed to. Logistics Officer (and Operations OiTicer) to carry out 98% of IMT members reponed to one person. the required actions. The 15% who responded that they The 2% reported that as Divisional Commander would report the proposal to the Operations Officer they reported to the Operations Officer and were all Division Commanders and therefore identified lncident Controller. the correct chain of command. These results demonstrate that the hierarchy of ICS, It is of concern that 23% of the respondents would go particularly in the field, is well understood and via their agency channels rather than the IMT and accepted. This result can be attributed in part to the logistics function. The IMT will never exhibit true level of training people have received in the last five cohesion and elliciency until the IMT carries out all years in lines of communication and chain of functions. Separate agencies duplicating logistical command. The.Rural Fire Service and NSW Fire arrangements is inherently inefficient. Brigades system reinforces this reporting structure with their brigade structures. Unity of Command Unity of Command is important - if every person on Duplication of effort ihe incident reports to one person, and receives one set Duplication of effort can cause inefficienciesin the of instructions, duplication and confusion is avoided. management of an incident, while a structure with This concept is being implemented well at fires in NSW: designated roles and responsibilities can remove the potential for duplication of effort. Although strongly The low reporting rate in the field may be due to linked with the concept of Unity of Command. the lack of a meaningful and useful objective. Over- duplication of effort can occur as a result of many whelmingly, the objectives in this study period were influences. During the inteniew, participants were asked broad and not 'action-centred'. Refresher courses and if duplication of effort was evident in the IMT, 80% exercises should stress the importance of Planning responded yes. 20% responded no. Table 12 outlines the Meetings and exercise the process of IAP development type of duplication that was evident. and briefing procedures.

Table 12 shows that duplication of IMT roles and tasks The process of reassessing objectives and strategies is is widespread. Overwhelmingly, the planning and not well understood by IMT members. In theory, the operations functions showed duplication of tasks (50%) planning meeting brings together the Section Officers including both sections carrying out separate mapping; to discuss and agree on strategies, and the lncident the operations officer not participating in the planning Controller has the responsibility for determining the process and the planning section writing the operations objective. The results from this research failed to plan; resource tracking being done by logistics. identify an understanding of these key responsibilities operations and planning; and having the operations and processes. officer and planning officer both tlying to fulfill the operations function. Although the IAP is being developed and distributed to the IMT, its usefulness or relevance is variable. The key Another strong trend (13% of respondents) alluded to issues are that the information is out of date, it does not operations being divided into agencies (such as the reflect the situation on the ground or what the field Deputy lncident Controller acting as operations olficer officers believe is required. Agencies are generally for their agency, and having operations officers for each adhering to the strategies outlined in the 1AP Specific agency). This may he a method to ensure that command training for the Planning section needs to address the is maintained and cooperation is facilitated, although it timeframes and processes involved in developing an LAP doesn't strictly adhere to the principles of Span of and the operational staff need to he specifically briefed Control. to provide field information to ensure the IAP is based on accurate intelligence. 14% of participants accepted duplication and inefficiency as a part of normal practice. During an Although pre-planning is occurring, the plans are not incident, the lncident Controller should manage the staff being used at incidents. A checklist of 'essential to ensure that staff are carlylng out their roles and that information' needs to be included in the Operations inefficiencies do not occur Plan for the use of the lncident Controller and planning section. Obviously these plans are not included in the Duplication of effort is occurring on a regular basis Operations Plan, but the checklist would ensure that during incidents. The interactions between the different imponant information in pre-plans are not overlooked. sections should be strong and open, if the The checklist could include: communication is not strong, duplication of effort may occur. Duplication of effort is inefficient, it can also Relevant legislation cause confusion and ill-feeling between agencies because Local Bush Fire Management Committee cooperation and trust are not being actively facilitated Operations Plan. through the ICS process. The reasons are not well Local Bush Fire Management Committee Bush Fire known, but may be linked to agency politics and the Risk Management Plan. tussle for control of an incident. Contact lists for all agencies Plans of Management for National Parks in the local Conclusions and Recommendations government area. 1. Management by Objectives Fire Management Plans for National Parks in the local Although the objective and strategies for the incident are government area. being developed by the IMT, they are not effectively State Forest Plans of Management. communicated to the field operations staff. An inter- . Depanment of Land and Water Conservation Crown agency standardised approach to briefing should be Land Fire management Plans. adopted to ensure that the objectives and strategies are Topographic maps for the local government area. communicated to the fireline in a consistent, timely and Vegetation maps for the local government area effective manner. The IAP should be used as the basis Environmental and community assets map for this briefing. Communication of the objective and strategies within the lncident Management Team is Preplanning also includes other factors such as fruitful reported as good. discussion in the BFMC. A list of items for discussion could include: training and experience of current key reactively assess Span of Control rather than assess it as IMT staff, resource availability updates, location of areas pan of a proactive management system. The actual of high fuel loads, fire trail capability, smoke sighting assessment of Span of Control is refreshingly flexible, arrangements, first response arrangements, location of with people focusing on many contributing factors such specific environmental and cultural assets. Although as the level of safety, the task at hand and the many of these issues are outlined in the Operations competency level of staff to cany out the task. The Plans, discussion prior to the fire season could be a 1:5 ratio is being used generally as a guide, not a rule. timely and useful process to raise the members' awareness and focus. Almost one quarter of people felt that they would go through their agency channels rather than a Unified command seems to be working well (i.e. there is IMTilncident ControllerlLogistics Section process to a common operations centre and common planning) but reduce or expand the Span of Control. This process common logistics is not occurring. Currently, for over would generate inefficiencies and confusion, and is alien half of the fires studied, the logistics function is to the concept of Span of Control in that coordination duplicated on an agency basis, consequently logistics is would not occur seen as ineffective and inequitable. A logistics team provided with adequate information on resources and Duplication generally has been reported to be occurring needs, can satisfy most requirements. One of the most as the norm rather than the exception during fires. prevalent responses to not-standardising logstics and The main form of duplication is occurring between the changeovers was the concept of 'different agencies, planning and operations sections. The reasons are not different needs'. As ICS was developed and adopted as a well known, but may be linked to inter-agency politics loo1 to reduce duplication and encourage coordination and the issue of in-control arrangements. For this of fire management, this attitude needs to be discussed reason, agencies may need to discuss these issues on and resolved in a pre-fire season forum at a local and both a local and statewide basis. Further training could Coordinating Committee level. occur to restate the roles of each of the operations and planning sections, and their links with other sections Reporting structures are not well understood. The and reduce the amount of duplication occurring during purpose of a hierarchy is that all aspects of an incident an incident. are being considered and addressed. If the hierarchy or reporting structure has a weak link, the likelihood of References potential failure or something being missed increases. AliMS 1994 lncident Control Syslem: The Operating Systcm oJ Training and exercises are integral to ensuring that AIIMS. AFAC Limited. people respond and perform in an agreed and Cowardin. D.H.. (1990, ICS - Think Regionally. Act appropriate fashion to incidents. Refresher courses and Nationally American Fire jouml. exercises need to reassert what the roles and Cowardin. D.H.. (1984) Action-Centred Management for ICS. Amcrican FireJournal, November 1984. responsibilities are and why the chain of command is Invin. R.L.. (1989). 'The incident Command System (1CSY the way it is. These exercises need to be varied. regular in Disaster Response - Principles oJ Preparation and Coordination, and involve all agencies that may participate in fire ed. E. Auf de Heide. Mosby, Miuri. management. Phelps, 8.. (1996), The lncident Command SystemJor Fire depanments: History oJthe merger oJ FGC and ICS. International 2. Span of Control Fire Service Network, lnc. The concept of Span of Control is deceptively simple. Riley. DJ., (1988) Know they neighbour - the key to unified command. USDA Fin Management Notes. Vol49 (1). In theory it provides a mechanism for coordinating the Rural Fire Service, (1999). Bwh Fire Cwrdinating Commitlee efforts of all areas involved through delegation of Policy 1/99: Managemen1 oJBush Fire Operations. NSW authority and responsibility Indeed, one measure of Coordinating Commitiee. Span of Control, Unity of Command, is being Rural Fire Service, (1998). New South Wales State Bush Fire implemented well, when everyone has one boss, they Plan (An Operational Bush Fire Management Plan) NSW know who they are, they know the lines of Coordinating Committee. communications and authority and they can perform Rural Fire Service. (1998b). Model District Bush Fire Management Plan Operations: punuanl to section 52 oJ the Rural their allocated tasks. Training is required to encourage Fires Act, 1997. Coordinating Committee 30 April 1998. the on-going maintenance of Span of Control during an Rural Fires Act (1997). incident. This training should focus on proactive and . .. . . ~~.~~.~~ predictive resource management, and be targeted at Jmifet Bean wm@v.&s for NNY National Park and Wildlife people who may fulfill the role of a Section Olficer and Service as Senior Ranger (Fire). She has also held positions in the lncident Controller. , NSW Rural Fire Service in planning mles in the Nwihern and Cenval Regiw. This paper k bared w a chapter horn her Master of As managers however, understanding how to facilitate ; Natural Resources thesis titledme implementation of the lncident Span of Control is a little more difficult. IMT members ' Control Spem in NSW. Email: bjtblunden(Dhotmai1mm inte~ewedreport on the overwhelming trend to L - ~ --. - ~ . @ Flood warning reviews in North America and ~uro~e:statements and sileke

John Handmer examines recent reviews of the flood warning systems in North America and Europe overviewing some of their issues, themes and gaps.

It would be difficult enough by distinct from concerns about the By John Handmer; Centre itself, but there are more and more limits of generaliibility of case for Risk and Community people potentially at tisk demanding study research. Safet)! Geospatial Science, warnings. Many are at risk for what RMlT Universit)! Melbourne we might call traditional reasons; If the national -let alone regional - -~-- that is living, working and traveling or local - contexts of the research This paper examines the many in flood prone areas. But an are very different to where it is being applied, then research results recent reviews of flood warning increasing number are at risk because they use such areas for may not be automatically systems in North America and recreation. This group is probably transferable. A large question mark Europe. My purpose is to set out not reached by warnings today with hangs over the applicability to the main themes and issues occasional disastrous consequences. Australia of research results from contained in these reviews. It is Separately, there is an emerging the US in particular (where most of the English language literamre also interesting and important to consensus that the natural phenom- enon of flooding is likely to worsen originates). Yet such research is establish areas and issues not with global warming (Handmer, usually quoted as if it was perfectly covered by the reviews, and I try Penning-Rowsell and Tapsell 1999). transferable to Australia or to identify these areas of silence. elsewhere. A similar comment may This paper examines the many A serious omission of the US and be made about the uncritical recent reviews of flood warning European material is the failure transfer of results within countries systems in North America and to rather different contexts, for to explicitly conceptualise the Europe. The purpose is not to example, the results of research warning task as being about compare these repons with the conducted in rural towns in the US enabling the communities, situation in Australia - although and Australia have typically been some comparisons are made - but enterprises and individuals at risk applied to metropolitan areas to set out the main themes and to take action to reduce their without consideration of the issues contained in these reviews. differences. There has been and risk and property losses. It is also interesting and important still is an assumption of universal to establish areas and issues not Processes are needed to ensure applicability - with the exceptions covered by the reviews, and I try that the needs of those at risk of indigenous communities and to identify these areas of silence. drive warning system design. easily identifiable neighbourhoods which appear to be distinctly Are reports from other different, such as the 'Chinatowns' Flood warning systems need countries relevant to of many major cities. Results from continual improvement just to Australia? long ago are even more problematic. maintain current levels of se~ce. Examining overseas experience Few would disagree that in addition Warnings are increasingly expected raises an obvious issue which is to all the other changes in our by those at risk and they are far too often ignored -that is the societies over the last few decades expected to be timely and accurate. question concerning the applic- our approach as individuals and There appears to be less tolerance ability of lessons and experience societies to risks and hazards has for what are viewed as mistakes - from elsewhere. This is quite changed dramatically, most even though we may regard errors distinct from the concerns raised famously as set out by Ulrich as inevitable given the complexities when the information from Beck in Risk society (1992). and inherent uncertainties surroun- elsewhere would be embarrassing ding the warning task. This is in or show the inadequacy of local This is not the place to examine effect a steadily rising standard. thinking and practice. It is also these problems - I simply want to indicate the need for caution, and to provide a useful analysis of environmentally damaging where suggest that the area is worth emergency response to the development proceeds in floodplain further investigation. European floods during the 1990s areas, and unlikely to produce resilient communities. Reports and other In addition, there are of course sources many projects, reports, web sites Enforcement problems raise and commentaries of relevance, for questions over sole reliance on a For North America the paper draws example see the Boulder site for the regulatory approach -an issue on recent reports from the US US (wwwcolorado.edufhazards) raised in reviews and interviews in Federal Government's National and for Europe, www.MITCH.ec.net, Canada (Emergency Preparedness Science and Technology Council or wwhrwallingford.co.uW Canada and Environment Canada EJective disaster warnings (2000); projects/RIBAMOD/. 2001). UK. USA. France and Sorensen's (2000) review of 'Hazard Poland. These problems suggest warning systems' completed as part Some key themes that approaches relying less on of the US second assessment of Examining many reports and papers enforcement and more on natural hazard research and policy: from many places unavoidably cooperation, as employed in New Kendall Post's (2001) review of means that any summary will be South Wales, may be more "Barriers to effective US alert and general. Here some major themes workable (May et. al. 1996). notification system"; Coping wiihJlosh have been selected that emerge from Jwds (2001) the results of a NATO the material as well as some The emphasis on workshop on the subject (Gmntfest distinctive omissions. and Handmer 20011, and Dennis technology The government warning reports Mileti4 summary volume on the Material on flood warnings can be reviewed for this paper generally reassessment of natural hazards in divided loosely into micro and emphasised the potential of modern the US, Disasters by design (1999). macro issues. There is much in technology and the need to apply it most reports on what can be termed For Europe there are numerous with more fewour. Moa of this the micro issues of warning - national reports and some Pan effort has been directed at message design, psychometric and European studies. The recent improvements to monitoring. demographic factors which affect evolution of warnings in the UK modelling and prediction systems warning response and so on. and establishment of the National (for example see As micro issues are very well Flood Warning Centre following the wwwM1TCH.ec.net) - this is hardly covered by the literature and show 1998 English floods and subsequent surprising given that most warning relatively little change over time, the enquiry (Bye and Homer, 1998) has reports are authored by technical emphasis here is on macro issues. been well covered elsewhere specialists. Apart from this area These concern broader policy and (Haggett 2002; Handmer et, al. much effort is going into large scale system design issues and the applications such as GDlN (Global 2000; Handmer 2001). The enquiry sharing of best practice. Areas of Disaster Information Network) and is mentioned in the text. There have continuing concern within the it is not always clear how this since been a number of further micro category include the need for: investigations following the floods effective processes for community would actually help local warnings of Autumn 2000 (Environment engagement; determining and or response - given that effective Agency 2001). Most of my delivering appropriate advice on management of rapid onset flooding comments on Europe will come depends initially on local warnings what action 10 take on receipt of a from a recent review 'Improving warning; and the identification of and response. All the current EU flood warnings in Europe' which high risk groups and ensuring that funded warning related research examined many European reports they get the warnings they need. projects are essentially about the (Handmer 2001), the NATO application of information and volume mentioned above (Gmntfest Warnings are very much part of communication technology - and Handmer 2001). the results of flood risk management, but are no although they may have other a major European funded study into substitute for effective risk manage- dimensions as well. This is climatic hazards (SIRCH or Social ment. The emphasis should be on important when we appreciate and Institutional Responses to the development of national that some agencies with warning Climate Change and Climatic (or continental) approaches to risk responsibilities claimed to he unable Hazards: floods and droughts.) and management. ElTective approaches to share key data between their own a workshop last September in in one area are often at present offices, let alone across national London which examined current offset by ineffective or non-existent borders. Apart from the use of the issues in European flood policy action elsewhere. This is inequitable internet, large scale fully operational (www.gs.rmit.edu.au/research/risk.h for those at risk, probably econ- examples of the successful applic- tm). Rosenthal and ['Hart (1998) omically inefficient, likely to be ation of modern technology in local The warning task has become harder because of our expanding use of Early warnings enable communities to reduce risk and properry loss. warning and response are rather They relate to the failure to An exception to this rather scarce. The NATO volume properly identify and address the negative assessment is the use of mentioned above contains one residual risk, they touch on issues the web as part of the interface example from Oklahoma which is of determining acceptable risk - between the science of prediction likely to be applied across the US a social and political process, and those at risk. Even here many (Crawford, 2001). This combines and the difficulty in achieving of the web displays do not inform the communication and access improved outcomes. In his review people of their own risk of potential of the web with traditional of warnings in the US. Sorensen flooding, and do not follow good weather monitoring and relies on a (2000) states that: "Better local practice in warning message sound hydro-meteorologcal management and decision making design. In particular, they do not network -something which is about the warning process are tell people what to do in the event under threat in many jurisdictions more critical than promoting more of a warning. An outstanding (e.g. the Canadian review). I know advanced technologies, although exception to this picture is the of no other large scale examples both would help." website of the UK Environment proven to be both cost-effective and Agency (wwwenvironment- performance enhancing. There are It may be that a real pessimist agencygovuk). Locally, the some more local cases, such as the would see that the concentration Traralgon flood warning system Denver metropolitan district which on information and communicadon has been designed to maximise uses a range of technology for storm technologies may make things community access to warning detection and warnings, and the worse by directing effort and information including via the Danish Hydraulic Institute's resources away from the real Bureau of Meteorology's Victorian FLOODWATCH system. problems and issues. We can see flood warning services web site that we have beuer and better (Kazazic at al. 2001). How does the emphasis on modem ability to monitor, detect and information technology match up predict hazards while using similar An urgent policy need may be to against the problems being faced by procedures to warn those at risk as agree on the technical specifications warning systems? I would assert were used thirty years ago. Why do for warnings via various new media that it matches poorly The main we wonder why warning response such as mobile phones, as suggested problems and complaints relate to and effectiveness as measured by by the US report on EJJective disaster human and institutional failures. outcomes has improved little? warnings (NSTC 2000). Key groups at risk tolls in Europe. France in particular warnings and response. Uncertainty Some attention in the repons has tried to implement warning over the timing and occurrence of the reviewed has been devoted to systems at camping grounds, flooding as well as the modelling of groups seen as particularly although there is a conflict between water spread, makes flood prediction vulnerable in the warning context. raising awareness of the hazard and dillicult. Although there are often These are usually seen as people the economic imperative of having complaints about the accuracy of who would have trouble responding people stay at a flood prone camp prediction -especially if it is to warnings. However, these groups ground. Organised groups appear around critical heights for road are relatively visible and include the to be generally aware of flood closures or levee failure - the real housebound, immobile, mentally ill, warnings although their risk related issue is failure to predict flooding and those who cannot hear or read decision-making processes may be altogether, or to communicate warning messages. poor. Like camp-grounds. they can warning messages to those at risk be regulated, targeted and warned before flooding commences. Far less auention is given to less and most would already carry Largely inseparable from this point visible but nevertheless at-risk communication equipment. One is uncertainty over the precise location groups. These include people who risk here is that a withdrawal of ofthe areas likely to be afjected. may not be reached easily by the insurance cover or heavy-handed particularly when dealing with rare normal warning communication regulation may lead to these groups events and unexpected phenomena mechanisms, such as: becoming more informal, less and pathways. Floods may affect visible and further outside formal All those mobile at the critical groups and areas that are not seen warning arrangements. warning time. as hazardous. Even the best hazard Tourists. When considering warning identification process will contain Business travelers (US reports). priorities we should focus on where uncertainties and assumptions, for Seasonal workers (US reports). they are likely to have the most example about what degree and Those who are socially isolated impact. There are two major areas type of flooding to include. such as the homeless and those here: where substantial property Frequendy, areas thought to be low- trymg to remain out of sight losses can be avoided with warnings or no-hazard may still be flooded in including undocumented people and appropriate response; and severe events. Warnings systems (those without permission to be where the risk of large death tolls is need to be designed to cover such in the country). high. To concentrate on the second areas even when they are not area - the groups most at risk identified in advance. Usually Nevertheless, these groups can appear to be organised groups of however, areas can be identified as usually be identified and located adventure recreationists, campers of very low, but nevertheless real, even if only by police patrols, with and vehicle occupants. About half hazard. The question then is how to the possible exception of the last of all flash flood related deaths in ensure their inclusion in warning listed. In practice, during a major the US are people in vehicles system coverage. flood event the emergency senices (Gmntfest and Handmer, 2001). may not have time for this task. Uncertainty over the impact of the Failure to warn organised hazard. Impact is related to the Another group is more problematic. recreational groups leads to timing and magnitude of the flood. This group consists of those who inquiries, lawsuits and in Europe A flash flood sweeping through a occupy high risk locations on a criminal charges. At present, the campground at night will likely casual basis such as bushwalkers, main group missed altogether by result in a high death toll. Flooding cross-country skiers and adventure warnings appears to be independent mid afternoon on a week day will recreationists. If we considered recreationists. To confirm this a gap cause maximum disruption in a recent well publicised flash flood analysis may be useful. major city, and so on. Another deaths in Europe we would add aspect of this is that late at night campers. This group can be sub- Handling uncertainty few people will receive or respond divided into those using official Although not generally set out to warnings with the result that the camp grounds, commercial or other explicitly in reviews and reports, impact could be much greater in organised groups, and independent uncertainty plagues most aspects of the absence of any damage adventure recreationists. flood warning. There are several reducing actions. interrelated dimensions to this, a The largest numbers at risk at few are discussed below. Perhaps more challenging is how any one time are campers and people respond to warnings, once organised groups of recreationists. Many pans of Europe and the US they have received and understood These two groups have contributed are subject to severe flash flooding them, and how this can influence significantly to flash flood death presenting testing challenges for the flood impact. An important issue here is uncertainty over can be confident that warnings are In Australia positive examples are appropriate action. Many warnings improving. Recent papers by provided by some of the regional do not provide advice on Rohrmann (1999) and Handmer offices of the Bureau of Meteorology, appropriate action. For some (2000) examine this issue and the which have worked to ensure that hazards, the advice provided may difficulties surrounding rigorous warning system improvements are increase the risk; for example, assessment. The NATO volume based on community needs evacuations from relatively secure contains discussion of some (Son~ber~et. al. 1999: Kazazic et. premises at the last minute for alternative approaches (Gmntfest bushfires, or through deep or fast and Handmer 2001). Some reviews flowing flood waters. deal with the issue implicitly only Warning system models discussed Here I simply want to highlight and proposed in the reports are Pre-event awareness some different approaches: essentially top-down or agency and education centred. This may he unavoidable, . Outcomes in terms of lives saved as the agencies generally have the Almost all post flood-disaster and property loss avoided. ultimate responsibility for warnings. reports advocate pre-event public The primary output of prediction but may perpetuate warning awareness raising and education. timing and accuracy dissemination and response However, this is no panacea. Assessment of each stage of the problems. If user needs and According to Sorensen's summary of warning process against targets priorities are overlooked or ignored the US experience (2000) "There is such as proportion of audience completely, it is dificult to see how no conclusive evidence regarding reached and time taken to reach warning outcomes will improve. whether or not a public education them. although system performance may or information program actually Satisfaction with warnings by show improvements judged by so makes a significant difference of those at risk. far - while it is clear that there is increasing human response to Warning system design and much room for improvement in warnings." His statement echoes function including the quality interagency coordination. There is much earlier findings published by and reliability of inputs. an ever increasing amount of the Illinois Department of . The principles and assumptions material put in front of the 'public'. Transportation in 1980. He goes on underlying design and operation. much of it more attractive (in the to say that the evidence for for example, is it based on the sense of being positive and enter- effectiveness is mixed, probably needs of those at risk assessed taining) and of more immediate because many programs are poorly through processes of community interea than flood warning engagement? designed or executed. Other reasons education. Furthermore, if material for this statement are that many The focus here is quite properly on does not address community needs assessments are poorly designed or warnings and warning messages it is unclear how it will achieve executed (see Rohrmann 1999): and issued in advance of the flood, but improvements. To address comm- in any case there is simply no much can also be done to reduce unity needs, these needs have to be guarantee of success in advertising, losses immediately after inundation known by those responsible for promotion, or in trying to change through salvage and appropriate warnings. Community engagement people5 attitudes or behaviour. treatment of flooded items. is also necessary to manage public This gloomy assessment is not a However, virtually no guidance is expectations of the warning system. available to those wishing to do reason for giving up, however. The At a more general level, engagement this. For businesses, continuity current British public awareness with the communities at risk is a planning provides a proven and preparedness program being fundamental part of the risk developed by their National Flood approach to damage limitation. assessment and management Warning Centre suggests one novel process of which warning systems approach to building awareness and Institutional design - form part. Similar comments could behaviour change using a strategic task focused be made concerning the commercial campaign over a ten year period. The material reviewed rarely sector Most warning systems also The program is based on the social mentioned the real purpose of involve the mass media, at least, marketing approach well developed warnings: that is to assist those at if not other private or autonomous in the field of health promotion risk. Where it yas mentioned, it groups, and these groups should he was in the context of micro issues pan of warning system design. Measuring success such as message wording. The This is an ongoing critical issue. fundamental problem of designing Another and very imponant aspect Without a clear and agreed warning systems so that they meet of an agency centred approach is the approach to assessing success, or community needs was ignored. way the links between the informal failure, it is not easy to see how we An exception may be the UK. and the ofTicial are ignored - the attempting to meet this need on a continental wide basis.

Partial exceptions to this general lack of material on process include the US NSTC (2000) recommendation for a public- private sector partnership to leverage government and industry needs, capabilities, and resources to deliver effective warnings. But there is no recommendation on processes to achieve this beyond suggesting that it could be in the form of a not-for-profit corporation. Separately, the report suggested that working groups of stakeholders should be established to support warning system improvements. (See also wwwpartnershipforpublic warning.org0. The inquiry following the British floods of Easter 1998 recommended some institutional changes including the formation of a national flood warning centre - which is now well established. Conclusions and implications Areas thought to be low risk should be included in warning system coverage. For the US, Sorensen summarises progress with flood warning over informal in this context being the last 20 years with the words: Processes for "not much improvement", although peoplek personal networks, implementation information sources and their he assesses that there is "some For a long time it has not been priorities. No official report improvement" for the prediction1 difficult to find statements on what discussed this issue, but European forecast component. This is a little a sound warning system should research highlights its importance harsh as the death toll appears to achieve, what components it should and potential, both for undermining have fallen and modest improve- contain, how messages should be and reinforcing official efforts ments have occurred even as the designed and so on. In 1986 (Tapsell et. al. 1999; Gilbert and warning task has become harder. Handmer and Ord reviewed the Gouy 1998: 24; Parker and It has become harder because of our literature to that date and found Handmer 1997). It is an area where expanding use of flood prone areas that such statements had been set investigation seems urgently needed. especially for recreation, and out since the 1970s. The recent US because of changes in society A cross-cutting institutional issue is and European material reviewed in People may be more difficult to the frequently mentioned weak legal this paper restates and adds to this warn as society becomes more position of warnings in common- literature, but does not set out atomised -and other features set law jurisdictions such as the UK, processes or procedures for out in Handmer (2000). This overall USA and Australia. But no solutions developing and implementing a negative assessment of warning are advocated in the reviewed warning system. (For an approach performance hides substantial reports. The problem is that legal used in Australia see Handmer, progress in many local areas and in uncertainty may make some officials Keys and Elliott 1999). Processes some countries such as the UK, as reluctant to provide information on are also needed for sharing well as the steadily improving flood preparedness and appropriate experience and ideas between the reliability of prediction and action on receipt of a warning. This various jurisdictions and agencies communication hardware. Sorensen appears to be less of an issue in the involved. In Europe, European also asserts that warning systems civil code countries of Europe. Commission funded projects are "have not been demonstrated to have any significant impact on Emergency Management Australia In Downing. T.E., Olsthoorn. A.and Toi. reducing damage to . .. private Institute workshop 'Flood warnings: S.J. (eds) Climate, change and risk. London: Routledge: 125-161. property or . . . economic status and trends". Mt Macedon, Illinois Depanment 01 Tansponadon.. disruption." However. Australian 9 April 2002. My thanks to those (1980). NotqyingJwdplain residents Vol 1: and UK research has long shown who provided comments at the An assessment of the iirerarure. Val 2: the property savings and economic workshop, and to Chas Keys of Annotated bibliography Division of Water benefits of successful warnings in the NSW State Emergency Senice Resources. Chicago. specific areas. The challenge and Jim Elliott of the Australian Kararic. E.. Baker, A.. Zimmerman. P and everywhere is to make such Bureau of Meteorology for Songberg. K.. (2001). Traralgonb flood warning system - community access. successes widespread and normal. constructive criticism. P~mdingsPlanningjor tk inevitable: Victorian flwdploin Mnnngem~nt Making success normal requires. References ConJermce Traralgon. Oaober 2001. among other things, that the Beck. U.. (1992). Risk wvirty - towards a May, P. Burby. R.. Ericksen, N.. Handmer. waming task be properly new malrrnity London: Sage. J.W. Dixan. J.. Michaels. S.and D.I. conceptualised as being about Bye P and Homer M.. (1998). Clrar 1998 Smith.. (1996). Envimnmmlal management enabling communilies, emerprises Joodr: Report by the independent review and governonre: intergovernmoltol team to the bwrd ojthr Envimnmcnt Agrnry and individuals to take action to appmher to hazards and runainability. Volumr I. Briaal: Environment Agency London: Routledge. reduce their risk and property 30 September Mileti. D.S.. (1999). Disasten by design: losses. The US and European Crawlord. K., (2001). The Killer Tornado A reorrersmrnt ojnatural hazards in the material generally fails to make this Outbreak of 3 May 1999: Applications of United Stater. Warhingtan DC: Joseph clear. It has been said for decades OK-First in Rural Communities. Henry Press. In Gruntlea. E. and Handmer. J.W. (eds) that the finest monitoring and National Science and Technology Council CapingwithJ7rrrhJoodr. Kluwer prediction systems possible are of no (NSTC). (2000). Effective Disaster Drabek T.E.. (1999). Understanding value if they do not serve this disaster warning response. Ihc Sorial purpose and they cannot serve it - Science Journal. 36(3): 515-523. Washington DC. by themselves. There are many Emergency Preparedness Canada and Parker. D.J. and Handmer. J.W. (1997). related issues in waming system Envimnment Canada.. (2001). Managing The role olunoflicial warning systems. oJan success, such as harnessing the Jwd hardand risk: report Journal of Contingencitr and Crisis independent pnnel. (Edited by Ashij Kumar. benefits of new information and Monogement. 60): 4540. Ian Bunon and David Elkin.) Emergency communication technology. Preparedness Canada and Environment Post. K.. (2001). Barnerr to eflective US inclusion of the informal and private Canada. alen and notification system. Presenratian at the Annual Hazards Workshop. Boulder sector, resolving the uncertainty of Environment Agency. (2001). Lessons Colorado. July 2001. legal liability, and the need to learned: Autumn 2OOOJlwds. UK Environment Agency: Bnstoi. Rohmann. B., (1999), Assessing hazard identify and engage those most at infomlion~communicationprograms. risk from flooding. In Australia, a Gilben. C. and Gouy. C.. (1998). Flaad management in France. In Rosenthal. U. Australian Psychoio@t. 33(2):105-112. group whose lives may be at risk and t'Han. P (eds): flwd response and Rasenthal, U, and t'Han. P. (1998). from flooding are the increasing crisis management in Wcrtrrn Europe: a flwd response and crisis management in numbers occupying low quality comporalive onolyris. Springer Verlag: Wmem Eumpe: a comparative anolyris. housing in riverside caravan parks. 15-55. Springer Verlag. Gruntfest. E, and Handmer, J.W. 2001). Songberg, K.. Baker. A, bazic. E. and Processes are needed to engage (rdr) Coping withJ7rrrhJ7wdr. Kluwer. Zimmemn, P, (1999). Tailoring with those at risk and to ensure Handmer. J.W. (2000). Are flood improvements to your flood warning system. Victonon flwd Management that their needs drive warning warnings lutile? 200C-2. ConJerenre Wangaratla: flwdplainr - risk system design and operation. Good wmasseyac.ru/-traumalissue~000- Uhandmerhtm and rewonlr. practice and experience needs to be Handmer, J.W. (2001). Improving flood Sorensen, J.. (2000). Hmrd warning shared on a regular basis - and this warnings in Europe. Envimnmrntal systems: review 01 20 years a1 progress. requires leadership. Australia may Hazards. 3:19-28. Natural Hazards Review May: 119-125. need to devote more attention to Handmer,J.W and Ord. K.. (1986). Tapsell. 5.. Tunstall. S.. Penning-Rowell. state or national leadership and to Flood warning and response. In Smith. E.C. and Handmer, J.W. (1999), The processes for continuous D.I. and Handmer. J.W (edr) fld hdhcfltcrs oJthr 1998 EasterJlwding in warning in Australia. Canberra: CRES. Bnnbuv and Kidlington. For the improvement, through engagement Australian National University: 235-257. Environment Agency, Thames Region. of communities and other Handmer, I.W., Kc*. C. and Elliott. 1.. Flood Hard Research Centre. Middlesex stakeholders critical to successful University. London. risk management. warnin& in Aunralia. Applied Geography. John Handmer. Centre for Risk and Acknowledgements 19: 179-197. Community Safety, Gempatiat Science, Handmer. J.W. Penning-Rowsell. E.C. RMll University. Melbwrne An earlier version of this paper was and Tapsell. 5.. (1999). Flooding in a Email: lohn.Handme~rmit.edu.au presented at a joint RMlT University/ warmer world: the view from Eumpe. Perception by Counter Disaster Personnel off the risks off bulk transport off dangerous g oods along the Brisbane-Gladstone transport corridor

This Southeast Queensland research team reports on the wide range of perceptions and responses of disaster personnel to a major dangerous goods road accident scenario.

Project Background By Childs, 1.R. W, Hastings, PA [School of Brisbane and Gladstone, located some 600 km apart on Humanities and Human Services, OUT Caneldine, the east coast of Queensland, are the two major heavy Beams Rd, Caneldine 40341 Carlisle, R. D industrial centres of Queensland. Both have significant [Chemical Hazards and Emergency Management (CHEM) Unit, Department of Emergency Services chemical industries, and there is considerable bulk (Oueensland)] and Powell, N[FI.Fire. El transpon of dangerous goods' along the Brisbane- Gladstone road-rail transport corridor. Over the past Counter Disaster personnel stationed along the decade, several road accidents occurred involving bulk- Brisbane to Gladstone road-rail corridor took part in transport vehicles carrying dangerous goods along this focus groups aimed at eliciting their perceptions of route, resulting in deaths, injuries and major losses of the hazards associated with the bulk transport of loads. Childs et. al. (2001) documented these incidents2 dangerous goods that occurs along this route. and broadly identified some of the factors along the Six groups, each representing a Disaster District, corridor that contribute to risk and vulnerability in discussed their responses to a major road accident relation to dangerous goods transport. scenario on a local stretch of the Bruce Highway The aim of the present research was to capture involving two fuel- carrying tankers and a resulting perceptions held by key emergency services personnel explosion (BLEVE). That is, a low probability, but (primarily responsible for local counter-disaster high-impact and rapid-onset hazard. A wide range of operations), government and community workers perceptions and responses to the scenario was noted located in several centres along the Brisbane-Gladstone both within groups and between groups, reflecting route regarding the risks of the bulk transport of differing hazard and risk perceptions, resource dangerous goods. The foci of the present study are to availability and mobility, and other geographical elicit from these groups: levels of awareness of factors, Initial management of the hazard. dangerous goods travelling along the Brisbane- establishing safe zones, effecting evacuation. Gladstone rouie, perceptions of the risks of managing traffic and dealing with casualties were potential emergencies involving such materials, and all variously raised as challenges to the emergency envisaged response and recovery sirategies for an services personnel and the frameworks of accident scenario. coordination and response under which they The research results presented here summarise the main operate. Some settlements along this transport elements of a report submitted to the Queensland corridor were identified as being notably vulnerable Department of Emergency Services (DES) in December to a dangerous goods accident because of their 2001 following a joint initiative between DES (CHEM proximity to the highway, and the relative Unit) and the Queensland University of Technology inaccessibility of the resources needed to cope (School of Humanities and Human Services). with such an event.

1 The term 'dangernus goods' is used in relation to chemicals considered to be ruificiently hazardous to require regulation of their tramponation under the Australm Dangerous Goods Code (the ADG Code). 2 In 2001, there were thm additional incidents at MI.Larcom. Miriamvale and Cabooltum resulting in two deaths. smaller towns. Although the Bruce highway now bypasses several of the larger centres (e.g. Bundaberg. Nambour and Caboolture) by several kilometers, the GLADSTONE highway remains the 'main street' for smaller towns such as Childers.

Qualitative analysis of the combination of risk factors (impact radius of potential dangerous goods emergencies, population potentially exposed, local geography and highway conditions and likely level of emergency resources) suggested varying levels of potential risk and vulnerability in the present dangerous goods context for several geographical regions along the route (Childs et. al. 2001). With the expected increase in the quantity of chemicals passing along the comdor commensurate with the projected population' and induarial growth in Southern and Central ~ueensland',there is a need to systematically assess risks and community vulnerability along the comdor with a view to contributing to the preparation of emergency senlces for dealing with any hazard arising from such transport. While the probability of a catastrophic accident is very low, the consequences of such an accident, should it occur in a built-up area, could be very severe. The potential risk from the use and transpon of dangerous goods throughout the region must, however, be balanced against the many economic, employment and other benefits that accrue from these activities.

Map 1: The BrisbaneGladstone Transpan Corridor Refer to Childs et. a1 12001) Methodology Table 1 for incident details. The broader context of the research was a disasted emergency risk-management framework, based on the The Brisbane to Gladstone Transport Corridor Australia/New Zealand Standard on Risk Management A range of dangerous goods travels via bulk tankers and (AS/NZ 4360: Risk Managemen0 and an approach to intermediate or smaller containers along the Brisbane to vulnerability assessment outlined by the Australian Gladstone transport comdor. These include petroleum, Emergency Management Institute (Hunter, 1996)'. This liquefied petroleum gas, liquefied ammonia, molten includes a description of community risk perception. sulfur, liquefied chlorine, concentrated hydrochloric While the availability of physical resources is an acid, compressed hydrogen, and sodium cyanide. important factor in coping with any disaster, the The primary road route comprises Brisbane's Gateway effectiveness of such resources is a function of the quality Motorway, the Bruce Highway (National Route 1) and of emergency preparedness and planning at the local feeder roads tolfrom regional centres and Gladstone. level. Preparedness and planning, in mm, is a function The rail link (North Coast railway) closely parallels the of nume'rous factors, but is anchored by risk perception highway, with the two routes being rarely separated by (eg. Young 1998; Zamecka and Buchanan. 1999). more than a few kilometers except in one section Community vulnerability to the hazards associated with between Maryborough and Gladstone (Map 1). the bulk transpon of dangerous goods, thus, depends in Both routes pass either through or near numerous the first instance on awareness and risk perception of settled areas, including northern Brisbane localities. emergency response personnel along the route, and on the Sunshine Coast Hinterland, regional centres and their perceptions of possible response and recovery

3 Population Trends and PrnrpectsJor @eenslond 2001, available from the Planning Information and Forecasting Unit. Depanment of Local Government and Planning. Queensland. 4 Mooted developments include a new alumina refinery at Ya-n; a magnesium production facility at Rockhampton (Stanwell), about one houri drive north of Gladstone: an ethylene dichloride plant at Gladstone. 5 Hunter5 (1996) model includes several stages- of risk evaluation and assessment: (i) descripmn or the hazard, the cammunitx the environment and the emergency services; (ii) analysis of interaction between the hazard. the community. the environment and the emergency services: (iii) assessment of community risk perception: (iv) ranking of vulnerabilities: and (v) compansan of risk ta existing risk criteria , . ~ .. . . ~ - ~ ~. ~ Redcliffe Burpengary Highway passes through built-up area with entry ramp and highway speed limits*.

- . ..~-~ . - - - Sunshine Coast Cooroy Highway passes near built-up area with entry ramp and highway speed limits*. I ~ - ~ ~ . -. - -~ ~ '1 1 Gympie Gympie Highway passes through extensive built-up area with intersections 1 and urban speed limits*. :.-- .- .~~ ~ - ~ ~ ~.~-. ~.~ ~.... ~- ! Maryborough Howard Highway passes through built-up area (small town) with intersections i and urban speed limits*. k--- . - -.. - -. .- - ...... ~. .. - ' I Bundaberg Gin Gin Highway passes through built-up area (small town) with intersections I and urban speed limits'. I- --- . I Gladstone Gladstone Feeder road from highway passes through extensive built-up area with ; South (Kinkora) intersections and urban speed limits*. ~~ ~ . - t kmlh; urban rpeed limits - 60-80 kmhr -- strategies in potential emergencies. The methodological discussion through a series of structured questions. challenge for the research was therefore to elicit and Areas of investigation and subsequent data-capture were capture information relevant to these factors. summarised under the headings of: awareness of dangerous goods passing through local areas; commun- Data collection ication; impact assessment; mobilisation of resources; Focus group discussions with appropriate emergency and location of resources. Discussions arising from the services personnel along the route were used to gather scenario were further directed by the facilitator to perception and response information concerning the consider the progression of responses from pre-incident, risks of dangerous goods incidents. At each focus group, through to event response, post-incident and recovery general questions addressing risk perception, and a local phases. Generally discussions took approximately two road accident scenario involving bulk dangerous-goods hours and were scheduled as part of a DDCG meeting. transport, provided the catalysts for discussion (see Police, Fire, State Emergency Service (SES) and local Map 1 and Table 1.).Given the scale of the proposed government institutions were well-represented. The scenario (see below), it was resolved to access Disaster Ambulance Service and medical authorities were District Control Groups (DDCG)"~~the project because represented at most but not all meetings. of their membership of senior emergency service personnel and other representatives (local government, Several information-gathering and recording strategies government se~ces,health professionals etc.). It was were employed by the researchers. One of the projecth later noted that less senior, operational personnel and chief investigators was present as an observer on each crews often attended the projecth focus group meetings, occasion to record discussion points. In addition, providing a practical perspective. Although access to, participants (identified by service-affiliation and not as emergency se~cespersonnel was gained via the DDCG, individuals) were asked to write their key thoughts on the committee itself was not the focus of the present formatted information-recording sheets provided study, rather the attending Emergency Services (structured as described above). Usually within three representatives were. days of each focus group, the research team met to identify the key issues that had emerged. Focus group meetings were held in the six non- metropolitan Disaster Districts along the Brisbane- The scenario Gladstone route (viz. Redcliffe, Sunshine Coast, Gympie, Maryborough, Bundaberg and Gladstone. In the case of a dangerous goods emergency or chemical disaster, the onset speed of the hazard is usually rapid Perceptions of the risks posed by the scenario and and, consequently, warning time for evacuations is most possible responses to these risks were sought from likely to be minimal or non-existent. Thus, in terms of focus-group participants. When necessary, the focus community vulnerability one is dealing basically with group facilitator (project research officer) stimulated pre-existing risk perceptions, the capacity for rapid and

6 As a result of the State Counter Disaster Organisation Act, 1975, Disaster Districts, Disaster District Control Groups (DDCG) and Local Governmen[ Counter Disaster Cornmiltees (LGCDC) were established in Queensland. Membership and functions of [he separate groupr/commitrees are ourlined in Counter Disaster and Rescue Serviiei (2001a: 2001b) comments are general by nature, and do not directly identify individuals, are they attributed to particular focus groups or participating organisadons. Some comments reported do not relate directly to the scenario used, but rather to response to dangerous goods emergencies in general. The descriptions below sometimes reflect perceptions of senior officers anticipating what crews attending incidents would do, and sometimes reflect the perceptions of those who were likely to be actually auending an incident at a given time. Attendance and participation in focus groups on the day was the determinant of sources of perception.

Awareness and risks of chemical hazards Participants were generally aware of a wide range of hazardous materials transported along the road and rail routes in bulk containers. Fuels (petroleum, diesel LPG), farm chemicals (fenilisers and pesticides), chlorine, ammonia, acids, explosives and cyanide were the most Participants thought the threat to their communities from the transport of bulk frequently nominated. From the knowledge of the chemicalsJdanoerous ooods was increasinq. researchers, these results indicate a reasonable picture of the types of dangerous goods moving along the route. Nevertheless, it was commonly reported as a cause for effective response by emergency services and the concern that information regarding quantities and community3 capacity to respond post-event. The latter timing of shipments was not easily available to local is particularly related to the resources of the emergency emergency se~cepersonnel. Focus-group participants services to evacuate post-event, and to cope with were overwhelmingly of the opinion that the threat to potential casualties and injuries resulting primarily from their communities from the transport of bulk the effects of fire, blast or toxic gas release. chemicals/dangerous goods was increasing. The written scenario presented to focus groups in the Responses to scenarios present research involved a collision between a petrol In all focus groups there was a degree of initial skept- tanker (20.000 litre) and an LPG tanker at a local icism by some individuals regarding the probability of intersection. This resulted in a substantial fire and a BLEVE' approximately 15 minutes later. Locations the scenario eventuating in reality. In most cases this varied from an isolated stretch of the Bruce highway skepticism dissolved (a) after the BLEVE video (Cairns 1987) was shown, and (b) after initial discussion near the small settlement of Howard to the busy northern fringe of the Brisbane metropolitan area established a recognition of the serious consequences of (Caboolture). Potential blast and danger zones were such an event, albeit at a low probability of occurrence. presented visually to the focus groups in the form of a There was no consistency of emergency response to the short video of a BLEVE incident, maps and air photos. scenario across the focus groups and, hence, no Under the scenario, the immediate consequences consistent indication that a rote response would following the BLEVE were 20 casualties. 40 severe generally be elicited from the invocation of a standard injuries and traffic banked up on the Bruce highway. set of operational procedures and instructions. The The focus groups were also asked to consider questions scope of the present research does not, however, extend of community recovery in the medium and longer term. to evaluating the envisaged responses in terms of the formal procedures, but simply notes this inconsistency Emergent issues from focus and seeks to contextualise it within overall emergency group meetings response and community vulnerabili~y. This research proceeded on the understanding that individuals would not be directly identifiable in the Initial risk perception and response reporting of results. Therefore, the following information Focus groups indicated that, in the initial stages of the is presented in a manner, and at a resolution. scenario (i.e. pre-BLEVE and during the explosive appropriate to that understanding. Only when phase) indicated that Queensland Fire and Rescue

7 BLEVE. Boiling Liquid Expanding Vapaur Explosion. Computer modelling carried out by the CHEM Unit. Queemland Depanment of Emergency Services, indicates that a ELEVE ofa 14 tonne LPG lanker would have injurious thermal eIfecls within a radius of 240 metres. 8 A video compiled by QFRS from a set of still photographs of the Cairns BLEVE in 1987. Service (QFRS) officers, rather than Police, would take the leadership role on the ground if present.

Two broad response patterns were evident in dealing with the pre-BLEW fire, should Queensland Fire and Rescue Services (QFRS) crews amve in time. One view was that if QFRS could get to the scene quickly enough (within minutes) and play water onto the LPG tanker to cool its temperature, a BLEW could be avoided. Some QFRS officers considered that taking such action, would lead to a high probability of injury or death to attending personnel placed in the danger zone. Nevertheless, an emergent view was that this approach may 'buy time' for evacuating the public. Furthermore, some QFRS participants felt that pressure from public and community expectations would mean that crews would attempt to fight the fire - even if the professional judgement suggested that to do nothing, except to evacuate the area, was the best action. Other QFRS officers made a much more conservative initial risk assessment and insisted that 1500m clearance he achieved from the accident site. One group commented that after conducting an evacuation there would be little more that they could do to modify the hazard.

Evacuation and establishing 'safe' zones Skepticism about the likelihood of the disaster scenario dissolved after participants were A critical question explored by all focus groups was how shown evidence of orevious events. to effect an initial evacuation of crowded areas at risk in the post-accident pre-BLEW period; for example being asked to intempt their business or activities and shopping centres adjacent to the accident scene, and to leave an area. State Emergency Service volunteers motorists in vehicles 'banked up' along the highway were noted to face similar issues. The solution was seen When the need to evacuate the public from the accident to he the declaration of an 'emergency situation' under vicinity was not raised by focus groups, the question the Public Safety Preservation Act, 1976' which gives was prompted by the facilitator. additional powers to police.

There was general acknowledgment of the potential for Establishing sale distances for the public (and the an initial lack of coordination among the emergency emergency senices) and maintaining these was seen as services in establishing safe zones and evacuating those another problematical issue. Hazard identification and deemed at risk. This was arguably driven by the risk perception on the part of emergency services unfamiliarity, speed and scale of the proposed incident. personnel at the scene clearly becomes critical in this Focus group participants were unaware of specific context. One focus group commented that when an evacuation plans under the circumstances of a rapid- evacuation zone had been established, the public would onset hazard such as the one presented. Given that the 'take the lead' from emergency service personnel as to available time for the required evacuation was very short the distances that should be maintained from an (i.e. 15 minutes under the scenario), participants incident. The problem was that in a situation where generally acknowledged that it would probably have to emergency senrice personnel have entered the danger be achieved by first-arriving QFRS or police officers. zone, the public may underestimate the safe distance required, and, without good crowd control members A concern of several police officers who had had of the public may enter the danger zone. experience with evacuations was the difficulty of managing the public under circumstances of danger. Discussion of the practicalities of achieving a successful Given the scenario, the comment was made that it was evacuation and controlling crowds under the scenario easy enough to stop traffic, but it may be quite difficult conditions raised some communication issues. One to persuade people to leave their vehicles unattended in interesting line of inquiry questioned how emergency order to evacuate an area. Police participants reported services personnel at the scene would rapidly and resentment from members of the public unwilling to authoritatively issue evacuation orders directly to the accept Queensland Police Service (QPS) authority when public. Do police officers possess public address systems

9 This can be invoked very qutckly by a commissioned Police officer, even if not present ar the scene. or hand-held loud hailers in their vehicles? Several aspects including hazard and risk identification, and police respondents conceded that such equipment may evacuation. In some regions and remote localities, it was not be readily available, given the rapid-onset hazard. suggested that the initial tasks of hazard identification (ie. recognising the potential for a BLEVE) and the Traffic control immediate response may fall to an auxiliary fire officer One of the main topics of discussion in several groups (e.g. Rural Fire Service). There was speculation that was how to manage gridlocked traffic caused by such an these operatives would not be adequately resourced, and accident on the highway This problem was particularly may not recognise the risk of a BLEVE. severe for the Burpengary scenario location (Redcliffe Disaster District focus group) because of heavy traffic One focus group raised the problems of establishing a just north of the metropolitan area. Relaytng information command post and effecting evacuations where incidents both among emergency service personnel and to the occurred in areas seniced by small or 'single officer' public (i.e. stationary gridlocked motorists), clearing police stations. Settlements along the Brisbane-Gladstone congestion on the highway, diverting traffic to side-roads comdor were specifically identified as being and evacuating casualties were seen as critical problems. representative of these contexts. Therefore, from the The possibility was raised that some critically injured perspective of resource accessibility, such small commun- victims may actually expire before they could be ities may he the most vulnerable to this tpe of hazard. evacuated because of delays caused by traffic congestion. Further discussion of the experience levels of emergency A suggestion was that this situation could be services personnel likely to attend the emergency scene ameliorated by the use of helicopters. In other less busy occurred at some meetings. This was particularly raised locations, however. the traffic control issue was regarded by police representatives in one region where there are as much less of a problem as drivers may have more high proportions of relatively inexperienced officers. room to manoeuvre and may have local knowledge to This was seen as a potential problem if junior officers take alternative routes. were left with the diflicult task of effecting a rapid Resource availability and mobilisation evacuation of members of the public from the high risk blast area (assuming such a zone had been defined). A key concern for most focus-group participants was the availability and mobilisation of adequate resources to This situation is exacerbated where senior officers would initially deal with the accident (and hazard), and then not be able to easily access an incident location due to the chaotic aftermath of the scenario. At all meetings, highway gridlock. QAS crews were also identified as emergency service representatives voiced strong concern likely to have inexperienced crew members in some over the accessibility of appropriate resources at the areas. QFRS cited their policy of maintaining a crew of incident locations, Issues involved the availability of: 'balanced experience' on fire trucks with no more than emergency service personnel (particularly experienced one junior (ie. <3 years experience) officer on an personnel), communications, medical aid, equipment, operational vehicle at any one time. water and foam. In terms of communication infrastructure, some Across all groups and scenarios, it was generally problems were noted with the mobile phone network accepted that off-duty personnel would have to be encountering 'dead spots' along some stretches of the recalled, and resources of surrounding localities highway A similar problem was also noted for dead- requested. Given a fast-onset hazard, the time taken to spots for 2-way radios in some areas. One group contact such personnel and for them to arrive at the commented that 'tmckies' with radios may provide an scene were significant issues. 'Next available' QFRS, opportunity to gain first-hand information from the QAS" and QPS resources could be up to 30 to incident scene in the early stages, and where access 45 minutes travel time away and it was not always clear problems existed for initial emergency senice vehicles. that even local services could reach the scene 'pre-BLEW Furthermore, highway traffic congestion Given the specific nomination of casualty numbers was deemed likely to impede the arrival of emergency within the scenario, access to, and mobilisation of senices, and senior officers,whose 'on site' presence emergency medical support was seen as critical. would be desirable in the circumstances. State Discussion of the treatment of patients with severe burns Emergency Service (SES) personnel, generally viewed as turned to the allocation of hospital beds. From the having potential roles in crowd control and traffic advice of medical personnel participating in the focus monitoring, could take an hour or more to mobilise, groups, the capacity of most regional hospitals (relevant even if locally available. to this study north of Brisbane) to deal with emergency burns victims appears to be limited to 3-4 beds. It was The availability of adequate and experienced personnel suggested that the further allocation of bums victims at the scene was viewed as potentially important to would most likely be handled from Brisbane, and

10 Queensland Ambulance Service involve patients being transported to Brisbane hospitals Further issues: roles of committees by air. One group speculated that this mode of patient The present research used Disaster District Control transportation could take several days if there were Group (DDCG) meetings to access emergency services multiple casualties. This group also raised the issue of personnel to investigate their perceptions of a 'tough decisions' having to be made in terms of dangerous goods transport incident and associated allocating relatively scarce medical resourced rapid-onset hazard response. The research focus was opportunities to the injured at the time of the incident. not specifically on critiquing the roles and operations of organisational structures, such as DDCGs or Local There was a strong opinion expressed that heavy Government Counter Disaster Committees (LGDCGs). machinery for site works should be available through although the relevance to the current research of their the local government channels. Provision of adequate separate functions is clearly recognised. Nevertheless. protective clothing (level 2 or 3 suits) for QFRS officers comments and observations made at focus group and auxiliary fire officers dealing with toxic releases in meetings suggest that a broader set of issues can be small towns was also raised. distilled in this context, and await further investigation. The following represents an attempt by the researchers Recovery. .phase to articulate these issues, based on information and Following considerations of immediate response to the observations gained from the focus group meetings and scenario, the facilitator directed discussions to the follow-up work. longer-term recovery phase. In recovery, there is an issue of differentiating the broader and longer-term roles of One issue to emerge from the research relates to the 'disaster management' and the shorter-term roles of roles of, and links between the DDCG and the LGCDC. 'incident management' that may directly involve The relative responsibilities of these committees is operational emergency services and SES personnel. established within State law and disaster plans (State Depending on the perception of 'disaster' vs. 'incident', Counter Disaster Organisation Act and the State Counter different roles and responsibilities within the disaster Disaster Plan). Nevertheless, faced with the practicalities management framework should be activated. This relates of responding to the scenario presented in the project to how well the various players in Emergency Services (a rapid-onset, high impact technological hazard) understand the impact that the event has on the relative roles and activities of the groups, as perceived community. The disaster management system only by personnel involved in response activities, were responds if the impact on the community cannot be sometimes blurred. Again, the researchers caution that adequately managed by local emergency services and this observation emerged from focus group discussions, existing community arrangements1'. Investigation of this and its investigation was not originally pan of the research design. aspect is beyond the scope of the current research, but may have implications for explaining some of the Opinion varied widely on the appropriate observations reported below. responsibilities of the DDCG and the LGCDC in the case of the specific scenario presented. The researchers noted Wider community dislocation varying degrees of willingness to take ownership of the Social dislocation ensuing from the scenario presented scenario situation across the groups. In some, it was was not generally raised as a planning issue by focus- suggested that DDCGs would be quickly activated, while group participants in this project. This may reflect a not in others. It was recognised that the event would need to prioritise the management of the immediate. unfold too quickly for either the DDCG or LGCDC to physical threat andlor result from perceptions of the convene, let alone play an effective role initially. The scenario and committee roles as indicated below. Only immediate response phase would most likely be dealt in one group did an individual raise the issue of social with exclusively by operational personnel. Later dislocation; the comment being that the affected response and recovery is therefore the time for potential community would be 'in shock'. Another group was responsibility for these committees. The DDCG was seen prompted to discuss longer term social recovery, but as playing a role in the recovery phase of a disaster in offered few specific ideas on the needs or management some focus groups, while in others members saw their of such. General uncertainty surrounded the issue of group as having no role at all in the longer-term organisational responsibility dealing with the wider recovery phase. community, particularly in the immediate and medium- term post-incident phase. In the longer term this The distinction in determining the roles of the two responsibility would be that of relevant government committees was contingent upon the perception of the departments and community organisations particular research scenario as being either a 'disaster' or kg. Department of Families, Red Cross, etc.) an 'incident'; the former being a larger scale and

11 Wayne Ripper. Director, Disaaer Operations. Counter Disaster and Rescue Services. Queensland Department of Emergency Services therefore a more serious event, particularly in terms of emergency services, the QPS and LGCDC maintain key the degree to which the community is able to cope. If an responsibilities. The question as to whether the scenario even1 is deemed to be a 'disaster' (i.e. beyond the of ihe presenl research was a disaster or an major normal coping capacity of the community), a key role is incident was therefore critical, yet the focus groups were activated for the DDCG, whereas if it were a 'major not consistent in categorising the scenario as either a incident', this may not be the case, and responding disaster or an incidenl.

I Recommendation 1: Awareness of chemical hazards I That opportunities be identified to improve the availability to local emergency services of information regarding the I transport of dangerous goods along the Brisbane-Gladstonecorridor. A major cause for concern was that information regarding quantities and timing of shipments of dangerous goods through communities was not easily available, and that hazards from such sources were increasing. A record of : average annual movements of bulk dangerous goods loads along transport routes could be established. An administrative system would be needed to collate and analyse the data relating to both road and rail transport.

Recommendation 2: Responses to scenario That community vulnerability to dangerous goods transport along the Brisbane-Gladstonecorridor be further investigated by extending existing hazard mapping programs to include bulk transport of dangerous goods. The hazard mapping program within the Department of Emergency Services could be developed further to integrate vulnerability associated with the hazard. Furthermore, due to the relative rarity of major dangerous goods incidents or disasters to provide experience, training based on mapped databases could become the primary mechanism for improving emergency response. Recommendation 3: Resource availability and mobilisation I That the causes of the wide variability found in response between the focus groups to the research scenario be investigated by considering factors such as local resources, location and training. Availability and mobilisation of resources were identified as major issues. Limited resources may be an unavoidable reality in small centres. An integrated strategy utilising the resources of a network of larger urban nodes along transport routes could improve rapid deployment of specialised resources to smaller centres when needed. More effective communication systems would reduce mobilisation times.

Recommendation 4: Evacuation II That community safety programs along the corridor include an education component dealing with appropriate responses by members of the public to emergencies arising from the transport of dangerous goods through communities. The need to protect public safety by prompt evacuation in the case of dangerous goods emergencies was well recognised. Public behaviour and acceptance of directives from emergency service personnel in such circumstances could be improved if public awareness of such hazards and appropriate behaviour strategies was increased. I Recommendation 5: Recovery i ,' That the need to assist communities recover from dangerous goods disasters be clearly recognised in the disaster ! management system. , While the need to pay close attention to longer-term community recovery from natural disasters is well-recognised 1 by the disaster management system, it is possible that this may be overlooked in the event of a major dangerous ; , goods incident. / Recommendation & Roles of committees I 1 That the interpretation of rapid-onset, dangerous goods emergencies as either incidents or disasters be further ! investigated. This would clarify the relative operational roles of Disaster District Control Groups and of Local , Government Counter Disaster Committees in responding to this type of hazard. I While the immediate incident response phase in the case of a dangerous goods emergency would be dealt with exclusively by operational personnel, perception of the role of the two levels of disaster management committees (Disaster District Control Group and of the Local Government Counter Disaster Committee) in the recovery phase would hinge upon whether the emergency was regarded as a major incident or a disaster. Further examination of these roles is needed to clarify this issue. L-~- ~ ~--- ~~~~~-~p~~~.-~ ~ At each focus group, questions addressing risk perception and a local road accident scenario involving bulk dangerous good transporl providw' the catalysts for discussion.

References The research could not have riroceeded without the caowradon Childs. I. R. W. Carlisle, R.D. and Hasrings. PA.. (2001). The and panicipation of the following groups and individuals in the Brisbane-Gladstone Transpan Corridor: identification of "sk and administration of focus groups. Their valuable assistance is vulnerability for the bulk transpan of dangerous goods. Aulralinn gratefully acknowledge& Journal ofEmcrgrnry Managemml. 15.4.54-57. Redclille Disaster District Control Group, Inspector Ken Childs, I. R.W. Hastings. PA.. Cailisle. R.D. and Powell. N., (2001). Benjamin (Chair) and Sergeant Andy Besant Perception by Counr~rDialer Personnel ojlhe Rkhr oJBulk lronspon of Sunshine Coast Disaster District Control Group, Inspector Neil Dangemu Gmdi Along the Bnslwne-Gladrtonr Trampon Comdol: Behm (Chair) and Sergeant Ian Tinker Research Repon. Queensland Depanment of Emergency Services, Gympie Disaster Diarict Contml Group. Inspector Vemnica 30~~. Cane (Chair) and Sergeant Pat Armstrong Counter Disaster and Rescue Services, (2001.4, The Quernrlond mler Managemenl Syslem Depanment of Emergency Services. Maryboraugh Duster District Control Group. Inspeaor Mal March. 12pp. Churchill (Chair) and Sergeant Chris Sidey Counter Disaster and Rescue Services. (2001b). Disarler Planningfor Bundaberg Disaster Distcia Control Group, Inspector Phil Lxal Govemmcnt Depanment of Emergency Sewices March. 8pp. Wardrope (Chair) and Sergeant Eruln Hoffman Federal Ofice of Road Safety. (199.3). Aurrralion Code Jor !he Gladnane Diter District Control Group. Inspector Gary Trampan of Dangernus Gwdr by Raad and Rail. Harland (Chair) and Sergeanr Dave Thomas Hunter. C.. (1996). A New Approach lo Emergency Managemenr. The research team also gratefully acknowledges: Education and Training in Aurrralia. Auslnrlian journol ojEmergenq Mr Jeff Shelberg, Area Manager, Queensland Ambulance Service. Managemenl. Spring. 12-19. Standard Association of Australia. (1997). HE76 Dangerous Gwds: Mr Tony Biu and Mr. Dennis Duquemin. Qld. Dept. of Main Inilial Emergency Rerponre Guide. Roads, far assistance with air photo resouices; Young. E.. (1998). Dealing with hazards and disarrerr: risk Ms Janelle Quelch (formerly Scientific Advisor, CHEM Unit) for perception and community participation in management. Auslmlian conducting quandlative risk modelling for the scenario: and Journal ojEmergency Monogemenl, 13. 2. pp 14-16. Mr Michael Walker, Queensland Transport, for in~tialadvice and Zamecka, A. and Buchanan. G.. (1999). Ihsarler Risk Monagemml. discussion on tranrpon issues. Queensland Depanment of Emergency Services. Mr Wayne Ripper, Director, Disaster Operadons. Counter Disaster and Rescue Services. Queensland Depanment of Acknowledgements Emergency Services, for advice and discussion on the final draft The present research has been supported by an Australian of this repan. Research Council grant. 199&1999. Dr Paul Barnes. formerly Manager, State Camrnuniry Safely. The researcherr gratefully acknowledge the suppon of the Queensland Fire and Rescue Services. Queensland Depanmeni of Emergency Services in makmg resources and expeniw available to the research. The participation of Mr Nick Powell F1.Fire.E. in facilitating focus ' For funher information please contact: i.childrQut.edu.au group meetings is gratefully acknowledged. Perceived chanae in risk off natural disaste& caused by gooobal warming Contrary to insurance company perceptions, Chris de Freitas argues that although the future state of global climate is uncertain, there is little scientific evidence to suggest any change will result in an increased risk of natural disasters.

r ~~ -- question arises as to the extent to which these , By C R de Freitas, School of Geography and 1 expectations are justifiable. Environmental Science, University of Auckland, PB 920 19, Auckland, New Zealand I Evidence versus perceptions of ~ -- worsening conditions Natural hazards causing disasters that lead to It is not uncommon to see media reports attributing the human suffering are as much a product of the occurrence of extreme climate events to global warming. social, political and economic environment as they The problem is that no matter how outrageous the tale it are of the natural environment. It follows, becomes the truth if it is told often enough, or at the very least shapes perceptions of shifting risk. According therefore, that the risk associated with natural to the United Nations Intergovernmental Panel on hazards is in part a social construct that, as Young Climate Change (Houghton el. al.. 1996, p. 173), (1998) has pointed out, is perceived differently by "Overall, there is no evidence that extreme weather events, or climate variability, has increased, in a global all of us and must be defined with this mind. sense, through the 20th century.." The 2001 IPCC For example, risk is defined by Emergency Report (Houghton et. al., 2001b. p. 5) states that "no Management Australia (1995) as the perceived systematic changes in the frequency of tornadoes, likelihood of given levels of harm. thunder days, or hail are evident ..." The increasing dollar cost of storm and other weather related events could be accounted for by a rise in the value of With this in mind, one major determinant of risk is the development and number of properties, especially in perceived trends in disasters. Given that a) weather tropical cyclone prone areas (Changnon et. al., 1997; hazards are the most significant natural hazards in Pielke and Lansea, 1998; Kunkel, Pielke and Changnon, Australia (Coates, 1998) and elsewhere (Downing et. al., 1999). In fact, there is great deal of research which. 1996; Dlugolecki et. al.. 1996; Kattenburg et. d.1996; taken together, suggests that extreme climate events may McCarthy et. al.. 2001) and that b) in the latter part of become both less frequent and less severe when the the 20th century insured losses have been unprece- planet warms. Some of this work is discussed here. dented (Dlugolecki et. al., 1996; Kattenburg et. a1..1996), the possibility of worsening trends have, understandably, Storms attracted the attention of the insurance industry In the Atlantic region, the number of intense humcanes There are a number of possible reasons for increasing declined during the 1970s and 19805, and the period losses: a) a greater concentration of people and high 1991-1994 experienced the smallest number of value property in vulnerable areas, mainly coastal; hurricanes of any four years over the last half century b) business processes have become more susceptible to (Idso et. al., 1990; Landsea et. al.. 1996; Bengtsson damage; or c) that changes have occurred in the et. al., 1996; Zhang and Wang 1997; Murphy and frequency and severity of extreme climatic events. Mitchell, 1995). There is also evidence from Europe that The last of these is in line with expectations of climate suggests a similar trend. For the period 1896-1995. change resulting from an enhanced greenhouse effect Bielec (2001) analysed thunderstorm data obtained at and this too has attracted the attention of the insurance Cracow, Poland, which is "one of the few continuous industry Dlugolecki et. al. (1996. p. 541) comment records in Europe with an intact single place of "It is a common perception in the insurance industry observation and duration of over 100 years." From 1930 that there is a trend toward an increasing frequency and onward the trend is negative, revealing a linear decrease severity of extreme climate events." The important of 1.1 storms per year from 1930 to 1996. Bielec (2001) Future qlobal warminq may lead to fewer and less intense N Niiio events. (Andrus. 2002) also reports there has been a decrease in the annual The most important energy source for extratropical number of thunderstorms with hail over the period of storms is the temperature difference between the tropics record, and there has been a decrease in the frequency and the poles. Most GCMs predict that the greatest ol storms producing precipitation greater than 20 mm. warming will occur over the high latitudes in winter with relatively little warming in the tropics and around Pirazzoli (2000) analysed storm surges, atmospheric equatorial latitudes. This implies reduced temperature pressure and wind change and flooding probability on variation since such variations result from air moving the Atlantic coast of France over the period from one latitude to another. Thus, according to these 1951-1997. He found that climate variability is predictions, the future contrast between the polar and decreasing. Specifically, the work shows that the equatorial latitudes will lessen, producing a weaker number of atmospheric depressions and strong winds gradient and fewer and less intense storms. Consistent that cause storm surges in this region are becoming less with this, the lPCC 2001 Summary for Policymakers frequent resulting in reduced frequency and severity of (Houghton. 2001b) notes that no significant upward coastal flooding. trends have been identified in tropical or extratropical Nguyen, and Walsh (2001) simulated the occurrence of storm intensity and frequency. hurricanes in the Australia region using a general circulation model (GCM) that assumes a tripling of the Floods and droughts atmospheric concentration of carbon dioxide. The Lins and Slack (19991, concerned about possible results showed that the numbers of hurricanes declined increases in floods and droughts accompanflng global in a greenhouse-warmed world and that the decline is warming, looked directly at stream flow from hundreds statistically significant. In a study of Atlantic of rivers in the United States. They state: hurricanes, Goldenberg et. a1 (2001) show that they The pattern indicates that basejlows are increasing occur in distinct multidecadal cycles and are linked to (which suggest that drought is decreasing), median or sea-surface temperature anomalies in the Atlantic averagejlow streamflow is increasing, but annual Oceanb main hurricane development region. Warm m~mumjlowsFncludingjloods) are neither anomalies are associated with increased major increasing nor decreasing. Hydrologically, the nation humcane activity; cold anomalies with suppressed appears to be getting wettec but less extreme." activity Goldenbera et. al. (2001) sumest- that increases in the,frequency oibig Atlantic hurricanes are due to Lins and Slack (1999) conclude that, as the surface air the natural variaiions in hurricane frequency and temperature of the globe gradually increased through intensity rather than global warming. 20th century, the conterminous United States became Air temperature USA China USSR ' Maximum January -0.19 -1.13 0.07 July -0.13 0.06 -0.02 Minimum January -0.26 -1.32 -0.37

8 July -0.19 - wetter but less variable at the extremes, where floods times the region had experienced several extreme and droughts occur. In a similar study, Molnar and droughts lasting from 60-70 years, many of which were Ramirez (2001) analysed precipitation and streamflow more severe than droughts of the twentieth century. In a trends for the period 194S1997 in the Rio Puerco Basin review of the subject. Woodhouse and Overpeck (1998) of New Mexico found no change in high-intensity have come to a similar conclusion. Gan (1998). using precipitation. data sets for the North American Midwest from the last century showed that warming is not associated with Working on the assumption that the best real-world increased occurrence of drought. Benson et. al. (2002) analogue of future global warming is past periods of studied sediments from Pyramid Lake. Nevada. They global warming. Nesje et. al. (2001) analysed a found that over the past 2700 years drought lasted from sediment core from a lake in southern Noway to 20 to 100 years, while droughts of the recent historical determine the frequency and magnitude of prior period have generally lasted less than a decade. floods in that region over the past thousand years. Similarly, Fritz et. al. (2000) used sediment cores to The results showed an extended phase characterised construct a 2000-year history of drought for a part of by very little flood activity coincided with the Medieval the Northern Great Plains of North America. Their Warm Period (AD 1000-1400). This was followed by a results indicated that droughts equal or greater in period of extensive flood activity that corresponded severity to those of the 1930s Dust Bowl were a with the period known as the Little Ice Age, which common occurrence during the last 2000 years. was characterised by lower air temperature, thicker Likewise, Stahle et. a1 (2000), using tree rings to and more long-lasting snow cover, and more develop a long-term history of drought over all of North frequent storms. America, found that drought during 16th century was There has been other research camed out using long- [he most extreme and prolonged in the past 2000 years, term data. For example, the results of research by far exceeding any drought of the 20th century, including Andrus el. al. (2002) confirm the findings of several the Dust Bowl drought. other studies that indicate the mid-Holocene was significantly warmer than it is currently Andrus el, al. Temperature extremes (2002) show a situation where a considerably warmer Coates (1998) has stated that heatwaves are the most climate than that of the present was apparently unable significant natural hazard in Australia in terms of loss of to sustain significant El Nino activity This demonstrates life. There is speculation that global warming will that future global warming may lead to fewer and less increase climate variability and thus the frequency of heat intense El Niilo events. waves. Michaels et. al. (1998) show there is little support for this, or for the popular perception that temperamres Focussing specifically on droughts, several studies have have become more variable. They examined daily examined the past 300 years during which global maximum and minimum temperatures from the United climate has been recovering cool conditions of the Little States. China and the former Soviet Union for day-to-day Ice Age and is bringing conditions closer those variability in January and July and most of the trends experienced earlier during Medieval Warm Period indicated declining variability (Table 1). conditions. These studies have shown convincingly that not only does global warming not produce more Several other studies have found that more warmth frequent and severe droughts, it does just the opposite. leads to more stable climate. Karl et. al. (1995) point out For example, in the United States, Cronin et. al. (2000) that an increase in the atmospheric concentration of using sediment cores from Chesapeake found that carbon dioxide should decrease temperature variability. conditions were wetter than normal and droughts Balling (1998) examined changes in the spatial uncommon during the Medieval Warm Period. At other variability of mean monthly and daily temperatures that temperatures in the Northern Hemisphere showed no statistically significant trend. If these forecasts are correct. variability in the data will shrink more than the predicted mean warming and give a distribution shown schematically in Figure 2.

Conclusion Global warming involves a scientifically realistic mechanism that links climate change to the concentration of greenhouse gases in ihe aimosphere. Although the future state of global climate is uncertain, there is little reason to believe that catastrophic change is underway. Moreover, there is little or no evidence to suggest any change will result in an increased risk of natural disasters caused by increased frequency and severity of climate extremes. Climate models suggest that increases in greenhouse gases are likely to give rise to a warmer and wetter climate in most places. Generally higher latitudes would warm more than lower (equatorial) laiitudes. This means milder winters and coupled with increased atmospheric carbon dioxide, which is food for plants, it also means a more robust biosphere with more forest, crops and ground cover for more animals and people.

References Andrus, C. F T. Crowe. D. E. Sandweis. D. H.. Reitz, E. J, and Ramanek. C. S., (2002). 'Otolith 6"0 record of mid-Holocene sea surface temperatures in Peru'Science. Val. 295, pp. 150&L511. Balling, R. C. Jr. (1998). 'Analysis of daily and monthly spatial variance campanenls in historical temperature records.' Physical Geography. Val. 18. p. 544-552. Balling. R. C. Jr.. Michaels. PJ, and Knappenberger, P C.. (1998). 'Analysis of winter and summer warming rates in gridded temperature time series.' Climale Research. Vol 9. p. 175-181. Bengtsson. L. Botzet. M, and Erch. M... (1996). 'Will Greenhouse Gar-induced Warming Over the Next 50 Years Lead to a Higher Frequency and Greater Intensity of Hurricanes? Tellu, Vol. 48A. pp. 57-73. Benson. L.. Kashganan, M.. Rye. R.. Lund. 5.. Paillet. F, 5mwt.J.. Kesler, C.. Mensin~.5.. Meko. D, and Lindamm. S.. (2002). 'Iiolmene rnuldrcd~land rnult~ccn~ennuldroughts dimm~ Nonhcrn Ol~fxma2nd Vwadl (ha~rrwryS(tm(r Rrnpx,. Val 21. pp. 659482. Bielec. Z.. (2001). 'Long-term Variability of Thunderstorms and Thundersmrm Precipitation Occurrence in Cracow, Poland, in the Period 18961995.' Atmospheric Resronh. Vol. 56. pp.161-170. have occurred during the historical climate record. His Changnon. S.A.. (1999). 'A Rare Long Record of Deep Sail research showed that, overall, the spatial variability in Temperatures Defines Temporal Temperamre Changes and an temperature anomalies has declined, and that the Urban Heat Island.' Climatic Change. Vol. 42. pp. 531-538. interannual variability in temperature anomalies is Changnon. S. A. (Chnngnon. D., Fosse. E. R., Hoganson. D. C.. negatively correlated to mean hemispheric temperatures. Roth. R. J. and Totsch, J. M.. (1997). 'Ellects al recent weather Balling et, al. (1998) and Michaels et. al. (1998) both extremes on the insurance industry: Major implications for the atmospheric sciences.' Bulletin 01th~Amencan M~teorolo@ol show that as the atmosphere warms, the month-to- Society. Vol 78, 425. month variability also declines (Figure 1). Coates. L.. (1998). 'Deaths and ENSO: Fate. Chance or Change? Karl et. al. (1997) state that GCMs predict temperatures Natural Hazards Quarterly. Vol 4. No. 4, pp. 2-3. will be confined to a tighter range. This is confirmed by Cronin, T.. Willard, D.. Karlsen. A. Ishman, S.. Verardo, 5, McGeehin,J., Kerhin. R.. Holmes. C., Colman. 5. and Easterling et. al. (1997) who found that most of the Zimmerman. A. (2000). 'Climatic vatiability in the eastern United increase in global temperatures has been occumng States over the past millennium from Chesapeake Bay sediments.' during the winter and at night. Summer maximum air Geology. Vol. 28. pp. 3-6

Post Disaster Surveys: experience and methodology

David King examines and questions research methodologies used in disaster studies in Australia.

The Centre for Disaster Studies was able to maintain its By David King, Director of the Centre for role of carrying out immediate post disaster studies Disaster Studies, School of Tropical Environment through the introduction of Emergency Management Studies and Geography, lames Cook Universit)! Australia3 Post Disaster Grants Scheme in the mid 1990s Townsville. (Fleming 1998). The centre had been re-established in 1994 with a completely new group of researchers who Rapid response post disaster studies take place had had no previous involvement in disaster research. immediately after a disaster has occurred, so the Involvement in post disaster studies thus provided rapid researcher carrying out the study needs to have a clear experience. and North Queensland provided no shortage methodology and research aim as soon as the disaster of events. The first study carried out by the new centre happens. The question raised by this type of research is was not actually a disaster declaration. Cyclone Gillian whether or not there is a right way of doing it, or at never eventuated, but it was the first time in a number least a standard methodology This question has of years that a major city, Townsville, had recieved a concerned researchers in the Centre for Disaster Studies cyclone warning. Thus the Bureau of Meteorology was at James Cook University since we initiated a fresh interested in learning how the community had emphasis on the social impact of catastrophes in the mid responded to its warnings. This began a very successful 1990s. This paper will begin by illustrating our own relationship between the centre and the Bureau of experience in post disaster research to show the range of Meteorology Most of the succeeding post disaster problems encountered in this kind of research, and the studies drew on funds contributed by the three agencies general findings and issues that the studies raised. of Queensland Department of Emergency Senices, the Following the summary of the centre's research. the Bureau of Meteorology and Emergency Management paper will go on to examine what other researchers have Australia. Thus between 1997 and ZOO1 the Centre been doing in the field of post disaster research, and to auempt to classify these types of studies to conclude whether or not there is a standard or typical approach and method. Fleming (1998) wrote an early review of the Australian post disaster program, prompting an ongoing evaluation as studies eventuated.

Sending rapid response teams to examine the impact of a hazard immediately after the event, has been a role of the Centre for Disaster Studies since it was kt established in the 1970s. The Centre was originally established in 1979 following a decade of concerted disaster research initiated aher the devastating impact on Townsville of Cyclone Allhea in 1971, and Tracey on Darwin. The creation of a centre following a serious natural disaster mirrors similar initiatives in emergency management and hazard research in other parts of Australia. The shock of a catastrophic event prompts a determination to 'do something' to be better prepared 'next time'. Inevitably in Australia there have been sufficient 'next times' over the succeeding years to justify the continuation of centres and institutes once established. While the Centre for Disaster Studies was initially concerned primarily with understanding the physical causes and impacts of hazards, the emphasis Examining the impact of a disster immediately after the event has been a role for the since the mid 1990s has been more concerned with the Centre for Disaster Studies since it was first established in the 1970s. social and community impacts of disasters. Townsville 1997 Cyclone Gillian Questionnaire, warnings & behaviour Cloncurry 1997 Flood Questionnaire, interviews Cairns 1997 Cyclone Justin Longitudinal Questionnaire Mareeba & short survey questionnaire lnnisfail Townsville 1998 Floods (Syd) Household Telephone, Business & Mag. Island surveys interviews Burketown 1998 Floods Questionnaire, interviews Normanton Karumba Innisfail 1999 Cyclone Rona Flooded buildings survey Cairns - Barron River suburbs Evacuation questionnaire Port Douglas & Mosrnan Warnings & behaviour questionnaire Wujal Wujal Warnings & impact interviews Broorne 2000 Cyclone Rosita Questionnaire, interviews Cairns 2000 Cyclone Steve Longitudinal Questionnaire Mornington Island 2001 Cyclone Abigail Questionnaire, Interviews Total 17113 5 2 9 9

carried out 17 separate studies in 13 different towns and contradictory demands being placed on the researcher communities (neally all in Nonh Queensland) following such that research objectives are watered down and 9 separate hazard events. These are listed in Table 1 outcomes must be sensitive to the politics of the sponsoring organisations. Fleming (1998) illustrated Partners and research needs these issues of agency control of research and As the emphasis of the Centrek research changed, so also manipulation of results to ensure a final product that did the role of post disaster studies, although the impact satisfied the organisation. However, the greatest threat to on people was always paramount even when small research centres like the Centre for Disaster understanding the process of the hazard dominated the Studies, is for the research initiative to be taken entirely research. While post disaster surveys in the 1980s were away from universities and shared out amongst funded from the Centrek own budget, university government agencies, each with a narrow area of re-structuring in the 1990s has resulted in all of the responsibility The very nature of disasters, in Centre's post disaster studies needing to be funded demanding a totally integrated response from all directly from government agencies and departments. The agencies, runs counter to such a trend and reinforces the Centre has thus had to work in much closer partnerships need for multi-organisational collection and sharing of with organisations outside the university, and has had to research data. Thus an independent post disaster study be more responsive to their needs, rather than engaged complements the specific information gathering of in purely academic research. In microcosm this is an agencies that are responsible for mitigation, response example of a national and international process, whereby and recovery. government funding agencies are demanding more Because post disaster surveys are carried out in direct focussed or applied research from universities. While that liaison with funding partners, those partners, govern- shift in focus and funding has prompted a vigorous ment agencies and local government, play a direct role debate about academic freedom and the ongoing need in formulating the research question. Some of these for pure research, this has been less of a problem for the questions remain fairly standard across a number of Centre for Disaster Studies because its research has studies, but much of the information gathered is specific always been applied to the practical problems of dealing to the severity, timing and location of the event, as well with and learning from disasters. as local issues. Some of these local issues are strongly Pan of the reality of that process of applied research is political or become political controversies later. This is the need for the researcher to work with several funding the nature of a disaster as a catalyst of crisis and change. agencies at once. These can affect the independence of Political controversies cannot be ignored in post disaster the research and may result in competing or even studies but in working with government agencies and the public it is a tension that requires sensitivity in disaster researchers but have switched to the impact of a communicating those issues. While the partner agency catastrophe when it has been close to home. may influence or define the research questions, inevitably the research design is primarily driven by the By drawing on a pool of researchers in a region that is disciplinary background and previous experience of the regularly impacted by predictable natural hazards the researcher, as well as relating to the broader literature on Centre for Disaster Studies has been able to attend a research methods. If this background does not fit with large number of hazard events over a short period. All the needs of the sponsoring agency the researchers may may be described as relatively low impact events in that find their work ignored, or their services dispensed with loss of life was minimal and structural damage was (Fleming 1998). either concentrated or at least did not amount to total devastation. The 9 events have all been floods or The partner agencies have roles that directly relate to the cyclones, or a combination of both, with repetition of management of the disaster event. They have clearly events in the same places. Thus as well as relating North defined responsibilities and procedures in monitoring or QueenslandiNorth Australian experiences to the broader warning of a hazard, or in responding to the crisis both literature, the Centre has also had the opportunity to during and immediately after the event, andlor cross relate its own post disaster studies, especially managing the clear up and recovery They may therefore within the same communities and even individuals. cany out their own surveys in the form of needs assessments for disaster victims and communities, and Post disaster experience they will very likely assess their actions at some period From all of these studies we can summarise seven main after the hazard in a formally structured de-briefing. groups of impacts or issues. These are: 1. the unequal Post disaster studies carried out by the Centre for distribution of the impact; 2. loss of services during the Disaster Studies and similar researchers are not of this event; 3. a lack of expectation of the impact; 4. late or type, and will attempt to avoid being in the way during minimal preparation; 5. community or neighbourhood the most immediate post disaster period. Instead the response; 6. confusion concerning warnings and the purpose is to capture a bigger picture of the disaster that media; and 7. a level of resilience. will contribute to the de-briefing and attempt to relate the experiences of this disaster to those of other In all of the disasters we have studied, only a portion of the community experienced severe loss or impact. These catastrophes, most often as they have been reported and were places that were especially physically vulnerable. analysed in the academic literature. The rapid response Some such as the Black River Settlement outside post disaster study must be comparable to other studies Townsville, or housing along the terrace of the of its kind, hence the need to use or be aware of the Cloncurry River, should never have been there in the research methodology ol previous work. It would be Iirst place. Apart from responsibility of local extremely uselul if there were guidelines available to the government, the residents 01 these places were clearly sponsoring agencies that could mitigate against excessive ignorant of the hazard on their doorsteps. Other or unwarranted influence over the researcher developments, such as the trend to enclosure of spaces However, the post disaster study must still take place as underneath houses (often as granny and teenage flats), soon as practicable after an event to record peoples' equally exhibits an attitude that the flood hazard is no experiences and memories while they are still fresh, but longer a threat (King 1998. Goudie 6s King 1997). equally without adding to the stress. The results of the The loss of emergency services and utilities during a study also need to be available for de-briefing sessions hazard has already been illustrated. Apart Irom the which may be some week; or at most a few months after Townsville flood, this has occurred in all natural the event (depending on the severity of the disaster and disasters. Generally people expect that they will lose the agencies involved). This means that most post power and water for a while, but it is a more serious disaster studies will necessarily be rapid sulveys. The oversight for emergency service operations to be located researchers must also be able to respond quickly, work in the more vulnerable parts 01 the city (King 1997, with partner agencies and produce useable results in a 1998). This has been an historical trend, where these short time frame. Some professional teams ol rapid se~ceshave been sited in a central location, which in response researchers exist, but generally government the case of the old city centres is most likely to be in the agency lunds will not support permanent employees vicinity of the wharf and sea front. Thus apart from the whose only role is to study a disaster, and once an sheer size of a major disaster in a city, the police, fire, individual is involved in the management or response ambulance etc. may not be able to get out of their work, it is no longer possible to cany out an buildings, let alone provide widespread assistance. independent big picture analysis. The rapid responders have included many academic researchers and university The most common response from people who have staff who become involved in local events as part of experienced major loss, was surprise and disbeliel, often their wider research. Many have not been primary backed up from community and personal knowledge. TvDe of immn varies enormouslv so survev suesfions musf also suit various laces and events that the river had never risen so high before, or the friends and neighbours. Clearly everyone in the floodwaters had never been so extensive. This is usually community has to know how to deal with the hazard, quite true, for any individual, but the devastating because the reality is that during and immediately after natural hazard is a predictable process at the State level. an event, many thousands of people are going to be People who experienced severe loss of property, actively involved in providing assistance. experienced that loss precisely because they never expected it (Goudie & King 1997). All of the post disaster studies contained questions on warnings and messages from the authorities. The media The disbelief is compounded by a universal lack of transfers messages, so that part of regular preparation is adequate preparations, whether for flood or cyclone. to have a working battery radio. Prior to and during an or at best hurried and minimal late preparations. Part of event both television and radio stations relay cyclone this derives from stoicism, part from a desire to he in and bad weather warnings. The technical language of control and not to be panicked. The result is that people warnings has caused some confusion, but the Bureau of end up out of doors once the strong wind has already Meteorology has responded by simplifying its messages. started and debris becomes airborne, or as flood waters There is also controversy over the use of sirens, and the rise, they are out in deep water moving belongings. location and use of emergency shelters. But after all people and pets. This has happened in many instances disasters the greatest criticism has been against the in the dark. In remote communities the lack of media, for inconsistency in the timing of broadcasts of preparation has resulted in a widespread lack of food. messages, and for either exaggerating or playing down a necessitating expense airlifts (Berry 1997). threat. There is an expectation gap between the public In all disasters people reported checking on or helping and the media, in which commercial television receives their neighbours. There were tales of genuine bravery the greatest criticism. Despite advances in and risk, some of which made good media stories. Most communication technology, remote communities in risk taking and rescue could have been avoided, though, North Queensland occasionally still fail to receive any if people had acted earlier, or had never built in such warning at all, as transmitters fail, or remote area vulnerable locations (King 1998). In the remote broadcasts are made from very distant locations where communities people expect to rely on their neighbours, there is no knowledge of local conditions. This was the but are unlikely to request assistance until the last experience of Wujal Wujal aboriginal community on minute. In the cities. the numbers of people needing Cape York Peninsula as Cyclone Rona passed virtually help runs into thousands, so that reliance has to be on overhead (Cottrell et. al. 2001). Finally there is a level of resilience inherent in Shortly after the main community survey had been communities, that emerges from the interviews and administered to a sample of 700 households, cyclone responses of participanls. Natural disasters such as Justin crossed directly over the northern beaches. Then floods and cyclones in Northern Australia are seen as in 2000 cyclone Steve crossed in the same place. Thus pan of the pattern of life and seasons. A lack of physical post disaster studies following Justin and Steve, re- preparedness is countered by a higher level of mental administered the same awareness and preparedness preparation, or perhaps risk acceptance. The disaster is questionnaire, with some modifications, to the same usually primarily an economic one as damage and crop sample of households. While there was some change in loss devastates a region, and the response to this is to residence, many of the same people participated in the clear up, rebuild, re-plant and get on with life. sequence of studies enabling a measurement of change over time, and in response to subsequent experience. Some questions were common to all surveys, especially those concerning warnings. However, with just the three The longitudinal surveys were significantly different floods the type of impact varied enormously because the from the much shorter rapid appraisal method surveys places and events themselves were very different. The carried out after other disasters. While the shorter Cloncuny flood was a river inundation that severely surveys generated comparative data, the repeated damaged dwellings in a pan of the town, while the issues community surveys give depth and much greater in the Gulf floods were primarily the problems that understanding of the processes at work in the related to isolation, damaged infrastructure and allied community. However, it was chance that provided the health concerns. Townsvillc's flood was a high rainfall opportunity to resurvey the same community. event that inundated the whole town, temporarily Generally it will be the rapid appraisal type of survey isolating almost every household. Similarly the cyclones that will have to sutlice. Thus it has been an were different and by the time of cyclone Rona we appropriate point at which to examine the experiences decided to do completely different surveys in different and methods of other researchers who have been places that had been impacted, according to the local involved in post disaster studies. issues. In Innisfail river flooding had been a major issue so the post disaster survey looked at the impact on Review of post disaster methodology houses that were in the flood zone. In Cairns the Tables 2 and 3 summarise an analysis based on 130 post population of two riverside suburbs was evacuated at disaster reports primarily available in the libraries of the night, so the survey examined the evacuation experience. Australian Emergency Management Institute at Mount In Douglas Shire (Port Douglas and Mossman), as well as Macedon and on [he web site of the Natural Hazards Wujal Wujal aboriginal community to the north, the Research Applications and Information Center at the issue was the rapid onset of the storm and the adequacy University of Colorado. Emergency Management or inadequacy of warnings. Thus the studies in these Australia modelled its post disaster grants scheme on the communities concentrated on those issues. Colorado scheme. However, a few of the reports that were examined came from outside both of these Most post disaster surveys were necessarily short, schemes. All dealt with immediate post disaster studies. consisting of rapid appraisal method questionnaires, although some of them very usefully made comparisons either administered face-to-face or by telephone, and of experiences across several events. backed up by inteniews of community leaders, key informants and experts or oflicers responsible for These studies have been classified in Table 2 according components of the response. to the place of study and the type of hazard. The Australian studies were carried out by Australian based The most useful surveys, though, have been the researchers, but most of the other non USA and USA longitudinal community surveys of the Cairns northern studies were carried out by United States based beaches suburbs carried out by Linda Anderson-Berry researchers under the National Science Foundation These began in 1996 as a TCClP (Tropical Cyclone Quick Response Scheme, which unlike its Australian Coastal Impacts Program) project to examine awareness equivalent allows funds for researchers to travel outside and preparedness for cyclones and storm surge. The the USA. Outside the Caribbean, most American northern beaches were selected as a relatively new outer researchers had an existing research link, or ethnic suburban area in a vulnerable location along the coast. connection with the overseas disaster site. The first survey was independent of any hazard event (there had not been a cyclone warning in the area for six Table 2 has sub categorised studies according to the years). The initial survey drew on earlier awareness and type of hazard that was involved in the disaster. In fact preparedness questions that had been used in Townsvllle most of the studies were primarily concerned with during the 1970s following Cyclone Althea, but the societal response to a catastrophe, rather than being a northern beaches survey instrument was greatly study of a particular type of hazard. The hazard is not expanded and generated extensive data. It was also the centre of the disaster, but rather the impacted complemented by suweys in schools of grade five and community. The hazard is the framework, but it nine children. inevitably configures aspects of response, mitigation and warnings. Besides, the disasters are powerful events reviews were of the same hazard but as it unfolded in that we name and construct within the framework of the different locations, such as El NiAo for example. character and form of the hazard itself. Case studies Cyclones (Hurricanes), floods and earthquakes Case studies of disasters were used as examples to dominate, and with tornadoes and bushfires account for the vast majority of hazards. All necessarily benefit from examine Emergency Management issues and policy post disaster study because of their predictabilily and analysis, as well as repons that were primarily the need for the population to be physically and concerned with mitigation. Physical assessments menlally prepared for a hazard event. Table 3 classifies of disaster impact used methodologies that were the main research methodology used in these studies appropriate to the discipline, such as engineering or geomorphology These were entirely case studies of All 130 reports that were reviewed for this paper specific events, but there was little or no involvement concerned experiences that were gleaned directly from of human participants involved in the disaster. The actual disasters. However. 25 were reports or aim of these studies was to add to knowledge of the bibliographies that either examined a number of disasters process of the hazard itself and in some cases the to draw comparative conclusions or they were annotated strengths and weaknesses of built structures. bibliographies. Most of these have been classified as reviews in which a general analysis of previous hazards The majority of the quick response studies were case summarised findings and experiences. Three of the studies of a specific disaster, with a broad aim of

Region1 Australia USA Central South Japan Pacific South & Europe Total 1 Hazard America & America South (inc. Caribbean East Asia Turkey

Flood 2 16 1 2 21

CycloneMurricane 3 14 7

Bushfire 5 3

Earthquake 1 5 I ! Tornado 7 1 Drought 2 2 1 Blizzarril Ice Storm 3 1 1 Volcano I Landslide I Severe Storm 1 1 ! 1 Plane Crash 2 2

Massacre 4 4 I Chemical Soill 2 1 3 I Bus Crash 1 1

Ship Loss 2 2 I General & Other 2 1 1 29

Total 14 57 10 2 2 4 7 9 130 4 Source: Emergency Management Australia. Natural Hazards Research and Applications Center University of Colorado. Bureau of Transport and Regional Economics Adamson. 5. 1997. Allen R.D. & Ram. W. 1998, Benight ,CC. 1996, Benight. C.C. & Harper, M.L. 1997. Blanchard-Boehm, D. 1997, Bush. D.M. et al 1996. Butler, D.R. 1997. Caporale. R. 2000. Carley. K. & Harrold. J. 1993,CoarseyRader. C.V. 1995. Comfort. L.K.. 1996 & 20W. Dow, K. & Cuner, S.L. 1997, Drexher, K.D. & Abueg. F.R. 1995, Dymon, U.J. 8 Borcoe, F.P. 1996, Edwards. B. et. al. 2WO. Enarson. E. 2WO. Fiwher, H.W. ill. 1999, Gant D.B. 1996. Hapke. H.M. 2WO, Jones. R.T. 1993, Kraure, G. 8 Dyer, C.L. 2WO. Langley. A. &Jones, R.T.2000, Legates. D.R. 8 Biddle. M.D.1999, MrEntire, D.A. 1998& 1999, Mitchell. W.A. 1996, Montz, B.E. &TobinG.A. 1997. Neal, D.M. 1995. O'Brien. P.W. & Payne. 1. 1997, Paul, B.K. 1995, 1997. 1999. Peraz-Lugo. M. 1999. Perry, J.B. et. al. 1996. Pine, J.C. 1999. Rao. et. al. 1985. Rodrigue, C.M. & Rovai. E. 1998. Rorsi. 1. 1998, Rodilsky, J.L 1999, Rubin. C.B. et. al. 1985. Sank D.N. 8 Kaiser. C.F. 2000. Schmilden, T. & K. &Om, Y. 1996. Schmidiin. T. & King. P.S. 1997, Schmidlin, T. et. al. 1998, Smith. Emm 8 North. C.S. 1998, Sylves. R. 1996. liefenbacher, J.P. et. al. 2WO. Tobin. G.A. 8 Montz. B.E. 1997. Tollinger, M.L. & Dixon, D. 1999. Waelde, L.C. et. al. 1998, Wilson. J. & Oyola-Vemaiel. A. 1998. Note: 25 reports were comparative analyses or annotated bibliographin of multiple evens and hazards. Assessment Review Management Study I case study 10 2 4 2 3 na' Questionnaire 1 48 Interviews Secondary Sources I Post Trauma 1 Methodology ! Observations I j General Analysis Focus Groups Economic Analysis

Source: Emergency Management Australia. Natural Hazards Research and Applicationr Center University of Colorado. Bureau of Transport and Regional Economics Adamson. 5. 1997. Allen. R.D. & Rorse, W. 1998, Benight. C.C. 1996. Benight. C.C. & Harper, M.L. 1997, Blanchard-Boehm. D. 1997. Bush. D.M. et. al. 1996 & 1999, Butler, D.R. 1997. Caporale. R. 2000. Comfon. L.K. 1996 & 2000, Coursey-Rader, C.V. 1995. Dow. K. & Cutter, S.L. 1997. Drexher. K.D. & Auberg, FR. 1995. Dymon. U.J. & Boxoe, F.P. 1996. Edwards. 8. et. al. 2000. Enarson. E. 2001, Fisrher, H.W. Ill. 1999. Gamble, D.W. 2000. Gant, D.B. 1996. Hapke. H.M. et. al. 2000, Kraure. G. &Dyer, C.L. 2000, Langley, A. & Jones. R.T. 2000. Legates, D.R. & Biddie, M.D. 1998. McEntire. D.A. 1998 & 1999. Mitchell. W. A. 1996. Montz, B.E. &Tobin. G.A. 1997, Morrow, 8.H. & Ragrdale. A.K. 1996. Neal. D.M. 1995. O'Erien. P.W. 1997, Paul, B.K. 1995. 1997. 1999, Perez-Lug0 M. 1999. Perry. J.E. et. al. 1996. Pine. J.C. 1997. Radrigue. C.M. et. al. 1998, Rarri. 1. 1998. Rozdilsky. J.L. 1999, Sattler. D.N. &Kaiser. C.F. 2000. Schmidlin, T. & Ono. Y. 1996. Schmidlin. T. & King, P.S. 1997. Schmidlin. T.W. et. al. 1998, Sylver, R. 1996, Tiefenbacher, J.P. et. al. 2000. Tobin. G.A. and Montz, B.E. 1997, Tollinger, M.L. & Dixon. D. 1999. Waelde. L.C. et. al. 1998, Werner, L.S. et. al., 1998, Wilson, J. & Oyola-Yemaiel, A. 1998. Note. There were care studies in each of the reoarate methods and case studies also constituted a method themrelves. assessing the impact, response and behaviours of The questionnaire is often supplemented by some other panicipanls. The studies carried out by the Centre for source of data and information. The most common Disaster Smdies fall into this same broad category, and sources are interviews and secondary data. In some altogether they form the bulk of post disaster studies. cases these have been the entire survey method, especially where the impact is small or remote. Key It is in this kind of post disaster study, where the informant interviews may include those whose role is to researcher is approaching the disaster to find out what respond or manage, as well as members of the public. happened and to search for lessons that may contribute Several reports used the term 'snowballing', a technique to mitigation and Emergency Management, that a variety whereby the interviewee refers the researcher on to of approaches may be employed. The most commonly other informants. Where the populadon is not known used research tool is a rapid appraisal questionnaire and a proper sample cannot be drawn, this technique is administered face-to-face or by telephone in the majority highly effective as links will usually go in diverse of studies, or by drop off and pick up where the directions. Some researchers used tape recorded researcher has a longer time in the field. Face to face or inte~ews,but these can seem intrusive and the telephone surveys are necessarily short and thus can majority (like newspaper reporters) still rely on field only cover a brief set of issues. However, after a disaster notes and log book. many people are still excited or shaken by the experience and are frequently willing to talk to the Secondary sources are records made by management inteniewer for an extended period. These extra agencies, databases such as the census, local government anecdotes can contribute to key informant type records and databases etc. A number of researchers interviews and provide valuable insights. made extensive use of textual analysis from newspapers and reoorts at the time of a disaster. News reooners Drop-off and pick-up surveys require a face-to-face invariably exaggerate and sensationalise stones, but they introduction and thereby achieve a much higher return will often capture the main issues even if numbers and rate than a mail out. They also allow for many more. names are not always accurate. and much more complex, questions. Their shoncoming is the need to prepare much more substantially before Observations were generally used as a backup to a main entering the field, which requires a sound knowledge survey method such as interviews and quesdonnaires. of the community and the likely impact of the disaster For very experienced researcher5 observation techniques before designing the suwey instrument. are valid, especially if categorised from a checklist of The devastating natural hazard is a predictable process at the State level. issues, impacts and questions. In particular, observations participants in any one study. For example some studies are used to contribute examples and to underscore looked at police post trauma experience, some at statements made by participants. Researchers may also teenagers, some at the elderly, women, medical staffetc. be participant observers. This has been the experience of The researchers were mainly psychologists, with some staff of the Centre for Disaster Studies and clearly was (probably) psychiatrists who engaged in more direct the case for several of the researchers involved in the medical analysis. Generally these studies were either quick response studies. The researcher loses some more longitudinal, with surveys close to and then objectivity (which is not necessarily an essential repeated at some distance from the event, or they were characteristic) but participant observation fills in the entirely distant from the event, in some instances by a linkages between issues and gives powerful insights into number of years. As ihe aim of these post disaster vulnerability and resilience. studies was to understand the longer term impact on peoples' mental health or state of mind, the greater Only three studies used focus groups as a method of distance from the event was an important component of data collection and analysis. This research method some of the studies. Most used highly technical can be extremely powerful, but may be inappropriate standardised post trauma tests and questionnaires. From immediately after a disaster and can be extremely the methodology statements these standardised survey difficult to organise. Only two studies carried out instruments did not appear to have been significantly an economic analysis, and both were done by skilled modified for the type of location or hazard. specialisis using economic analytical techniques designed for disasters. Clearly this area is one Very few of the case study post disaster surveys used where more research needs to be done to contribute both the general approach of questionnaires and to mitigation. inte~ewsetc.. as well as post trauma specialist surveys. They were either one or the other - quite Post trauma studies distinct groups of researchers and surveys. Furthermore Entirely separate from the bulk of the rapid response the post trauma methodologies were oriented to specific surveys were specialist studies that employed post psycho-social models or theories. The general case trauma methods. These were concerned with all groups study quick response surveys approached a broad range of pariicipants, emergency managers, response and of models and concepts as a structure lo ihe rescue personnel and those who had sullered loss or methodology, depending upon the disciplinary trauma in the disaster, although not usually all background of the researcher The difference between these two types of post disaster This opportunity is more limited in Australia, although studies, the post trauma studies and the general quick disaster education programs are on the increase, and a response case studies, significantly separates the greater number of younger researchers are entering the methods employed. The post trauma studies are less field both in universities and in government likely to have to take place immediately after the event departmenls. and thus may not contribute to the de-briefing. Research A small flaw in post disaster studies is their snapshot of a into understanding how people deal with and recover moment in time. A much deeper understanding of the from trauma may contribute to Emergency Management. impact and hehaviour of a community is enhanced mitigation and our understanding of vulnerability and through longitudinal studies, or at the very least, return resilience. While these studies are primarily individual visits at a time significantly after the event. This responses, they may also aid our understanding of opportunity will also develop as more researchers and community and the individual in the community. more research interest become focussed on under- The methodology is also standardised and frequently standing disasters. This is particularly likely as the former takes place in a controlled or clinical.environment. emphasis on response has shifted to longer term By contrast, the quick response case studies are recovely, and most importantly, to mitigation, education primarily fieldwork based and are much more hazard and preparedness. These policy directions have undoubt- and location specific. A variety of methodologies are edly been influenced partially by the experiences of post used to gather information, but questionnaires and disaster studies and the desire to contribute knowledge inte~ewsare the dominant methods. Results and and understanding to long term mitigation and lessons learned contribute directly to an understanding prevention. Thus the developing policy emphases and of individual and community behaviour in response to a the experiences of post disaster surveys reinforce one disaster, and provide insights into ways in which another in moving towards the bigger picture - the long Emergency Management and mitigation may be term and mitigation. While a snapshot in time, post enhanced. They generally do not extend beyond the disaster studies are a record of history and provide us the immediate post disaster period because that is the way opportunity to learn from that history so that we may in which most post disaster studies are funded. avoid repeating past mistakes. Researchers who have been involved in a number of It is reassuring to find that one's own work complements post disaster studies are able to discern common trends that of a wide range of similar studies and fits into a and patterns, hut this has also been achieved by third category of research methodology common to many party researchers who have analysed experiences lrom of the others. This paper began by reviewing our own multiple events and reports. post disaster studies in order to illustrate the diversity of evenls and situations and research problems. in Conclusion summarising these studies and drawing common However, a limitation to a number 01 audies is that they conclusions and issues, the point is that the synthesis of are too isolated, too location and hazard specific, or often crude and rapid fieldwork provides us with a even too discipline or model specific. Some researchers, bigger picture of disaster experiences that helps us in possibly coming to an inter disciplinary disaster study understanding vulnerability and resilience, and in for the first time, explain their models and devising effective mitigation strategies. It is a common methodologies as though they have never been tried or criticism of geographers that they carry out fieldwork examined before, or who regard the event they are and subsequently try to work out what it all means. studying as unique. Generally though, post disaster Rapid response surveys, many also carried out by studies employ common methods, drawing most geographers, frequently do the same thing. Unclear aims significantly on either standardised post trauma and vague research questions are clearly a problem of methodology, or on rapid appraisal methodology many post disaster studies, but the immediacy and uncertainty of the situation frequently leaves little choice A significant contrast between Australian and USA but rapid fieldwork. Consequently it is crucially studies is the ability of the United States to fund important that the research methodology is legitimate research in other countries. A strong case could be made and replicable. In the studies that have been examined lor Australian researchers to carry out studies within the in this paper this is almost entirely the case. Standard immediate Asia Pacific region, especially where a social science methods were used within a framework 01 significant Australian aid response is involved. The the diversity of disciplines, the towards overall research United States quick response studies have been running aims, and the constraints of speed and sensitivity for a much longer period and have developed as time towards victims and managers. has passed. A tradition of such studies in Australia will The full lin of references which accompanying this article also very likely develop over time. An important role has appear on the EMA website at w.ema.govau/ajem/ been played by research students in the US, who have worked with more senior researchers and gone on to David King may be contacted at Email:Da~d.King~jcuCUedu.au lead educational and research programs of their own. The Po0itics off Emergency Response and Recovery: PreOiminarw 0bservations sn ~aiwank~92~Earthquake

Prater and Wu from the Hazard Reduction & Recovery Center, Texas A&M University reinforce the importance of planning and organisational structure in disaster response and recovery.

---~- - and an epicentre at Chichi Township, Nantou County. Prater and Jie-Ying Wu, It jolted people awake at 1:47 am on the morning of Reduction & Recovery Centec Texas A&M September 21. 1999. Approximately 2,471 people, out UniversiM College Station, TX, U.S.A. of a population of 22,048,356, were killed by the ~ resulting building collapses. An additional 8.739 were On September 2 1, 1999 the 921 or Chi Chi injured; of these, 723 were seriously injured. It is earthquake struck the centre of Taiwan. The estimated that 51,925 housing units were destroyed, and earthquake's magnitude was ML 7.3. and MW 7.7 54.376 units were partially destroyed. The estimate of and the epicentre depth was 8 km. The earthquake total economic losses due to the eanhquake is about $14 billion'. or 3.3% of Taiwan? GDP hit at 1.47 am, and left about 2,471 dead and 11,305 seriously injured. Over 10,000 buildings An event of this scale in a small country undergoing collapsed and direct losses were estimated at rapid social, economic and political changes can have US$14 billion. This paper describes some of our noticeable effects. This paper is based on a study in preliminary findings on a National Science which data was gathered primarily on emergency Foundation project conceived in order to study the response and secondarily on disaster recovery organisational and institutional response to 921. Project staff used elite inteniew methods to gather data concentrating on linkages between organisations on emergency response efforts at the local, county, and and different levels of government. The study area central government levels. lnteniew data were covered the two most affected counties, two supplemented with documents produced by different smaller towns and Taipei. Data were gathered from responding agencies, media reports, and other reports. Our efforts were centred on the two counties most interviews, government documents, and media alTected by the earthquake: Nantou and TaiChung. reports. Our research questions centred on the We also studied two townships, one in each county: agenda status of earthquake preparedness, the role Puli and TungSi. Some interviews were conducted at of the military in disaster response and recovery, the the central government level in Taipei. This paper role of the private sector in disaster response and describes some of our preliminary findings, focusing on recovery, and the effects of local government the relationship between the disaster response effort capacity on disaster response and recovery. and the Taiwanese government system. The next section The authors gratefully acknowledge support from reviews the theoretical questions involved and describes the propositions to be examined. A presentation of the the American National Science Foundation (CMS data and an analysis of how it addresses these 0085056) and assume sole responsibility for all propositions follows. statements and conclusions. Theoretical issues There is a large and growing body of literature on the Introduction importance of agendas in policy change (Schattschneider The 921 earthquake was the strongest to hit Taiwan for a 1960. Bachrach and Baratz 1962. Kingdon 1995, Cobb century, with a magnitude of 7.6 on the Richter scale and Elder 1972, Baumganner and Jones 1993). Scholars

1 All monetary amounts are given in US dollars, at the rare of 32NTB to 1 have used this literature to help us understand why it is so difficult for decision-makers to concentrate more of societyk resources on disaster prevention. It is common for societies to live through a repeated disaster cycle of inadequate mitigation and preparedness leading to high levels of disaster impact and an unsatisfactory emergency response (May and W~lliams1986. Birkland 1997, Prater and Lindell2000, Olson 2000). The 921 earthquake provides a valuable opportunity to examine the generalizability of these ideas by applying them to events outside the United States. From this literature, we derived the expectation that disaster preparedness had low agenda status in Taiwan before the earthquake. This would have had the effect of limiting the resources and influence of the agencies charged with disaster response. thus negatively influencing emergency response at all levels of government.

The other issues addressed in this paper are drawn from the literature on governmental centralization and decentralization. There is a vast literature on the subject, but there is as yet no widely accepted definition of what type of government can be called centralised and what can be called decentralised (Hutchcroft 2001). By most measures of governmental centralization, Taiwan is a relatively centralised system. For example, numerous scholars have used financial indicators of centralization (Pickvance and Preteceille 1991, Smith 1985, Nickson 1995, Manor 1999). In Taiwan, there is a national level law governing the allocation of government revenues and expenditures that gives revenue from the most productive taxes directly to the central government. Some of this revenue is then disbursed to the lower levels of government, but this share is less than half overall, and in some cases local governments only receive 10%. Municipal, county and township governments are allowed to collect taxes on some activities, and 'special' taxes can be enacted by local level governments. Local governments have to share some of this revenue with the central government however, and are often reluctant to enact politically unpopular 'special' taxes. Local government budgets are thus dependent on subsidies from the central level to meet basic payrolls.

Another indicator of centralization is the location of the planning process (Sivanna and Azzii 1996). Taiwan has national legislation covering urban and regional planning. All townships are required to develop urban land use plans and emergency management plans, which are to be checked by the higher levels of government for compliance with national legislation. In some cases, townships do not have the resources to develop their own land use plans, and so must hire consultants to do them, or get help from the central government to develop them.

The location of basic governmental functions and the amount of discretion local government agencies have Auenda status of emeruenw manauement was low before the earthouake. in managing these functions is another indicator of centralization (Page and Goldsmith 1987. Manor 1999). Findings Here, too, Taiwan is quite centralised. For example, there are national standards dictating the number of 1. The agenda status of emergency management was low firefighters. fire trucks, and emergency medical units before the earthquake, which had a negative eject on cities must have per unit of population. All law emergency response operations. enforcement is centralised, with policy and aaffing As is often the case, emergency management did not decisions made at the national level by the National get much attention before the earthquake. Taiwan had Police Administration. adequate building codes, but enforcement was not There has been little research on the relationship always strict. During periods of building boom such as between the levels of governmental centralization and the early 1990s there was a tendency by contractors to emergency response systems. Two early works explicitly ignore some construction regulations, and by govern- addressed this issue. Anderson (1969) reported that ment agencies to overlook this practice because it allowed the supply of housing and office space to be centralised systems tend to produce higher levels increased rapidly The rapid increase in office and of military involvement in emergency response. residential space led to lower prices, which was a In centralised systems, local governments are often quite politically popular accomplishment. weak. The military is often the best-organised and best- equipped institution in a country, and is called upon to Land use planning was another potential mitigation use its resources and personnel in the national crisis policy that looked better on paper than in practice. provoked by a large-scale disaster. Thus we expected to There were rules about where construction was see an active militaly presence in Taiwanh response to allowed, but enforcement was not uniform Townships the 921 earthquake. varied widely in their levels of capacity, and local officials did not always understand the risks their McLuckie (1975) found that decentralised systems are jurisdictions faced. There was an over-reliance on associated with an increased role for the private sector in mitigation coupled with a lack of understanding of emergency response, reflecting a wider distribution of disaster vulnerability. One of our informants in Puli, a resources throughout society and greater openness in city badly damaged by the quake, said "Nothing ever decisions about resource allocation. He found that happens here. We did not expect such a tragedy" This specialised resources were often called in from outside in spite of Puli's mountain location, near an earthquake the impact area in both types of systems, but such fault and in a landslide zone. resources were more likely to be under local civilian Many emergency management functions in Taiwan were control in decentralised systems. In addition, he found given to the National Fire Administration (NFA) under regions with greater resources were more likely to act in the Ministry of the Interior (M01) and to local fire an autonomous fashion than poorer regions in the same departments. However, other functions were allocated country McLuckie's work led us to expect that the on the basis of hazard, with the Ministry of Economic private sector would not be an important actor in the Affairs (MOE) having responsibility for floods and 921 response, and that TaiChung county, the more pipeline explosions, the Ministly of Transportation and prosperous county in our study, would be more likely to Communication (MTC) having responsibility for act quickly and independently of the central government transportation accidents, and the M01 for fires. than the poorer county government of Nantou. earthquakes, and typhoons. In theory, the institution in charge of each type of hazard is to cover all four phases Based on this literature review, we have the following of emergency management (mitigation, preparedness, propositions to examine: response, and recovery). In practice however, there has Disaster preparedness and emergency management been some specialisation, with the NFA under the MOI had low agenda status in Taiwan before the concentrating on preparedness and response including earthquake. This low status is expected to limit the search and rescue (SAR) for floods, and the Water resources and influence of the agencies charged with Resource Development Administration of the MOE disaster response, negatively affecting emergency doing flood warnings. The Central Weather Bureau of response at all levels of government. the MTC provides warnings for typhoons, as well as An active military presence in Taiwan's response to running the island's network of seismographs. The the 921 earthquake is expected. Planning and Building Administration of the MOI does work on earthquake mitigation, and sheltering is the The private sector is not expected to have an responsibility of the Ministry of Education. imponant role in the response to the 921 earthquake. TaiChung county is expected to act more quickly and The basic structure of emergency management in Taiwan independently of the central government than revolves around the Central Hazards Mitigation Council Nantou county. (CHMC), and the NFA plays an imponant role in this None of the sovernmenn or institutions olanned for situations where- buildinos were damaoed

Council. The CHMC is aided by an Implementation budgeted to run the CHMC or the CHMC', Committee and a Technical Advisory Committee. There lmplementing Committee because there are no are no paid personnel in any of these units, and there is personnel assigned to these units. The NFA has the no budget for these units. All administrative work is to responsibility of managing the paper work involved in be done by the NFA. The missions of the Council are to running the CHMC and its lmplementing Committee. make hazard mitigation policy, develop and promote a lts budget during the fiscal year '98-99 was about hazard mitigation plan, make and promote an $30,531,400, or 0.8% of the national budget. emergency management plan, and approve the establishment of the Central Disaster Emergency Officials at the central and local levels stated Operations Center in case of emergency. repeatedly that many agencies did not take their disaster-related responsibilities seriously, and were During an emergency, a group is assembled at the unprepared to respond when needed. This lack of national level in the Central Disaster Emergency preparedness was evident at all levels. The NFA and its Operations Center located in the NFA headquarters in associated local fire departments were the only Taipei. This group is composed of both the CHMC and government agencies that had consistently and the CHMC lmplementing Committee. These people are systematically worked on disaster preparedness. executives from many different ministries and govern- When the disaster occurred, the fire departments mental jurisdictions. The Implementing Committee is found themselves dealing with the SAR functions for under the coordination of the head of whatever which they were trained, and also issues of supply, ministry is in charge of that particular category of sheltering, and organisation for which they were events (flood, fire, earthquake, typhoon, etc.). One ill-equipped. It took some days for the other agencies problem is that many of these members of the CHMC involved to learn what was needed and begin to fit into Implementing Committee are themselves Ministers, or their disaster roles. Most city level agencies had no rank nearly as high, and are unwilling to 'take orders' familiarity with the disaster plans the local fire from an equal. departments had prepared, even though they had responsibilities under the provisions of the plans. Another problem with this organisational structure is that none of the participants in the CHMC are full-time Legislation passed by the Legislative Yuan in 1995 paid professionals in emergency management. Their mandated the development and staffing of an EOC by duties as members of the CHMC or the CHMC's each county government, but few have done this yei. lmplementing Committee can therefore conflict with Many fire departments already have EOCs, and run the their duties in their fulltime positions. No funds are emergency call system (119 in Taiwan) already, so local governments often use their EOCs and rely on their earthquake caught everyone unprepared. There were expertise during emergency situations. For example, the problems in the response due to this lack of preparation, fire department in Taichung County has a detailed although improvised solutions were quickly emergency response plan that is revised twice a year to implemented with some success due to the high reflect changes in personnel. Before the earthquake, no capacity of the central government and the prompt other county departments had requested a copy, but arrival of foreign aid. afterwards all departments needed one to find out what their responsihilities were. Disaster drills are held 2. The military is expected [o play an important rnle in annually, but many departments had never participated emergency response operations. in one and had little idea of what was involved in emergency response. After the earthquake they relied on Taiwan's government has a close relationship with the advice from the fire department in order to quickly military because of historical circumstances since the develop their own disaster operation procedures. KMT moved to the island in 1949. In addition. all males (with a few excepdons for family reasons) must spend The day-to-day operations of most departments have two years in military service. Therefore, it would not be little to do with emergencies and many had ignored surprising to see the military playing an important part their emergency responsihilities, believing that "it's the in disaster response. Fire Department's job." Many agency personnel are only assigned to emergency operations part time, so The military is included in the central government's emergency management functions suffer from a lack of emergency management planning, hut this planning is dedicated personnel. The local government executives not nearly so thorough and detailed as, for example, the are theoretically in charge during emergencies, but many US. Federal Response Plan coordinated by FEMA. The know little about emergency management. The lack of Minister of Defense is a member of the CHMC, together an all-hazards approach has limited the understanding with many other ministers, chairmen, mayors, and of the many common elements of disaster response for other high level officials. Emergency duties of each any type of event and the bureaucracy's willingness to ministry and agency were not spelled out in any detail invest in developing disaster response capacity. before the earthquake.

One reason the fire departments in Taiwan do not have On the day of the earthquake. President Lee issued a very high status administratively is that until 1998 the instructions for emergency response and disaster National Fire Administration was a division of the relief. The only instruction to deal with the military National Police Administration. Other departments established a policy allowing men whose families were occupy more central locations in the governmental disaster victims to substitute three months of national structure and have better access to the executive and to service for their regular two year period of military resources. The fire department in TaiChung County is service. That same day, Premier Siew established a set of trymg to become directly responsible to the mayor, policies for disaster assistance and relief. The first of which will further raise its status. these put the MOI in charge of "immediate and on-site relief assistance." The MOI was to be assisted by the The consensus among our informants was that the Ministry of Defense (MOD) and the Public earthquake had changed peoplek attitudes and habits. Construction Commission. More detailed instructions There had been some small incidents like typhoons to the MOD were issued five days later (personal during the interval between the earthquake and our interview, Jan 2001). These instructions gave official interviews, and the level of response from all local approval to activities that had, in some instances. government departments had remained higher than it already taken place. The armed forces were directed to was before the quake. assist with disaster relief and reconstruction "until all such tasks were completed." and to put their Emergency planning for disasters due to natural hazards equipment at the disposal of the civil authorities in general occupied a low position on the public agenda. (Information Division. 1999). and earthquakes in particular were not addressed. In both counties that we studied, any emergency In fact, the military had already deployed many men planning and drills that had been done were focused on and a great deal of equipment to help out with smaller, more common events like typhoons and multi- emergency response. At 2:30 am on the 21 September, casualty traffic accidents. No agency we spoke to had the Commander of Hang-Sun base in metropolitan done any planning for earthquakes. None of the Taipei issued orders for all the military services to begin volunteer SAR teams had any earthquake rescue disaster rescue operations . By 4 am, military units were training. The hospitals' plans were focussed on incidents moving to the impact area. The armed forces set up four with 30 or fewer casualties. None of the governments or Field Command Centres in Nantou County, TaiChung institudons had done any planning for situations in County, Taipei County and the City of Taipei (Urban which their buildings were damaged. Thus, the 921 and Housing Development Department, 2000). The Community Recuperation Centres helped minority communities with their recovery needs, and helped local churches to develop camcitv for communirv serwte.

According to our interviews, military personnel and heavy equipment for debris removal (Information supplies arrived in some places as quickly as three hours Division, 1999). after the earthquake. In other locations, the military response was a little slower. Few base commanders 3. The private sector is not expected to play an important seemed to wait for orders from the central authorities role in emergency response and recovery before acting. The central government order gave official . . . .~ - -- - . ~~ ~~ standing to what was already happening in the disaster In contrast to many recently democratised countries, impact area. Taiwan has a lively and outspoken civil society In addition, the economy is relatively healthy, with a Despite the speed with which the base commanders large and expanding base of privately owned and offered their men and equipment, the soldiers did not managed companies, ranging from small shops to fairly always play an active pan. This might have been large players in the high-tech industry The private because of a lack of training, but there were reports of sector contributed in many ways to disaster response. soldiers waiting for instructions before digging into piles and is maintaining a high level of activity during the of rubble to look for victims. recovery phase.

Taiwan's armed forces are located in many small bases One group that played a prominent role in the response scattered throughout the island. This increases their phase was the International Association of Search and availability for purposes of the State, such as crowd Rescue of the Republic of China. This group began in control, suppression of anti-government demonstrations, 1981 with 15 members. and now has about 10.000 and disaster reliet They are better supplied than the members. They are organised into local teams of local civilian governments, and better organised than 50-90 people each. There are eight paid staff in the most civil authorities. These factors contributed to the central office, and at most one each in the division notable presence of the armed forces in response to the offices. Members pay annual dues of about US$60 per 921 earthquake. By 28 September, they had completed year. They also receive subsidies from local businesses 90,000 supply and evacuation trips, over a thousand and religious groups. This group was very active during flights, and provided thousands of tents, blankets. the disaster response period. They were one of the few vehicles, food and water rations, as well as much needed groups that had some experience with rescues from collapsed buildings because the previous year they had avoided to minimise energy consumption. The architects the chance to learn SAR techniques from a North were asked to use natural ventilation and design features American USAR team during the response to a building to keep the schools comfortable. collapse in Taipei County. Other religious groups also participated in the relief Other examples of private sector activity include effort. The Presbyterian Church in Taiwan adopted a religious organisations like the Tz' Chi Foundation, a social services approach. They have developed a Buddhist group founded by a Taiwanese nun in 1966 network of Community Recuperation Centres (CRCs) in (Liu 2001 and personal interview Jan. 2001). Tz' Chi the hope of making "the best use of donations from has an active disaster relief program under its Mission of church members as well as the general public" and Charity, and has participated actively in international extending support for reconstruction up to four years disaster relief effortssince 1991. Since it is a (Huang and Chen 2001). The CRCs had two missions: decentralised organisation and many members have to help minority communities with their recovery needs disaster experience, members were able to organise and to help local churches develop their capacity for themselves quickly into teams in their neighbourhoods community service. The CRCs offered the same type of after the earthquake hit. By 3 am, they had set up four counseling and emotional support services as Tz' Chi, stations near the collapsed hotel building in Taipei. plus day care for children and senior citizen home care. providing direct aid for victims, emotional and help with nablgating the bureaucratic maze for acquiring psychological support for victims and rescuers, support government assistance to rebuild, and economic for rescue teams, and an information office. They development projects. These centres have filled a critical immediately set up a vegetarian kitchen in cooperation role in Nantou County, where there were only six with other religious groups that were serving non- professional social workers to serve a population of vegetarian food. 544,762, approximately one for every 90 thousand people. TaiChung County was somewhat better off with By the second day, officials from the Taipei branch office 32 social workers for a population of 1,479,105 had arrived in Nantou County with tools and a (approximately one per 50,000 persons). generator, as well as other supplies. Local members had begun relief efforts as a group immediately after the The CRCs acted as coordinating centres, helping quake, digging their uniforms out first in order to victims gain access to needed governmental wn?ces and identify themselves as Tz' Chi members. On the 3rd day services provided by NGOs. Many of the workers were after the quake. Tz' Chi founder Master Cheng Yen hired locally, and had little or no professional training. visited the area and decided that the most immediate Some problems arose in the implementation of the CRC need was for emergency housing. Tz' ChiS resources plan because it had not taken sufficient account of the were immediately focused on acquiring large numbers of role local churches and pastors play in governing tents, and then small, quickly built temporary housing church activities. The programk structure has been units. They also gave immediate cash gifts of $625 to revamped to include more input from the local pastors the families of the deceased and $156 to the injured. and congregations, which is hoped to increase the over and above the cash benefits from the government. systemk effectiveness.

The next need Tz' Chi addressed was rebuilding the Local businesses were very active in responding to the schools. Over 20% of the school and university disaster as well. Although the potential contribudons of buildings in Taiwan were damaged. In the non-urban businesses were overlooked in emergency planning areas of TaiChung and Nantou counties alone. before the earthquake, many donated food, water, heavy 140 schools were damaged (Soong, Yao and Lin. 2000). equipment, tools, and other materials to the relief effort The MOE appealed for help from non-governmental without waiting to be asked. EVA Air heard that organisations (NGOs) with the task of rebuilding the refrigerated containers were needed to serve as schools. Private organisations were allowed to choose temporary morgues and donated several containers to which schools they would rebuild, and the government Taichung County (Personal interview,Jan. 2001). The took on the rest. Many school principals came to Tz' Chi government did not pro~ldeemergency acquisition to request help in building new schools because the procedures until the Emergency Decree published by the private sector had a reputation for faster action than President four days after the earthquake. In spite of this. government recovery programs. Tz' Chi has created every local government we interviewed reported Project Hope to oversee the school construction and has excellent cooperadon from local businesses, and few undertaken the reconstruction of 53 schools. instances of price gouging.

Architects were asked to develop a unique design for Important evidence of the strength of civil society in each of the schools, using local materials and taking the Taiwan is the level of donations to the special fund for physical site and local culture into account during the earthquake relief. This fund, which was set up by the design process. These schools are being built of steel- government on September 23rd to accept donations reinforced concrete, and the use of air conditioning is from overseas and from inside Taiwan received more than $431,500,000 from Taiwan and overseas (921 as a full-time organisation based in Tsong Shin Village in Foundation, 2000). There were many other private Nantou County (seat of the former Provincial accounts collecting money for earthquake relief as Government). While this Commission disbursed well, and total donations have been estimated at government funds, another group, the 921 Earthquake $903 million (Chen 2001). PoliUcal appoimees of the Recovery Foundation was set up to manage the donation Executive Yuan were required by the Prime Minister to fund. Business and academic leaders were appointed to donate one month of their salaries to this fund, but we this foundation by the central government. The new found no other elldence of coercion. The amount raised administration changed some personnel and has tried to for this fund showed that Taiwan is not a poor country, make the awards process more transparent so that relatively speaking, and that the people felt a great deal people would have conlidence that their donations were of solidanty wth the earthquake5 victims. being spent wisely There are thus multiple sources of aid for \lctims, and those who are not assisted by one The vibrancy of Taiwan5 civil society stems at least in group can go to another. pan from h~storicalfactors. While certainly authoritarian. the Republic of China under Chiang Kai 4. TaiChung Counly is expected to respond independently Shek and his son Chiang Ching Kuo was more open ofthe central government and more quickly than than many authoritarian societies. The government, for Nantou County. geopolitical reasons, found it necessary to present a facade of democracy, and over time this generated the Central government planning and control of finances expectation of democracy among the masses. In the end, has not served to even out disparities in the capacities of the authoritarian regime had to allow for political the different local governments in Taiwan. The many expression of dissent, which further emboldened the differences between the two counties in this study opposition and empowered the citizenry (Rigger 1999). include differences in economic base, topography, settlement patterns, local government capacity and In addition to the political reasons for a strong civil characteristics of the earthquake itself. society, there are institutional factors involved as well. Many government agencies rely to a large extent on First, the two counties have very different economic volunteers to fill out their ranks and perform needed bases (see Table 1). Nantoub economy is based on tea, services. This is especially notable among two Iirst- fruit, and tourism. There is little industry, and what responder agencies: police and fire. The volunteer fire little lhere is tends to be based on the local agricultural fighters and police senre regular shifts and in many products, such as the breweries that make wine from cases are as well trained as the professionals. Other locally grow plums. Before the earthquake, there was a agencies also rely on volunteer assistance, as is the case thriving tourist industry centred on the picturesque with the understaffed Ministry of Social Work office in Sun Moon Lake, a favourite honeymoon spot. Other Nantou County and its collaboration with the cities and villages were popular destinations for Presbyterian Church in Taiwan. Thus. Taiwan5 Taiwanese seeking a bit of outdoor activity One inhabitants are accustomed to collaborating closely with interesting result of this is the large number of their government on service delivery, although they are volunteer SAR teams active in the county, specialising in very politicised and do not hesitate to criticise mountain rescue. Some villages like Chi Chi have drawn government policy increasing numbers of tourists because of Local landmarks and historic buildings, especially temples, For these reasons, the third proposition does not hold that were damaged in the earthquake. true for Taiwan. The private sector has contributed significantly to the response and recovery efforts. TaiChung county has a more diversified economy, In addition, many of the private sector resources relying on farming to some extent, but with more involved remained under the control of local manufacturing than Nantou and an active seapon. There governments, NPOs, and the owners of businesses. is an extensive small business sector that is dominated A special committee called the 921 Earthquake Post by manufacturers of mechanical equipment. Disaster Recowry Commission was set up by the Executive Yuan on 27 September. The Commission was Topography has a strong influence on the settlement reorganised in June of 2000 by the new administration patterns of both counties. 83% of Nantou5 land is

County Agriculture Industry Service Annual Household Income Nantou County 23.6% 30.7% 45.8% $26.837 Taichung County 7.0% 48.9% 44.1% 127,350 National Average 8.3% 37.2% 54.5% $27,783 County Population t of County Employees Employees per 1.000 5an Bernadino. CA 1.418.380 Santa Clara. CA 1,497577 Bexar. TX 1.185.394 Dallas, TX 1,852,810 Taichung, Taiwan 1,481,407 Kern. CA 543,477 San Joaquin. CA 480.628 El Paso, TX 591,610 I Travis. TX 576,407 Nantou. Taiwan mountainous, and only 17% is arable and suitable for governments formerly received large subsidies Irom the occupation (Personal interview, Jan. 2001). Villages are provincial government, but these were eliminated when scattered along the rivers that flow through narrow, lush the provincial level of government was dissolved in June valleys crowded with small farm plots. The population is 1999. The subsidy from the central government does low (544,7621, as is the population density compared to not make up the shortfall, so counties frequently have to the rest of the coumry, and Nantou is the only county in rely on bank loans to meet payroll, and most have cut Taiwan that does not border the sea. Like rural areas in staff and other expenses. Some counties have not been many countries, Nantou county is losing its young adult able to meet their loan obligations, and are not yet out population as the young people leave to find education of fiscal trouble. and work in the cities. The two county governments differed substantially in TaiChung County extends a long finger into the their capacity to handle a disaster. Nantou was island's central mountain chain, and the land slopes understaffed and underfunded to a much greater degree quickly into the broad coastal plain of the island's than TaiChung. It is difficult to make direct western coast where the bulk of Taiwan's population comparisons to the United States, because the local lives. The population density is very high, and government system in Taiwan is very different from that townships and villages are frequently not separated by of the U.S. For instance, many functions such as any noticeable open space. Farms are small and education, police and fire that in the US. are controlled intensely cultivated, with up to three crops a year from by the cities or other administrative units are controlled each field. Even in the small towns, any open space not by the county or central government in Taiwan. Table 2 devoted to city parks is cultivated, and buildings of compares the level of county employment in a high several stories line the narrow streets, with commercial services state with a large population (California), and a uses on the ground floor and residences above. Only a low services state with a large population (Texas), to the small part of the county is as picturesque as Nantou, two Taiwanese counties we studied. The California but sections of two national parks fall within the counties have much larger staffs, and even in Texas, the county's eastern border. This section of the county more populous counties have larger staffs than their supports a small tourist industry. counterparts in Taiwan. Given the fact that county governments are given greater responsibilities in Taiwan The level of stafling in Taiwan's county governments than in the US, it is readily apparent that the county depends on two factors: the target figures set by the governments are understaffed. central government and the reality of local budgets. Stafing levels of police, fire, health and environmental Indeed, neither county government we studied was protection departments are determined by centralised prepared to handle a disaster as big as the 921 bureaucracies, although consultation with local earthquake. There was evidence of emergency response magistrates is increasingly important because of the planning, but only in the fire departments, which were emphasis on democratisation. Elementary and secondary supposed to coordinate with other departments. Other school teachers are county employees. About 70% of all departments ignored their disaster management tax revenue goes to the central government and the responsibilities until after the 921 event. As one remaining 30% returns to the local governments informant said: %fore the earthquake, nobody read the (counties, townships. villages) through a complex plan. After the earthquake, all the departments wanted a formula of tax collection and disbursement. Local copy." Many departments had not participated in the annual drills, so they had no familiarity with disaster operations or their role in disaster response. In no case was there any recovery planning, or planning for damage assessment of any kind. Disaster response planning had concentrated on typhoons, which are fairly frequent (on average 2.5 per year). These give plenty of time for warning the population, although evacuation is generally not attempted. No one had planned to respond to a large earthquake in spite of the numerous known fault lines (52) and frequent earthquakes (over 80 significant quakes during the past century). It was felt that the issue was adequately addressed by mitigation practices such as land use planning and building codes. None of the many volunteer and professional SAR groups had trained for earthquake rescues, concentrating instead on training for mountain, landslide, or water rescues.

I In some ways, the experiences of Nantou and TaiChung Before the earthquake there was a thriving tourist industry in Nantou. counties during and after the 921 earthquake were very different. For one thing, the epicentre of the quake was in Nantou, the poorer of the two counties. In fact, aid these problems are related not only to differences in to TaiChung County was slow in arriving because the government capacity, but also to differences in the extent of damage there was not understood at first. counties' economic bases. There are several reasons As in many disasters, clear information on damages did Nantou is having more difficulty recovering than not come in to the central government's EOC quickly TaiChung. First, TaiChung had less damage prop- enough to provide data for eflicient resource allocation. ortionately In Nantou County, almost 40% of the The flow of information to the central government EOC housing stock was destroyed or damaged, while was hampered by systemic problems (lack of prior in TaiChung County, only about 10% was destroyed coordination among responding agencies) and by or damaged. physical damage to communications systems, roads, Second, TaiChung is a wealthier county with a more and bridges. diversified economy The damage to Nantou's tourist The network of seismic sensors quickly pinpointed the attractions was serious, and well publicised. For some eanhquakeh epicentre near Chichi Township in Nantou time, visitors stayed away not only because of the County, so aid was immediately routed there. damage they saw on television broadcasts, but also Unfortunately, it took two to three days for the central because they feared aftershocks. Third, Nantouk county government to realise the extent of damage and government building was destroyed. The county casualties in TaiChung County. This led to problems government was struggling to serve higher than normal with resource allocation, and townships that were hit demands while in temporary quarters that were not hard in TaiChung County felt that they were neglected. designed for government functions. Their ofices were In fact, aid reached both areas at about the same time. still located in a stadium two years after the disaster. and both areas had some isolated villages that were hard Fourth, Nantou's county magistrate was not as eflective, to reach until the roads were reopened. For the first few and has been under investigation for conuption. He was days, what aid came in was femed in on helicopters that in custody while we were in the county for interviews. were also used to transport the injured out to hospitals We heard two perspectives on this. The local outside the damage zone. government people defended him, saying he had only All respondents described the classic problems with tried to help move things along faster by doing away disaster response (Dynes 1974): poor assessment of with some of the red tape. Central government needs, misallocation of personnel or resources. personnel said that he was in trouble because he had convergence of outside resources, and poor only allocated 13% of the aid sent to him for communications due to earthquake damage. In addition, distribution by the end of the first year after the disaster. they suffered from confused lines of authority due to In contrast, the TaiChung County government organised poor planning at the central level that was transferred its Recovery Committee before the Executive Yuan down to the local governments, and an almost total lack established the 921 Earthquake Post-Disaster Recovery of urban search and rescue training and equipment. Commission in the former seat of the provincial government, Tsong Shin Village in TaiChung County on Differences in local government capacity began to October 29th. The TaiChung County Recovery show up as recovery tasks were initiated. Some of Committee has had a fairly good working relationship affecting emergency response at all levels ol government. An active military presence in Taiwan's response to the 921 earthquake was expected. The private sector was not expected to have an important role in the response to the 921 earthquake. TaiChung county was expected to act more quickly and independently of the central government than the Nantou county government.

The first and second propositions were supported by the evidence in this case. Low agenda status had negative effects on disaster response at all levels of government. From the national level to the smallest township we studied, everyone we interviewed stated that no agency other than the fire depanments had done any serious ulanninr- for disaster response, let Both of the countries studied had serious difficulties with emergency response alone the other phases of emergency management. because the magnitude of the event was larger than anyone had envisaged. Emergency management functions ended up in the domain of a minor administration, the NFA, that had with the townships, because the County magistrate is only recently gained independence from the National considered to be a good manager who can delegate Police Administration. Disaster response suffered responsibility as needed. because of the NFk low bureaucratic status. This finding suggests that disaster research, much of which Fifth, some of the most intractable recovery problems has been conducted on North American communities. affect Nantou much more than they do TaiChung. For may also apply to other countries. instance, ground displacement, up to 11 meters horizontally and 10 meters vertically (Loh and Lee As was expected in a highly centralised system, the 2000) has resulted in one of the most serious recovery military played a large pan in disaster response. This issues in the agricultural areas, especially in Nantou was not seen as a negative by most of our informants. County Much of the agricultural land has shifted to a who appreciated the high level of skill the military significant degree and must be resurveyed. There were brought to key issues of transport and housing in the not enough people qualified to do this large amount of immediate aftermath of the earthquake. Military base surveying quickly so outside help was needed. commanders did not wait for orders from the central To compound the dilficulties, land ownership is not command to intervene, but responded immediately, clear in many cases. There are two reasons for this. based on their reladonships to the communities where The first involves the aboriginal people who have they were located. occupied Taiwan for hundreds of years. With the arrival of the Han and Hakka Chinese, the aboriginies were The third proposition was not supported in the case of pushed up into the mountainous interior but have never Taiwan's 921 earthquake. The private sector in Taiwan been granted clear title to lands that they have occupied did have a key role in disaster response that continued for generations. The second problem results from into the recovery period. The private sector includes inheritance issues. Some land has been informally businesses and non-profit organisations such as subdivided among heirs without clear title to any parcel churches or volunteer SAR teams. These organisations ever being established. Now both these groups of people were prominent in both the disaster response and have problems establishing their ownership of land. This recovery phases after the 921 earthquake. This result has affected their ability to receive government benefits may be related to the social mobilisation and for damages or loans to rebuild their houses and replace tremendous growth of civil society that occurred all over crops damaged by eanhquake-triggered landslides the world in the 1980s and 1990s. Taiwanese society (Urban and Housing Development Department 2000. was affected by this wave of social mobilisation. personal interview Jan. 2001). especially after the legalisation of opposition parties in 1989. The relationship between centralization of Summary and conclusions government and a weak private sector may have We began this paper with four propositions for weakened since McLuckie's seminal research was examination: completed in the 1970s. Disaster preparedness had low agenda status in The fourth proposidon was partially supponed. Both of Taiwan before the earthquake. This low status was the counties we studied had serious difficulties with expected to limit the resources and influence of the emergency response because the magnitude of the event agencies charged with disaster response, negatively was larger than anyone had envisioned. Moreover, as noted in connection with the first proposition, disaster Chen Lmg-Chun 2OJI .A D~uwonon ihr Charactenat~s. preparedness was chronically low throughout Taiwan. Impact, and Emergency Korponstr of Chi-Ch, Eanhquakc . Cobb. Roger W.. and Charles D. Elder. 1972. Participation in This minirnised the initial differences between the American Politics: The Dvnamicr of Agenda-Building. Boaan: counties. However, differences in local government Allyn and Bacon. capacity and in the scope of damage have affected the Dynes. Russell R. 1974. Organisational Behavior in Disasters. Columbus. OH: Disaster Research Center long-term recovery processes. Huang. Chiao-Hsing, and Liang-Chun Chen 2001.7he Making ol a Post-Earthquake Community Care Network-Experience of the In summary, this research has reinforced what we know Presbyterian Church in Taiwan.' pp. 35-1-35-10. Proceedinp: APEC Workshop on Dissemination of Disaster Mitigadan about the importance of planning and organisational Technologies for HumanLstic Concerns. Taipei. Taiwan: June 18-21. structure in disaster response and recovery. Without Hutchcroft. Paul D. 2001.Tentralizarion and Decentalizatian in adequate structures in place and intensive planning for Administration and Poliocs: Assessing Territorial Dimensions of Authority and Power' Governance: An International Jaumal of all hazards, disaster response is bound to be confused Policy and Administration 14:23-53. and appear chaotic to the affected populations. This Information Division. Taipei Economic and Cultural Olfice in New perceived inadequacy may have had some bearing on York. 1999. "Imponant Relief Measures by the Executive Yuan lor the September 21 Earthquake." the results of the March 2000 elections, in which the httpJl\~~~.tai~i.org/wha1~new/quake928hrm. KMT lost the presidency Recent scholarship has Kingdon. John W. 1995. Agendas. Alternatives and Public Policies. addressed the issue of disasters and regime change 2nd ed. New York: Harper Collins Liu, King-Pong. 2001. "Lessons irom Tz' Chib Earthquake Relief (Olson 2000). This case illustrates political effects short Programs." pp. 33-1-3?-13. Proceedings: APEC Workshop on of regime change that can follow a disaster Dissemination ol Disaster Mitigation Technologies far Humaniaic Concerns. Taipei. Taiwan: June 1521. Although some of our informants stated that they did Lah. Chin-Hsung, and George C. Lee. 20OO."Geolagy and Tectonics of Taiwan." pp. 5-12. The Chi-Chi. Taiwan Earthquake not believe the problems with earthquake response of September 21. 1999: Reconnaissance Repon. MCEER Technical caused the KMTs defeat, the party understood the Report MCEER-00-0003 danger. This is illustrated by changes in the amount of Manor, James. 1999. The Political Economy of Democratic Decentraliiadon. Washinnon. D.C.: World Bank. the benefits for victims' families, which was increased by the president from $ 6,250 to $ 31.250 for the dead and from $ 3.125 to 16 6,250 for those seriously injured. New York: Plenum Press. McLuckie. Benjamin F 1975. 'Centralization and natural disaster Another instance of policy change in response to response A priliminary hypothesis and irnerpretations." Mass political pressure was in the relaxation of the rules for Emcrgencier. 1. 1-9. declaring a residence 'totally collapsed' as opposed to Nickson. R. Andrew. 1995. Local Government in Latin America. Boulder CO: Lynne Rienner Publishers. 'partially collapsed.' Many benefits hung on this Olson, Richard 5. 2000,"Toward a Politics of Disaster: Losses. distinction: a reduction of military senice, access to Values, Agendas and Blame." lnternotionolJoumal oJMarr temporary government employment, extra points for Emergerules and Dtsaster 18:265-287. Olson. Fichard 5. and Vincent T. Gawronski. 2000. The 1985 anyone taking the college entrance examination, access Mexico Eanhquakc Disaster: A 'Critical Juncture'?" Presented at to the health care system without the normal the 25th Annual Hazards Research and Applications Workshop. copayments, low interest loans for rebuilding or buymg Boulder CO: July 9-12. Page.,, . Edward C, and Michael I. Goldsmith. 1987. 'Centre and a new house, and a reduced income tax rate (interviews I.,ca.~y Fun.l~mr xi<,,snd dlr:rzmn " pp 1-1 1 ~n Edwrd Jan. 2001). It is not surprising that the process of P=de 2nd Zttchxl J Cddm~th(Eds J Centr~land Local damage assessment became highly politicised and Government Rrlmons London LK Sage I'uhluuons Lld Ptrkvancc. Chnr and Edmond Prrtocrllle 1991 'Canclurm vulnerable to the influence of guanxi (relationships). Towmli a compantn.e andg,~~of axe r~rurtunngandlocal power' pp 197-209 in Chn, Pcknnrc md Kdmond l'rr~r,zlllc Whether the government's response to the 921 Edr 5u'c R~stru.~.~nntmd" L:aI P~wrLocdon LK Blddlcr Lt.! earthquake cost the KMT the presidency is uncertain, Urban and Housing Develapmenr Department. Council lor Economic Planning and Development. 200O"Response and although the evidence is suggestive. This project has Recovery to the 921 Earthquake." Internal repon. clarified the ways in which existing political and Prater, Carla 5.. and Michael K. Lindell. 2000."Politics of Hazard governmental institutions can affect disaster response and Mitigation.' Natural Hazards Revlew 1:73-82. Rigger. Shelley 1999. Politics in Taiwan: Voting for Democracy recovery. Future research will compare disaster response New York: Routledge. in Taiwan to cases in Japan (1995) and Mexico (1985). Schattschneider. E. E. 1960. The Semiravereign People: A Realistb View of Democracy in America. New York: Halt. Rinehan and Winston. References Simnna. N and Abdul Azk. 1996. 'An ove~ewand conclusions.' 921 Foundadon. 2000. http:/lu?vw92lfundargtw pp. 278-288 in Decentralized Governance in Asian Countries. Anderson, William A. 1969. "Social structure and the role of the Thousand Oaks O\:Sage Pulications. military in natural disaster' Sociology and Social Research. 53, Smith. 8. C. 1985. Decentralization: The Ternlarial Dimension of 242-252. the State. London UK: George Allen and Unwin Ltd. Bachrach, Peter, and Monon Barau. 1962. "The Two Faces of Soong. Tsu T.. George C. Yao, and ChiLChang Lin. 2000. 'Critical Power' American Political Science Revlew 56:947-52. Facilities: pp. 2941. The Chi-Chi. Taiwan Earthquake ol Baummnner. Frank R. and Bwan D. lones. 1993. Agendas and Sepwmber 21. 1999: Reconnaissance Repon. MCEER Technical lnstabLty in Amencan ~aliti&.Chic&: ~niversityof Chicago Press. Repon MCEER-00-0003. .- -- -~ -- -- ~ ~ .~ .. . . . Birkland. Thomas A. 1997. Alter Disaner: Agenda Setting. Public Carla Prater and Jie-Ying Wu may be contacted at Policy, and Focusing Even&. Washington. D.C.: Georgetown Email: cadaOarchone.lamu.edu University Press. , -- . . Bali bombing Urban search and rescue EMA was called into acdon soon after the Australian The first meeting of the newly formed National Urban response to the Bali incident commenced. The initial Search and Rescue Working Group (NUSARWG) was task assigned by the Department of Foreign Affairs and held in Sydney in October. Replacing the former Trade (DFAT) was to coordinate the reception in National Urban Search and Rescue Steering Committee, Darwin of victims evacuated by the Australian Defence NUSARWG aims to promote the strategic development Force and to coordinate their on-movement to hospitals of urban search and rescue capabilities nationally. in the south. EMA chairs and supports the working group EMA completed its task within three days of the comprising of representatives from each State and incident. Following an offer by the Australian Territory, the Police Commissioners Conference, the Government to treat critically injured Indonesian Convention of Ambulance Authorities, the Australian citizens in Australia, EMA was then tasked to liaise with Council of State Emergency Services, the Australasian the DFAT doctor in Bali to coordinate the movement of Fire Authorities Council and the Australian Disaster Indonesian victims to Australian hospitals. Only five Medicine Group. The Ministry of Civil Defence and patients were transported to Australia. Emergency Management and the Fire Service represents A further task followed with EMA being asked to assist New Zealand. DFAT with the repatriation of bodies to Australia. EMA For junher injormation conlacr: Trevor Haincs also provided community recovery advice to a number Ph: 02 6266 5169, mail: [email protected] of Commonwealth agencies and facilitated effective linkages between Commonwealth and State agencies Fiji exercises involved in the Australian response. EMA participated in two emergency-management- EMA was greatly assisted in the performance of these related exercises in Fiji during October. tasks by emergency management personnel in States The first, Exercise SUVEQ 2002, a Fiji Ministry for and Territories. Regional Development initiative, was designed to test For further information contact: Rod McKinnon the National Disaster Plan. A magnitude 6.5 Phone: 02 6266 5328, email: [email protected] earthquake, in the ocean off Suva, causing a tsunami Papua New Guinea volcano that subsequently impacted on the Suva City region eruption assessment was the base scenario of the exercise. The eruption of Mount Pago in the West New Britain The second was a combined Australian Defence Force Province of PNG during August threatened to impact (ADF) and Republic of Fiji Military Forces (RFMF) heavily on the health and welfare of local communities. Humanitarian Assistance (HA) and Disaster Relief (DR) exercise called Exercise Longreach 2002. The purpose of At the request of the PNG Government, the Australian the exercise was to practise the ADF and the RFMF in Agency for International Development (AusAID) the conduct of a HAiDR operation following a request conducted an assessment of the situation to provide advice to the PNG National Disaster Committee and the for assistance from the Fiji Government. Provincial Disaster Committee on the possible provision Steve Banks. EMManager Regional Technical of assistance by the Australian Government Assistance, participated in both exercises in an advisory Under a Record of Understanding between EMA and capacity. At the request of the Fiji Ministry for Regional AusAlD for the delivery of emergency management Development, EMA also sponsored the participation of services, EMA participated in the assessment mission. Urban Search and Rescue Team Leader. Mr Trevor At shon notice, EMA despatched Steve Banks, to make Owen, from the Melbourne Fire and Emergency a number of recommendations to AusAlD on the Services Board in Exercise SUVEQ 2000. priority lor the provision of assistance to the affected For further information contact: Steve Banks communities at the scene. Phone: 02 6266 5505, ernail: rteven.banks@ernagovau Forjurlhcr injormation contn: Steve Banks Phone: 02 6266 5505, email: [email protected] I PLANNING & OPERATIONS continued APEC earthquake seminar Aviation accident preparedness workshop In September. David Morton from EMA participated in In Oc~ober,EMA co-sponsored a Pacific Island Region a seminar in Chinese Taipei conducted under the Aviation Accident Preparedness Workshop in Nadi. Fiji. auspices of the APEC Earthquake Response Other sponsors were the Disaster Management Unit Cooperation Program for Energy Supply Systems. He from the South Pacific Applied Geowience presented a paper outlining Australiak emergency Commission. the Association of South Pacific Axlines. response arrangements and gwing details of a proposed Qantas and Air New Zealand. EMA sponsored project to establish a national The aim of the workshop was to enhance the capability emergency risk management methodology and of the Pacific Island Region to manage aviation benchmark for utilities. accidents. Thirty five representatives from airlines, The meeting developed guidlines for improving the airport operators. National Disaster Management earthquake performance of energy supply systems and Offices, police and other government departments from for preparation and implementation of effective 10 Pacific countries participated in the workshop which emergency response and restoration plans. Progress was was deemed a success by all participants. also made in establishing effective post-earthquake cooperation mechanisms between member economies. Forfunher information contact:]oanne Laurence Phonc: 02 6266 5618, ernail: joanne.laurenc~@cma~gmau Forfunher information contact: Dave Morton Phone: 02 6266 5325, ernoil: dnnd.morton@ernagouau

I

DEVELOPMENT GROUP Emergency Management Volunteers in Action photographic competition Australian Safer Communities awards The Emergency Management Volunteers in Action The Anorney-General the Hon Daryl Williams AM photographic competition concluded on 27 August QC MP presented the National Safer Communities Awards to the winning entrants at a ceremony held ai culminating in an awards ceremony at the National Parliament House on 9 October 2002. Over 100 people Press Club. The competition attracted 46 entries from in attended the event that generated considerable media and freelance photographers in the Professional media attention. Stream, and another 76 entries from volunteer The awards recognised excellence and innovation in organisations and individual volunteers in the emergency management. Nationally, sixty-seven entries Volunteer Stream. The Attorney-General presented were received with the State and Temtory winners awards to the following photographers: progressing through to the national finals. The award Professional stream winning earies were representative of the breadth of Winner: 'Devastation'. Angela Trapani, emergency management and covered such things as flood Shepparton News, Victoria (see front cover of this awareness, kitchen fire safety, bushfire behaviour edition of AJEM). modelling, arson prevention, recovery management at the Highly Commended: 'Rescued Possum', Jo-Anna local government level, clifl safety, volunteer training, Robinson. Adelaide Advertiser, South Australia research and the prevention of exotic animal diseases. Volunteer stream In his speech, the Attorney-General encouraged other . Winner: 'Biting the Bullet', Brenton Ragless, Country emergency management authorities, government and Fire Services, South Australia private corporations to adopt the ideas generated by Highly Commended: 'Frontline', Didier Moutia, the awards. St John Ambulance Australia, New South Wales; and Forfurthrr injormation contact: David Winterburn . 'Rescue'. Glenn Alderton, State Emergency Phone 02 6266 5009, mail: [email protected] Senice, Queensland Forjurther injomtion contact: TrevorJenner Management of cyclone risk Phone 02 6266 531 7, email: [email protected] The cyclone season in WA, NT and Queensland lasts nearly six months (November to April) with communities Emergency Management Volunteers in the cyclone area vulnerable to attack by high winds, The newly formed Australian Emergency Management heavy rain and/or storm surge. Given the growing Volunteers Forum met for the third time recently to population, the increasing value of assets and the continue to address the issues and recommendations development of industry in northern Australia, the arising from the Volunteers Summit. The FON~now has management of cyclone risk in Australia is very a distinctive logo and currently establishing a web site, important. However, arrangements to manage cyclone/ is planned for public access on 1 December 2002. storm surge risk vary across the three Stategemtones and there is little coordination between the risk Forfinher injomation contact: David Winterburn management agencies. Following a proposal from WA. Phone 02 6266 5009, email: [email protected] EMA has initiated a project on the management of This is EMA cyclone risk. The aim of the project is to improve coordination between the cyclone risk management EMA has produced and distributed the companion agencies and to produce Guidelinesfor the Management document to the EMA 2002-2005 Corporate Plan. oJCyclone Risk in Australia. To progress the project a This is EMA. The publication presents corporate workshop will be held in Darwin on 12-13 November to information about EMA, relationships within the consider current cyclone risk management practice and Department, the operational environment, and trends how best to approach the development of guidelines. in the emergency management sector. Forjunher injormation contact: Peter Amold Forfunher infomation contact Rob Camcmn. Phonc 02 6266 5496, emnil: [email protected] Phone: 6266 5408, cmnil: [email protected] Education and Training Group School Education The Education and Training team have been revising. The Disaster Education in Schools section of the EMA updating and enhancing the delivery and assessment web site is designed specifically a5 a teacher resource an< system and learning and assessment strategies to comply student research capability on issues relating to natural with the new national industry competency standards and man-made disasters in Australia. A collahorative from the Public Safety Training Package and the project between three states and territories, the project Australian Quality Training Framework Standards for includes the involvement of a team of teachers from bott Registered Training Organisations (ANTA 2001). primary and secondary levels and a range of subject In 2003 the lmtitute will still deliver the nationally areas, reviewing the site and making recommendations agreed content of our current programs. However, the for its improvement according to a set of criteria. learning and assessment strategies will now be based on The results of their findings will be available this year five national industry competency standards from the and development work on the web site will begin next Advanced Diploma in Public Safety (Emergency year. The new School Education officer. Neil Barker, Management) together with a number of accredited who will take up his position at Mt Macedon in early courses for which national competency standards are February 2003 will manage the project. Neil will be the not yet available. This approach is consistent with fourth teacher who has worked with EMA as pan of the feedback from the Training Needs Assessment and the Teacher Release to Industry Program (TRIP), which is a Evaluation for Impact Study conducted by the joint venture between the Victorian Employers' University of Melbourne in late 2001. Chamber of Commerce and Industry (VECCI) and the The Institute is also developing a Recognition of Prior Victorian Department of Education & Training (DE&T). Learning (RPL) strategy including a bridging program to lhe current School Education Oficer: Greg Moore, can be contactm enable those who attended EMA courses from 2000 at Mt Macedon until 6 December 2002. Contact details kill remain (when the Public Safety Training Package was the samefor next year For mallerr relating to schools, teachers an6 studens, telephone 03 5421 5242, or email whwled@emagovau. introduced) to be asswed against the competency standards from the PS Training Package and receive Scanlon wins Charles E Fritz Award recognition. There will be no charge to participants for this program. Joe Scanlon. Director of Emergency Communications Research Unit at Carleton University in Canada is the Further information can be obtained by contacting Judy Parker 2002 winner of the Charles E Fritz Award for career on [email protected]. achievements in the Social Science Disaster Area. Joseph Scanlon has a Bachelor of Journalism degree and Detailed information regarding reviews of the a graduate Diploma in Public Administration from programs offered at the Institute can be found at the Carleton and a Master of Arts (Politics) from Queen's EMA website: www.ema.gov.au or by contacting the relevant course coordinator below. University. After a career in Journalism - he was both Washington and Parliamentary correspondent for the Emergency Risk Management Program Toronto Daily Star - and a brief career in politics - he was Further injormtion con be obtained by contocting the executive assistant to the late Hon. Judy LaMarsh, he Course hrdinaror. Marh Scillw, [email protected] joined the faculty of Carleton5 School of Journalism. He Sewed the School as its Director from 1966 to 1973. Exercise Management (ExMan) Further injonnation can be obtained by contacting he Since becoming Director of ECRU. Professor Scanlon has Course Coordinator. Colin Fqord, [email protected] been a regular lecturer at the Canadian Emergency Preparedness College and lectured for 20 yean at the Emergency Coordination Centre Management Canadian Police College and for ten years made regular Furher injormntion can be obtained by contacting [he visits to give seminars for the British Emergency Planning Course Cwrdinnror. Bany Dcon, [email protected] College in Easingwold. Yorkshire. Recently he has also Course in Evacuation Management lectured in Australia. Austria. England. Holland, Hungary, Funhrr inJonnotion on thls mure can bc obtainrd by contacting the Ireland. Spain. United States as well as Canada. Course hrdi~t0r.Michael Dichinron, [email protected] Between 1994 and 1998. Professor Scanlon was Competency Standards Review President. Research Committee on Disasters, International Further injonnatwn can be obtainrd by contacting Morgety Webstel: Sociological Association. He is now both past president [email protected] and general editor of a series of books the Research Graduate Certificate in Emergency Management Committee is producing with Routledge. London. Funhcr in~nnotioncon be oblnined by contorting The Charles E Fritz Award is named after an early Dianne Cooper. [email protected] pioneer in social science disaster studies and is given for Recovery Management significant and numerous contributions to the disaster Changcs in he Recovcty Management Course my area, including research, publications, policy develop- be obtained by contacting tht Come Cwrdinatot ment as well as providing input into the professional Bruce Grq [email protected] development of the field. It is given in recognition of major and notable career achievements of a lifetime. The timetable for EMA activities covering the first half of ZOO3 can also be found at the EMA website Emergency Management Australia congratulates Joe www.ema.gov.au on receiving this recognition. CONFERENCE DIARY

Date: February 6-9.2003 INTERNATIONAL Location: Orlando. Florida. North America >ate: January 6-9.2003 ritle: 2003 International Disaster Management .ocation: Charleston, South Carolina. North America Conference: 'Disaster 2003" itle: Coastal GeoTools - 2003 Detail: The organizers have issued a call for )etail: Possible focus areas include hazard mitigation, presentations. land use and communitv development, Enquiries: John Todaro, Director of Education, Florida remote sensing, watershed planning and Emergency Medicine Foundation. Florida other harards-related topics. Abstracts are due College of Emergency Physicians, July 19, 2002 3717 South Conway Rd. Orlando, Florida. 32812.7607; (407) 281-7396. dxt.17; inquiries: Mark Jansen, NOAA Coastal Services Center. fax: (407) 281407; email: jtodaroOfemf.org. 2234 South Hobson Ave. Charleston. SC 29405. 241 3; (843) 740-1200; Host: Florida Emergency Medicine Foundation.

email: Geo.ToolsOnoaa.gov --~-- iponsor: National Oceanic and Atmospheric Date: 5-8 Feburary, 2003 Administration (NOAA), Coastal Services Center (CSO. Location: Portland. Oregon. North America ritle: The Earthquake Engineering Research Institute (EERI) Annual Meeting >ate: January 8-10,2003 Enquiries: EERI, 499 14th Street Suite 320, Oakland, CA mation: Catholic University in Washington 94612-1934; Tel: (510) 451-0905; fitle: Terrorism: Threats, Tactics, Training and Fax: (510) 451 5411. Email: 33riOeeri.org. Technology, Website: httpJtwww.eeri.Org/ letail: Transit Terrorism, terrorist attack profiles, terrorism early warning and intelligence Date: 11 February, 2003 fusion, terrorist training, terrorist tactics, terrorist targeting, emerging technologies, Location: OFC Ontario. North America threat of cyberterroris, and newest aviation litle: 2003 Fire Coordinators' Conference security threat information. inquiries: www.terrorism.com Date: 12-16 February, 2003 Location: Baltimore. Maryland Date: January 24-28.2003 Title: 7th World Congress on Stress, Trauma and scation: St. Petenburg, Florida, North America Coping: Crisis Intervention in a Changing fitle: Weapons of Mass Destruction Conference World 2003: Preparedness Through Partnership Detail: Call for presentations and paperr is available Enquiries: Gregory Wans: (727) 39&9482; at the website: hnpJtwww.icisf.or@/dp.pdf email: Gregory.Wans2Omed.va.gov; Enquiries: Shelley Cohen; (410) 750-9600; httpJtwww.va.govhvmd/. email: [email protected] iponror: University of South Florida College of ~ost: International Critical Incident Stress Medicine, Veterans Health Administration, Foundation (ICISF). One and twoday Emergency Management Strategic Healthcare precongress workshops will be offered Group. Florida Department of Health - February 12-13.2002. Emergency Operations, Brooke Army Medical Center, Pinellas County Emergency Management and Federal Emergency Date: 13-15 February, 2003 Management Agency. Location: Christchurch. New Zealand. Australiasia Title: Pacific Conference on Earthquake Engineerin! Date: January 27,2003 Detail: Abstracts due by April 2002 Location: London. United Kingdom Enquiries: Conference Secretariat Conference Office, Title: Emergency Lighting - Design and Practice Centre for Continuing Education. Canterbu~y Private Bag 4800. Christchurch, New Zealand; Enquiries: Fire Courses Secretary, Mid Career College, Tel: 64-3-3W2534; Fax 64-3-364-2057. PO Box 20, Cambridge CB1 SDG UK Email: peeOcont.canterbury.ac.nz counes@mid-~areer-college.ac.uk; tel: (01223) 880016; Host: New Zealand Society for fax: (01223) 881604. Earthquake Engineering. Date: 22-26 Februaty, 2003 Date: 12-21 March, 2003 Location: Washington DC, North America Location: Madison. Wisconsin. North America Title: National Emergency Management Association litle: Disaster Management Workshops (NEMA) 2003 Mid-year Conference Enquiries: Don Schramm, email: Enquiries: Registration materials will be available in December 2002. Contact: NEMA; [email protected];or Jennifer (859) 244-8162; email: [email protected] Orter, Department of Engineering Professiona Development. University of Wisconsin- Date: 23-26 February, 2003 Madison, 432 Lake Street, Madison. WI 53706; Location: Providence, Rhade Island. North America 1-800-462-0876or (608 262-4468; Title: International Disaster Recovery Association fax: (608) 263-3160; (IDRA) Annual Meeting email: [email protected]. Details: The theme for the 13th annual conference is Sponsor: University of Wisconsin - Madison, 'readinerr, resilience, recovery and Department of Engineering Aprofessional reassessment of all topics focus on telecom Development contingency planning. Enquiries: IDRA (508) 845-6000; email: 20030idra.com Date: 16-23 March, 2003 Date: 24-28 February, 2003 Location: Kyoto, Shiga &Osaka, Japan Location: Las Vagar, Nevada, North America Title: Third World Water Forum Title: International Erosion Control Association Enquiries: Secretariat of the 3rd World Water Forum. 34th Annual Conference and Expo 5th Floor 2-24 Kojimachi Chiyoda-ku, Enquiries: International Erosion Control Association. Tokyo 102-0083, Japan; PO Box 774904, 1355 5. Lincoln Ave., tel: +81-3-5212-1645; fax: +El-3-5212 1649; Steamboat Springs. CO 80477-4904. jponsor: World Water Council. Date: 3-6 March, 2003 Location: Rotterdam. The Netherlands. Europe Title: International Conference on Advances in Flood Forecasting in Europe WESTERN AUSTRALIA Enquiries: BobvanKappel; email: bob.vankappel~ldelh.nl Host: WUDelft Hydraulics and the Joint Research Date: 27-28 March, 2003 Center of the European Commission. .ocation: Perth. Western Australia Date: 4-8 March, 2003 litle: Demystifying Emergency Management: Location: New Orleans. Louisiana, North America warts and all Title: &sociation of American Geographers (AAG) inquiries: Scott-Andrew Smith, Annual Meeting email [email protected] Details: Includes a session on hazards-related isues. Deadline for submisstions is September 25. 2002. Enquiries: Jayajit Chakrabarty. Department of Geography, University of South Florida, 4202 East Fowler Avenue, SOC 107, Tampa. CALL FOR PAPERS - AUSTRALIA FL 33620; email: [email protected] Date: 8-12 March, 2003 Date: 10-12 September, 2003 Location: Reno, Nevada. North America .ocation: Canberra. Australia ntle: 2003 National Disaster Medical System National Convention Centre. (NDMS) Conference We: Australian Disasters Conference Enquiries: Contact NDMS. Tel: 001110018 1800-USA- Safer Sustainable Communities - 2003 NDMS, and press the 'star' key. )exription: The 2003 Australian Disasters Conference Email: [email protected]. Committee invites the Australian emergency Host: National Disaster Medical System management community and other Date: 10-15 March, 2003 prospective participants to submit abstracts for the presentation of papers. !ocation: Puerto Rico. South America Title: Dissertations Initiative for the Advancement inquiries: To register your interest and to ensure of Climate Change Research (DISCCRS) you receive further information and Symposium regular updates via our mailing list jponsors: American Geophysical Union, American email the Conference Coordinators Meterological Society. American Society of [email protected] Limnology an Oceanography. Ecological or phone Liz or Katie on (02) 6232 4240. Society of America and Whitman College hquiries: To register with DISCCRS, or for complete symposia information including eligivility deadlines and how to apply contact C Susan Weiler; (509) 627-5948; email: [email protected]; or web site: http~/aslo.orgldisccr~disccrsposter.pdf. BOOK REVIEW With Spirit and Courage: the extraordinary life of Paul Featherstone By Paul Featherstone with Ian Heads. Publisher: Pan Macmillan Australia Recommended Retail: 830.00

current world-class status and his formation of the Specialty Casualty Access Team (SCAT) in the eighties. The feats of courage and endurance described in the book illustrate why Featherstone advocates SCAT members need to be multi-skilled. SCAT access operations often need medical and physical capabilities for feats that are not only extraordinary but also often heroic. He says that training and the ongoing tests of physical and mental endurance are crucial to SCAT operational effectiveness especially given the unpredictability of access circumstances such as pounding seas, cliff ledges or caves. Spirit and Featherstone also talks of the human qualities necessary for a rescue paramedic. He believes paramedics need confidence to place their lives in their colleagues' hands because in their Courage profession, the likelihood is that they inevitably will. With so many additional factors working against the The Extraordinary Lifc of volunteer like violent weather, poor visibility, tight spaces and PAUL FEATHERSTON unstable surfaces a common theme emerging from this book is the importance of teamwork. With spirit and courage: the extraordinary life of Paul Reviewed by Giovanna di Natale. Reference Librarian, Featherstone is the autobiography of rescue paramedic. Emergency Management Australia. Paul Featherstone. who recentlv received worldwide attention as the rescuer of Stuart Diver, the sole survivor of Australia's Thredbo landslide. Featherstone takes the reader inside some of the most dramatic moments in Australian life in the last 30 years Safer Sustainable Communities - 2003 includinq the Thredbo landslide, the Granville train disaster The 2003 Australian Disasters Conference hosted by and the Nyngan floods. Emergency Management Australia (€MA)has been In this inspirational and thoroughly captivating book. rescheduled. The Conference will now be held in Canberra from 10-12 September 2003. Featherstone tells of the spirit and teamwork that he believes makes Australians special. 'Feathers', as he is EMA recognises that the success of the Conference will rely heavily on the quality of speakers and the material presented. known in paramedic circles, describes the national sense The Conference Committee invites the Australian emergency of team spirit he encountered when the volunteer team management community and other prospective participants to extricated Stuart Diver from the crush of concrete and submit abstracts for presentation at the conference. rock at Thredbo. Conference planners expect the length of each presentation will be 30 minutes including time for questions/discussion. Recounts of exhilarating access operations in the book The Conference Committee seeks abstracts for papers reveal the natural and human elements that may work addressing one or more of the Conference themes and against a rescue attempt. Featherstone talks of how his streams listed below. Papers will be selected based on aircraft was inverted due to the unpredictable turn of a innovation, relevance to the conference themelstreams and author's expertise, and clarity of reasoning criteria. water tornado direction during a sea rescue of lost fishermen. He also tells of a surfer who placed a rescue The theme for the Conference is 'Community Safety is Everyone's Business'. In this context, the Conference mission in jeopardy because he wanted to have his streams are: surfboard winched up with him when stranded in Working with Communities rough seas. Understanding Risk and Uncertainly He also describes the many dangers of rescue missions Consequence Management (Response and Recovery) including descending from a helicopter winch into a Sustainability Critical Infrastructure Protectiollllhe Role of Business violent sea storm during an ink black night, encountering Information and Communication a black snake while plucking a victim from the Nyngan . floodwaters and crawling under a potentially unstable For a copy of the Initial Announcement and Call for Abstracts brochure, confacf Susan Stevens. train to access an injured victim. phone: 02 6266 5005, As well as his historical recollections of some of Australia's email: susansfevenst3ernagorau; or the Conference Secrefanat. major disasters. Featherstone follows the institution of the nu- Einstein & Edison, phone: 02 6232 4240, s... Australian Paramedic Unit from its inception in 1976 to its "..... ,..., -.., email: enqui@einsteinandedison,com.au a..+..s~. e