Effective all-hazard warning system development and community resilience: Results from ongoing multi-hazard research in New Zealand, USA and Australia

Graham S. Leonard & David M. Johnston, GNS Science, Lower Hutt, New Zealand

Douglas Paton, Department of Psychology, University of Tasmania, Australia

GNS Science

New Zealand’s legislative and legal frameworkOrganizational roles and linkages

• The Ministry of Civil Defence and Emergency Management (MCDEM) has a statutory role operating the National Warning System, communicating multi-hazard warning messages to regional CDEM Groups. Government agencies, including GNS Science via Geonet, supply early warning messages.

• MCDEM also has a role providing guidance and resources for public notification, education and response to warnings.

• Regional CDEM Groups have an obligation to warn the publicand minimise the impact of emergencies, through a range of actions including warnings. Groups include local government, emergency services and health agencies.

• GNS Science also conducts social research into specific arrangements for end-to-end warning systems that maximise effective response to those systems.

GNS Science

New Zealand end-to-end all-hazard National Warning System

PTWC

MetService

GeoNet

Govt. Departments

MCDEM CDEM Group Notify?

Yes

Pre-plannedwarning receipt

Pre-plannedcriteria

Notification system(s)

Pre-plannedmessages, protocol(s)

System2

S ystem3

Syst em1

System4...

Different for different hazards

All must be pre-planned

: early warning provider, message path, receipt methods, notification criteria,

warning messages, notification protocols (inc luding notification system(s) used)

Resilient community empowered and capable

of making appropriate response decisions -

requires development of resilience indicators such

as self-efficacy, and provision of all

components of effective warning systems through to education, signage, maps, simulation exerc ises, and

evaluation of these

Effectivew

arningsystem

Early warning

system

Direc t warning supplied / sought

External warning supplied / sought

Internal hazardmonitoring, warning

generation, confirmation,notification proceedures

GNS Science

Geonet and GNS Science hazard / social research

• Monitoring of New Zealand earthquakes, volcanic activity and coordination of PTWC warnings. Development of local/regional-source tsunami monitoring.

• Generation of multi-hazard warning messages for national and regional emergency management.

• Natural hazard and risk research.

• Research into effectiveness of warnings; including practical actions and resilience motivators to improve this – the focus of this presentation.

GNS Science

Current social research evaluating and working to improve effectiveness of warning system response

New Zealand: National Warning System improvements, national tsunami risk and preparedness review, volcanic & tsunami warning response evaluation projects (national and specific projects at Auckland, Ruapehu, Napier), Auckland region public notification needs analysis.

Australia: flood warning awareness (New South Wales), wildfire warnings.

USA: Tsunami and lahar warning planning and education (Washington state).

GNS Science

An evolving model of components of an effective end-to-end warning system

1. EARLY WARNING SYSTEM: Hardware, electronics, communications and planning necessary to effectively detect a hazard, generate warning messages and transmit them to at-risk regions (including any use of public notification hardware).

2. PLANNING: Decision-making tools: thresholds, evacuation routes and maps, inter-organisational relationships and communication channels.

3. COOPERATION, DISCUSSION AND COMMUNICATION: Pre-planned and exercised communication between central government agencies, local emergency management agency staff, scientists and community representatives. Renewal of contacts must be regular and permanently sustained, to overcome common high staff turnover.

4. EDUCATION: Public education, staff training, maps, and signs.

RES

EAR

CH

: (Sc

ienc

es o

f mon

itorin

g sy

stem

s,

sour

ces,

pro

paga

tion,

inun

datio

n, ri

sk

asse

ssm

ent;

and

of a

war

enes

s, a

nd e

ffect

ive

resp

onse

5. EXERCISES: Scenario development and simulations — table-top and full, with observation and feedback.

EFFE

CTI

VEN

ESS

EVA

LUA

TIO

N

Presented in Webb (compiler), 2005, Review of New Zealand tsunami preparedness, GNS SR 2005/162

GNS Science

Mt Ruapehu volcano warnings research,New Zealand

GNS Science

Resourcing and technical issues

• In New Zealand there is always a compromise between warning needs and effort / finance available. We have found that empowering local groups and the community gives a high return.

• Early warning hardware options are evolving rapidly, planning for upgrades and the inclusion of mobile technologies is ongoing. Hardware options must be within resource constraints and are only as effective as the rest of the components of an end-to-end system. Adequate resourcing for these components is as important as the hardware itself. [e.g. New Zealand National Warning System upgrades; national tsunami review].

• Regular ongoing testing and exercising is essential to reliability.

• Technology that addresses social behaviour is essential to high levels of effective response [e.g. Auckland region].

• No single end-to-end option suits all hazards and end-users – a national template must be locally tailored involving locals [e.g. New Zealand and Washington coast, USA].

GNS Science

Assessment of Auckland region notification needs: multiple channels for multiple hazards and end-users

GNS Science

Sustainability and maintenanceCapacity building and training

• Effective end-to-end warning systems are only possible if all of their components are sustained in the long term.

• Sustained maintenance (both financial costs and effort) must be planned for in the long term and have local buy-in. These efforts will at some point in time eclipse any start-up costs, this must be allowed for at the outset. [e.g. Ruapehu and Auckland research].

• Local warning dissemination and response agencies, and community groups, must be involved in planning of the multi-hazard system. Local system steps must be tailored to the local needs. Capacity and training at a local level are then much more easily sustained. [e.g. Washington coast, USA].

• Resilience requires correct action at the time of warning, which goes beyond awareness and intention to act. Current research is looking at motivators of action, and promoting the resilience indicators that predict action – awareness, self-efficacy and intention [e.g. Auckland, Australian wildfires].

GNS Science

Challenges, benefits, and synergies• The main challenge is developing and sustaining all

components of an effective end-to-end system indefinitely.

• Initiatives that motivate social change towards improved resilience operate over decades [e.g. Auckland resilience project]. Political and funding cycles are much shorter.

• Regular exercising of our national multi-hazard system in the context of end-to-end warnings develops multi-organisational synergies, relationships and shared understanding.

• The benefit of a regularly exercised national multi-hazard system is a wide awareness of the mechanism and nature of the system, of a range of warning scenarios and messages, and of the correct response to them. Exercising is essential for evaluation and education [e.g. Ruapehu, Napier].

GNS Science

Lessons learned

• All links in the end-to-end chain must succeed to get close to ‘full’response to a warning.

• The key to effective response to hazard warnings is a resilient community motivated to learn, understand and respond correctly.

• Warning notification that takes into account social behaviour(provides warning message details directly) is essential to a high response rate in a multi-hazard warning environment.

• Notification of a range of hazards to varied end-users requires multiple path-ways, including all of ‘natural warnings’, via institutions to ‘those in their care’, via community/organisational networks’, via ‘third partyhardware’ and ‘via warning-dedicated hardware’.

• The existence of early warning system hardware alone does not necessarily promote effective response to it. Exercising the full end-to-end system is essential to evaluate effectiveness.

GNS Science

1960 Chilean tsunami, Gisborne, New Zealand

GNS Science

Specific recommendations

• Plan all steps of the end-to-end warning chain in cooperation, allowing for similarities and differences across multiple hazards, involving the community in the planning of other steps. [e.g. Ruapehu, National Warning System]

• Improve community resilience - motivate appropriate actions in response to warnings for all hazards. Develop resilience indicators to monitor awareness and likely levels of action.

• Exercise the system regularly, varying the hazard or using compounding hazards. [e.g. New Zealand national warning system exercises and evaluation; Ruapehu; Napier; tsunami exercise in May 2006]

• Exercise extreme situations that will strain the system and response. Involve many interest groups, in realistic exercises.

• Feed back evaluation of exercising into improved planning

GNS Science

• Develop risk, preparedness and warning effectiveness evaluation measures and apply them in an ongoing way [e.g. NZ national tsunami review – reviewed multi-hazard components that apply to tsunami too; Australian flood warning awareness evaluation]

1050

200500

8001200

17002200

3000

Auc

klan

d W

est

Man

ukau

Wes

tN

ew P

lym

outh

Wai

take

re W

est

Nel

son

Wha

ngar

eiIn

verc

argi

llP

oriru

aW

aita

kere

Eas

tK

apiti

Tim

aru

Nor

th S

hore

Auc

klan

d E

ast

Man

ukau

Eas

tW

haka

tane

Taur

anga

Dun

edin

Low

er H

utt

Nap

ier/H

astin

gsC

hris

tchu

rch

Wel

lingt

onG

isbo

rne

0

500

1000

1500

2000

2500

3000

3500

4000

Deaths

Return period (years)

0% 20% 40% 60% 80% 100%

Pre-event mitigation/ reduction

Who is responsible for regional warning dissemination

Who in Group receives warning

Where warning comes from

Proportion of time can receive warning

How warning is received

What is done with warning

Who is contacted once warning received

Tsunami warning hardware

Tsunami warning hardware testing

Arrangements for getting info to public

Group Plan addresses local & distant sources

Sub-group-level plans (ie. Territorial Authorities)

Tsunami warning decision preplanning

Inundation or evacuation mapping

Mention evacuation arrangements in Group Plan

Evacuation decider/ manager

Role of Police

Role of Fire Service

Role of other key organisations

Role(s) of Territorial Authorities (if more than one)

Arrangements for giving "all-clear"

Tsunami warning SoPs

Media/ communications plans

Arrangements with radio station(s)

Communications with other agencies

Tsunami training

Awareness/ monitoring research

Tsunami warning public education

Tsunami-related signs exist

Tsunami warning simulations

Early Warning-response planning

Co-operation, discussion

& communications

Education, staff training, maps & signage

Early Warning System(Regional components)

CDEM Groups with this written arrangement (%)Generic written arrangement that may beapplied to tsunami

Generic in plan - may be more specificannex

A written arrangementtsunami-specific

REG

ION

AL-

LEVE

L C

OM

PON

ENT

OF

EFFE

CTI

VE W

AR

NIN

G S

YST

EM

Regional-level written arrangements, as reviewed from Group plansand documents beyond Group plans

(in Group Plans - all very short - a few lines)

New Zealand tsunami deathscompared to return period

GNS Science

Successful warning is not the existence of an

early warning system alone,but effective response to it

17 June 1996 Mt Ruapehu eruption, New ZealandContact:Dr Graham LeonardGNS ScienceP.O. Box 30368, Lower Hutt, New ZealandPh: +64-4-5701444, Fax +64-4-5701440www.gns.cri.nz