www.lirneasia. net Early warning systems in disaster-risk reduction Rohan Samarajiva & Nuwan Waidyanatha IDRC-CIDA workshop 12 July 2007, Moratuwa, Sri Lanka
ww
w.li
rnea
sia.
net
Early warning systems in disaster-risk reduction
Rohan Samarajiva & Nuwan Waidyanatha
IDRC-CIDA workshop12 July 2007, Moratuwa, Sri Lanka
ww
w.li
rnea
sia.
net
Agenda
Disaster risk reduction: what must be done The LIRNEasia-Sarvodaya community-based approach to
early warning Pilot project research design
Different communication technologies ERP [Emergency Response Planning] training v no training Organized villages vs. less organized villages
Findings Efficacy of Hazard Info Hub at SCMDC Reliability of technologies Reliability of systems Training Organizational development Women
ww
w.li
rnea
sia.
net
Physical world wherehazards occur
Symbolic worldwhere action
originates
Mediatedinterpersona
l
Sri Lanka on December 26th, 2004
With a few insignificant exceptions,a world without ICTs
One in 500 died in Sri Lanka without one minute of official warning
ww
w.li
rnea
sia.
netPhysical world where
hazards occurSymbolic worldwhere action
originates
Mass media
Mediatedinterpersonal
Warnings
Warnings
How ICTs should be deployed
More time to run; more lives saved
ww
w.li
rnea
sia.
net
Who should act?
Early warning is a classic public good Government must supply
Early warning is based on incomplete, probabilistic information and judgment Government must take the responsibility
75% of tsunami warnings in the Pacific are false; false warnings can be dangerous
Community preparedness (e.g., “tsunami ready” designations in the US) is vital
But if government appears unready to act . . . Second-best options must be explored
ww
w.li
rnea
sia.
net
Early warning chain (standard form)
CitizensNational early warning center
First responders
Media
ww
w.li
rnea
sia.
net
Early warning chain (community based; applicable to Last-Mile HazInfo project)
ERP1
National early warning center
VillagersSCDMC
ERP4
ERP3
ERP2ICT GuardiansFrom domestic &
international sources
ERP1
National early warning center
VillagersSCDMC
ERP4
ERP3
ERP2ICT Guardians
ww
w.li
rnea
sia.
net
Pilot project design, 32 villagesWith ERP Training No ERP Training
Sarv
odaya S
tage 1
, 2,
3
VSAT Urawatha(Galle)
MoPNidavur(Batticalo)
FxPThirukadalar(Trincomalee)
AREAMoratuwella(Colombo)
MoPMeddhawatha(Matara)
MoP Thambiluvil (Kalmunai)
FxPOluville(Kalmunai)
AREA Maggona(Kalutara)
AREA + RAD Modarapallassa(Hambantota)
AREA + FxPWathegama North(Matara)
AREA + MoPPalmunnai(Batticalo)
Control VillageAbeyasinghepura(Ampara)
AREA + RADThondamanar(Jaffna)
AREA + FxPKarathivu(Kalmunai)
AREA + MoPMunnai(Jaffna)
Control VillageModara(Colombo)
Sarv
odaya S
tage
4
VSATModaragama(Hambantota)
MoPDiyalagoda(Kalutara)
FxPPeriyakallar (Batticalo)
AREAPanama North(Ampara)
MoPSatur-kondagnya(Batticallo)
MoPSamodhagama(Hambantota)
FxPIndivinna(Galle)
AREA Brahamana-wattha(Galle)
AREA + RAD Kalmunai II(Kalmunai)
AREA + FxPSamudragama(Trincomalee
AREA + MoPValhengoda(Galle)
Control VillageMirissa South(Matara)
AREA + RADVenamulla(Galle)
AREA + FxPKottegoda(Matara)
AREA + MoPThallala South(Matara)
Control VillageThalpitiya(Kalutara) AREA: Addressable Radio for Emergency Alerts, Class B configuration of WorldSpace System
MoP: Java enabled Mobile Phone, Dialog-Microimage innovation MiDews application
RAD: Remote Alarm Device, Dialog-University-of-Moratuwa Innovation
FxP: CDMA Wireless Fixed Phones with 1xRTT functions, Sri Lanka Telecom
VSAT: Very Small Aperture Terminals coupled with Internet Public Alerting System Innovative-Tech & Solana Networks
ww
w.li
rnea
sia.
net
Research questions
What technology is best for reaching the community?
What emergency response plans are best for saving lives of villagers? How important is training? How significant is the level of organization in the
village?
How does the intervention affect women? Will the ICTs be integrated into everyday
life?
ww
w.li
rnea
sia.
netCDMA Fixed Phone GSM Mobile Phone Remote Alarm Device
Addressable Radios for Emergency Alerts
Very Small Aperture Terminals
ICTs used in reaching communities
ww
w.li
rnea
sia.
net
Reliability is the difference between the time it takes HIH Monitor to issue the CAP message and the time the message is received by the ICT Guardian.
HIH-Monitor
ICT GuardianRelay Alert
ERP CoordinatorsDissemminate
Report Status
Acknowledge
Resolve CAP Alert()
Activate ICTG ERP()
Activate Community ERP()
Relay Results
Activate HIH ERP()
Download Alert()
T0
T1
T2
t0
t1
t2
Reliability is the difference between the time it takes HIH Monitor to receive message and the time the message is received by the Community.
Reliability of technologies & systems
ww
w.li
rnea
sia.
net
Early warning chain (community based; applicable to Last-Mile HazInfo project)
ERP1
National early warning center
VillagersSCDMC
ERP4
ERP3
ERP2ICT GuardiansFrom domestic &
international sources
HIH
ww
w.li
rnea
sia.
net
Reliability of HIH Monitor Tasks
1.00
0.39
0.81
0.97
0.69
Task
Reli
ab
ilit
y
0.06090.7725HIH Monitor
Message Relay Process
VarianceAverage
Expected value = 95%
Download was quick because the simulation was staged within a defined three-hour time period. Acknowledgement was rare, possibly because they knew it was a drill. Event of Interest step was time consuming (2-page form) and needs improvement. Approval was instantaneous because the SCDMC Director was on premises during drills. Issue of alert took time because Each ICT required its own interface to issue the alert.
Evaluation of Hazard Info Hub (HIH) actions
ww
w.li
rnea
sia.
net
What is being measured . . .
difference between the time it takes HIH Monitor to issue the CAP message and the time the message is received by the ICT Guardian
ww
w.li
rnea
sia.
net
Coupled ICTs performed better than individual ICTs with the AREA + FXP combination being most reliable. Of the stand alone devices AREA is the most reliable with 13 of 16 working; The community had difficulty in installing and maintaining the MiDews Java applet on their MOPs and 7 of 10 participating MOPs failed; RADs were not tested as stand alone devices. FXP and MOP failed in North-East due to services being deactivated due to war.
Reliability of ICT Configurations in Last Mile Hazard Warning System
0.50
0.31
0.17
0.89
0.670.64
0.50
AREA FXP MOP AREA+FXP AREA+MOP AREA+RAD None
ICT Configuration in Community
Rel
iab
ility
Reliability of technology connecting HIH to ICT Guardians (inner circle)
ww
w.li
rnea
sia.
net
Early warning chain (community based; applicable to Last-Mile HazInfo project)
ERP1
National early warning center
VillagersSCDMC
ERP4
ERP3
ERP2ICT GuardiansFrom domestic &
international sources
ww
w.li
rnea
sia.
net
What is being measured. . .
difference between the time it takes HIH Monitor to receive message and the time the message is received by the Community
ww
w.li
rnea
sia.
net
Reliability of systems (outer circle)
AREA alone is unidirectional. Stand-alone AREA, RAD, and VSAT score low because alerts received in English only; all other ICT configurations accommodate three languages. Using a combination of FXP or MOP coupled with one-way Satellite communication device AREA is most effective because it
- Provides larger geographic coverage with Satellite covering GSM and CDMA shadow areas- Increases bi-directional capabilities of communities to report local events to the Hub- Language diversity reduces ambiguity in messages
Effectiveness of ICT Confiurations in Last-Mile Hazard Warninig System
0.08
0.39 0.43
0.09 0.07
0.7 0.69
0.13
ICT Configuration in Community
Eff
ecti
ven
ess
ww
w.li
rnea
sia.
net
Reliability of ICT w.r.t Training Regime for ICT Guardians
0.96
0.991.00
0.990.99 0.99
0.930.94
0.990.97
AREA FXP MOP RAD VSAT NONE
ICT Name
Reli
ab
ilit
y
Trained Untrained
Reliability of ICT w.r.t Training Regime for ERP Coordinators
0.740.720.83
0.630.63
0.87
0.71
0.580.67
0.59
AREA FXP MOP RAD VSAT NONE
ICT Name
Reli
ab
ilit
y
Trained Untrained
All 28 ICT Guardians received training. Only 16 of 32 Community ERP Coordinators received ERP Training. ICT Guardians were coached during live-exercises; random events over longer period would show different results. ERP Coordinators were also coached therefore results are not accurate. Still could not reach the required level.
Effects of training of ICT Guardians & ERP Coordinators
ww
w.li
rnea
sia.
net
Effect of organizational level of villages
Reliability of ICTs w.r.t Organizaional Level
0.29
0.43
0.17
0.28
0.130.13
0.46
0.17
0.09
AREA FXP MOP RAD VSAT NONE
ICT Name
Re
liab
ility
Stage 4 and 5 Stage 1, 2, & 3
Organizational capacity is unimportant in the adoption of existing technologies such as Mobile and Fixed Phones. This is also evident from the Control Villages, which used their own Mobile Phones and Fixed Phones. Organizational capacity is important when adopting new technologies such as AREA. Unfortunately RAD was not tested in less organized villages. It is easier to organize disaster preparedness activities in organized villages.
ww
w.li
rnea
sia.
net
Distribution of Male vs Female in Hazard Mitigation
Adult Male28%
Adult Female
72%
Adult Male Adult Female
72% of adult participants were female because the simulations were conducted between 0900 and 1200hrs when most men were out of the home. The women showed enthusiasm and willingness to participate in disaster management activities. High participation of children occurred because the mothers brought their children along as they could not be left at home alone.
ww
w.li
rnea
sia.
net
Integration of ICTs into everyday life
Expected to get data from Diaries that had been provided From use records
Both failed because Dairies were not maintained Use was very low
In some cases, District Sarvodaya Offices has reclaimed the phones for use in urgent IDP activities
Appeared that people thought we did not want them to use the ICTs for other purposes and did not volunteer information such as listening to BBC on AREA sets
ww
w.li
rnea
sia.
net
Conclusions
Critically important to get SCDMC running as a 24/7 operation and to have disaster risk-reduction embedded as a central element of the worldview of all at Sarvodaya Need to integrate management of all disaster projects Think beyond project horizon (has already started
through strategic planning process) Robust early warning for community-based approach
(as opposed to direct-to-citizen approach) requires Bidirectionality of media, for local reporting and
acknowledgement No stand-alone AREA sets CAP compliance and ability to send long messages in
multiple languages AREA sets in combination with other media
Mobile has potential, after software problems solved, but script limitations make it a distinct also-ran
ww
w.li
rnea
sia.
net
Conclusions
A community-based approach requires Training, both for ICT Guardians and for ERP
Coordinators Higher organizational levels
No conclusive evidence, but training through periodic simulations may be more effective than conventional training
No conclusive evidence, but integration into everyday activities may not be very important in the case of AREA+
No conclusive evidence, but women are unlikely to be marginalized from community-based approaches in Sri Lanka
ww
w.li
rnea
sia.
net
Project-specific conclusions
Pilot projects on truly novel problems require imagination and flexibility Example: change from strict adherence to “no-
training” condition in the face of demands from villages
Substitution of ham radios by other technologies
Those closest to the action know best Project Director and IDRC Task Managers gave the
main players the freedom to run the project