Mobilizing ICTs for early warning: Lessons of the 2004 tsunami Rohan Samarajiva.

Mobilizing ICTs for early warning: Lessons of the 2004 tsunami

Rohan Samarajiva

World changing (?) disasters

1755 Lisbon earthquake, tsunami and fires (killing people worshipping in churches) 100,000 deaths Contributes to displacing religion from center of

intellectual life 1883 Krakatoa volcanic implosion and tsunamis

35,000 deaths First disaster covered with help of undersea

telegraphy 2004 Indian Ocean tsunami

200,000+ deaths & USD 4.5 billion in damage First Internet mediated natural disaster

Physical world wherehazards occur

Symbolic worldwhere action

originates

Mediatedinterpersona

l

Physical and symbolic worlds, absent

linking technologies

Physical world wherehazards occur

Symbolic worldwhere action

originates

Mass media

Mediatedinterpersonal

Warnings

Warnings

The physical, the symbolic & their linking through ICTs, simplified

Ideal form of hazard detection & monitoring and warning system

Physical world wherehazards occur

Symbolic worldwhere action

originatesMass media

Mediatedinterpersona

l

Warnings

Information & communication technology & institutions

Warning CenterHazard detection & monitoring system

Physical world wherehazards occur

Symbolic worldwhere action

originatesMass media

Mediatedinterpersona

l

Warnings

Physical world of hazards, symbolic worlds, link technologies & institutions that may/may

not work

Information & communication technology & institutions

Warning CenterHazard detection & monitoring system

Physical world wherehazards occur

Symbolic worldwhere action

originates

Mediatedinterpersona

l

What happened in 10 out of 12 Indian

Ocean countries* on December 26th, 2004

* Excluding Kenya and Seychelles, which had the advantage of 7 hours

How the world knew of the 2004 teletsunami Earthquake was known within minutes of its

occurrence (0059 UTC) at Hawaii at Pacific Tsunami Warning Center Pallekale seismometer, and therefore by Sri

Lanka Geological Survey & Mines Bureau• But none interpreted it as causing a teletsunami

World’s best tsunami experts knew only when they read news reports from Sri Lanka, probably LBO report filed at 0333 UTC (next talk), or AFP reports filed at 0346 UTC in Colombo

Why was Aceh’s scream not heard, unlike Sri Lanka’s? Aceh

Telecom infrastructure was poor to start with, but was destroyed by earthquake and tsunami

Civil war no journalists Sri Lanka’s east coast

Cease fire of 2002 had allowed 100,000 + phones to go in

Most of infrastructure survived Journalists free to report

ICTs matter, but . . .

But institutions matter even more

ICTs are necessary conditions Institutions are the necessary

conditions

Why did the institutions fail?

Despite completion of the International Decade for Natural Disaster Reduction in 1999, conferences, workshops, training courses . . .

Why are they failing even now? With limited exception of Thailand, the

tsunami affected countries have not got their act together even now, six months later

Disaster warnings are classic public goods, subject to market failure

Non rivalrous Non excludable

Will be undersupplied by the market But government failure too

Disaster warning systems

Made up ofHazard detection & monitoring system

• Regional/international (e.g., tsunami & cyclone)

• Local (e.g., dam breaches, landslides)

Warning & alert dissemination system

Hazard information systems

Regional and international systems subject to problems of intergovernmental coordination Thailand and India still squabbling

National and local systems subject to government failure caused by Mindsets that privilege relief & recovery over

preparedness and warning Multiple stakeholder problems Results of actions falling outside electoral

cycles –Anthony Downs

But government failure is not found in all countries . . . Possible the problem is limited to “predatory

states” -- Peter Evans Driving force is rent seeking

• Partly to get re-elected (Downs)• But venality too

• Relief and recovery emphasis fits better with rent seeking

Empirical evidence? Kaufmann, Kraay and Mastruzzi (2005)

Government effectiveness?

Control of corruption?

No clear correlation, but worth further investigation Maldives, Thailand and Sri Lanka, countries

that suffered significant losses, surprisingly high in percentile rankings

Malaysia had highest/2nd highest relative scores but gave no warning

Kenya had low scores but managed to issue warnings Distance and time

Use of percentile ranking by Kaufmann et al. blurs gap between relatively less badly governed and well governed countries

In conclusion . . .

ICTs a necessary condition for warning ICTs, but also institutions Governance must be improved, but no need

to improve all of it at once Public-private partnerships in Bangladesh:

• “The cyclone of 1970 took the lives of 300,000 people but the cyclone of the same intensity of 1991 killed 138,000 people, and the cyclones of 1997 and 1998 resulted in only 127 and 6-7 deaths respectively”

Good governance: the bulwark that saves lives

For more . . .

http://lirneasia.net Samarajiva, R. (2005, forthcoming).

Mobilizing information and communications technologies for effective disaster warning: Lessons from the 2004 tsunami, New Media and Society