Thomson Reuters Foundation, trust.org (TRF)
By Laurie Goering
When is the right time to warn of an impending flood?
As the skies open up with heavy rain and the water rises dangerously behind the Nangbeto hydropower dam in rural Togo, local authorities face a tough decision: When do you raise a warning for the flood-prone villages below and approve funds to set up relief efforts? What if nothing happens and you’re accused of wasting money? Or if you’re not fast enough and people die?
A new virtual reality game created by the Red Cross/Red Crescent Climate Centre puts users in the shoes of decision makers, letting them decide whether or not to ring an alarm bell, stamp papers for aid delivery, and load supplies into a relief truck – all from a virtual hill overlooking the green valley and the surging waters of the dam.
Based on three years of real hydrological data from the tiny flood-prone West African nation, the experience aims to help users get a sense of how hard decision-making can be – and suggests how creating better prediction models and pre-authorising aid might save money, time and lives.
The game also points to a potential innovation in dealing with disaster risk: Using rainfall and flood records to build a computer programme that learns from past experience. Such a programme potentially could predict flooding more precisely than people, and could be set up to send out automatic flood warning alerts, for instance, when trigger points are passed, Suarez said.
That, combined with fast emergency funding decisions when triggers are met, could help speed aid to threatened communities before losses mount, and reduce frustrating hang-ups in response time.