Big data for resilience: realising the benefits for developing countries

Insurance has a role to play in mitigating economic losses caused by natural disasters

'Big data' is defined as 'an ecosystem made up of the combination of three factors: digital data from sources as diverse as satellites and mobile phones; the capacity to analyse and use that data, and the people who produce, analyse, and/or use the data. The concept of big data goes well beyond the datasets themselves, regardless of their size.' (Data-Pop Alliance, 2015).

Under the SHEAR Programme, DFID, NERC and ESRC funded eleven case studies and pilots to explore the links between big data and resilience.

A synthesis report, combining an extensive review and analysis of the available academic and policy literature and the findings from the eleven studies, was funded by DFID. The synthesis report was developed by a team of researchers under the umbrella of Data-Pop Alliance of the Harvard Humanitarian Initiative (HHI), MIT Media Lab, and the Overseas Development Institute (ODI).

Big data synthesis report

The synthesis report sheds light on this rapidly changing area by highlighting the growing body of empirical work that explores ways in which big data has been used to increase resilience. The report explores the opportunities, challenges and required steps for leveraging this new ecosystem of big data to monitor and detect hazards, mitigate their effects, and assist in relief efforts.

Case studies

Four themes emerged in the eleven case studies and pilot projects that DFID with NERC and ESRC commissioned to explore the links between big data and resilience.

Building resilience through crowdsourcing

Early flood detection for rapid humanitarian response: harnessing big data from near-real-time satellite and Twitter signals (Jongman et al.).
Increasing resilience to natural hazards through crowdsourcing in St Vincent and the Grenadines (Mee and Duncan).
Inclusiveness in crowdsourced disaster response (INCROWD) (Roth and Luczak-Roesch).

Using mobile network data to understand actions, behaviours, and attitudes

Mobile network data and climate resilience: analysis of Cyclone Mahasen in Bangladesh using de-identified data of five million phones in the Grameen phone network (Bengtsson et al.).
Big data for flood resilience in East Africa (Iliffe et al.).
Leveraging mobile network big data for disaster risk reduction: minimising harms and facilitating access (Samarajiva and Lokanathan).

Improved statistical methods for defining disaster risk

Landslide susceptibility mapping in data-poor environments (Cheng).
Big data for tsunami hazard warnings in India (Guillas).

Big data and communication technologies for awareness raising and disaster relief and recovery

Mobile-based disaster risk monitoring system: an innovative approach to enhance community-led disaster preparedness in Uganda (Kiragga et al.).
The potential of big data to encourage long-term and preventative disaster risk reduction behaviours: evidence from Cochabamba, Bolivia (Sou).
Big data in disease disaster management in developing countries: a mobile phone data-use framework (Cinnamon et al.).