We need disaster data to save lives and guide recovery

I have been horrified by the heartbreaking news and images coming out of Türkiye and the Syrian Arab Republic after the deadliest earthquakes of the century left tens of thousands of people injured and killed.

These terrible earthquakes have been particularly devastating given the sheer scope of destruction left in their wake. But unfortunately, catastrophes like these are far from rare. Over the past decade, natural disasters have led to the death of nearly half a million people around the world.

Although we might not be able to prevent many of these disasters in the years to come, we can mitigate their impact by saving lives and making recovery efforts more effective.  And for that, we need timely and cost-effective data at high levels of granularity.

Here are three ways we can ensure such disaster data are available in crisis situations:

1. We must pursue both the fundamentals and the frontier

Government data—such as data from censuses, surveys, civil registration, and administrative systems—continue to provide the foundation for critical disaster-related statistics. At the same time, new technologies and data sources produced by private entities—like mobile phone usage, social media activity, online queries, crowdsourcing platforms, and remote sensing technologies—can help us save time, increase accuracy, and enhance precision, which are of utmost importance in emergency situations.

In order to build toward effective data-driven disaster reduction systems, we need to repurpose private data and combine them with public data to better respond to emergencies. 

We have seen such efforts succeed before. For example, after the 2014 earthquake in Napa, California, disaster-related semantics were extracted from Twitter by a machine learning algorithm and matched with geolocation data to promptly assess the scale and impact of the disaster. Similarly, mobile location data provided by a private company have been used to understand population movement patterns and recovery trends after historical disasters in Mexico City and Mumbai.

We are only scratching the surface of what is possible through the integration of different types of data, but the impressive results achieved so far compel us to continue these explorations.

2. We need to invest in our collective “data readiness”

The combination of innovative data sources, cloud platforms, and the fast and efficient processing made possible by these technologies, is enabling local, national, and international policymakers to understand the scope and scale of disasters, estimate the number of affected people, and comprehend the situation anywhere around the globe from reliable, objective, third-party data in mere minutes to hours. This is crucial in crisis situations where every second counts.

Last summer, the World Bank’s Development Data Group was asked to conduct a satellite-based rapid damage assessment for floods in Assam and Meghalaya, India, while the floods were still ongoing. Using synthetic aperture radar data and some cloud processing based on code we made publicly available, we were able to quantify the scale of the affected area and the number of people impacted in less than two hours to help guide rapid and appropriate action.

Much of this starts with investing in what I call “data readiness” – that is, data collection and dissemination systems, data governance, data use and literacy, and the preparation of foundational information before disaster strikes. To enable data readiness more broadly, we need to form new collaboratives for innovation, as well as join forces to invest in countries’ data capacity and data systems so they can better anticipate and respond to disasters. 

3. Our efforts around disaster data need to be grounded in value, equity, and trust

As we are exploring how we can use these new data sources through collective partnerships to mitigate disasters and protect the most vulnerable among us, we need to be keenly aware of the need for strong data governance and safeguards against data misuse.

Building well-functioning, data-driven disaster reduction systems is only possible when countries can use and reuse both public and private data, and leverage synergies between them to create more value.  We also need to ensure that every person and every country can benefit equally from data. Creating trust in the integrity of the data system by staying vigilant against potential harms, including cybercrime and discrimination, is equally critical.

Fundamentally, we need to work toward what the World Development Report 2021: Data for Better Lives calls a “new social contract on data,” guided by the principles of value, equity, and trust.

As a warming climate increases the frequency and intensity of natural disasters, we have to be better prepared than ever to save lives and livelihoods, particularly in lower-income countries. Such an extraordinary challenge requires an extraordinary and collective response. And with the right data, we can get it right.