Disaster Maps

In the wake of a disaster, Disaster Maps share real-time information with response teams, helping them determine things like whether communities have access to power and cellular networks, if they have evacuated, and what services and supplies they need most.

Features

Privacy

Disaster Maps use statistical techniques to maintain individuals' privacy. For example, we only share privacy-protected information and also add up data points in a given area (called a map 'cell' or a 'tile') to prevent re-identification. If there are only a few individuals in an area, we also smooth populations across tiles, meaning that we average the number of people in a given area with nearby areas, making it even harder to re-identify anyone.

Timeliness

Disaster Maps can be generated within 24 hours of a natural disaster — much faster than comparable tools— and update daily as the situation on the ground unfolds. This allows agencies to respond to changing circumstances in evacuations, connectivity, or supply needs.

Global Reach

Since the launch of Disaster Maps in June 2017, we have generated data for 100+ natural disasters, helping to guide response efforts around the world. During the 2018 hurricane season, our maps informed active disaster recovery in India, Guatemala, Indonesia, the Philippines, California, North Carolina, Florida, and other disaster-affected locations around the world.

The most provocative part of this data, from Direct Relief’s point of view, seems to lie in quickly getting a read on how people are actually behaving during emergencies, rather than how we suspect they ought to behave. That’s often the difference between intending to do good and accomplishing it.”

Andrew Schroeder, Director of Research and Analysis, Direct Relief

Who Uses Disaster Maps

International agencies

International agencies and UN organizations like UNICEF and the World Food Programme use Disaster Maps data to guide their local deployments to disaster-affected areas and support local governments in their response efforts.

Domestic non-profits

Domestic organizations like the American Red Cross, SEEDS India, and Humanity Road use Disaster Maps to support their local communities, track evacuations, and route supplies to the areas that need them most.

Universities and researchers

Universities and researchers use Disaster Maps to analyze how disaster-affected populations are using social services, whether they evacuate based on official orders, and how social ties affect their resilience after a disaster.

This Facebook population map shows the shift in people during the Camp Fire from the town of Paradise to the nearby town...

A woman receives services from SEEDS India following the floods in Kerala. SEEDS used Facebook Disaster Maps to better...

A man receives supplies from a Direct Relief worker in Puerto Rico. Direct Relief uses Facebook Disaster Maps to better...

A child receives a respiratory mask in the context of a wildfire.

A Medical Professional explains the supplies in a Direct Relief kit to a colleague.

A Direct Relief staff member in Puerto Rico oversees the distribution of supplies in Puerto Rico.

More Less

Methodology

Facebook Disaster Maps show where disaster-affected populations are located, how they are moving, and whether they have access to cellular networks and power. All data is privacy-protected— that is, it doesn't connect to a person’s name or any other identifying information. To protect individual privacy, data sets are aggregated across time and space in the following ways:

Temporal aggregation

While timely data is needed during a disaster, aid organizations are unable to process and respond to new inputs in real time. For this reason, we share data at regular intervals, usually every 24 hours.
Spatial aggregation

We group map areas into 360,000 square meter tiles, or to local administrative boundaries, in order to gather overall insights about groups of people in a given area, rather than providing any insights about individuals.
Spatial smoothing

For each map area, we compute a weighted average of the population value in the tile with the values in neighboring tiles. This local averaging results in a map with a smoother, clearer signal, while preserving privacy.

Learn more

Case Studies

This study by Luxembourg Institute of Socio-Economic Research and Universite Catholique de Louvain uses Facebook users’ mobility to study the daily evolution of cross-border movements of people during the Covid-19 pandemic.

Disaster Maps

Cross-Border Mobility Responses to Covid-19 in Europe: New Evidence from Facebook Data

This study investigates the relationship between disaster damage, sheltering-in-place, evacuation-related mobility, and contagion following Hurricane Zeta in Southeastern Louisiana and The Wildfires... This study investigates the relationship between disaster damage, sheltering-in-place, evacuation-related mobility, and contagion following Hurricane Zeta in Southeastern Louisiana and The Wildfires in Napa and Sonoma Counties, California, known as the Glass Fire. The study draws on data from the county subdivision level and mapped and aggregated tallies of Facebook user movement. Findings suggest associations between disaster damage and higher rates of COVID-19 cases.

Disaster Maps

COVID-19 To Go? The Role of Disasters and Evacuation in the COVID-19 Pandemic

This study investigates the relationship between disaster damage, sheltering-in-place, evacuation-related mobility, and contagion following Hurricane Zeta in Southeastern Louisiana and The Wildfires in Napa and Sonoma Counties, California, known as the Glass Fire. The study draws on data from the county subdivision level and mapped and aggregated tallies of Facebook user movement. Findings suggest associations between disaster damage and higher rates of COVID-19 cases.

On April 9th, 2021 the La Soufriere volcano in the north of the island of St. Vincent erupted, spewing heavy ash and smoke across the island and much of the surrounding region. An estimated 16,000... On April 9th, 2021 the La Soufriere volcano in the north of the island of St. Vincent erupted, spewing heavy ash and smoke across the island and much of the surrounding region. An estimated 16,000 persons in the most heavily affected area were evacuated from the island and another estimated 2,000 persons left their homes. Direct Relief conducted a space-time hotspot analysis of population changes based on Facebook's Disaster Maps aggregated population data between April 10th and April 17th, which indicates the magnitude of the population reduction in northern St. Vincent, and the relative shift to southern communities.

Disaster Maps

Volcanic Explosion in St Vincent and the Grenadines – Facebook Population Mobility Analysis from April 10th through 17th, 2021

On April 9th, 2021 the La Soufriere volcano in the north of the island of St. Vincent erupted, spewing heavy ash and smoke across the island and much of the surrounding region. An estimated 16,000 persons in the most heavily affected area were evacuated from the island and another estimated 2,000 persons left their homes. Direct Relief conducted a space-time hotspot analysis of population changes based on Facebook's Disaster Maps aggregated population data between April 10th and April 17th, which indicates the magnitude of the population reduction in northern St. Vincent, and the relative shift to southern communities.

Understanding the population composition and distribution of a region affected by a major natural disaster is vital for the allocation of resources to communities in need and critical to inform... Understanding the population composition and distribution of a region affected by a major natural disaster is vital for the allocation of resources to communities in need and critical to inform mortality estimates. The utilization of social media traces, coupled with mobile phone data, could provide live estimates of post-disaster population changes in disaster-affected areas.

Disaster Maps

Quantifying the dynamics of migration after Hurricane Maria in Puerto Rico

Understanding the population composition and distribution of a region affected by a major natural disaster is vital for the allocation of resources to communities in need and critical to inform mortality estimates. The utilization of social media traces, coupled with mobile phone data, could provide live estimates of post-disaster population changes in disaster-affected areas.

View All Case Studies

In the News

How Data is Helping Facebook Provide Disaster Management Maps livemint Using Data for Good - Video Open Data Science Facebook and Da Nang City partner for Disaster Response Lao Dong What is the partnership between Danang City and Facebook about? Thanh Nien Centers look to new Facebook tools for help during disasters AP Facebook Tool Helps Charities Find Evacuees Who Need Help US News and World Report Lifeboats Thrown by Social Media The Hindu Business Line How Facebook Has tripled its disaster maps partnerships Devex How Facebook’s Disaster Maps is helping aid organizations serve people affected by florence Fast Company

Get Involved

Please email the Disaster Maps team if you're interested in becoming an NGO partner.

Disaster Maps

Features

Privacy

Timeliness

Global Reach

Who Uses Disaster Maps

International agencies

Domestic non-profits

Universities and researchers

Methodology

Temporal aggregation

Spatial aggregation

Spatial smoothing

Case Studies

Cross-Border Mobility Responses to Covid-19 in Europe: New Evidence from Facebook Data

COVID-19 To Go? The Role of Disasters and Evacuation in the COVID-19 Pandemic

Volcanic Explosion in St Vincent and the Grenadines – Facebook Population Mobility Analysis from April 10th through 17th, 2021

Quantifying the dynamics of migration after Hurricane Maria in Puerto Rico

In the News

Get Involved