Good Data: Helping in Times of Need
Normally, this blog focuses on kinds of tools and projects being used in academia. But in light of recent events, it’s important to note just how the tools and skills that the Digital Scholarship Lap is helping students to cultivate can be translated and applied into real-world situations.
Below, we’ve highlighted several important efforts which feature data science and digital media to assist in the recovery efforts after Hurricanes Harvey, Irma, and Maria, and the recent earthquakes in Mexico. And each one features applications or tools that you can learn how to use in the DSL.
When Hurricane Harvey slammed into Houston, emergency dispatchers said: “Call for rescue. We aren’t checking Twitter.” Survivors, unable to get through to 911, did what came naturally — they turned to social media.
We saw their cries for help, and we answered. In one hour, we launched an interactive, live updated map so volunteer boat rescuers knew exactly where to go.
The story behind CrowdRescue’s origins gets right to the heart of what this blog post is about: the way data science can be utilized to organize people, resources, and responses in times of crisis. Innovative solutions to problems are being created every day as people realize the power of data both to assess impact as well as to spread awareness.
CrowdRescue’s way of contributing to the recovery efforts over the past few weeks has been to crowdsource data through social media and information self-reported via a form on their website. This data is entered into a database with the help of Fulcrum, interpreted with Tableau, and visualized with GIS mapping software from Esri, resulting in the map above. Information crucial to the recovery efforts in Puerto Rico is transmitted to rescue and humanitarian teams in support of their efforts. As CrowdRescue says, they are “helping the helpers.”
Among the many kinds of work and products Google creates is the Google Crisis Response division. This section of the Google empire is focused on humanitarian work, especially in times of great public need, such as natural disasters or areas engaged in conflict that puts large populations of people at risk. In the team’s toolbox are a number of Google products or developments that can cooperate to assist in response and recovery efforts. In addition to Google Person Finder, which helps people affected by or even displaced by crisis events update or make contact with others, Google uses their Fusion Tables data management application to organize information about resources, shelters, areas of need, and more, and with Google Sites, they can create an individual Crisis Response page for each incident.
And, of course, there are the Google Maps and Google Earth tools, which help to display in geographic and real-time terms the extent of a particular event’s devastation and needs. For the recent tropical storm and hurricanes in the Gulf of Mexico and the Caribbean, as well as the earthquakes in Mexico the team created individual maps to better help understand the damage and identify each area’s specific needs. Right now, the Harvey map (above) is displaying lists of the hazards and road closures that still need to be addressed while the Irma map allows users to see areas affected by the storm in aerial images taken in the days after Irma had passed.
The map for Hurricane Maria, however, is displaying shelter locations as well public advisory or warning alerts. The situation in Puerto Rico and the other islands affected by the storm is impeding the collection of data that could help better understand the extent of the damage and need, which is reflected in the lack of information on the map so far. Data can be of great help in times of need like this, but there has to be a way to collect it. Until recovery efforts can re-establish some power to the island, or crisis response teams can start to communicate information back, understanding what is needed will be a slow and difficult process.
Not every tool or application created in the wake of the recent disasters has been by organizations. A group of current and former MIT and Harvard students have collaborated across Mexico and the US to create Manos a la Obra. The website has two focuses: Offers and Needs. On one tab, it displays offers of assistance, detailing location and contact information, what the person is able to offer (food, water, a place to sleep, manual labor, etc.), and on the other, requests for help along with the same important information. The information is submitted using a Google Form, and then sorted into a publicly accessible Google Sheet. It uses Google Fusion Tables and Google Maps to plot and display the data. And the relevant coding information has been made available on GitHub, which means that this can be replicated — by the creators or by others — in times of future need.
Beyond the work it’s allowing people to do in Mexico as they begin recovering from the damage and losses of the earthquake, this resource is notable for its grassroots origins. Necessity, as they say, is the mother of invention. These men and women, professionals and amateurs, saw that there was a need for a tool to connect people in the wake of the earthquake.
And so they created one.
Lastly, the Catholic Charities USA Disaster Operations Map helps responders assess which areas are most vulnerable when a disaster hits and what sort of intervention or assistance might be a priority. The mission of the CCUSA is to provide “disaster relief assistance and [they] wanted to create a map to better target mitigation, preparedness, relief, and recovery projects in order to best serve communities that are both at greatest risk for disasters and most overlooked or outright excluded from federal assistance during disasters.”
The map encodes a geographically-defined population’s Social Vulnerability Index according to criteria which include socioeconomic status, household composition, minority status and language skills, as well as housing and transportation situation. It then cross-references the area’s risk factors (according to the ATTOM’s Natural Hazard Housing Risk Index) for events like hurricanes, floods, earthquakes, wildfires, tornadoes, and hail. With these two assessments, response teams can anticipate what kind of help or intervention an area might need in a time of crisis and send specific resources and personnel to assist residents. One neighborhood might need more interpreters or bilingual responders because of a high number of residents not fluent in English, while a neighborhood across the affected area might need more specialized medical equipment and transport vehicles because the population includes many individuals with disabilities, for example.
DataKind, a group that combines data science and analytics with organizations for social change and social good, then created the map interface with Mapbox, an open source location data platform, using vector tiles converted from a shape file.
The tools and skills that the Digital Scholarship Lab is helping students to cultivate are truly world-changing. And if you wonder how that could be, just take another look at these websites and tools above.