Data-Driven Hope: The August 4th Tragedy in Beirut



It was a calm Tuesday afternoon, my fiance and I had just come back from work and put on our favorite TV show. We were sitting at our dining table, and I noticed it shook. “Please stop shaking your leg!” I said to my fiance, and he replied, “It’s not me, I think it’s an earthqu—” and that was the end of it. The noise that followed was loud; deafening – it was as if the earth was agitated and roared back. I lived fifteen minutes away from the Beirut port, and when I opened my eyes, all of the glass in our apartment was shattered and strewn across the floor. The fallout was apocalyptic. We had stopped thinking about ourselves. We made it, and we were still alive. We were thinking about our loved ones.

The largest non-nuclear explosion of the twenty-first century was the enormous port explosion that shook Beirut on August 4, 2020 and was caused by the improper storage of ammonium nitrate in the Beirut port. More than 200 people were killed through this negligence, over 6,000 were injured, and more than 300,000 people had to leave their homes. The overall damage from the blast ran in the billions of dollars. 

In the past, this chemical has caused major industrial disasters. An explosion at an ammonium nitrate plant in Oppau, Germany, killed 561 people and was heard hundreds of kilometers away in 1921. In 2015, the detonation of approximately 800 tonnes of ammonium nitrate in the Chinese port of Tianjin killed 173 people.

For the people of Lebanon, new social, economic, and political realities arose on August 4. Since October 17, the nation had been in a state of open uprising against the ruling class over a variety of issues, mostly economic and political. Many of these issues were exemplified by the Beirut blast itself and the ruling class’s response (or lack thereof) to the port explosion. The protests briefly resumed after the explosion. More than 300 people were reported missing on this day. An Instagram feed called “Locate Victims Beirut” was created a few hours after the blast to post pictures of the missing. The page immediately attracted over 70,000 followers, many of whom hurriedly offered information that might be of assistance or sent in pictures of loved ones who had not yet been located. About 40 victims identified were being operated on at the American University of Beirut Medical Center (AUMBC). In a sense, social media platforms became the main method of locating missing persons after the explosion. As a data scientist, I thought, is this the only source of information we have?  During disasters (explosions, disease outbreaks, floods, and other crises) real-time data can definitely aid in saving lives, but only if governments and response agencies know how to use and store it. If Lebanon had the infrastructure, data storage capacity, and a solid data governance paradigm, locating missing people would have been much more accurate, and indeed faster. Some examples of how this would work are outlined below. 
"As a data scientist, I thought, is this the only source of information we have?"
AI-Powered Surveillance Cameras
Finding missing people has been revolutionized by using security cameras working together with Artificial Intelligence (AI). Both the public and private sectors in many high-income countries now have the means to work together to locate missing people using facial recognition software. AI allows security cameras to set up alerts for when a missing individual was last seen on camera, providing vital information about where to look for them. An organization or city can now link to an infinite number of cameras as surveillance systems move closer to being cloud-based. A global collaborative network that will help monitor busy public spaces in real time is being created with the aid of AI, which would have required a lot of people, time, and money to create otherwise. If Beirut had these types of AI-enhanced cameras, missing people would have been found much faster. The facial recognition map of Central Asia is shown below. As it indicates, in Lebanon, facial recognition is banned.
Visual Capitalist

Using Cell Tower Data to Track a Missing Person’s Location

Mobile phones transmit a radio frequency signal to the “nearest” tower whenever a call is placed, according to proponents of cell tower position tracking. To determine which tower the phone was connected to at a certain time, investigators can examine phone data. An example is present in the image below. The investigators will conclude that the phone and, presumably, the user of the phone were inside the range of that tower at the time the call was placed once they have identified the specific tower that enabled a specific call. Once locating a range where a missing person is located, the search process is much more focused, as the illustration below shows.
Research Gate

Locating People Through Credit Card Transactions

Banks in Lebanon imposed restrictions on foreign currency withdrawals to $200–$300 per week as a result of demonstrations that began in 2019 in response to the financial crisis. Additionally, cash withdrawals from those accounts were stopped entirely when the COVID-19 pandemic started, and a cap was placed on local currency. As a result of monetary policies implemented by the Central Bank, people are currently unable to withdraw money even from local currency accounts. However, the immediate problem that cardholders are currently facing is the widespread rejection of their cards by businesses. All of Lebanon’s gas stations ceased accepting credit cards at once a few months ago. Then, in an act of unity, supermarkets and food stores imposed a 50/50 policy: they would only accept 50% payment by card and 50% payment by cash. Hence, tracking people by credit card use is quite impossible in this situation. But, in an alternate dimension, credit card data would have been quite useful to track financial activity and thus link people to their last location before the blast, hence tightening the search radius to find them.

Closing Thoughts

While salient privacy concerns exist around all the technologies discussed to find missing persons, they are, in all likelihood, here to stay. Therefore, developing privacy-protecting frameworks that prevent ongoing surveillance but facilitate access to data during specific times and contexts (i.e., utilizing their power for good) is an ideal solution. The disaster in Beirut (and many more that will undoubtedly occur worldwide) demonstrate the ability of these data to aid in various crisis scenarios. It is unfortunate that Lebanon (and many other countries with limited financial resources) do not, or cannot, employ either of these tactics. However, I have faith that we will someday be able to store and use these data sources in safe ways, as I believe in the power of our youth to provide innovative solutions and their drive for change. 

About the author: Yasmine Hamdar is a Computational Social Scientist with Data-Pop Alliance. She is Lebanese, and currently resides in the United Arab Emirates.


Project Report

Segundo Informe Nacional Voluntario de Guinea Ecuatorial 2024

El Segundo Informe Nacional Voluntario de Guinea Ecuatorial 2024 recoge el impacto

Journal Article

Unveiling Local Patterns of Child Pornography Consumption in France Using Tor

Child pornography—better known as child sexual abuse material (CSAM)—represents a severe form

Project Report

Stratégie Nationale des Données du Sénégal – Résumé

En 2023, DPA a fourni un soutien technique à l’élaboration de la