A disaster waiting to happen

A deteriorating oil tanker in the Red Sea contains over 1M barrels of oil, and it's expected to spill any day now. Our models show it would be a public health catastrophe, but it's not too late to prevent it.

Like Comment
Read the paper


The Safer is a deteriorating oil tanker off the coast of Yemen containing over 1.1M barrels of oil, threatening an international environmental and humanitarian catastrophe. Through our modeling, we found that if the Safer were to spill, it could disrupt food, fuel, fisheries, clean water access for millions while also producing a substantial amount of air pollution.

Experts familiar with the situation believe the Safer could spill at any moment, yet no efforts have been made to offload the oil. Negotiations over the Safer have repeatedly stalled, as parties involved have been engaged in a years-long conflict. The motivation behind our study was that perhaps the potential impacts of a spill from the Safer aren’t so clear – it’s obvious that a spill would harm the environment, but it’s not so obvious that it could lead to millions of people across multiple countries losing access to food and clean water. Our hope, then, is that by quantifying what is at stake in terms of human health, international actors might be more compelled to act with urgency.

Average surface oil concentration of 1,000 simulated spills in the winter (a,b,c) and in the summer (d,e,f). Columns denote progress of the 1,000 spills after one week (a,d), two weeks (b,e), and three weeks (c,f). Colored contours represent percentiles of average surface concentration over 1,000 simulated spills, and can be interpreted as the expected surface concentration relative to other grid cells in the exposed area. Shaded region represents the area within which approximately 90% of spill trajectories are expected to fall. Blue dots represent desalination plants.

Behind the paper

The idea for this study was born at the Refugee and Asylum Seeker Health Initiative (RAHI) Symposium, housed in the University of California, San Francisco. I attended a panel session on Yemen and COVID-19, which discussed how the pandemic has affected a country already undergoing what’s considered the world’s worst humanitarian crisis, marked by blockade-induced famine, water-borne illness outbreaks, forced displacement. The speakers gave a grim assessment of the situation in Yemen – Dr. Aisha Jumaan detailed the state of healthcare and infrastructure collapse and Dr. Amir Mohareb discussed the lack of resources available to handle outbreaks of COVID-19. Dr. Mohareb briefly mentioned the threat posed by the Safer at the end of his talk, and I realized this was something that I could model with my public health and data science skillset. I started coding up the study that afternoon.

Upon getting some promising preliminary results, I sent some drafts and outlines out to some of my advisors, mentors, and collaborators for feedback. Shortly after, Dr. Fatima Karaki, director of RAHI, put me in touch with Dr. Jumaan and Dr. Mohareb – their deep knowledge and experience with the subject matter helped to meaningfully connect the work to the people who stand to be affected by the spill. In the end, we had assembled an impressive team of collaborators including domain experts, epidemiologists, physicians, environmental engineers, data scientists, and myself.

It wasn’t easy working on the paper given our self-imposed time constraints. We set an aggressive deadline for ourselves to submit the paper – otherwise, what if the Safer spilled before we were able to finish? I didn’t have time to do all of the sub-analyses I had in mind, so we only kept the ones that were most pertinent to the interests of public health. How many people will go without food or water? How many fisheries will close down? How many will be at risk of hospitalization from the air pollution? The research questions we didn’t keep had less to do with health and more to do with environmental or economic impacts, so we scrapped them to keep a more focused message.

Despite the time constraints, the results we found along the way emboldened us to see the project through to the end. Through our models, we found there’d be millions without food or water, and that the key ports through which aid arrives would be shut down. Modeled air pollution was substantial, though not nearly as catastrophic as the food and water shortages. We were surprised to find that oil spill cleanup, even when assuming unrealistically optimistic conditions, was virtually useless in mitigating the health impacts of the spill. Given the dire implications of our results, we knew this was something we’d want to publish quickly and to a broad audience.

Fortunately, the review and publication process was an unusually positive experience. Nature Sustainability was the only journal we submitted our work to, and within two rounds of review, we received constructive comments that substantially improved the quality of the paper. The editors and reviewers respected the urgency of our work, providing fast turnaround times. It was a pleasure to work with them throughout the process.


Normally, this is the part where one might talk about potential future research directions. Could we follow up with those extra analyses we scrapped, or should we do more studies on this issue? I suppose we could, but I honestly don’t see the point. By now, it’s exceedingly clear that the Safer poses an extreme public health risk to millions in the region, and that the issue needs to be remediated. Plus, it’s already obvious the spill will have environmental and economic consequences, so I don’t think a few more studies would qualitatively change our assessment of the matter.

Instead, the bottleneck for change here is political, not scientific. I hope to use my voice as a scientist to continue communicating the potential harms of this impending spill. The situation with the Safer is highly politicized, and in an era where merely advocating for public health is considered a political stance, I hope the discourse will center the real human lives at risk here. I welcome any discussion or questions about our study, and I’m grateful for any coverage that it may receive, whether positive or negative. At the end of the day, the most important issue is not about us or our study; it’s about the people who stand to suffer from this impending spill. The people of Yemen in particular deserve so much better than the suffering and injustices they’ve had to endure, and I hope our work can play some role, however small, in bringing this preventable catastrophe to light.

Benjamin Huynh

PhD Candidate, Stanford University