Conflicting realities: mapping discrepancies in migrant deaths data on the Atlantic route 

Guest post by Maurice Stierl and Marta Sánchez Dionis. Dr Maurice Stierl is a senior researcher at the Institute for Migration Research and Intercultural Studies, Osnabrück University. His book ‘Migrant Resistance in Contemporary Europe’ was published by Routledge in 2019. Marta Sánchez Dionis is a migration and human rights researcher currently serving as an Advocacy Officer at SOS MEDITERRANEE. The views expressed in this article are her own and do not reflect those of SOS MEDITERRANEE. 

In recent years, the Atlantic Ocean has seen an increase of precarious migration, with more and more boats departing from the coasts of Senegal, Mauritania, or Morocco  to reach Spain’s Canary Islands. As the number of people attempting this journey has risen, so too have estimates of migrant deaths at sea, reaching unprecedented levels. Yet, the true scale of loss remains uncertain. As states do not provide any systematic documentation of migrant deaths at sea, non-governmental and international organisations have set out to fill this ‘knowledge gap’ by counting the dead. However, their findings often diverge sharply, producing conflicting realities of loss at sea and raising pressing questions not only about how such estimates are produced but also what these discrepancies may mean for activist and humanitarian responses, policymaking, and broader perceptions of migration across the maritime border.  

Documenting the scale of dying and disappearing at sea is a tremendously complex task, given that many boats disappear without ever being found, in what have been termed ‘invisible shipwrecks’. Actors trying to understand what happened to a particular boat at sea have to search for traces and piece fragments of information together. Family members and friends of the disappeared, media sources, activist networks, local witnesses, or government agencies may provide some information, but rarely do they offer the full picture. As a result, actors seeking to count the ‘missing’ and dead must corroborate various, at times conflicting, accounts about a specific boat that left from a specific location at a specific time, carrying a specific number of people, and never arrived. 

Given the difficulty of the task, it may not be surprising that existing attempts to count deaths and disappearances at sea come to different conclusions. What may surprise, however, is the extent of the discrepancies in the data. In our research project at the University of Osnabrück, we have examined different ‘data practitioners’ in the Atlantic region. What their figures show is a growing disparity between estimated numbers of deaths.  

The International Organization for Migration (IOM), which started counting ‘missing migrants’ in 2014, highlights a dramatic increase in deaths along the migration route to the Canary Islands, where the number of deaths recorded by the organisation rose from 202 in 2019 to 877 in 2020. In 2021, IOM estimated 1,126 deaths, and by 2024, the toll had risen further to 1,142 – the highest number documented so far. Yet, other available data paint a much deadlier picture.  

Caminando Fronteras, a Spanish NGO that began counting along the Atlantic route in 2015, speaks of 1,832 deaths in 2020, a figure that grew to 3,939 the year after and reached the harrowing toll of 9,757 in 2024. In comparison, IOM’s estimate of 1,142 deaths in 2024 represents a remarkable difference of 8,615 lives lost. To some extent, these conflicting findings reflect the varying methodologies used by the different data practitioners. While IOM primarily relies on media reports to document deaths in the region, Caminando Fronteras gathers data from survivor testimonies or reports from family members and communities in countries of origin and transit.  

Caminando Fronteras’ December 2024 report made headlines in Spain and internationally, prompting diverse reactions. A day after its release, the president of the Canary Islands’ regional government, Fernando Clavijo, cited its figures in a letter urging the European Commission to place the Atlantic migration route among its priorities in terms of migration policy in 2025, highlighting the unprecedented ‘migration crisis’ facing the archipelago. In a recent meeting with the EU Commissioner in Strasbourg, Clavijo referenced Caminando Fronteras’ data once more, calling for increased Frontex involvement along the coasts of Senegal, Gambia, and Mauritania in a bid to ‘save lives’.   

The Spanish NGO’s data on migrant deaths in the Atlantic have also created a heightened sense of urgency among non-governmental search and rescue organizations. While for the past decade, search and rescue efforts by the ‘civil fleet’ have focused on the Aegean Sea and the central Mediterranean Sea, reports on the rising death toll in the Atlantic have prompted several of these organisations to look westward and consider possibilities of engagement.  

Among them is the Humanitarian Pilots Initiative, a Swiss non-governmental organisation that has carried out monitoring flights in the central Mediterranean Sea. In cooperation with Sea-Watch, the initiative has spotted more than 1,000 boats in distress between the shores of Libya and Italy since 2017. Now, in response to alarming reports of escalating deaths in the Atlantic, a light twin-engine aircraft operated by the organization will soon embark on an exploratory mission to search for migrant boats at risk of capsizing or disappearing into the vastness of the ocean.  

By extending their operations to the West African coast, south of the operational area of the Spanish sea search and rescue agency Salvamento Marítimo, the humanitarian pilots open a new chapter in the efforts of civil society actors to prevent loss of life along Europe’s sea borders. Yet the extreme discrepancies in existing death counts raise challenging questions regarding how and where such interventions should take place: given the vast distances of the Atlantic route, with boats taking up to ten days to reach the Canaries, where should rescue or monitoring assets be placed and which maritime regions prioritised?  

Existing data practices offer little help. While IOM’s Missing Migrants Project records every presumed shipwreck in its public database, the locations of shipwrecks provided on its maps are often little more than rough estimates, as the project acknowledges. Caminando Fronteras, meanwhile, roughly identifies maritime regions where it suggests that most shipwrecks have happened, but it does not share more detailed information on its data practices and methodologies, hampering attempts to better understand or corroborate its findings. 

Thus, and somewhat ironically, while data practitioners set out to visibilise that which often remains under the public’s radar – ‘invisible shipwrecks’ and the uncounted and unaccounted loss of life at sea – their own data prompt considerable uncertainty. In our research project, we have spoken with key actors in Senegal, Spain and Germany, all of whom were struggling to come to terms with the different estimates and what these discrepancies may mean for their work. While some shook their heads in disbelief about the possibility of an annual death toll nearing the mark of 10,000, others considered it plausible.  

When we asked Ruben Neugebauer, member of the Humanitarian Pilots Initiative, what he made of the existing death counts, he emphasised that data discrepancies were indeed a challenge for their operational design: “Of course, it is a big question for us why the figures on the death toll differ in the way they do. IOM’s methodology is clear, they only count clearly documented cases, so we can assume that the actual number of fatalities is higher. At the same time, the figure by Caminando Fronteras seems extremely high. Given this discrepancy, we see our first deployment also as an assessment operation to gain more insights and knowledge.” Remarkably, not the lack of data but the existence of different sets of data compelled the Humanitarian Pilots to consider their intervention as a ‘fact-finding’ mission.  

How, then, should we deal with these discrepancies in data? While we do not have conclusive answers, we need to begin by acknowledging the contestedness of data. As Science and Technology Studies scholarship has long emphasised, data and statistics are not innocent. Though often presented as objective or factual, they do not simply describe a coherent and singular reality ‘out there’. Indeed, as the varying death counts for the Atlantic show, data can produce multiple, and conflicting, realities.  

There is thus also a productivity, or performativity, to data production, meaning that depending on which statistical counts are believed, different borders become the deadliest in the world. For IOM, their statistics reinforce their claim that the central Mediterranean Sea is the deadliest border globally, while Caminando Fronteras’ statistics make the Atlantic route “the most lethal in the world”.  

Moreover, there is a consequentiality of data. Data can shape which border spaces become sites of urgent intervention. In the case of Caminando Fronteras, its statistics have played a crucial role in creating a sense of urgency among both governmental and humanitarian actors to engage in the Atlantic region.  

For us, the contestedness, performativity, and consequentiality of data regarding the death toll in the Atlantic raise many questions regarding how data is produced and how researchers, policymakers or rescue actors should assess data discrepancies and navigate the conflicting realities they create. We also wonder about the question of responsibility, with which we want to end: if data is indeed never innocent, what responsibility comes with the production of statistics on migrant deaths and everything they set into motion? 

Funding:
Maurice Stierl received funding by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – SFB 1604 – 501120656. 

Any comments about this post? Get in touch with us! Send us an email, or find us on LinkedIn and BlueSky.