Model traces ship AIS blackout zones in conflict-affected Black Sea waters

Ships that vanish from the Automatic Identification System (AIS) in the waters off Abkhazia and Eastern Ukraine may be heading to ports, and a new study offers a statistical way to guess where. The research, published in EPJ Data Science, analyses more than 27 million AIS transmission signals recorded in the Eastern Black Sea during 2018, a region shaped by the protracted conflicts surrounding the coastal breakaway territory of Abkhazia in Georgia and the occupied territories in Eastern Ukraine.

Long AIS transmission gaps are a persistent problem for maritime domain awareness. Vessels can switch off their transponders or lose coverage, and in conflict-affected coastal environments such gaps may obscure movements that analysts would want to detect. The study treats these prolonged gaps not as missing data to be discarded, but as events to be modelled, asking whether plausible port-calling opportunities can be associated with the periods when a vessel’s signal disappears.

Technically, the authors combine a Discrete Port Calling Opportunities Model (DPCOM) with a grid-based statistical framework. DPCOM evaluates candidate trajectories that could explain a given transmission gap, identifying which nearby ports a vessel could plausibly have reached while dark. The grid-based framework then maps where anomalous behaviour clusters spatially, allowing the researchers to evaluate the distribution of AIS transmission loss across the study area rather than examining gaps in isolation.

The results indicate that the trajectories selected by DPCOM are strongly associated with statistically significant anomaly hotspots, and that recurring patterns of AIS transmission loss can be identified in the 2018 Eastern Black Sea data. In other words, the model’s preferred explanations for signal gaps align with the places where anomalous behaviour concentrates, suggesting the approach captures something real about vessel behaviour in the region rather than random noise.

Beyond the specific findings, the study is presented as a reproducible framework for analysing prolonged AIS transmission gaps. Because it relies on historical AIS data and a documented statistical pipeline, other researchers can apply the same method to different regions or time periods, which the authors argue highlights the value of large-scale behavioural analytics for maritime domain awareness in conflict-affected coastal environments.

The scope of the work is worth keeping in mind. The dataset covers a single year, 2018, and a single region, the Eastern Black Sea, so the findings describe patterns in that setting rather than a global picture of AIS dark activity. The model identifies plausible port-calling opportunities and statistical associations; it does not establish what any individual vessel actually did during a transmission gap, and the study is a modelling exercise rather than an operational enforcement tool. As an early-shared, peer-reviewed accepted article, it is also subject to further edits before the final version of record. Even so, for analysts tracking shipping in contested waters, the work offers a transparent, data-driven starting point for interpreting the silences in a vessel’s track.

Subject of Research: Statistical modelling of AIS transmission gaps and port-calling opportunities in the conflict-affected Eastern Black Sea

Article Title: AIS-derived port-calling opportunity modelling in conflict-affected waters

Article References: van Ginkel, M. (2026). AIS-derived port-calling opportunity modelling in conflict-affected waters. EPJ Data Science. https://doi.org/10.1140/epjds/s13688-026-00699-2

Image Credits: AI Generated

DOI: 10.1140/epjds/s13688-026-00699-2

Keywords: AIS, maritime domain awareness, Black Sea, Abkhazia, port calls, transmission gaps, anomaly detection, DPCOM, vessel behaviour, geospatial analysis, conflict zones, EPJ Data Science