A Vessel Traffic Service (VTS) plays a central role in maritime traffic safety. Regulations are given by the International Maritime Organization (IMO) and Guidelines by the International Association of Marine Aids to Navigation and Lighthouse Authorities (IALA). Accordingly, VTS facilities utilize communication and sensor technologies such as an Automatic Identification System (AIS), radar, radio communication and others. Furthermore, VTS operators are motivated to apply Decision Support Tools (DST), since these can reduce workloads and increase safety. A promising type of DST is anomaly detection. This survey presents an overview of state-of-the-art approaches of anomaly detection for the surveillance of maritime traffic. The approaches are characterized in the context of VTS and, thus, most notably, sorted according to utilized communication and sensor technologies, addressed anomaly types and underlying detection techniques. On this basis, current trends as well as open research questions are deduced.

Climate change is expected to increase the frequency of simultaneous crop failures across major producing regions, raising concerns about global food security. Whereas previous studies have mainly focused on production impacts, the resulting socio-economic consequences remain less explored. Here, we assess how climate-driven multi-breadbasket failures translate into systemic risks for global food prices and regional consumption. Driving the agent-based agricultural trade model Agrimate with probabilistic production failure scenarios for historical (2006–2015) and +2 ∘ C climate conditions, we show that climate change-driven increases in simultaneous production failures across global breadbasket regions lead to substantial increases in the likelihood of world market price spikes. The probability of exceeding a historical 90th percentile peak in bi-weekly prices nearly doubles for wheat and increases by factors of 2.70 and 2.45 for maize and soybean, respectively. These dynamics translates into increases in conditional consumption at risk of 0.24, 0.71, and 1.20 million calories per capita per year ( Mcal / capita / year ) for wheat, maize, and soybean, respectively. Historical regional consumption risks are highly heterogeneous, ranging from negligible impacts to over 20 Mcal / capita / year across all crops and exceeding 100 Mcal / capita / year in high-impact regions. These patterns reflect structural characteristics of the food system: production concentration, import dependencies, and small supplier base amplify risks, whereas higher stock-to-use ratios provide a buffer. In export-oriented regions, risks to domestic consumption can increase when, in the absence of protective policies, surging world market demand redirects supplies abroad. Climate change further compounds these structural vulnerabilities, leading to regional increases in consumption risk of up to 6 Mcal / capita / year in the most exposed areas. Together, these findings point toward the need for a coordinated policy agenda: diversifying production and trade, strengthening storage management, and implementing early-warning systems.

The global supply chains for major food staples, including wheat, rice, soy, and maize, are significantly reliant on few chokepoints, predominantly situated within the maritime network. Grains traded at international markets are produced in a small number of breadbasket regions. This geographical production concentration has a substantial impact on the degree of reliance on these maritime chokepoints. It has been demonstrated on multiple occasions in preceding years that ports and shipping routes are susceptible to disruption as a result of extreme weather or political conflicts. Here, we analyse short-term risks to global and regional food security arising from chokepoint disruptions. To this end, we have developed a model to construct global supply chain networks, incorporating different types of roads, inland waterways, railways, and maritime shipping lanes, with different ship types. Additionally, the model accounts for different types of logistic infrastructure that are important for crop transportation, like ports and railway stations and borders, with their subsequent costs and waiting time. The resulting networks are validated by different explicit examples of known crop transport routes. For the impact modelling, first, a transport cost matrix is calculated within the network, from and to every global Admin-1 region, which is done with a lowest-cost Dijkstra algorithm. Secondly, a crop trade matrix from and to every Admin-1 region is calculated. This is done by aggregating real-world trade data from country-level to Admin-1 level with a cost-based gravity-model that includes different types of consumption and the transport cost matrix. Thirdly, the lowest-cost-path between all regions that trade with each other is calculated, through the same Dijkstra algorithm as in step 1, and multiplied with the amount of trade from the trade matrix.We assess risks to food security arising from factual as well as counterfactual scenarios, including single and multi-chokepoint disruptions. The outcomes of the different scenarios are compared with the baseline scenario, in which no chokepoint is deactivated. The study quantifies (i) how many people are affected by, and (ii) how much additional transport costs arise from alternative routes due to a disruption of a chokepoint, per crop.The implemented supply chain network model provides a basis for understanding the implications of disruptions to global food security caused by chokepoint disruptions, highlights strongly affected ‘hotspot’ countries, and establishes the foundation for dynamic modelling of food insecurities. The model is developed for a fast computation of disruption analyses in big networks and will be available freely after final development.