Ship Operation Mode Recognition and Carbon Emission Inventory Composition Based on AIS Data

With the booming development of the global shipping industry, port areas have become significant sources of atmospheric pollutants and greenhouse gas emissions, posing serious threats to air quality and public health. Therefore, establishing carbon emission inventories for precise assessment has become crucial. However, traditional emission inventories predominantly rely on static "top-down" statistical methods, which suffer from limitations such as coarse spatiotemporal resolution and inaccurate operational mode characterization. This study aims to identify ship-specific fine-grained operational mode in port areas throught multi-dimensional features from AIS data using the K-means clustering method. Based on this, we construct high spatiotemporal resolution emission inventories for different container ship types according to emission factors. The results indicate that Panamax container ships emit the highest levels of pollutant gases, with greeehouse gas emissions from ship maneuvering modes reaching 4,294,812.3 tons. These findings highlight the urgent need for effective measures to combat port air pollution and promote sustainable shipping development.