Back to top

Maritime Reporting Systems


MarineTraffic



www.marinetraffic.com is a website that heavily relies on AIS data. Operated by MarineTraffic, a maritime intelligence company, it includes the Live Map service, that displays the current position and navigational status of vessels that bear an AIS transponder. AIS messages are received by the large MarineTraffic network of terrestrial receivers as well as by satellite stations (SAT-AIS) provided by satellite-AIS providers (e.g., ORBCOMM, exactEarth and Spire).


Vessels appear on the map using different colours, that correspond to their type (e.g., tankers, cargo vessels, passenger vessels, etc.). Different shapes are also used corresponding to the navigational status of a vessel (e.g., a circle appears when the vessel has stopped). Both the vessel type and navigational status and speed are contained in the AIS messages. This enables users to filter the vessels that appear on the map, as shown in the image below.



Each vessel has a dedicated page with relevant information derived from the processing of AIS static and dynamic messages. For example, the vessel details page of the vessel named HABIBA M, contains information about its MMSI, call sign, flag, age, dimensions, etc. This information is obtained by the AIS static messages that this vessel has transmitted. Also, the current position of the vessel, its current navigational status (e.g., underway using engine), origin and destination, speed are also available. This information is obtained from the dynamic AIS messages that the vessel has currently transmitted.

As a simple exercise, after studying this chapter of the book (Maritime Reporting Systems) and especially the description of the AIS static and dynamic messages, you can visit the MarineTraffic Live Map and select a vessel of your choice. Then visit the vessel details page and see all relevant information about this vessel. Try to match the fields of AIS as described in the book, with the information displayed on this vessel. You will notice that, in the vessel details page, there is more information than what is included in the AIS messages. This information comes from (i) further processing and analytics and (ii) correlating the AIS dataset with other datasets to have a more in-depth understanding of the vessel behaviour.




Density Maps



Density maps is another layer on the MarineTraffic website that shows the density of the vessel track per vessel type for one whole year. This view enables users to spot the higher density areas at a glance, while restrictions for specific vessel types can also be used in order to display the density map per vessel type (e.g., tankers).

The added value of density maps is that they support the improved understanding of vessel traffic, through providing an overview of vessel behaviour either at a regional or global scale. Analysing density maps as they evolve over time also supports understanding traffic changes and pattern distribution.

The term “vessel density” has several co-notations and thus is used with several meanings in this domain. Therefore, vessel density can refer to:
  1. the average number of vessels detected within a defined geographical area (spatial grid) in a given timeframe;
  2. the average number of crossings within a defined geographical area (spatial grid) in a given timeframe (often also referred to as “vessel traffic density”);
  3. the total vessel presence times within a defined geographical area (spatial grid) in a given timeframe;

Calculating vessel density based on number of AIS messages
(source: EMODNet EU Vessel density Detailed method).

There is a considerable difference in the methods used for the creation of density maps according to the definition used, including calculations based on the number of vessel positions detected, the number of vessel tracks, their length crossing a given area and many more variations.

For example, the EMODnet method used for density maps, consists of “reconstructed ship routes (lines) from the ship positions (points), by using a unique identifier of a ship. A line is created for every two consecutive received positions of a ship. For each line, length and duration (note that it is possible to calculate duration because each AIS message comes with a timestamp) are calculated and added as attributes. The lines obtained can then be intersected with a cell representing a unit area. Because each line has length and duration as attributes, it is possible to calculate how much time each ship spends in a given cell over a time period by intersecting the line records with the cell.”. In the EMODnet method, density is expressed in hours per square kilometre per month, since vessel trajectories are reconstructed from received position points. This method assumes that there is a unique identifier available for each vessel. On the other hand, EMSA’s density maps count the number of routes that cross each cell of the same grid.

The most commonly used dataset for density map calculation is the Automatic Identification System (AIS).

Students can follow the step by step guide available here to create density maps based on the EMODnet method.

Further analysis on AIS data


Let us now see an example of how new information can be derived from further processing of AIS data. In the vessel details page of HABIBA M, there are a couple of charts that show how its speed and draught change through time. These calculations are derived from a large-scale analysis on historical AIS data and allow us to identify uncommon events in a vessel’s behaviour, i.e., big changes in a vessel’s speed and draught, that may trigger alerts.



Estimated Time of Arrival (ETA)


One of the fields of AIS messages is the estimated time of arrival (ETA) to the reported destination, which is reported by the vessel’s crew and is often not accurate. However, ETA can be calculated using the navigational information of the vessel (i.e., the speed), without relying on the ETA field of the AIS messages. For this reason, both the reported ETA and Calculated ETA are included in a vessel details page.

Integration with other sources


In order to obtain a more accurate picture of a vessel’s behaviour and of the maritime domain, in general, we need to integrate AIS data with data coming from other sources. For example, weather data can be correlated with AIS data. The image shows the weather layer enabled on the MarineTraffic map, that shows the speed and direction of winds worldwide. Correlating AIS data with weather data enables us to know the weather conditions under which a vessel is sailing. The current weather conditions in the area where a vessel currently sails (i.e., its latest known position are reported in AIS) are also available on the vessel details page.


What is even more interesting, is a vessel’s general navigational behaviour with respect to the weather. The chart below shows the speed of the vessel HABIBA M with respect to the speed of wind, compared with the speed of other vessels of the same type under the same weather conditions (wind direction with respect to the vessel, and speed). Is there anything interesting to observe?



Example Dataset



You can find a sample AIS dataset here. To visualise the data, you may perform the following steps:

  1. Download and open QGIS.
  2. Select a base layer to be used as a background map. For example, you an add a XYZ Tiles layer from OpenStreetMaps using this URL: https://tile.openstreetmap.org/{z}/{x}/{y}.png
  3. Go to Layer-> Add Layer -> Add Delimited Text Layer to add each CSV.
  4. Select Apply.

Then, you will see all the AIS positions contained in each file displayed on the map. You will notice that each layer on the map is displayed with one colour per layer (i.e., one per CSV file). To be able to identify the positions for each vessel, you can choose to use a different colour for each vessel, by following the steps described below:

  1. Right click on one of the AIS layers.
  2. Select the “Symbology” option.
  3. Select “Categorised” in the first drop down menu.
  4. Select “mmsi” in the “column” drop down menu. This means that AIS positional data will be grouped by mmsi and each group will be assigned a different colour.
  5. Select “Classify”, “Apply”, and OK.

After completing the steps described above, you will be able to identify the positions of each vessel, as shown in the image.


Another use case could be to group the vessel positions per vessel type, and use a different colour for each vessel type. To do this, follow the steps described above, but select the “ship_type” column instead of the “mmsi” column. Visualising the trajectories of the same vessel type using the same column could help us identify patterns/similarities in the movement of vessels of the same type within a certain area.


Research using Maritime Data



The research department of MarineTraffic focuses on getting the most out of maritime data by developing tools and techniques that perform large-scale data integration, multi-sensor fusion, and AI-powered analytics. Examples of projects using multi-sensor fusion, AI, and Big Data techniques for actionable maritime intelligence can be found on the MarineTraffic Research website. For example, Big Data techniques on historical AIS data were employed in a recent study that measured the COVID-19 impact on global trade.

Also, the following video describes the Anomaly Detection tool for MarineTrafffic, a data-driven approach which relies on Big Data and AI technologies to identify uncommon behaviour in vessels (e.g., route deviations, proximity events, navigation in shallow waters, etc.).