Reference metadata describe statistical concepts and methodologies used for the collection and generation of data. They provide information on data quality and, since they are strongly content-oriented, assist users in interpreting the data. Reference metadata, unlike structural metadata, can be decoupled from the data.
This table contains experimental data of quarterly estimates on vessels arriving in the EU main ports, based on data sources provided by the European Maritime Safety Agency (EMSA).
The data are presented for the EU aggregate.
Vessels arriving in the EU main ports have been estimated by Eurostat by using an autoregressive integrated moving average model with exogenous variables (ARIMAX). The model is based on Eurostat’s quarterly data of previous quarters and EMSA data sources (automatic identification system (AIS) and other administrative data).
The data are presented by vessel type according to the annex VI Type of ship classification (ICST-COM) of the Directive 2009/42/EC on statistical returns in respect of carriage of goods and passengers by sea.
Data collection on maritime transport, particularly on vessel traffic of sea-going vessels, on voyages undertaken wholly or partly at sea. Incoming traffic of vessels that embarked/disembarked passengers or loaded/unloaded goods.
3.4. Statistical concepts and definitions
The concepts used in this domain are the following:
Port
A place having facilities for merchant ships to moor and to load and/or unload cargo or to disembark and/or embark passengers to or from vessels, usually directly to a pier.
Statistical Port
A statistical port consists of one or more ports, normally controlled by a single port authority, which is able to record ship, passenger and cargo movements.
Reporting Port
A statistical port for which statistics of inward and outward maritime transport flows are compiled.
Main port
A main port is a statistical port which has annual movements of no less than 200 000 passengers or recording more than one million tonnes of cargo. For ports selected on the basis of only one of these cargo or passenger criteria, detailed statistics are required only for that transport.
Seagoing vessel
Floating marine structure with one or more surface displacement hulls.
In the context of the Directive, sea-going vessels are vessels other than those which navigate exclusively in inland waters or in waters within, or closely adjacent to, sheltered waters or areas where port regulations apply.
For more information on the concepts and methodology, please consult the Reference Manual on Maritime Transport Statistics on Transport/Publications/Additional Methdological publications section on the website of Eurostat.
3.5. Statistical unit
The data on maritime vessels used in the domain are collected at port level.
3.6. Statistical population
The statistical population is maritime vessel above 100 Gross Tonnage entering EU main ports.
3.7. Reference area
The data are available for the EU aggregates (calculated excluding Czechia, Luxembourg, Hungary, Austria and Slovakia that are not having maritime ports).
3.8. Coverage - Time
Data are available for the period since 2017.
3.9. Base period
Not applicable
Number of vessels
Quarterly data
6.1. Institutional Mandate - legal acts and other agreements
The dissemination is based on the terms of Commission Decision 2001/423/EC on arrangements for publication or dissemination of the statistical data collected pursuant to Council Directive 95/64/EC on statistical returns in respect of carriage of goods and passengers by sea.
6.2. Institutional Mandate - data sharing
Not applicable
7.1. Confidentiality - policy
Regulation (EC) No 223/2009 on European statistics (recital 24 and Article 20(4)) of 11 March 2009 (OJ L 87, p. 164), stipulates the need to establish common principles and guidelines ensuring the confidentiality of data used for the production of European statistics and the access to those confidential data with due account for technical developments and the requirements of users in a democratic society.
7.2. Confidentiality - data treatment
As mentioned in Article 4 of Commission Decision 2001/423/EC, "the highest level of detail in which data may be published or disseminated is the level of port to and from maritime coastal area. The Commission may however publish at more aggregate level if the quality and/or completeness of information are not appropriate in such detail."
8.1. Release calendar
Release calendar is available on Eurostat website
8.2. Release calendar access
The release calendar of Eurostat is available at this website.
8.3. Release policy - user access
In line with the Community legal framework and the European Statistics Code of Practice, Eurostat disseminates the European maritime statistics on Eurostat's website with professional independence in an objective, professional and transparent manner in which all users are treated equitably. The detailed arrangements are governed by the Eurostat protocol on impartial access to Eurostat data for users.
Documentation of the methodology for estimating data on vessels traffic is available as an annex.
10.7. Quality management - documentation
The quality of the estimated data is described in the document available in the annex further below.
11.1. Quality assurance
Quality checks are performed each time data are received from reporting countries. Reports are produced presenting the results of these checks and are sent to the concerned countries. When quality issues are encountered, countries are required to investigate the problems and to send back to Eurostat corrections or new revised datasets as soon as possible.
Meetings of the Working Group on Maritime Transport Statistics (with participants from all reporting countries) are regularly organised in order to discuss improvements in the quality and timeliness of the statistics.
The results of the estimation method are compared with the provided statistics and analysis was made to determine better suiting estimation method. It will be subject of regular re-evaluation and adjustment when needed.
Additional information can be found below in section 18.4.
11.2. Quality management - assessment
The compilation of European maritime transport statistics is based on the legal framework described in section 3.1, which defines the scopes, definitions of variables and classifications to be used by all reporting countries. Additional clarifications and principles for harmonisation of methodology and compilation practices between the reporting countries are described in the Reference Manual on Maritime Transport Statistics (available under ‘Additional methodological publications’). In addition, there are regular discussions on possible improvements in the quality and timeliness of European maritime transport statistics in the meetings of the Working Group on Maritime Transport Statistics.
The latest quality review of maritime transport statistics was performed in 2022-2023. The user satisfaction surveys results for transport statistics show that there is an increased user satisfaction with the timeliness of these statistics. However, it was recommended to continue improving the timeliness of these statistics. The estimates produced by Eurostat in cooperation with EMSA aim at providing early access of the users to such data.
12.1. Relevance - User Needs
Users, and especially policy makers, need these statistics at an earlier time than the current release date (one year later after the end of the reference year) to allow capturing timely changes that may arise in vessel traffic due to punctual events, such as closure of Suez Canal, pandemics, taxation, etc.
User needs are monitored in different ways. Eurostat regularly organise meetings of the Working Group on Maritime Transport Statistics, where main stakeholders are represented (including representatives from other Commission services, the European Maritime Safety Agency, ports, ship-owners). Eurostat is regularly in contact with the Commission services (especially DG MOVE). Eurostat also participates to meetings of statistical nature organised by other relevant organisations (such as UN, ITF, ESPO and EMSA). Contact with users is also regular via the user support service of Eurostat: this activity concerns, in addition to the previous mentioned users, other categories of users such as national administrations, enterprises, researchers, universities, citizens, journalists and so on.
12.2. Relevance - User Satisfaction
Eurostat conducts regular user satisfaction surveys. User requests and comments are regularly received through Eurostat’s user support service.
The estimates presented here are experimental and Eurostat intends to receive users’ comments by disseminating these data.
12.3. Completeness
The EU countries report maritime statistics according to the legal acts.
The estimates of the EU aggregates cover all Member States that report maritime statistics.
13.1. Accuracy - overall
The data are estimates and their overall accuracy is considered to be good at EU level, where results are considered reliable.
The results are based on an estimation method (ARIMAX) combining several data series (Eurostat dataset and two datasets of EMSA). The precision of the estimates of the total number of vessels for the period 2017-2019 is calculated between -6% and 4%, based on the deviation with the data reported to Eurostat (available data in July 2026). The precision of estimates of the total number of vessels for the period of 2020-2024 varies mostly in the range of -5% and 7%, with three quarters having larger margin of differences, such as 14% (2021Q1), 10% (2021Q4) and 16% (2020Q2) (available data in July 2026). The accuracy for the different types of vessels varies, specifically for passenger and cruise vessels.
Analysis based on initial estimates done for 2015-2019 are available in the document in the annex. To be noted that meanwhile improvement in the filtering of MARINFO data was done by excluding double counting of vessels (anchoring).
The data sources are in general considered highly reliable, e.g., due to the administrative obligations (SafeSeaNet) and exhaustive method of observation (AIS records based on vessels transmitters).
13.2. Sampling error
Not applicable
13.3. Non-sampling error
Coverage
The detailed quarterly data collection is only mandatory for main EU and EEA ports, handling more than 1 million tonnes of goods or 200 000 passenger movements annually.
Measurement
The use of “unknown codes” instead of specific codes for type of vessel can be a factor which will reduce the accuracy of the maritime statistics on the more detailed levels. In addition, it reflects in the vessel classification table between Eurostat and EMSA data sources.
Processing
Results are estimates based on a certain method (ARIMAX).
Classification by type of vessel
The correspondence of the type of vessels between Eurostat and EMSA’s classifications may lead to certain misclassification of vessels. There is as well a small share of missing information on the type of vessel (less than 1% of the total number of vessels for each EMSA dataset at EU level), that could have some influence on the accuracy of the estimates.
For more information, please refer to the document in annex.
14.1. Timeliness
Quarterly estimates on maritime vessels are planned to be disseminated within few weeks after the reference quarter.
14.2. Punctuality
In general, the reporting countries are able to respect the legal deadlines. Exceptions are limited to specific or abnormal situations.
The AIS data follow a process of continuous updating.
15.1. Comparability - geographical
Geographical comparability is obtained via the application of common concepts and validation rules. The harmonised interpretation of the methodology is enhanced by the organisation of regular meetings of the Working Group on Maritime Transport Statistics, where all reporting countries are represented.
15.2. Comparability - over time
There are no major changes in the methodology used to collect and process the data in this domain since the first period of estimated data in 2017.
15.3. Coherence - cross domain
The estimated data on number of maritime vessels entering the main EU ports are coherent with data in maritime domain (mar) at the EU level as the growth rate trends of EMSA data sources follow the trends of the data reported to Eurostat. More information is available in the document in the annex.
15.4. Coherence - internal
The intra-dataset quality checks are performed on statistics provided to Eurostat. Estimates are considered of good internal coherence. Some possible explanations of discrepancies might be aspects related to the use of multiple data sources, for example, need of further improvement of vessel classification table between Eurostat and EMSA data sources. More information in the document in annex.
There are no additional costs for the countries and no additional burden on respondents / data providers.
17.1. Data revision - policy
To further specify the general Eurostat revision policy, the revision policy has been established for maritime transport statistics. Revisions of data are taken into account and processed when provided by the national competent authorities.
17.2. Data revision - practice
The revision practice effectively corresponds to the revision practice of the maritime transport domain (mar) and as listed under sub‑concept 17.1 (data revision – policy).
All reported errors (once validated) result in corrections of the estimated data.
Revisions may be made by data providers as more complete information becomes available or as a result of quality checks.
These quarterly estimated data are in general provisional.
18.1. Source data
Data on maritime vessels entering the main EU ports are collected by the national competent authorities in the reporting countries using a variety of data sources, such as port administration systems, national maritime databases, or shipping agents. These data are provided to Eurostat as dataset F2 of the relevant legal act Directive 2009/42/EC.
EMSA provide data from two sources:
SafeSeaNet (SSN) is a vessel traffic monitoring and information system. It was set up as a network for maritime data exchange, linking together maritime authorities across Europe. It enables European Union Member States, Norway, and Iceland, to provide and receive information on vessels, vessel movements, and hazardous cargoes.
MARINFO dataset is built on Automatic Identification System (AIS) data from several commercial sources. The port calls are identified using algorithms and AIS data.
18.2. Frequency of data collection
The estimated data on vessel traffic are quarterly available.
SafeSeaNet data cover notifications provided by each EU and EFTA Member States and exchanged via the SafeSeaNet system. These exchanges are based on Directive 2002/59/EC as amended. The data are available in number of vessels as well as by ship class.
The vessels included in SafeSeaNet are ships above 300 GT, with IMO or MMSI number assigned, engaged in seagoing navigation (excluding inland waterways vessels).
MARINFO dataset is built on Automatic Identification System (AIS) data from several commercial sources. The port calls are identified using algorithms and AIS data. The vessels included are those above 100 GT, having an IMO number assigned, including ships engaged in inland navigation and ports activities. The data are collected and processed by the commercial data provider. The data provider reviews all vessels every day and does not carry out selective updating, thus following a process of continuous updating.
18.4. Data validation
Quality checks are performed by Eurostat each time data on maritime transport statistics are received from reporting countries. Reports are produced presenting the results of these checks and are sent to the concerned countries. When quality issues are encountered, countries are required to investigate the problems and to send back to Eurostat corrections or new revised datasets as soon as possible.
Quality checks performed on the SafeSeaNet (SSN) data of EMSA: microdata is collected and processed at national level according to an agreed data quality criterion. In addition, upon submission, EMSA performs independent data quality validation checks to reduce missing reports from Member States. The average number of missing reporting is below 1 %.
In addition, before running the estimation method, preliminary coherence data checks are performed on data series of the SSN and MARINFO to prevent any major quality issue.
18.5. Data compilation
ARIMAX is a statistical method used for forecasting time-series data. It predicts results, such as the number of vessels, based on past data and additional influencing factors. For these estimates, the ARIMAX model uses historical data from the Eurostat F2 dataset and EMSA data sources (as exogenous variables). The data were grouped by vessel type and period at EU level to create time-series for prediction. The weights assigned to the exogenous variables (vessels of SafeSeaNet and MARINFO) are determined for each type of vessel to ensure that the specific impacts of external factors on forecast outcomes are accurately recognized. For each vessel category, distinct models are developed, reflecting the unique characteristics and trends of each type in the forecasted results. The models are systematically updated with each new set of data to ensure the accuracy and relevance of the forecasts.
The use of the latest EMSA data in this method in addition to Eurostat previous quarters statistics, give assurance on capturing changes in vessel traffic (port calls) timely that may arise due to certain economic events.
This table contains experimental data of quarterly estimates on vessels arriving in the EU main ports, based on data sources provided by the European Maritime Safety Agency (EMSA).
The data are presented for the EU aggregate.
Vessels arriving in the EU main ports have been estimated by Eurostat by using an autoregressive integrated moving average model with exogenous variables (ARIMAX). The model is based on Eurostat’s quarterly data of previous quarters and EMSA data sources (automatic identification system (AIS) and other administrative data).
The data are presented by vessel type according to the annex VI Type of ship classification (ICST-COM) of the Directive 2009/42/EC on statistical returns in respect of carriage of goods and passengers by sea.
10 August 2026
The concepts used in this domain are the following:
Port
A place having facilities for merchant ships to moor and to load and/or unload cargo or to disembark and/or embark passengers to or from vessels, usually directly to a pier.
Statistical Port
A statistical port consists of one or more ports, normally controlled by a single port authority, which is able to record ship, passenger and cargo movements.
Reporting Port
A statistical port for which statistics of inward and outward maritime transport flows are compiled.
Main port
A main port is a statistical port which has annual movements of no less than 200 000 passengers or recording more than one million tonnes of cargo. For ports selected on the basis of only one of these cargo or passenger criteria, detailed statistics are required only for that transport.
Seagoing vessel
Floating marine structure with one or more surface displacement hulls.
In the context of the Directive, sea-going vessels are vessels other than those which navigate exclusively in inland waters or in waters within, or closely adjacent to, sheltered waters or areas where port regulations apply.
For more information on the concepts and methodology, please consult the Reference Manual on Maritime Transport Statistics on Transport/Publications/Additional Methdological publications section on the website of Eurostat.
The data on maritime vessels used in the domain are collected at port level.
The statistical population is maritime vessel above 100 Gross Tonnage entering EU main ports.
The data are available for the EU aggregates (calculated excluding Czechia, Luxembourg, Hungary, Austria and Slovakia that are not having maritime ports).
Quarterly data
The data are estimates and their overall accuracy is considered to be good at EU level, where results are considered reliable.
The results are based on an estimation method (ARIMAX) combining several data series (Eurostat dataset and two datasets of EMSA). The precision of the estimates of the total number of vessels for the period 2017-2019 is calculated between -6% and 4%, based on the deviation with the data reported to Eurostat (available data in July 2026). The precision of estimates of the total number of vessels for the period of 2020-2024 varies mostly in the range of -5% and 7%, with three quarters having larger margin of differences, such as 14% (2021Q1), 10% (2021Q4) and 16% (2020Q2) (available data in July 2026). The accuracy for the different types of vessels varies, specifically for passenger and cruise vessels.
Analysis based on initial estimates done for 2015-2019 are available in the document in the annex. To be noted that meanwhile improvement in the filtering of MARINFO data was done by excluding double counting of vessels (anchoring).
The data sources are in general considered highly reliable, e.g., due to the administrative obligations (SafeSeaNet) and exhaustive method of observation (AIS records based on vessels transmitters).
Number of vessels
ARIMAX is a statistical method used for forecasting time-series data. It predicts results, such as the number of vessels, based on past data and additional influencing factors. For these estimates, the ARIMAX model uses historical data from the Eurostat F2 dataset and EMSA data sources (as exogenous variables). The data were grouped by vessel type and period at EU level to create time-series for prediction. The weights assigned to the exogenous variables (vessels of SafeSeaNet and MARINFO) are determined for each type of vessel to ensure that the specific impacts of external factors on forecast outcomes are accurately recognized. For each vessel category, distinct models are developed, reflecting the unique characteristics and trends of each type in the forecasted results. The models are systematically updated with each new set of data to ensure the accuracy and relevance of the forecasts.
The use of the latest EMSA data in this method in addition to Eurostat previous quarters statistics, give assurance on capturing changes in vessel traffic (port calls) timely that may arise due to certain economic events.
Data on maritime vessels entering the main EU ports are collected by the national competent authorities in the reporting countries using a variety of data sources, such as port administration systems, national maritime databases, or shipping agents. These data are provided to Eurostat as dataset F2 of the relevant legal act Directive 2009/42/EC.
EMSA provide data from two sources:
SafeSeaNet (SSN) is a vessel traffic monitoring and information system. It was set up as a network for maritime data exchange, linking together maritime authorities across Europe. It enables European Union Member States, Norway, and Iceland, to provide and receive information on vessels, vessel movements, and hazardous cargoes.
MARINFO dataset is built on Automatic Identification System (AIS) data from several commercial sources. The port calls are identified using algorithms and AIS data.
Quarterly
Quarterly estimates on maritime vessels are planned to be disseminated within few weeks after the reference quarter.
Geographical comparability is obtained via the application of common concepts and validation rules. The harmonised interpretation of the methodology is enhanced by the organisation of regular meetings of the Working Group on Maritime Transport Statistics, where all reporting countries are represented.
There are no major changes in the methodology used to collect and process the data in this domain since the first period of estimated data in 2017.
