Eurostat, the statistical office of the European Union

Unit E1: Agriculture and fisheries

The following variables are recorded with regard to aquaculture production:         

'Species' means the species of aquatic organisms identified using the international 3-alpha code as defined by the FAO (ASFIS List of Species for Fishery Statistics Purposes). Individual species are grouped in aggregates according to their taxonomy and living habits. These aggregates are specified in the International Standard Statistical Classification of Aquatic Animals and Plants (ISSCAAP) and indicated in the ASFIS list.

'FAO major areas' means the geographical areas as defined by the FAO (CWP Handbook of fishery statistical standards). The FAO major areas covered are: 27 'Northeast Atlantic', 37 'Mediterranean and Black Sea', 34 'Atlantic Eastern Central', 5 'European inland waters', 1 'African inland waters'.     

'Cultivation method' includes ponds, cages, tanks and raceways, enclosures and pens, recirculation systems, others and not specified. For molluscs also on-bottom and off-bottom systems.           

'Aquatic environment' distinguishes the water types fresh water and salt (sea and brackish) water.           

The methods and water types are defined in Annex I of Regulation (EC) No 762/2008.

Commercial aquaculture production (NACE class A3.2) destined to human consumption (except all uses for aquatic plants).

According to Regulation (EC) 762/2008, aquaculture production means the output from aquaculture at first sale. For animals, only production for human consumption is considered, while ornamental or functional species as well as feed production is excluded. By contrast, aquatic plants will be recorded notwithstanding their final use.

The production volume is expressed in tonnes live weight [TLW] of the product. This weight includes all shells and bones. Data for the economic value of the production are reported as unit price in national currency per tonne [NAC_T]. The production of hatcheries and nurseries is reported in numbers and expressed in millions. Data on the structure of the aquaculture sector are expressed in thousand cubic metres, hectares or, optionally, 'metres rope length' according to the method.

Aquaculture production facility.

Ireland and Italy added the following specifications:

-        Ireland “Aquaculture production facility. This is either a landbased facility or location within a coastal bay where a business applies a specific aquaculture practice upon a stock of a specific species”.

-        Italy: “For the territorial census the reference is the enterprise, a legally recognizable figure, with a company name, regularly registered with the Chamber of Commerce, characterized by a productive system composed of human, technical and commercial resources mainly used, in the case considered, in aquaculture activities.

The company carries out its activity in one or more plants, the focal point of the survey, within which are located the breeding structures, equipped with their own logistical and territorial individuality, in turn possibly made up of different infrastructures in based on which they can be divided into different production units. An enterprise can therefore have one or more plants.

The technical and production data are usually referred to individual plants, as these can have very different characteristics from each other”.

All commercially active production facilities of aquatic animals or plants.

Many countries report data on aquaculture from a period earlier to the entrance into force of Regulation 762/2008. Figure 1 illustrates the time coverage of national aquaculture statistics[1]:

not applicable

The aquaculture statistics are based several types of data sources. The most common ones are censuses (59%) and administrative data (25%). Expert estimates are the third most common data source type (12%) followed by sample surveys (4%) (Figure 2).

The countries reported that they use in total 51 different data sources to fulfil the reporting obligations set by Regulation (EC) No 762/2008. The situation by country is illustrated in Figure 3.

The number of sources per country varied from one source (e.g. CZ, FI) to six sources (UK), the average being 2 sources per country. The figures are not always necessary fully comparable as it is not always easy to share the same perception of what a data source is (e.g. for expert estimates). A detailed list of data sources is in Annex 1[1].

3.1.1.     Census

Censuses are the most numerous data source for aquaculture statistics (30). In total 26 countries carry out censuses for aquaculture statistics.

The most common mean of collection of census data are the postal (34%) and electronic questionnaires (31%) followed by telephone interviews (17%). Face-to-face interviews (10%) and other options (8%) are used to a lesser extent (Figure 4)[1].

Table 3 specifies the names of censuses per country as well as the population sizes and the surveyed entities (which are essentially either production facilities or enterprises).

3.1.2 Sample Survey

Sample surveys are carried out only in Germany and Spain. The postal questionnaires are the most common mean of data collection for sample survey data (40%) (Figure 5).

Table 4 summarises the sample surveys conducted in the two countries (surveyed entity, sample size and sampling basis).

Table 4 Name, sample size and sampling basis of sample surveys

3.1.3.     Administrative sources

Overall, 12 administrative sources are utilised in 11 countries for aquaculture statistics. Most of these sources are updated on annual basis (60%), 30% continuously and 10% monthly (Figure 6).

The names of the administrative data sources are listed in Table 5. 

Table 5 Name of the administrative data sources by country

The countries have estimated the proximity between the definitions in the administrative register and the ones in the EU regulation. It is judged “very good” by 30% of countries, “good” by 40%, “medium” by 10% (information not available in 20% of countries).

 The main differences between the administrative source and the statistical definitions are the following:  

-        Belgium: “Veterinary units can differ from production facilities. Activities are more extended than the production for human consumption.

No production data are available, only a summary of the identification and registration of the enterprises is available."

-        Slovenia: “In addition to data on aquaculture production by commercial entities our register contains the data on aquaculture production for own consumption, too”.

3.1.4.     Expert estimates

Expert estimates are used for aquaculture statistics in six countries (Table 6). Typically, they are used for complementing other data sources (surveys and/or administrative data). The experts are often working in governmental fisheries/statistical organisations both at national and regional levels, in research organisations and producer organisations.

Table 6 Names of expert data sources by country

All countries stated that they collect data annually as requested by the Regulation (except for AQ5: structure of aquaculture sector: every three years).

List of reported data sources

Quality management is composed of quality assurance and quality assessment of data.

not applicable

The analysis of the accuracy takes into account the monitoring of the overall accuracy together with the analysis of the sampling and non-sampling errors. The latter is furthermore broken down into coverage error, measurement error and non-response error. The following paragraphs illustrate the results provided by the countries; nevertheless, more details on the accuracy and reliability of the aquaculture data can be found in Section 6 of the national quality reports.

Germany applied the relative standard error; Spain indicated that an “other” method was used.

not applicable

not applicable

Census

The over-coverage is very rare in aquaculture censuses. Belgium, Bulgaria, Germany, Greece, Spain, Cyprus, Latvia, Lithuania, Malta, the Netherlands, Slovakia and Finland declared that the impact is “none”. Estonia, Austria, the United Kingdom and Norway declared that the impact is "low”. The rest of the countries did not reply this question.

The following further information was provided:

-        Austria: for calendar year 2017 c.a. 40%;

-        Hungary: 0%

-        The United Kingdom: ZERO: Scotland: Annual Fish Farm Production Survey & Annual Shellfish Farm Production Survey.

Inclusion of some non-commercial (environmental research) enterprises: England & Wales: Site visits of authorised Aquaculture Production Businesses + Northern Ireland: Annual Aquaculture Production Business Survey. Applies to AQ4 only, and effect minor (estimated at <5%).

Sample survey

Administrative data

No recrd of over-coverage rate 

not applicable

not applicable

6ensus

Sample survey

not applicable

not applicable

Census

Sample survey

not applicable

not applicable

not applicable

Countries could report if they had conceptual changes since the last quality report necessitating data revisions. Only Romania indicated that conceptual changes occurred, without though explaining the main reasons for the change. On the contrary, Cyprus, Slovakia and Iceland provided some explanations for the other reasons why the data have been revised.

Table 14 Data revision – conceptual changes occurred since the last quality report

More information on the accuracy is available in section 6 of the quality reports.

0

Regulation (EC) 762/2008 sets up the data transmission calendar for aquaculture statistics

not applicable

not applicable

The punctuality of the data transmissions is assessed by Eurostat on the basis of the date of the received transmissions in EDAMIS. Table 16 shows the availability and punctuality of aquaculture datasets for reference year 2017[1]. The table also reports the status of the delivery of the quality report.

Based on the analysis of the Punctuality it can be pointed out that:

·     Bulgaria: 2017 data for AQ2a and AQ4 were transmitted with 18 days delay; AQ5 data were transmitted with almost 6 months delay (179days).

·     Spain: 2017 data not delivered for AQ3, AQ4 and AQ5. The ones for AQ2A and AQ2B were transmitted 1 month after deadline.

.    France: all data 2017 were transmitted but datasets AQ3, AQ4 and AQ5 were transmitted with approximately 1 month delay (31, 32 and 51 days respectively).

·     Italy: 2017 aquaculture data were transmitted one month after the deadline.

·      Romania: 2017 data for AQ5 were not transmitted; datasets AQ2A and AQ4 were transmitted 2 weeks and 1 month after deadline respectively.

·     United Kingdom: 2017 data were transmitted with approximately 2 months delay

Several quality reports required thorough review, while Poland and Romania did not transmit the final revised version of their quality reports (AQ6).

The table reports the delay in days from the expected deadline. When the cell is left blank it means that there was no delay recorded (with the exception of AQ6 Quality report where, if the cell is left white, it means that the national quality report has been validated and sent for publication.

not applicablesee section 7.2

Coherence and comparability are assessed in two different ways: internal comparability as length of the time series and coherence against other data sets.

The analysis of the comparability over time is assessed by the length of comparable time series for aquaculture statistics (Table 17). Ireland explained in detail the reason for the breaks in the series occurred.

After a cross-check analysis between the timing indicated in section 8.2.1 “length of comparable time series” and section 2.8 “coverage time of the NQR” it can be pointed out that some countries indicated inconsistent dates. For instance in EL,  FI, UK and NO, the starting date of comparable time series is earlier than the indicated starting date of aquaculture statistics (Figure 16)[1].

not applicable

see section below

not applicable

not applicable

not applicable

Not available.

not applicable

not applicable

not applicable

not applicable