FISHERIES DATA COLLECTION FRAMEWORK

FISHERIES DATA COLLECTION FRAMEWORK The DCF Reporting and Implementation Cycles and the Data End-user Feedback Cristina Castro Ribeiro 2015 Report EUR xxxxx xx

European Commission Joint Research Centre Institute for the Protection and Security of the Citizen (IPSC) Contact information Castro Ribeiro Cristina Address: European Commission, Joint Research Centre (JRC), Institute of the Protection and Security of the Citizen (IPSC), Maritime Aairs Unit G03, Via Enrico Fermi 2749, 21027 Ispra (VA), Italy) E-mail: cristina.ribeiro@jrc.ec.europa.eu Tel.: +39 0332 78 9329 JRC Science Hub https://ec.europa.eu/jrc Legal Notice This publication is a Technical Report by the Joint Research Centre, the European Commission s in-house science service. It aims to provide evidence-based scientific support to the European policy-making process. The scientific output expressed does not imply a policy position of the European Commission. Neither the European Commission nor any person acting on behalf of the Commission is responsible for the use which might be made of this publication. JRC97428 EUR xxxxx xx ISBN xxx-xx-xx-xxxxx-x ISSN xxxx-xxxx (online) doi:xx.xxxx/xxxxx Luxembourg: Publications Office of the European Union, 20xx European Union, 2015 2

CONTENTS Summary... 4 Rational:... 4 The END-USER s Feedback... 5 The Feedback from the Data End-users in 2015... 6 Suggestions to streamline the end-user feedback for the future... 16 What content should the end-user feedback have and where to address these contents in the IT PLATFORM?... 16 THE DATA Transmission It Platform... 17 References... 18 3

SUMMARY Every year the Member States Data Collection (DCF) Annual Reports and Data Transmission to the data end-user are evaluated by the STECF regarding the (a) the execution of the national programs approved by the Commission; and (b) the quality of the data collected by the Member States. For that, aside of the Annual reports on the activities developed by each MS, also the data end-user should provide feedback on the data received, and in specific identify the problems these data might have. This is a process that imports to be seen and used as positive input to streamline the collection and transmission of data and to foster incrementally higher quality standards to the EU MS data. For such to happen, it is also of importance that the feedback received from the end-user to be of enough detail to allow a proper assessment and the identification of the linkage between the data gap and the reason for such gap on the source. In this report the conclusion from STECF EWG 1510 on the ways to improve the annual end-user feedback are further explored and detailed with the aim of supporting future exercises and in specific to aid on the dialogue between the Commission and the DCF data end-users on this regard. RATIONAL: The Community framework for the collection, management and use of data in the fisheries sector, Reg. (CE) 199/2008 (DCF), with the detailed rules of application laid down by the Commission Regulation (EC) No 665/2008 of 14 July 2008 defines that yearly the Member States should submit by electronic means an annual report of the implementation of the national program of the year before, to present the annual execution of the program and specifying the outcomes of the planned actions. Alongside with this submission, also the transmission of the DCF data done by the MS to the data end-users, taken place in the year before, should be yearly reported. The report on the data transmission is mutually done by the MS and by the data end-users. The MS report is part of the MS Annual Reports (AR). Equally, the end-user should provide annually the feedback on the quality of the data they have received and in specific any issue that might have arose from these data Data transmission Issues (DT). The assessment of these two elements (AR and DT) on regard to its scientific validity should be done by the Scientific, Technical and Economic Committee for Fisheries (STECF) in accordance with the provision of the Reg(CE) N. 199/2008, article 7. Therefore, every year, the STECF dedicates one of its Expert Working Group (EWG) to access the MS Annual reports and the data transmission issues. The last STECF Expert Working Group (EWG) took place in Gdynia, Poland, in June 2015. This was the STECF EWG 1510 on the Evaluation of MS AR and Data Transmission Issues. Apart from the regular assessment of the AR and DT, this EWG was also requested to carry out an analysis of the Annual Report and data transmission exercises in view of identifying feedback to be provided to end-users in order to improve the way in which they provide data transmission feedback to the Commission in future (Tor 3.1). The findings by the EWG 1510 on this regard have been submitted to the STECF plenary and conclusions were drawn. Regarding the data transmission in 2014 and the format/content of the feedback received from the end-users, the STECF has concluded as follow: STECF concludes that the online platform for DT issues should continue to be used and improved by the EWG suggestions (section 7.1.2). Considering the various problems with the evaluation of DT issues identified by the EWG, STECF urges the Commission to review and amend the formats and procedures used for the end-user feedback on DT in dialogue with the end-users, taking the suggestions compiled by the EWG (section 6.1 and Annex 6) into account. For the future Commission foresees to change its approach of monitoring Member States' implementation by giving priority to preventing cases of non-compliance and finding early remediation to failures. This is foreseen to be 4

achieved through giving more attention to systemic issues leading to failures to submit data, and through closer cooperation with end-users regarding their feedback on Member States' delivery of data. (COM(2015) 294 final) The present report builds on the findings and conclusion from the STECF and adds to these the technical details to aid on the dialogue between the Commission and the end-users in support of the foreseen improvements. THE END-USER S FEEDBACK Reg. 199/2008 defines end-users as bodies with a research or management interest in the scientific analysis of data in the fisheries sector. In 2015 the EWG 1510 was requested to assess the feedback for the transmission of scientific data from the Member States (MS) to the following end-users: DG MARE Directorate-General for Maritime Affairs and Fisheries GFCM General Fisheries Commission for the Mediterranean (GFCM) IATTC Inter-American Tropical Tuna Commission ICCAT International Commission for the Conservation of Atlantic Tunas ICES International Council for the Exploitation of the Sea IOTC Indian Ocean Tuna Commission JRC Joint Research Center RCM Regional Coordination Meetings WCFCP Western and Central Pacific Fisheries Commission But how this feedback does enter in the DCF Process? The image below depicted how the implementation of the DCF takes stock of the end-user feedback and how the implementation and reporting cycle are interconnected. 5

FIGURE 1 - THE ANNUAL IMPLEMENTATION AND REPORTING CYCLES OF DCF. THE INPUT /OUTPUT FLOW BETWEEN CYCLES. One of the main points from the end-user feedback is that it provides insight and highlight issues that can be used in the near future to streamline and improve the MS activities in support of the production their data. However, for such to be possible, a clear identification between the final data products, those data the end-user receives, and the activities to produce them as explained in the national programs and Annual Reports, need to be possible. This requires the feedback about these data to be thorough and objectively informed. THE FEEDBACK FROM THE END-USERS IN 2015 The number of issues the EWG was requested to assess in 2015 is of 813 raised by 9 different end-users as shown in table below. In 2014, STECF EWG 1407 was addressed with 429. Even though both STECF EWGs accomplished the task of assessing the data transmission issues they have been addressed with, some of those issues were loosely evaluated and impossible to comment due to lack of detail. Therefore no actions were recommended or possible situation were remedy since then due to the lack of understanding between the parts involved. 6

TABLE 1 - SUMMARY TABLE OF THE DATA TRANSMISSION ISSUES ADDRESSED TO THE STECF EWG1510 FOR ASSESSMENT. (IN STECF1513) End-user Data Call Nr.data Issues DG MARE Seabass/Effort 3 GFCM Task 1.2, 1.3, 1.4 and 1.5 38 IATTC 1 ICCAT T1FC: Fleet Characteristics, T1NC: Nominal Catches, T2CE: Catch and Effort and T2SW/CAS: Catch-at size 32 ICES Expert groups: AFWG; HAWG, WGBIE, WGBFAS, WGCEPH, WGCSE, WGDEEP, WGHANSA, WGMIXFISH-ADVICE, WGNEW, WGNSSK, WGWIDE 512 IOTC Coastal fisheries, Long-line fisheries and size data 3 JRC Aquaculture, Effort, Fleet economics, Mediterranean and Black Sea and Processing Industry 159 RCM Baltic, North Atlantic and North Sea and Eastern-arctic 58 WCPFC Total 7 Total 813 Important to this analysis is also the characterization of how the process of gathering this feedback from the endusers has happened in 2015 and the instances that are insofar involved. For the 2015 exercise for the first time an IT platform has been developed and used in support of this process. Main goal of this new tool is to facilitate exchange of information on data transmission issues between MS, End Users, STECF, MARE; ensure transparency on the process and permit a proper storage of the information flow along the years. The IT platform was developed by the Joint Research Center, JRC IPSC, under it Administrative Arrangement with DG MARE. This IT solution uses functionalities of the same Liferay portal used for the Data Collection and STECF websites and is accessible from the following address: http://datacollection.jrc.ec.europa.eu/compliance. The access to the platform is provided with the data collection websites credentials (user name in lower caps). For the MS these are the same as the credentials used for other interactions with EU/JRC-DCF web tools. For the data End-users specific log in has been created for this specific purpose. Figure 2 presents two images of the It platform, one in consulting mode and the other in editing mode. The management of the accesses to the platforms is done with different profiles containing specific rights & roles. These are: MS see their country and edit MS comments, DG MARE sees all, edits DG MARE assessment/decision, STECF EWG sees all, edits STECF comments/assessment, End users see the issues they raised; edit End user comments. 7

FIGURE 2 - VIEW OF THE IT PLATFORM. IMAGE ON THE LEFT SHOWS THE LOOK AND FEEL OF THE PLATFORM ON THE CONSULTATION MODE, WHEREAS FIGURE ON THE RIGHT SHOWS THE EDITING MODE. Even though the platform was implemented early 2015, not all end-users have made use of it while providing their inputs to the Commission. The feedback from most end-users, except JRC and DG MARE, were received in different formats and then converted and inserted in the IT platform either by DG Mare or by JRC staff. CONCLUSION FROM STECF 1513 (STECF EWG1510) The STECF EWG 1510 was requested to assess the Data transmission issues regarding the year 2014 and feedback given by the end-users, as well as comment on this feedback, with the view to improve the way in which end-users provide data transmission feedback to the Commission in future. From this analysis the concluding comments from the EWG are: 1) Timeliness: End-users should automatically create a date stamp when MS responded to the data call, to prevent discrepancies between submission date reported by MS and date stamp reported by end-user. 2) End-user feedback to MS: end-users should specifically appoint the failure to the MSs referred to in the comment, and send the comment solely to the relevant MSs, to prevent unnecessary burden of responses for other MSs. 3) Formulation of end-user feedback: The data call originator (e.g. end user/working group and/or stock coordinator) should be involved in the creation and evaluation of data transmission, to specify on what data was requested but not provided, the impact this had on the assessment and what action would be required to remedy the gap. It is the responsibility of the chair of each Working Group or the end user to ensure that accurate information on data gaps are clearly highlighted in the working group report. 8

4) Data call in relation to data collection and international agreements: End-users should be aware that wish lists for data not covered by MSs data collection under the DCF neither covered by international agreements the European Commission is part (e.g. requirements of more stringent sampling plans, sampling of the slippage fish, species not covered neither by DCF or RFMOs, species requested on a lower taxonomic level that requested under DCF, etc ) are not data transmission failures and that requirements of modification of the DCF should be discussed and addressed to the Commission and RCMs. (note: This can be sought as part of the end-user input for a future end-user driven DCF as it is foreseen the recast DCF will be.) 5) Data omissions: If a MS has informed the end-user that due to issues beyond their control they are unable to collect certain data, and in spite of this communication, the end-user continues to request the data, then only in the first year this can be announced as a data transmission failure, and should not be repeated in following years. Data should not be further requested from the MS in the ensuing years. 6) Persistent non-compliance: Repeating year after year non-compliance on the same issue by a given MS needs to be addressed. Some kind of procedure should be implemented in order to prevent persistent noncompliance. On the other hand, it is encouraging seeing if a given MS improves from year to year, actually putting an effort in complying with the guidelines. There should be a way for the STECF/COM to have a way to check next year if issue for not collecting data has being remedied by MS. With the aim to prepare a more in-depth analysis of end-user feedback in 2015, to best support improvement of feedback in the future, the first five points addressed in STECF EWG 1510 were translated into five questions; answers to each of these questions per end-user were prepared. Since point number six goes beyond the scope of this exercise this was not translated into a specific question and therefore the individual end-user feedback will not be assessed for this specific point. This is however a point related with the data transmission platform that in some extent is tackled if the IT platform is well used and the data issues well informed. The five questions are: 1. Timestamp for the data reception available? 2. MS concerned identified? 3. Data & variables, type of issue and issue severity, identified? 4. Issues go beyond DCF requirements? 5. Issues repeatedly arose? The answers to these five questions and for each end-user are compiled in table 2. 9

TABLE 2- ANALYTICAL PRESENTATION OF THE DATA END-USER FEEDBACK ON THE MEMBER STATES DATA SUBMISSION DURING 2015 REGARDING THE 5 MAIN POINTS IDENTIFIED BY THE STECF EWG 1510. Enduser DG MARE 1. Timestamp for the data reception available? No 2. MS concerned identified? Yes GFCM Yes Yes 3. Data & variables, type of issue and issue severity, identified? Data & variables Type of issue Data groups were identified but not the variables. The issues identified were on catch and effort data but without identification the variables requested, if days at sea, fishing days, volume/value of landings, etc. The type of issues was identified according to the following categories and as foreseen in the database: quality/timing/coverage. Severity The impact the issues had on the meeting/working group/final output were identified according to the categories in the database. Data & variables Type of issue Data groups were identified but not the variables or even the species and gears for which the data are missing. Example: 44% complete - no data on total effort, catch or landing value, discard value, bycatch value of fishing periods, no data on catch/landing value and CPUE/LPUE value of main associated species The type of issues was identified according to the follow categories and as foreseen in the database: quality/timing/coverage. 4. Issues go beyond DCF requirements? Not Applicable GFCM requests different levels of aggregation compared to DCF. This has been putted forward from MS as a justification for the missing data. 5. Issues repeatedly rose? Not Applicable. Ad-hoc data call. Not applicable. The yearly data to be provided in each calendar year is set to be for the reference year n-2. 10

Enduser 1. Timestamp for the data reception available? 2. MS concerned identified? IATTC No Yes ICCAT No Partly; Since the correctness of this identification has been challenged (Denmark and Netherlands were identified as missing data countries, however it seems they have not activity in these kind of fisheries). 3. Data & variables, type of issue and issue severity, identified? Severity The impact the issues had on the meeting/working group/final output were identified according to the categories in the database. Data & variables Type of issue Neither the data groups nor variables missing were identified. The only data issue raised states no data for longliners. The type of issues was identified according to the follow categories and as foreseen in the database: quality/timing/coverage. Severity The impact the issues had on the meeting/working group/final output were identified according to the categories in the database. Data & variables Type of issue Most of data and variables identified, however for some issues the species missing data are not clearly identified) e.g. large tuna species. Several issues related with possible lack of quality very vaguely described, eg. questionable data quality for all fleets and gear groups Part of the issues vaguely described. Issue regarding the possible lack of data quality/coverage need to be further informed. Not enough the statement questionable data quality. 4. Issues go beyond DCF requirements? No No 5. Issues repeatedly rose? The yearly data to be provided in each calendar year is set to be for the reference year n-1. Therefore this shouldn t be applicable. However it s not clear whether the issues raised do respect only the transmissions in 2014 of 2013 data. The yearly data to be provided in each calendar year is set to be for the reference year n-1. Therefore this shouldn t be applicable. However it s not clear whether the issues raised do respect only the provision of data in 2014 of 2013 data. 11

Enduser ICES 1. Timestamp for the data reception available? No 2. MS concerned identified? Partly; Since in some circumstances the issue doesn t fit at all the MS addressed. e.g comments from WGWIDE addressed repeatedly to several no relevant MS. IOTC No Yes 3. Data & variables, type of issue and issue severity, identified? Severity No information on the impact the issues had on working groups/meeting or even the data were afterwards used. Data & variables Type of issue Part of the issues addressed with complete information. However some issues related to quality lack detail. Examples: anb-78ab: Accurate estimates of growth parameters are needed to facilitate the development of an analytical assessment. The types of issues were identified according to the following categories and as foreseen in the database: quality/timing/coverage. Issue regarding the possible lack of data quality/coverage need to be further informed. Not enough the statement that quality is insufficient if any suggestion or benchmark is proposed is not satisfactory. Severity The impact (severity) of each issue in the working-group/meeting work/final outcome identified according to the scale in the IT platform. (low/medium-lowmedium) Data & variables Data was identified; however the variables missing in specific were not always identified. 4. Issues go beyond DCF requirements? Some comments are outside the DCF scope, even suggesting the enlargement of the DCF with complementary data collection (surveys, new species not included in Decision EU/2010/93 or concerning recreational fisheries, etc.). These are important issues but instead they should be addressed to the proper fora (either Commission or RCMs/RCGs) and not separately to MS. No 5. Issues repeatedly rose? Not clear if the issues are only about the provision of 2013 data or if more years are included in the analysis. Not clear if the issues are only about the provision of 2013 data or if more years are 12

Enduser 1. Timestamp for the data reception available? 2. MS concerned identified? JRC Yes Yes 3. Data & variables, type of issue and issue severity, identified? Type of issue The type of issues was identified according to the follow categories and as foreseen in the database: quality/timing/coverage. Severity No information on the impact the issues had on working groups/meeting or even the data were afterwards used. Data & variables - The identification of the data type and variable are mostly identified; however still lack of detail in the reference years/areas or gears and fleet segments prevent a full comprehension. Example: National level capacity data provided for the active fleet only; data on inactive vessels missing, inconsistent with fleet segment level data. Impact: Suggests incomplete coverage of the DCF data - Several issues regarding quality are not sufficiently informed to allow a proper assessment. Example: Effort data and Catch data (Landings, Discards) are inconsistent: large effort values in some years-areas-gears are accompanied by very low or no catches at all. Type of Several issues were not classified on 4. Issues go beyond DCF requirements? No 5. Issues repeatedly rose? included in the analysis. Yes. Several issues had been already identified in the feedback from previous years. These issues have been addressed to the MS for reaction, in the past. This repetition should be avoided. 13

Enduser 1. Timestamp for the data reception available? 2. MS concerned identified? RCM No Yes 3. Data & variables, type of issue and issue severity, identified? issue their nature or were misclassified. Timeliness issues mismatched as quality issues. Example: Considerable amount of data submitted on 08/09/2014, well after the deadline- QUALITY Severity The severity of each issue in the final outcome identified according to the scale (low/medium-lowmedium). Additionally to the severity of the issue, the specific impact each data issue had in the meeting/work was identified. Data & variables Type of issue Data very generically identified. It lacks the specification of the variables/species/fleets. Most issues classified as coverage reflect technical issues that might arise due to differences in the way data is extracted. It seems several technical issues are not actually a result from MS data delivering but instead issues with the regional database. Example: the number of species in age samples in the RDB differed between before and after the extraction of sample records with no information from the RDB (England) Severity The impact these data gaps might have in the RCM output haven t 4. Issues go beyond DCF requirements? Several issues related with MS willingness to submit data to Regional databases due to data confidentiality issues. Also several issues that derive from technical constraints in the database that might not be directly linked to an issue at the source. 5. Issues repeatedly rose? 14

Enduser 1. Timestamp for the data reception available? 2. MS concerned identified? WCFCP No Yes 3. Data & variables, type of issue and issue severity, identified? been identified. Data & Data gaps not always sufficiently variables identified, data groups and variables missing; Example: no data for longliners (minimum of 5% is required) ; no size data provided by any fleet segment (operational data/aggregated data?) Type of Issues classified according to its issue nature (coverage, quality and timeliness). Though only coverage issues have been raised. Severity The impact each issue might have had on the end-user work has not been identified neither translated in terms of its severity. 4. Issues go beyond DCF requirements? The issues identified might go beyond the DCF scope. Example: data on branchlines between floats. This variable is not yet listed as a DCF variable; doubts may arise. 5. Issues repeatedly rose? Not clear if the issues are only about the provision of 2013 data or if more years are included in the analysis. 15

SUGGESTIONS TO STREAMLINE THE END-USER FEEDBACK FOR THE FUTURE In the sequence of what has been present along the previous section of this report, there are two main actions that will need to take place in the future to foster higher quality to this annual exercise. One is on how best describing each data issue and where these contents should be addressed in the IT platform; this is described in the following point. The second action is about the optimisation of the IT platform. This IT Platform has proven to be of major relevance and however the tool itself isn t thoroughly explored in this report, the main conclusion from STECF EWG about the needed enhancement are here included. These conclusions on the specific enhancements were added to the last point of this report. WHAT CONTENT SHOULD THE END-USER FEEDBACK HAVE AND WHERE TO ADDRESS THESE CONTENTS IN THE IT PLATFORM? TABLE 3 - CONTENTS OF THE DCF END-USERS FEEDBACK AND HOW TO DESCRIBE THEM IN THE IT PLATFORM Information Content IT Platform field Identification of the end user End user designation. <end-user> A dropdown list facilitates the identification of the data end-user. The end-user identification is connected with the log-in credentials. year The year the data call was <Year> launched. In case of a RFMO, the year in which data was submitted. In principle it will always be the year before the exercise takes place. Member State concerned Identification of the Member State the issue is addressed to. If more than one MS is involved, then the issue should be repeated as many times as the number of different MSs. <Country> Clear identification of the data call Data calls: enter the name of the data call to allow for a proper identification of the issues, referring to the data call document (eg official letter). Example ICES data call on VMS data to OSPAR (VMS-OSPAR). RFMO should identify the form/task of the statistical/scientific data program the issue is related to. Detailed information about missing data In order to allow for a proper identification of the data, the following information should be included : the data group e.g Effort data/landing data/biological data, etc. <id_call> <data> 16

Issue Type of issue Information Content IT Platform field the variable(s) e.g. fishing days/ days at sea, length data, age data. the reference year(s): 2012 the fleet segment/fishery: the specie. A concise description of the issue <issue> should be added. If it is a data gap or a delay in transmission, that should be noted together with the identification if such event had impact on the working group or final data product. If the issue is related to low coverage or questionable quality of data, the problem should be described and the benchmark to attain in future identified. Severity of data failures A proper classification of the issue must be entered. The severity of the data failure should reflect the impact it had on the working group or output. It should be supported by proper identification. <Issue_type>: A dropdown list facilitates the identification of the category (possible values: Quality; coverage, timeliness and unknown) <Issue_type>: A dropdown list facilitates the identification of the severity (possible values: high, low, medium, medium-high) THE DATA TRANSMISSION IT PLATFORM The new tool, implemented to support the exercise of collecting feedback on the data transmission issues and the follow up of the process therein, got a very positive feedback from the MSs, STECF pre-screeners and some end-users. Main comments were that it is efficient and easy to use. The tool allows end-users to fill in apparent data failures in the platform, standardizing the presentation of data failures by all end-users and guarantees to have a complete description of the problems in data transmission. However, it has been only used by the pre-screeners and not by the STECF Expert Working. For the tool to be of use by the group in the future, some enhancements are still needed. EWG recommendations for the future: The IT tool must allow the selection of issues filters based in every of the column with relevant information, and not only select issues based on DG Mare Decision, Country, Year, Issue Type and Severity. Also to ensure a proper comparison of the issues, the comments and assessments, the user interface must allow the visualization of several lines at the same time. The look and feel of the interface should be similar to a spreadsheet. 17

REFERENCES Scientific, Technical and Economic Committee for Fisheries (STECF) Evaluation of 2014 MS DCF Annual Reports & Data Transmission (STECF-15-13). 2015. Publications Office of the European Union, Luxembourg, EUR 27410 EN, JRC 96975, 287 pp. http://stecf.jrc.ec.europa.eu/documents/43805/1002766/2015-07_stecf+15-13+- +Evaluation+of+2014+DCF+AR+and+DT_JRC96975.pdf European Commission. 2015. Proposal for a REGULATION OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL concerning the establishment of a Union framework for the collection, management and use of data in the fisheries sector and support for scientific advice regarding the Common Fisheries Policy (recast) (COM(2015) 294 final). Scientific, Technical and Economic Committee for Fisheries (STECF) Evaluation of 2013 MS DCF Annual Reports & Data Transmission (STECF-14-13) 2014. Publications Office of the European Union, Luxembourg, EUR 26811 EN, JRC 91550, 257 pp. http://stecf.jrc.ec.europa.eu/documents/43805/768107/2014-07_stecf+14-13_evaluation+of+2013+ar+and+data_jrc91550.pdf 18

European Commission Joint Research Centre Institute for the Protection and Security of the Citizen (IPSC) Contact information Castro Ribeiro Cristina Address: European Commission, Joint Research Centre (JRC), Institute of the Protection and Security of the Citizen (IPSC), Maritime Aairs Unit G03, Via Enrico Fermi 2749, 21027 Ispra (VA), Italy) E-mail: cristina.ribeiro@jrc.ec.europa.eu Tel.: +39 0332 78 9329 JRC Science Hub https://ec.europa.eu/jrc Legal Notice This publication is a Technical Report by the Joint Research Centre, the European Commission s in-house science service. It aims to provide evidence-based scientific support to the European policy-making process. The scientific output expressed does not imply a policy position of the European Commission. Neither the European Commission nor any person acting on behalf of the Commission is responsible for the use which might be made of this publication. JRC97428 EUR xxxxx xx ISBN xxx-xx-xx-xxxxx-x ISSN xxxx-xxxx (online) doi:xx.xxxx/xxxxx Luxembourg: Publications Office of the European Union, 20xx European Union, 2015 19

Europe Direct is a service to help you find answers to your questions about the European Union Freephone number (*): 00 800 6 7 8 9 10 11 (*) Certain mobile telephone operators do not allow access to 00 800 numbers or these calls may be billed. A great deal of additional information on the European Union is available on the Internet. It can be accessed through the Europa server http://europa.eu. How to obtain EU publications Our publications are available from EU Bookshop (http://bookshop.europa.eu), where you can place an order with the sales agent of your choice. The Publications Office has a worldwide network of sales agents. You can obtain their contact details by sending a fax to (352) 29 29-42758. 20

XX-NA-xxxxx-EN-N JRC Mission As the Commission s in-house science service, the Joint Research Centre s mission is to provide EU policies with independent, evidence-based scientific and technical support throughout the whole policy cycle. Working in close cooperation with policy Directorates-General, the JRC addresses key societal challenges while stimulating innovation through developing new methods, tools and standards, and sharing its know-how with the Member States, the scientific community and international partners. Serving society Stimulating innovation Supporting legislation doi:xx.xxxx/xxxxx ISBN xxx-xx-xx-xxxxx-x 21