September 2008 Due to a change in policy at the national submitter level, all DATRAS data products are now publicly available without restriction. Extract from IBTS Report 2007 8.1 Data policy During the previous meeting in 2006, the IBTSWG reviewed the new ICES data policy that allows open access to all data stored at ICES, and was implemented at May 1 st 2006. It was concluded that in general the IBTSWG is positive towards an open data policy as it will encourage use of data. However, the group found that the policy does not take into consideration the problems that an open policy could create, with main concerns being that data can be misinterpreted by users not knowing the survey, gear and changes made during the overall time series, and that the existing database contains some errors that have not as yet been corrected. The IBTSWG therefore recommended in 2006 that each institute should discusses the issue internally in order to impose ICES to implement the access levels that the group proposed in 2005. This proposal covers the groups concerns and at the same time opens up access to the data. Each nation discussed the open access policy within their institute. All institutes now conform to the ICES open access policy, except for the Netherlands institute IMARES who defines access procedures for its survey-data submitted to ICES. Their policy is based on the access levels that the IBTSWG proposed in 2005 and in short says that the use for all ICES-related work is without access limitations, but the use for work that is not related to ICES can be restricted, depending on species and aggregation level. A detailed description of the IMARES data access policy is given in Appendix 5. If ICES gets a request for the supply of restricted data, IMARES should be informed about the request, and the release of the data has to be approved (maximum of 2 week notice) before the request is granted. Extract from IBTS Report 2006 9.1 Data policy The Working Group reviewed the new ICES data policy. In general the Working Group is positive towards an open data policy as it will encourage use of data. The group evaluated the use of IBTS data in 2003 (IBTSWG report CM 2003/D:05) and found that the IBTS data already was extensively used to the benefit to the scientific community and the use of the data is expected to increase in the coming years. However, the group found that the policy did not take into consideration the problems that an open policy could create for the data providers as expressed in previous IBTS reports. The main concerns being that data can be misinterpreted by users not knowing the survey, gear and changes made during the overall time series, and that the existing database contains some errors that have not as yet been corrected. The group fears that, if all data are available in the public domain, data suppliers may have to spend large amounts of time on verifying the results and interpretation of studies undertaken outside the ICES community. This can be especially problematic if a misinterpretation is published or publicised, where it is not transparent how they have reached their conclusions. The group also raised the question; who is responsible for refuting any claims based on a misinterpretation of the data, ICES, the IBTSWG or the individual institutes? With an open policy the group also fear that data could be copied and published with a web front-end for downloading of the data other places than in ICES. This would result in more than one version of the DATRAS database on the Internet. It should therefore not be possible to publish the data with download facilities on the Internet unless it is in corporation with ICES, where ICES are providing regularly updates or direct link to the database. The discussion on open data access is not unique to ICES but is also taking place in relation to the new EU fishery data collection regulation. The IBTSWG therefore recommends that each institute discusses the issue internally and find out what legislations applies to data in their country.
The group would prefer that ICES implement the access levels that the group proposed last year. This policy covered the groups concerns and at the same time opened up access to the data. The group recommends that each institute accept this access policy. If the open access has to be accepted due to national and international legislations, the group still wants to be informed about who is using the data and for what purpose. This could be implemented in the next version of DATRAS. One possibility could be that before downloading data through the internet the person requesting data has to fill in a web form explaining what the data will be used for, where it will be published and some back ground information on the person who requests data. This information should be send automatically by email to the survey coordinator from each institute to keep them informed. One of the main reservations about making all data publicly available is the misidentification of species (see WD 3 in Annex 5). A way to resolve this could be to only release data for problematic taxa on a family or genus level, and only provide data for these taxa at species levels for requests from the ICES community. DATRAS does not currently have this capability, though it could be built in. On the DATRAS download page there is a health warning explaining what to be aware of when using the survey data. This health warning should be accepted before data are downloaded and it should be placed at the top of the download file in order to ensure that data users see and hopefully read the health warning. Extract from IBTS Report 2005 7.2 ICES Policy on access to DATRAS The protocol on access to the DATRAS database that was presented in last year s report was discussed and some minor alterations were suggested. IBTS agreed that access to various forms of IBTS data should be improved. Though IBTS were not able to agree a change to the current access policy at the meeting, and so that there would be no immediate change, they discussed potential policies and hope to propose a new access policy for 2006. For further information on access policy (e.g. forms for requesting data and health warnings for the data), see ICES (2004). To structure data access, three IBTS survey/area combinations were distinguished, the countries participating in these combinations and whether data were submitted to the database: COUNTRY NORTH SEA WESTERN (1) SOUTHERN DATA IN DATABASE Denmark England/Wales France Germany Ireland Netherlands Norway Portugal Scotland Spain Sweden Notes 1 This does not include the porcupine survey due to the Spanish data regulation. Within each of these survey/area combinations there was agreement on the data access policy. This is shown in the table below, which distinguishes four user categories:
1 ) Institutes that have supplied data to the database. 2 ) ICES Assessment Working Groups 3 ) Other ICES Working Groups 4 ) Public and other parties that request data, typically for research purposes. and the following three data types: 1 ) Standard maps and graphs: Per survey/area combination for all relevant ages of species for which assessments are conducted. Maps will show bubble plots indicating abundance per ICES rectangle or per haul. Time series of the indices and a graph showing the proportion of the age-groups will be generated. These will be available for 8 commercial species (cod, haddock, whiting, herring, sprat, mackerel, saithe and Norway pout). 2 ) Aggregated data: A query of the database using pivot tables. Based on these tables, plots and graphs can be made on an interactive basis. The minimum level of aggregation differs between survey/area combinations. ICES rectangle: IBTS in the North Sea, Skagerrak, Kattegat and the BTS in the North Sea, Channel and Irish Sea Stratum: IBTS western and southern divisions Subdivision and stratum: BITS Baltic Sea 3 ) Un-aggregated (raw) data. These are: catch (numbers at length and/or numbers at age) data on a haul-by-haul basis, and SMALK (Sex, Maturity, Age-Length-Keys) data per individual. Data access for the four User categories and Data types can be organized according to the following matrix, and includes Free access, password protected access and access to extracted data after granted request. TYPE OF DATA USER CATEGORY National Fisheries Institutes (data suppliers) ICES Assessment WGs Other ICES WGs Public and other parties Standard maps and graphs (8 commercial species) Free access Free access Free access Free access Aggregated data (by Free access Free access Free access Free access ICES rectangle) (4,5) Aggregated data (by ICES rectangle) for other species and times Password Password protected 3) protected (1) Receive data after request to ICES If request granted by ICES Raw data Password Password protected (3) protected (1) Receive data after request to ICES (2,3) After request granted by national contact person (2) Notes: 1 For those survey/area combinations that the laboratory contributes to, otherwise by request 2 If access to raw data is given, ICES (IBTSWG) and the national fisheries laboratories supplying the data retain intellectual property, hence if data are to be used for publications, authors must liaise with IBTS members to ensure that both data analysis and interpretation are appropriate
3 ICES Assessment WGs that use IBTS data should provide IBTSWG with feedback regarding the utility of the survey data for the species/stock in question, so that IBTS know which data are performing well 4 Data would be available for the following dominant species: gadoids (Gadus morhua, Melanogrammus aeglefinus, Merlangius merlangus, Trisopterus esmarki, Trisopterus minutus, Merluccius merluccius, Molva molva, Pollachius virens), flatfish (Limanda limanda, Hippoglossoides platessoides, Microstomus kitt, Pleuronectes platessa, Glyptocephalus cynoglossus, Lepidorhombus whiffiagonis), other demersal species (Eutrigla gurnardus, Lophius piscatorius, Lophius budegassa, Echiichthys vipera, Amblyraja radiata, Raja clavata, Scyliorhinus canicula) and certain pelagic fishes (Clupea harengus, Scomber scombrus, Trachurus trachurus, Sprattus sprattus). 5 Data will only be freely available from 1983, and excluding more historical data and the three most recent years data.
Extract from IBTS Report 2004 6.2 ICES Policy on Access to DATRAS The protocol on access to the DATRAS database that was presented in last year's report was discussed and some minor alterations were suggested. To structure data access, three IBTS survey/area combinations were distinguished, the countries participating in these combinations and whether data were submitted to the database (table ). Country North Sea Western 1 Southern Data in database Denmark England France Germany Ireland Netherlands Norway Portugal Scotland Spain Sweden Notes: 1 This do not include the porcubine survey due to the Spanish data regulation. Within each of these survey/area combinations there was agreement on the data access policy. This is exemplified in the table below, which distinguishes three user categories: 1. Public and other parties that request data, typically for research purposes. 2. ICES working groups 3. Institutes that have supplied data to the database. and three data types: 1. Standard maps and graphs: Per survey/area combination for all relevant ages of species for which assessments are conducted. Maps will show bubble plots indicating abundance per ICES rectangle or per haul. Time series of the indices and a graph showing the proportion of the age-groups will be generated. 2. Aggregated data. A query of the database using pivot tables. Based on these tables, plots and graphs can be made on an interactive basis. The minimum level of aggregation differs between survey/area combinations. * ICES rectangle: IBTS in the North Sea, Skagerrak, Kattegat and the BTS in the North Sea, Channel and Irish Sea * Stratum: IBTS western and southern divisions * Sub-division and stratum: BITS Baltic Sea 3. Un-aggregated (raw) data. These are catch (numbers at length and/or numbers at age) data on a haul-byhaul basis and SMALK (Sex, Maturity, Age-Length-Keys) data per individual. Data access per "User category" and per "Data type" can be organized according to the following matrix. F is the abbreviation for "free access", P for "password protected access" and R for "access to extracted data after granted request".
Data type User categories ICES WG 1 Data supplier 2 Public and other parties Standard maps and graphs F F F Aggregated data P P P/R 3 Non-aggregated (raw) data R P R 4 Notes: 1 ICES WGs will have access to data from only those survey/area combinations that are relevant for their recommendations and as such should be specified in those recommendations. 2 Data suppliers will only have access to data of those survey/area combinations to which the institute has provided data. 3 Per survey/area combination the members can decide whether individuals will have free access to aggregated data or only after request. If a request is granted, an extraction of the data will be made available 4 Access can be requested and if granted, an extraction of the data will be made available All data (aggregated or non-aggregated) are protected by passwords. Each institute delivering data to the database can suggest a username and password to ICES which will give them access to the data in those survey/area combinations they are (for a sufficiently long period) part of. ICES WGs will have a password that allows them access to aggregated data, raw data will be issued to the chair of the WG after request. Other parties can request access to the (both aggregated or non-aggregated) data through the ICES website. A standard form must be filled in to inform the institutes involved in the survey(s) on Who is requesting data, including partners in the research project The purpose of the data request Which data (at what aggregation level) are requested Confirmation that the ICES rules for acknowledging the data source will be observed Completing the form will result in a request to the relevant survey contact person of each institute involved with that survey/area combination and this person will be requested to reply to ICES within 14 days. If a contact person does not reply within this time limit, it will be taken as acceptance of the request for data access. When after 14 days no relevant data supplier has objected, ICES will extract the requested data from the database and make them available. As only France has submitted data for the southern division the above agreement has no consequences yet for this division. If more countries in this survey/area combination submit data a comparable agreement on data access may be drafted.