OPENNEP ACCESSSSECC Contribution to the Theme Section ���Large-scale studies of the European benthos: the MacroBen database��� MARINE ECOLOGY PROGRESS SERIES Mar Ecol Prog Ser Vol. 382: 225���238, 2009 doi: 10.3354/meps07826 Published April 30 INTRODUCTION Often marine biological data are the result of projects with a limited temporal and spatial cover (Floen et al. 1993). Taken in isolation, datasets resulting from these projects are only of limited use in the interpretation of large-scale phenomena. More specifically, they fail to be informative on a scale commensurate with the prob- lems of global change that humankind is confronted with (Costello & Vanden Berghe 2006). Individual stud- ies are restricted in the amount of data they can gener- ate but by combining the results from many studies, massive databases can be created that make analyses on a much-enhanced scale possible (Grassle 2000). �� Inter-Research 2009 �� www.int-res.com *Email: evberghe@iobis.org Addresses for other authors are given in the Electronic Appen- dix at www.int-res.com/articles/suppl/m382p221_app.pdf MacroBen integrated database on benthic invertebrates of European continental shelves: a tool for large-scale analysis across Europe E. Vanden Berghe1,2,*, S. Claus, W. Appeltans, S. Faulwetter, C. Arvanitidis, P. J. Somerfield, I. F. Aleffi, J. M. Amouroux, N. Anisimova, G. Bachelet, S. J. Cochrane, M. J. Costello, J. Craeymeersch, S. Dahle, S. Degraer, S. Denisenko, C. Dounas, G. Duineveld, C. Emblow, V. Escaravage, M. C. Fabri, D. Fleischer, A. Gr��mare, M. Herrmann, H. Hummel, I. Karakassis, M. Ke ��dra, M. A. Kendall, P. Kingston, L. Kotwicki, C. Labrune, J. Laudien, E. L. Nevrova, A. Occhipinti-Ambrogi, F. Olsgard, R. Palerud, A. Petrov, E. Rachor, N. Revkov, H. Rumohr, R. Sard��, W. C. H. Sistermans, J. Speybroeck, U. Janas, G. Van Hoey, M. Vincx, P. Whomersley, W. Willems, M. W8odarska-Kowalczuk, A. Zenetos, M. L. Zettler, C. H. R. Heip 1Flanders Marine Institute, Wandelaarkaai 7, 8400 Oostende, Belgium 2Present address: IOBIS Secretariat, Institute of Marine and Coastal Sciences, Rutgers University, 71 Dudley Road, New Brunswick, New Jersey 08901, USA ABSTRACT: We describe an integrated database on European macrobenthic fauna, developed within the framework of the European Network of Excellence MarBEF, and the data and data inte- gration exercise that provided its content. A total of 44 datasets including 465354 distribution records from soft-bottom macrobenthic species were uploaded into the relational MacroBen database, corre- sponding to 22897 sampled stations from all European seas, and 7203 valid taxa. All taxonomic names were linked to the European Register of Marine Species, which was used as the taxonomic ref- erence to standardise spelling and harmonise synonymy. An interface was created, allowing the user to explore, subselect, export and analyse the data by calculating different indices. Although the sam- pling techniques and intended use of the datasets varied tremendously, the integrated database proved to be robust, and an important tool for studying and understanding large-scale long-term dis- tributions and abundances of marine benthic life. Crucial in the process was the willingness and the positive data-sharing attitude of the different data contributors. Development of a data policy that is highly aware of sensitivities and ownership issues of data providers was essential in the creation of this goodwill. KEY WORDS: Macrofauna �� Biogeography �� Integrated database �� Data management �� Data rescue Resale or republication not permitted without written consent of the publisher

Mar Ecol Prog Ser 382: 225���238, 2009 Such data have never been of greater importance con- sidering the recent observation of major shifts of marine species due to global change. A recent report from the Marine Board of the European Science Foundation shows clear evidence of northward migrations of sev- eral marine-species populations from European seas (Philippart et al. 2007). Increasingly, vast thematic data- bases are being created, made possible by advances in computer technology. The Ocean Biogeographic Infor- mation System (OBIS), the data-integration component of the Census of Marine Life (CoML), assesses and ex- plains the diversity and distribution of marine life through a network of linked databases (Grassle 2000). The Marine Biodiversity and Ecosystem Functioning (MarBEF) initiative includes implementation and net- working of large-scale biodiversity research in Europe part of this forms a contribution to the OBIS network through its European node, EurOBIS. Within the Mar- BEF working group Global Patterns of Marine Biodi- versity across Ecosystems (Theme 1), an integrated database (MacroBen) on soft-bottom macrobenthos was compiled. The present paper describes the content, data integration, standardisation and functionalities of the MacroBen database. CONSTITUENT DATASETS Each dataset (Table 1) was archived and described in detail at the data centre of the Flanders Marine Insti- tute (VLIZ). Describing datasets in a standardised way made it possible to create a searchable metadata inventory, thus facilitating data discovery and sharing. These metadata include information needed to decide on the relevance of a dataset in a particular context, e.g. where and when the data were collected, what the intended use of the dataset was, how data were col- lected, who played a role in the collection and the management of the data, where the data are stored and in what format and under which conditions they are available. All metadata descriptions are publicly available through the MarBEF website (www.marbef. org/data/dataset.php). Archiving the datasets prevents them from being lost by ensuring the long-term integrity of the data. The Integrated Marine Informa- tion System (IMIS), the metadata system maintained by VLIZ (Cattrijsse et al. 2006), was used as a tool for metadata management this system is also used to manage information on behalf of the MarBEF network. A unique citation, brief explanation of the original dataset objective and derived publications from each of the contributing datasets are listed in Appendix 1, together with a link to the extended metadata de- scription available in IMIS. In Appendix 1, we also pro- pose a standard dataset citation including title, dataset collector(s) or custodian(s) (equivalent to author[s] or editor[s]), dataset owner (equivalent to publisher) and final year of sampling. We hope that this standard citation, or an equivalent one, will stimulate formal recognition of data sources in future re-use of the data. DATA INTEGRATION The MacroBen database integrates datasets on soft- bottom macrobenthic fauna from all European seas. A total of 44 datasets was collected, harmonised and integrated into a relational Access database. The inte- grated database includes both quantitative and quali- tative data on organisms and several abiotic para- meters ranging from the northeast Atlantic and the North Sea, Norwegian Sea and Arctic Ocean, the Baltic Sea and the Mediterranean and Black seas. One of the main difficulties in integrating and com- paring different datasets from various data providers is the harmonisation of the data 3 types of harmonisation were performed: (1) taxonomic, (2) geographical, and (3) sampling methodology. All geographical coordi- nates were converted to decimal degrees. All datasets in which information on the datum was available used the WGS84 coordinate system. We assumed that this was also the case for those datasets where this infor- mation was missing. Taxonomic names as recorded in the constituent datasets were matched with the European Register of Marine Species (ERMS). This authoritative taxonomic register provides a list of species occurring in the Euro- pean marine environment, defined as up to the strand- line or splash zone above the high-tide mark and down to 0.5 salinity (Costello et al. 2001). Spelling mistakes were corrected, information on gender and life stage removed and stored in the appropriate fields, and the taxonomic name as recorded in the dataset was linked to the name as included as valid in the ERMS. Many names in the constituent datasets proved to be omis- sions in the ERMS, and were, in consultation with the ERMS taxonomic editors, subsequently added to the ERMS. A further description of the ERMS can be found in Cuvelier et al. (2006), and on the MarBEF website (www.marbef.org/data/erms). Various sampling tools and methods were used over the different datasets. About 50% of the data was derived from sampling with Van Veen grabs. To a lesser extent the data were derived from sampling with a Smith-McIntyre, Ponar or APN grab, cores or other hand-operated devices. Two datasets provided data from museum collections. Overall, 33% of the data were qualitative in nature, while 30% contained not only specimen counts but also biomass data (Table 2). The sampling area from the quantitative datasets 226