The IMERS database or Integrated Marine Environmental Samples and Readings database was set up at the Flanders Marine Institute (VLIZ) in response to the growing need for a general repository that could store the data resulting from sampling campaigns in a uniform way. The database has been designed for storing measurement data based on sampling of water, sediment, suspended matter and biota. The measurement data, the so-called readings, are stored together with a maximum of information on the actual sampling event that lay at the basis of the measurement. This way, for example biotic readings can be queried together with the environmental readings that were registered during the same visit of a station.The design of the database can be split up into three major parts. The first part is based on the general concept of sample-taking and the main tables here are ‘Trips’, ‘Visits’, ‘Events’, ‘Samples’, ‘GranRecords’, ‘BioRecords’, ‘Specimens’ and ‘Readings’. The second part of the database structure is built around the table ‘ReadingTypes’. This part documents the reading and stores information on what parameter, unit, matrix is measured and what method is used. Each record in the ‘Readings’ table is linked to a readingtype. The third part of the database structure is built around the table ‘ReadingAdministration’. This part of the database stores information on the origin of a reading (file and data set), who is the author of the reading and what restrictions apply to it. All data can be traced back to their origin through a link with this table.All data submitted to, or collected by VLIZ are stored in this single database. In order to be able to group data coming from various sources and to be able to make data selectively visible, a system of context labeling was built into the database. The one-to-many relationship between the readingadministration and these context labels make it possible for each reading to be part of one or more contexts. The same technique, using contexts, was applied to the IMIS ...