Register for DFO's BioChem archive, query marine biological and chemical data from Canadian waters, and read the ZIP return.
Applies in: CA
BioChem is Fisheries and Oceans Canada's archive of marine biological and chemical data. It is not a public download catalogue and it is not a live dashboard: it is a query archive you register for, ask a question of, and receive a data package from. This guide covers the query route only — finding and extracting data that are already archived. Loading data into BioChem is a separate route, reserved for DFO data managers through the edit application, and it is not described here beyond that one sentence.
The archive holds biological and chemical data from department research initiatives and data collected in areas of Canadian interest, according to the Biological and chemical marine data: BioChem database page. The data are divided into two functional areas: discrete data, usually from water bottle sampling, including pH, nutrients, chlorophyll, and dissolved oxygen; and plankton data, usually from towed nets, including species counts and biomass measurements. The data support climate research, and the same page says they may also be used for other purposes, including improving understanding of ecosystem variability, ground-truthing satellite data, and supporting fisheries research. Before doing anything else, decide which functional area your question belongs to, because every later step — the query form, the output files, and how to read them — splits on that choice.
BioChem is the right archive when your question is about marine biology or chemistry in waters of Canadian interest and the data come from discrete samples or plankton nets. The Atlantic Zone Monitoring Program (AZMP) is a large source of the data in the archive: the technical report documenting BioChem says BioChem is the data archive for bottle and plankton data resulting from AZMP, and that since 1999 the program has been the source of a large proportion of data recently loaded to the archive. Other sources named in the report include international collaborative programs, universities, and the Sir Alister Hardy Foundation for Ocean Science continuous plankton recorder data.
Three checks belong before registration. First, the holdings are historical as well as current: as of December 2013 the archive contained records from more than 2,100 missions from locations around the globe, with the earliest discrete data record dating from 1921 and the earliest plankton record from 1914. Those figures are dated December 2013 — no newer holdings total was verified on the current BioChem page — so treat them as the last published count, not as today's size, and do not quote them as current. Second, check the service interruption notice on the current BioChem page before assuming a region is covered: Maritimes Region data (generally the Northwest Atlantic) are partly unavailable while DFO reviews them, and the workaround is an email request, not a query. Third, remember what BioChem is not: temperature and salinity appear as ancillary data, but BioChem is not the primary repository for physical oceanographic data, so a question that is really about temperature or salinity belongs in the physical-data archive, not here.
There is no self-serve sign-up page to click through. The current Biological and chemical marine data: BioChem database page states the registration step in one sentence: users must register to use BioChem by sending an email to biochem.xent@dfo-mpo.gc.ca. Once you have your account information, you may use the query application to extract data. That is the whole of the verified registration process on the live page: no form link, no stated turnaround time, no stated list of what to include in the email, and no stated eligibility test.
Two cautions follow from that thinness. The first is the conflict preserved in the FAQ above: the 2014 technical report describes a registration form on the website and says users are asked to provide email contact information and organization affiliation, with new users assigned to one of three levels — user, data manager, or super user — where a user can only query and view existing records and cannot modify any records. The live page supersedes the report on the method (email, not a form), but the report is still the only source that says what registration asks for, so include your name, email address, and organization affiliation in the email, and state the data you intend to query. The second caution is about the account levels: as an outside researcher you will be a query user, not a data manager. The edit application, which loads or edits data, is reserved for DFO data managers, and the current page says so explicitly. Do not ask for edit access, and do not expect the registration reply to grant it.
No fee is stated on the current page or in the 2014 report, and no processing time for registration is stated on either. If the reply asks for anything this guide does not mention — a form, a fee, or an affiliation check — follow the reply, not this guide, and recheck the BioChem page, since the registration route has already changed once between the report and the live page.
Open the BioChem Query application once you have your account information. The application queries either the discrete or the plankton functional area: pick the area first, because the criteria and the output format follow from it. Criteria are specified to identify the data of interest and to limit the size of the query return, and they can include items such as geographic area, date, and data type. A user manual is available from the query webpage that provides step-by-step instructions for building and submitting a query, according to the 2014 report — open that manual from inside the application rather than guessing at the form fields, since the live form is the authority on what it accepts.
Scope the query narrowly on the first attempt. A mission in BioChem is a group of sampling events — it may be a multidisciplinary oceanographic cruise from a large ship, nearshore sampling from a smaller vessel, or a collection of sampling events such as weekly visits to a station over the sampling season — and recent datasets carry a unique mission descriptor assigned by Oceanography and Scientific Data to assure uniqueness and standardize the format. If you already know the mission descriptor, the mission name, or the program (for example, an AZMP mission), use it to bound the query; if you do not, bound it by area and date instead, and keep the date range short. A query that asks for everything in a region over decades will return a package too large to work with and too mixed to interpret, and the archive's own guidance is to use criteria to limit the size of the return.
Submit the query and stop: there is no on-screen result. Submitted BioChem queries run off-line, and the user receives an email, usually within a few minutes, when the results are ready. The email gives a link to the DFO FTP site where the resulting compressed ZIP file can be retrieved, and the file remains available for seven days. Although the application and the email notification are bilingual, the query results as well as code table terms and definitions are in English only, so plan to work in English even if you submitted in French.
When the email arrives, download the ZIP file from the DFO FTP link promptly: the file remains available for seven days, and the report gives no way to re-download it after that window except by running the query again. Save the ZIP file itself, not just the extracted CSVs, because the query-criteria text file inside it is the only record of exactly what you asked for, and you will need it to reproduce the extraction or explain it in a methods section.
The ZIP package contains several files, and the data file is only one of them. The record of the query criteria, the collectors list (which can be a person, a project, a program, a published report, or other, and which may link to additional metadata in the DFO data inventory where it is referred to as the BioChem series identifier), the counts file giving the number of records associated with each data type or taxon code returned — with units and data format for discrete data — the Data.csv file containing the data and related metadata, the locations list of the sampling events returned (mission name, event ID, station name, latitude and longitude, start date and time), the missions list of all missions considered during the query, and, for plankton queries only, a samples list of the events and associated metadata of records in the Data.csv file. Read the counts and locations files before opening the Data.csv file: the counts file tells you which variables and taxa actually came back and in what units, and the locations file tells you where and when the samples were taken, which is the fastest way to spot a query that returned the wrong region or the wrong years.
The Data.csv file is not self-explanatory, and the report says so plainly: the lack of clarity is mostly related to the cryptic nature of the codes used for the different data types. The code tables must be consulted for full descriptions, and they are too voluminous to reproduce in a report — codes are regularly added, so any printed list would quickly become outdated. The Biological and chemical marine data: BioChem database page links a code tables archive of 17 text files containing database codes and their descriptions to help interpret query results and load data, and an entity relationship diagrams archive of 12 PDF files showing the database structure. Download both before trying to read a return, and keep them with the data: a data type name is often a concatenation of variable name, analysis method, and modifier, so a column labelled with a code such as a salinity datatype or an oxygen datatype only becomes a measurement once the code table is consulted. Units for data values are found in the Counts.csv file, not in the Data.csv file.
Every data value carries a quality-control flag, and the flags largely follow the GTSPP meanings: 0 means no quality control, 1 means the value seems correct, 2 means the value appears inconsistent with other values, 3 means the value seems doubtful, 4 means the value seems erroneous, 5 means the value was modified and now seems correct, 6 and 7 are unassigned, 8 means quality control was performed by the data producer, and 9 means the value is missing. Flag 8 is not a standard GTSPP flag. Higher-value flags take precedence over lower-value ones, so a value judged doubtful by one test and erroneous by a later test keeps flag 4. Filter or flag your analysis by these codes rather than silently including every row: a value flagged 4 is a value the archive itself calls erroneous, and a value flagged 0 has had no quality control at all.
Three known quirks of the archive belong in the working routine. First, some datasets have inadvertently been loaded more than once, because unique mission descriptors were not used from the start of data loading; personnel are working to eliminate these replications, and users should be aware of their existence when considering query results — so check the mission descriptor and event ID for duplicates before counting records. Second, for plankton data, the absence of a count for a taxon does not prove the taxon was absent: most plankton datasets do not include zeros in the counts field, and a taxon may have been grouped with others or assigned to a higher taxonomic rank, while some legacy datasets hold transformed values such as numbers per cubic metre rather than raw counts. Third, the taxonomic code table holds more than 5,400 entries as of December 2013, including names that are not individual species — non-taxonomic groups, groupings of species or life-history stages that are difficult to distinguish, and even non-living material such as oil, plastic, and sand — so verify the taxonomic name against the code table rather than assuming every row is a species count.
Cite the database in every publication or report that uses the data. The current Biological and chemical marine data: BioChem database page gives the citation as: DFO [current year]. BioChem: database of biological and chemical oceanographic data. Department of Fisheries and Oceans, Canada. BioChem database page. Database accessed on [insert date here]. In the reference list, also include the technical report: Devine, L., M.K. Kennedy, I. St-Pierre, C. Lafleur, M. Ouellet, and S. Bond. 2014. BioChem: the Fisheries and Oceans Canada database for biological and chemical data. Can. Tech. Rep. Fish. Aquat. Sci. 3073: iv + 40 pp. The 2014 report's own citation wording for downloaded records differs slightly — it cites the database with the older BioChem web address — so use the current page's citation as the operative one and include the report as the second reference, as the current page directs.
Keep the query-criteria text file from the ZIP package with the citation, because the citation alone does not say which slice of the archive was used. A reader who wants to reproduce the extraction needs the functional area, the geographic area, the date range, and the data types or taxa queried, and that record is in the package, not in the citation. If the data came through the Maritimes Region email workaround rather than a query, say so in the methods note and cite the email as the source of the data, since no query record exists for it.
The current [Biological and chemical marine data: BioChem database](http://www.dfo-mpo.gc.ca/science/data-donnees/biochem/index-eng.html) page states that users must register to use BioChem, and it states no eligibility restriction and no fee. The 2014 technical report that documents the database says BioChem was made available on the internet starting in 2008, open for use by the scientific community and general public, and it likewise states no fee. Do not assume a fee exists or that one does not: no fee is stated on either opened official source, so if a charge ever appears, it will be in the registration reply, not in this guide. The data support climate research, and the current page says they may also be used for other purposes, including improving understanding of ecosystem variability, ground-truthing satellite data, and supporting fisheries research.
Both wordings exist on official sources, and they do not agree, so both are preserved here. The current [Biological and chemical marine data: BioChem database](http://www.dfo-mpo.gc.ca/science/data-donnees/biochem/index-eng.html) page says users must register to use BioChem by sending an email to biochem.xent@dfo-mpo.gc.ca, and that once you have your account information you may use the query application. The 2014 technical report instead says anyone can use BioChem after first obtaining a username and password using the registration form on the website, and that users are asked to provide email contact information and organization affiliation. Follow the current page — send the email — because it is the live instruction; the form wording is from a 2014 report describing the system as it then worked. The current page does not say what to put in the registration email, so include your name, email address, and organization affiliation, which is what the report says the registration process asks for, and state what data you intend to query so the reply can be directed correctly. If the reply points you to a form, use the form; if it sends account information directly, use that.
BioChem holds biological and chemical data from department research initiatives and data collected in areas of Canadian interest, divided into two functional areas. Discrete data usually come from water bottle sampling and include pH, nutrients, chlorophyll, and dissolved oxygen; plankton data usually come from towed nets and include species counts and biomass measurements, according to the current [Biological and chemical marine data: BioChem database](http://www.dfo-mpo.gc.ca/science/data-donnees/biochem/index-eng.html) page. The 2014 technical report adds that most discrete records relate to water chemistry, grouped into classes such as nutrients, pigments, organic matter, and oxygen, and that most plankton records are zooplankton, primarily arthropods and chordates such as fish larvae. What it does not hold as a primary archive: physical variables such as temperature and salinity are included as ancillary data when available, but BioChem is not the primary repository for those types of data. Do not go to BioChem for a temperature or salinity time series as the authoritative source — query it for the chemistry and biology, and treat the physical columns as supporting context for the bottle or net sample.
A query is built in the [BioChem Query](https://inter-j02.dfo-mpo.gc.ca/bcq-bcr/) application, which queries either the discrete or the plankton functional area, with criteria such as geographic area, date, and data type to identify the data of interest and limit the size of the return. Submitted queries run off-line, and the user receives an email, usually within a few minutes, when the results are ready; the email gives a link to the DFO FTP site where the resulting compressed ZIP file can be retrieved, and the file remains available for seven days. Although the application and the email notification are bilingual, the query results, code table terms, and definitions are in English only. The ZIP package contains a record of the query criteria, a collectors list, a counts file giving the number of records per data type or taxon code with units and data format for discrete data, the Data.csv file holding the data and related metadata, a locations list of the sampling events returned, a missions list of all missions considered during the query, and, for plankton queries only, a samples list. The Data.csv file looks very different depending on whether the query was directed to the discrete or the plankton functional area, and the report's Annex V gives the field-by-field guide to reading it. Keep the whole ZIP: the counts and locations files are what make the data file interpretable, and the query-criteria text file is the record of exactly what you asked for.
Start with the code tables, not the data file. The data types are codes, and a data type name is often a concatenation of variable name, analysis method, and modifier — for example, a salinity column may carry a datatype such as Salinity_Sal_PSS_Cal1 and a dissolved oxygen column such as O2_Winkler_Auto1, with the code tables giving the full descriptions. The [Biological and chemical marine data: BioChem database](http://www.dfo-mpo.gc.ca/science/data-donnees/biochem/index-eng.html) page links a code tables archive of 17 text files containing database codes and their descriptions, to help interpret query results, and an entity relationship diagrams archive of 12 PDF files showing the database structure. Units for data values are found in the Counts.csv file, not in the Data.csv file. Every data value carries a quality-control flag, and the flags largely follow the GTSPP meanings: 0 means no quality control, 1 means the value seems correct, 2 means the value appears inconsistent with other values, 3 means the value seems doubtful, 4 means the value seems erroneous, 5 means the value was modified and now seems correct, 6 and 7 are unassigned, 8 means quality control was performed by the data producer, and 9 means the value is missing. Flag 8 is not a standard GTSPP flag. Higher-value flags take precedence over lower-value ones, so if one test judged a value doubtful and a later test judged it erroneous, flag 4 is retained. Two interpretation traps from the report matter for plankton data: most plankton datasets do not include zeros in the counts field, so the absence of a count for a taxon does not prove the taxon was absent — it may have been grouped with others or assigned to a higher taxonomic rank — and some legacy datasets hold transformed values such as numbers per cubic metre rather than raw counts, so check the filtered volume, depth strata, and split fraction before calculating standardized abundances.
The current [Biological and chemical marine data: BioChem database](http://www.dfo-mpo.gc.ca/science/data-donnees/biochem/index-eng.html) page carries a service interruption notice: DFO identified content-related issues after an internal review of Maritimes Region BioChem data, and the affected data will be unavailable until further reviews are complete. DFO says it will make the data available again to BioChem query users as it reviews each dataset and addresses the issues. If you require data loaded by Maritimes Region — generally the Northwest Atlantic — email DataServicesDonnees@dfo-mpo.gc.ca, providing details including geographic coordinates and temporal limits, and DFO says it will review requests and provide data whenever possible. Do not treat a query that returns nothing for the Northwest Atlantic as proof that no data exist: check the interruption notice first, then email the data services address with your coordinates and date range. No end date for the interruption is stated on the page, so recheck the page before assuming the data are back.
No registration turnaround time, registration fee, query size limit, query manual contents, current holdings total, Maritimes Region interruption end date, or edit-application procedure for outside users is stated in this guide, because none was verified on an opened official page. Figures and claims cut during drafting: any current count of missions, records, or taxa beyond the December 2013 figures carried as dated, since no newer holdings total appears on the current BioChem page; any description of the query form's fields beyond geographic area, date, and data type, since the form itself was not opened past its landing page; any fee or processing time for registration, since neither opened source states one; and any claim about which DFO regions' data are currently available beyond the Maritimes Region notice, since the page states no other interruption. The registration conflict is carried as both wordings rather than reconciled: the live page says email, the 2014 report says a website registration form. The citation conflict is carried the same way: the live page's citation is the operative one, and the 2014 report's older web address is not reproduced. Recheck the BioChem database page before registering or querying, since the registration route has already changed once and the Maritimes Region interruption has no stated end date.