The Ocean Observatories Initiative (OOI) is a science-driven ocean observing network that delivers real-time data from multiple coastal and open-ocean arrays to address critical science questions regarding the world’s oceans. During each array service cruise OOI performs hydrographic sampling to evaluate and validate the deployed instrumentation. Service cruises are conducted annually for the open ocean arrays and every six months for the coastal arrays. This repeat sampling has created time series of valuable physical, chemical, and biological information. Water samples from Niskin bottles on CTD casts are analyzed either on the ship or in onshore labs to measure oxygen, salinity, nutrients (nitrate, nitrite, silicate, phosphate, ammonium), chlorophyll, and the carbon system (pH, dissolved inorganic carbon, total alkalinity).
FIGURE 1
OOI launched a collaboration in 2023 with the Biological & Chemical Oceanography Data Management Office (BCO-DMO) to make OOI water sampling data available via the BCO-DMO website and ERDDAP server. BCO-DMO curates publicly available research-ready oceanographic data in accordance with FAIR data principles. Advantages of distributing OOI data through BCO-DMO include concatenation of the cruise-by-cruise data into a single dataset with a Digital Object Identifier (DOI) and provisioning through ERDDAP, which provides both human and machine-to-machine interfaces. The BCO-DMO Dataset pages include descriptions of sampling and processing methods, and README files for each cruise.
Currently BCO-DMO has data from the OOI Station Papa Array (Gulf of Alaska, annual cruises over 11 years), Irminger Sea Array (North Atlantic, 10 years), Southern Ocean Array (SW of Chile, 6 years) and Argentine Basin Array (South Atlantic, 4 years). You can access the datasets via this direct link or from the BCO-DMO home page: Click on Projects, then search for “OOI Discrete CTD and Water Sampling Cruise Data”.
FIGURE 2
Figures 1 and 2 provide an example of the concatenated datasets using 10 years of data from the Irminger Sea Array. A Python script (implemented in a Jupyter Notebook available in https://github.com/WHOIGit/ooi-on-bco-dmo ) was used to access the data from the BCO-DMO ERDDAP server, extract variables of interest, apply available quality control (QC) flags, and visualize the data. Figure 1 shows profiles of selected variables for successive cruises to give a sense of the depth-time data coverage. Note that the sample depths are relatively sparse since the OOI sampling goal is to validate instruments on the moorings rather than collect comprehensive profile data. Figure 2 represents profile variability over time by an overplot color-coded by year.
Even though constrained to “Acceptable” QC flags, some of the values plotted appear to be outliers, indicating the need for the user to consider further data quality assessment. Note that Discrete README files within the BCO-DMO dataset and CTD Cast Logs on OOI’s Raw Data Archive provide useful information. For example, the low values of oxygen in 2021are noted as inconsistent with oxygen from the CTD cast, whereas the high values of salinity in 2015 appear to be real, associated with a salinity maximum observed by the CTD. Since creating the Jupyter Notebook, data for two of the Irminger Sea cruises in OOI’s Raw Data Archive have been updated (including Nitrate for the 2021 cruise); these updates will be in the next version of the Irminger Sea dataset on BCO-DMO.
For additional Python scripts to explore OOI Discrete CTD and Water Sampling Cruise Data as distributed by BCO-DMO, for example to plot a discrete parameter against its corresponding CTD sensed parameter, see notebooks available in https://github.com/WHOIGit/ooi-on-bco-dmo/tree/main/notebooks .
