BOOS (Baltic Operational Oceanographic System)

The Baltic Sea is surrounded by nine countries: Denmark, Estonia, Finland, Germany, Latvia, Lithuania, Poland, Russia and Sweden. The Baltic Operational Oceanographic System (BOOS) comprises various national governmental agencies that are responsible for environmental observations, modelling operations and production of forecasts, services and information for the marine industry, the public and other end-users. BOOS is built on different systems. The goals of BOOS are to [12]:

1. Improve and further establish services to meet the requirements of environmental and maritime user groups
2. Coordinate, improve and harmonise observation and information systems, where necessary
3. Increase the quality of and harmonise user-oriented operational products
4. Decrease the production costs of public products and services by sharing the workload
5. Co-operate with HELCOM (Helsinki Commission) and other relevant bodies with the aim to avoid duplication of work and to maximise mutual assistance
6. Identify new customers for operational oceanographic products
7. Further develop the market for operational oceanographic products
8. Develop BOOS pursuant to the GOOS principles
9. Provide high quality data and long time-series required to advance the scientific understanding of the Baltic Sea
10. Provide data and forecasts to protect the marine environment, conserve biodiversity, and monitor climate change and variability
11. Improve and further establish services to meet the requirements of environmental and maritime user groups

The development of BOOS was supported with the start-up financing from the EU THEMATIC NETWORK project “Programme for a bAltic network to assess and upgrade an oPerational observing and forecAsting System in the region” (PAPA, EVR1-CT-2002-20012). PAPA’s main objectives are to demonstrate constructive ways to implement a basin wide operational oceanographic system according to the ambitions of GOOS:

1. Build the basin-wide network for ocean monitoring and forecasting, linking all the Baltic countries, broadening and strengthening the existing network of national institutions
2. Identify the gaps in the monitoring systems in the region and in the capability to measure, model and forecast the ecosystem, taking stock of current RTD projects and of the EuroGOOS and BOOS activities
3. Build capacities for expertise in the setting up and running of observing platforms, in managing data, in modelling and forecasting the ecosystem
4. Design an effective observing and forecasting system, inter-comparing experiences and standardising practices, towards the coordinated upgrading of the observing and forecasting capabilities in all Baltic countries
5. Raise awareness on the benefits of ocean forecasting at local, regional and global scales, involving stakeholders and disseminate PAPA results and products.

During the PAPA-project a password protected FTP-box system was developed for data exchange between member institutions. It is a decentralised system where each member puts data it wants to exchange in its own FTP-box, where it can be collected by the other members. Therefore, each institute is free to put what data they want in their own box in whatever format they like. All partners are obliged to setup an FTP-box (read only by the other partners) for publishing operational data, metadata, disclaimer, header and format information. The only restriction is the use of the ASCII format. By using the ASCII format, individual datasets are readily available for all partners (e.g. real-time data and the inventory of the QA database are available via the web). However, resulting from the SEPRISE project, a number of regional data centres have been given responsibility for collating the various in situ datasets from various FTP-boxes in the BOOS region. Each regional centre specialises in a particular parameter. These collated datasets are placed in a single FTP-box per regional centre. The institutes who are responsible for collating regional data within BOOS are listed in the Table below

This section outlines the currents, temperature and salinity profiles, water level, and waves data formats for BOOS.

SMHI are responsible for collecting currents data from FTP-boxes in the BOOS region. This compiled currents dataset is stored in a single ASCII file and disseminated via a central FTP-box. A single row of data in the ASCII file stores currents data from all stations for a single instance of time. The data format for each column within a row is illustrated here:

1 YYYYMMDDhhmm → data and time

2 currents direction → degrees at depth x for station 1

3 currents direction → degrees at depth x+1 for station 1

4 currents direction → degrees at depth x+2 for station 1

.

? currents speed → cm/s at depth x for station 1

? currents speed → cm/s at depth x+1 for station 1

? currents speed → cm/s at depth x+2 for station 1

.

.

? currents direction → degrees at depth y for station 2

? currents direction → degrees at depth y+1 for station 2

? currents direction → degrees at depth y+2 for station 2

.

? currents speed → cm/s at depth y for station 2

? currents speed → cm/s at depth y+1 for station 2

? currents speed → cm/s at depth y+2 for station 2

.

.

Date and time is recorded in the first column of each row (YYYYMMDDhhmm format, e.g. 200706171600). The time zone is always UTC. For the first station, columns two, three, four, etc. stores a series of currents direction values (degrees) for a particular depth, where each columns measurement is associated with a depth value referenced in the current’s readme file. Following the currents direction values is a series of currents speed values (cm/s) for the first station. Again, each of these measurements is associated with a depth value referenced in the current’s readme file. For all subsequent stations, respective measurements are stored in successive columns and associated with a depth value referenced in the current’s readme file. The code for erroneous or missing data is -999.0000.

BSH are responsible for collecting temperature and salinity (T/S) data from FTP-boxes in the BOOS region. This compiled T/S dataset is stored in two ASCII files: one for sea temperature and another for sea salinity. This data is disseminated via a central FTP-box. The format for both ASCII files is:

Long Lat StatName Time depth:value depth:value ….

Representing longitude, latitude, station name, data and time (YYYYMMDDhhmm), and a series of water depth and associated measurement values.

BOOS water level observations from tide gauges are collected by DMI via FTP-boxes from the various BOOS partners. Sixty days of data from each observation station is stored in a single individual ASCII file and disseminated via a central FTP-box. However, the frequency of the data is the original data sampling frequency provided from the original data provider. The format of the file is:

YYYYMMDDhhmm wl

where the left hand value represents date and time (e.g. 200704192300) and the right hand value represents water level in metres (e.g. 0.6). Missing water levels data is represented with the value -9.99. Water level is referenced to national reference levels.

This information was not available during the writing of this report.

Developed during the PAPA project, a metadata database was setup on a server at the BSH which can be accessed by the BOOS members. The database is password protected and the access information was first announced at the BOOS/PAPA meeting in May 2004 in Norrköping, Sweden. The database interface has two main features: one is a map based overview function and the other consists of parameter based queries.

PAPA database on fixed stations

The queries return tables containing the desired information and thematic maps can be produced interactively. This so called ‘PAPA-OBS database’, contains metadata of operational stations or measuring sites and the ‘QA database’, the quality assurance database, containing information on sensors, accuracy, calibration and data processing procedures used in the BOOS community. They are linked to an interactive web site where the information can be retrieved. This database can be accessed at:

http://gotland.bsh.de/papa/db_start_page.html

Following the results of the PAPA project, the upgrade of an operational observing and forecasting system in the Baltic Sea was established using common protocols for quality control to obtain high quality data. They are based on the evaluation of basic data pre-processing procedures and existing quality assurance (QA) protocols. The main objective of quality control for operational data is to ensure data consistency for each measured parameter within a collection of operational data. One of the most important goals must be to assure that the quality and errors associated with the data are transparent to the user. Therefore, additional quality assurance procedures have to be applied and metadata must be collected regularly as quality assurance information. Data quality assurance information notifies users of the data about the way in which they were gathered, checked, and processed. It also provides information about the types of algorithms used, the kinds of errors that occurred, and how these errors were corrected or flagged. These common protocols include all steps, which are necessary to collect quality controlled operational data:

1. Inspection of sensors / instruments at a laboratory by using reference standards
2. Regular reference measurements during a measuring period
3. Calculation / conversion algorithms
4. Overall data transmission
5. Data processing
6. Storage and flagging (including all quality assurance procedures)

For BOOS, three classes of common protocols for operational real-time measurements have been defined:

1. Best performance QA protocol for operational oceanographic measurements with ten steps to get high accuracy quality controlled datasets
2. Good performance QA protocol for operational oceanographic measurements with eight steps to get good accuracy quality controlled datasets
3. Minimum performance QA protocol for operational oceanographic measurements with six steps to get minimum accuracy quality controlled datasets

Additional information on BOOS quality assurance is available in the PAPA report “Design of common protocols for data collection, transmission and quality control” which is available at this link:

http://www.boos.org/papa/doc/PAPA%20-%20Report%20on%20quality%20assurance.doc

Next Table summaries BOOS quality assurance protocols.