SEA-SEARCH = Guidelines and Practices / MAST: Guideline for better Practice in Data Documentation

MAST: GUIDELINE FOR BETTER PRACTICE IN DATA DOCUMENTATION
March 1997, MAST Data Committee

INTRODUCTION

This guideline as well as the "Guideline for Project Data Management" is written to help teams to implement the "Code on Data Management in MAST Projects".

The guideline is presented in the form of a checklist of actions which goes into some detail but does not intend to be comprehensive. Appropriate choices have to be made by each project regarding its proper data documentation. Comprehensive documentation for your data may be less simple than expected. Therefore a word of caution. Put yourself in the position of the possible user. It would be a pity if your data are ignored once because nobody trusts their quality due to lacking documentation.

The objective of compiling data documentation is

to ensure that data are consistently well described,
that the quality and limitations of the data are apparent, and
that there is sufficient information available to assess the suitability of the data for a particular task.

The end result of good data documentation is to contribute to usable data sets of known quality by accompanying them with supporting documentation.

For the purpose of this document, data production is considered in two parts.

Data Acquisition - which includes the collection, processing and analysis of raw material or raw data up to the point where data is generated; and
Data Processing - which involves the further manipulation, processing or enhancement of this generated data.

Please note that not all aspects of the guideline will apply for all data types and users are advised to use as appropriate. Please consider that data appear in different forms in different projects - data may be analog readings, numeric, charts, images, samples, specimen etc, and data may generated in the laboratory, from a physical model, or gathered in a field experiment.

CHECKLIST ON DATA DOCUMENTATION

Data Acquisition

Measurement/Sample Collection

Describe the equipment used. Give name, gear code and details of its deployment as appropriate. If possible include references for the method/gear being deployed.
Describe the measurement platform and describe the techniques used for positioning the platform and equipment.
Give details of calibration of the equipment used, document details on sampling frequency, list experts involved.
Describe environmental conditions as appropriate.
Any limitations associated with using the chosen sampling strategy should be documented.

Sample Processing, Sample Analysis

Document how the sample was preserved.
State whether sample was processed in-situ, shipboard or laboratory. Give the name of the Laboratory where samples were processed .
Describe different stages in sample processing (including instrumentation used) from treatment of sample from when it is collected through to final processing of sample to the point where data is generated (details of the methods/instrumentation used should be given with full references).
Describe any instrumental error corrections made during the course of processing the sample.
Describe the expected precision, accuracy or reproducability of the used methodology and the limits of detection.
Give details if the methodologies employed in sample processing have been validated.
Comment on any limitations associated with used sample processing techniques.
Use references to published literature where ever possible in order to shorten and compact the information.

Data Processing

Data Processing/Analysis

Describe clearly the different stages in processing and analysing the data including reduction algorithms, statistical analysis, etc. Details of the methods used should be given with full references.
Describe algorithms and computer programmes used.
Give details of the degree of precision, accuracy and reproducibility of the data processed/analysed where known.
Give details if the methodologies employed in data processing/ analyses have been validated.
Please comment on any limitations incurred in using these sample processing techniques
Use references as appropriate - give full bibliographic reference.

Quality Control and Assurance Information

Quality control and assurance exercises can be carried out at each of the data generation stages as described above. Therefore for each stage, describe and include documentation that will give information on the quality of the data. That is:
Describe your cruise properly in a Cruise Report.
Describe exercises carried out to standardise the data or to calibrate your data.
Describe any data validation exercises carried out in response to the results of calibration and intercalibration as well as comparison with standard methods.
Give details of the detection of limitations, gaps or errors in position and time.

Data Formats
When compiling data files, it is essential that the contents of a data file can be clearly understood by a subsequent user and that the essential information such as station numbering, references to position, time and depth are always clearly identifiable. All data formats⁽¹⁾ must be well described to ensure that there are no misunderstandings and that the data will be usable by the receiver.

⁽¹⁾ There are certain documents existing that give general guidelines for formatting and desciption of certain data. For example, the IOC/GETADE document (Intergovernmental Oceanographic Commission (of UNESCO) "Formatting Guidelines for Oceanographic Data Exchange" gives general guidelines for the formatting and description of certain basic oceanographic data (see: http://www.ices.dk/) for recommendation of available codes to be considered.