 |
| Titel |
Data Publication and Quality Control Procedure for CMIP5 / IPCC-AR5 Data |
| VerfasserIn |
Martina Stockhause, Michael Lautenschlager, Heinke Höck, Frank Toussaint |
| Konferenz |
EGU General Assembly 2011
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 13 (2011) |
| Datensatznummer |
250047921
|
|
|
|
|
|
| Zusammenfassung |
| Climate data and especially data of climate projections of Earth System Models play a key
role for the investigation of climate change impacts and for the derivation of mitigation and
adaptation strategies. The user community grows from the original climate research
community into a heterogeneous climate community with diverse requirements for data and
data related services.
The next assessment report of the Intergovernmental Panel on Climate Change
(IPCC-AR5) will provide a large dataset of high quality for earth system and impact
researchers as well as for decision makers. The climate projections are defined within CMIP5
(Coupled Model Intercomparison Project Phase 5, cmip-pcmdi.llnl.gov/cmip5/). Data
is served via the distributed data infrastructure of the Earth System Grid (ESG,
www.earthsystemgrid.org). Thus data has to be described by metadata and fulfill reliable and
uniform quality standards.
Data passes different types of Quality Controls (QC) connected with different access
policies. Three different QC data levels are distinguished:
QC level 1: CMOR2, ESG conformance of data and CIM conformance of
metadata,
QC level 2: WDCC conformance and subjective controls, and
QC level 3: STD-DOI data publication.
For QC level 1 the completeness of metadata and the well-formattedness of data are checked.
Data assigned QC level 1 is visible for all. The access is restricted to the data owner and
specific scientists announced by him.
Within the quality checks for level 2 metadata and data are checked still separately.
Metadata is controlled by a scientist for plausibility and consistency. The expense of the data
checks is shared between the three ESG Federation partners: PCMDI (www-pcmdi.llnl.gov),
BADC (British Atmospheric Data Centre, badc.nerc.ac.uk), and WDCC (World
Data Center Climate, wdc-climate.de). The Quality Assurance (QA) procedures
includes consistency between data header and data and of statistical quantities against
pre-defined ranges. Data of QC level 2 is accessible under the risk of possible data
changes.
The QC procedure of level 3 follows the rules of STD-DOI data publication
(www.std-doi.de), based on which an international regulation is currently derived within the
DataCite consortium (datacite.org). Data checks consist of double-checks (using results of
performed QCs level 1 and 2) and cross-checks between data and metadata. In addition to this
technical QA (TQA) supervised by the WDCC, the scientific QA (SQA) is performed by the
data author. After receiving the final author approval for the content of the core metadata and
the description of the SQA procedures, the data is published, i.e. assigned the persistent
identifier DOI. Data of QC level 3 / assigned a DOI is no longer a matter of change and
accessible to all registered data users. Data is then uniquely citable in scientific publications
by its DOI.
The granularity of the STD-DOI data publication has to be suitable for citation within
model comparison studies. This is the experiment, i.e. the aggregation of all data
belonging to all ensemble runs performed for a specific climate projection. Optionally, a
peer-reviewed publication in a data journal like ESSD (Earth System Science Data,
www.earth-system-science-data.net/) or alike for a larger data aggregation (coarser DOI) is
supported by the WDCC, providing the opportunity for the scientists to receive additional
scientific credits for their data publication. |
|
|
|
|
|
|