GMOS-FR data products

This page summarises the data products and data that describes the datasets (metadata) offered by GMOS-FR and related to mercury atmospheric data collected at sites coordinated by French research units (IGE UMR5001, GET UMR5563) within the international GMOS/GEO-GOS4M networks and long term observation experiments carried out as part of the French IPEV GMOStral-1028 programme and other national french and/or international programs (LabEX, LEFE, ANR).

GMOS-FR data comply to international standards

GMOS-FR data and data that describes the datasets (metadata) are Findable, Accessible, Interoperable, and Reusable, so called FAIR principles. In practice, the FAIR principles aim at giving the user sufficient tools to understand the meaning of the data before and after downloading them.

GMOS-FR follows International Standard Operation Procedures as described in AMNET and GMOS international networks.

Data product levels

GMOS-FR data Portal has four main levels of data and data products: these are represented by raw data and the numbers 0, 1 and 2. Only GMOS-FR levels 1 and 2 data are available and downloadable.

Raw data

Raw data are information directly obtained from human measurements or automated sensors that have not undergone any transformation. They may provide quantitative or qualitative information about physical variables of the environment and may be of various forms, such as text, CSV files or physical samples.

Level 0 data

Level 0 data are data in physical units either directly provided by instruments or converted from engineering units (for example, mV, mA, Ω) to physical units from raw data. Level 0 data may have been filtered by a quality check (for example, thresholds). Description of L0 is different according to the type of data considered (mercury atmospheric data from continuous measurements and monitoring sites, mercury in precipitation, in snow…). L0 data are data that are generated as intermediate steps in the data processing of levels 1 or 2 data preparation, and for this reason they are not handled as persistent data and not shared outside the GMOS-FR PI. The Level 1 data are used for internal quality checks, as in communication between principal investigators.

Level 0 mercury atmospheric data (issued from continuous measurements and.or monitoring sites)

After data acquisition (©Tekran instruments corporation), atmospheric raw data sets are processed for quality assurance using a workflow and associated software developed in IGE in 2012 (ex-LGGE research laboratory). The workflow applied, albeit with some different peculiarities, is closed to the one, in the framework of GMOS FP7 program with G-DQM (see D’Amore et al., 2015).

L0 level data is divided in L0A and L0AM levels data.

L0A level corresponds to an intermediate QA/QC flagged dataset with information on data validation after workflow step one application ; Tekran raw data and maintenance/instrument reports being both the input files data. L0A level operational phases could be divided in two sub-steps (1a and 1b).

– Step 1a runs an automated process that filters the raw data. The system compares the dataset against 43 potential flags corresponding to 43 criteria that specifically refer to the three operation phases related to GEM/TGM and RGM/PBM concentrations calculations, desorption cycles (DES – only for RGM/PBM), and CAL (see Tables 1–4 in D’Amore et al., 2015). The flags are grouped in three sets: valid, warning and invalid. Each flag refers to a specific condition, or criterion, that the system checks in order to screen the Tekran data. Each raw observation is individually flagged depending on the result of each corresponding criterion and returns, as a temporary output, a flagged dataset.

– Step 1b consists to automatically include all field notes, anomalies, routine controls and spare part changes implying corrections and invalidations of data regrouped in the 1a flagged dataset step.

L0AM is a second intermediate QA/QC flagged dataset including data records clarification to L0A dataset, prior to their full approval, by the site’s monitoring program responsible.

This data records clarification step (1c) allows the manual flagging data invalidation in addition to those invalidated by automatic processes (1a and 1b), using known criteria defined by the site manager according to his knowledge and experience of the field concerned.

This last and third step of the L0 level validation processes (L0A and L0AM) allows access to level L1, which corresponds to the data fully QAed/QCed, to the initial temporal acquisition resolution.

Level 0 for other dataset

Work in progress (mercury in precipitations, in snow, ancillary parameters)

Level 1 data : QAed/QCed observational data (original time resolution acquisition)

Level 1 data are the quality-checked GMOS-FR dataset following international SOP, with data in original time resolution acquisition, ready to be distributed through the GMOS-FR data portal. Data L1 users have to keep in mind that, even if L1 data provides high-quality checked dataset, this one may be improved for more elaborated data product, depending of the study scope. Data L1 users are invited to directly contact the GMOS-FR web portal PIs (Dr O. Magand – olivier.magand@univ-grenoble-alpes.fr , Dr A. Dommergue – aurelien.dommergue@univ-grenoble-alpes.fr ) as well as other contacts possibly related to the data used (see in contact section in the selected dataset) in the framework of the use of the data and preparation of a research paper (see data policy and contact section).

Level 2 data : Qualified observational data as elaborated product

Level 2 data are data products that are based on L1 GMOS-FR observational data can be presented at the GMOS-FR data portal as an elaborated product. Data L2 users have to keep in mind that, even if L2 data provides high-quality checked dataset, this one may, as for the L1 one, be improved for more elaborated data product, depending of the study scope. Data L2 users are invited to directly contact the GMOS-FR web portal PIs (Dr O. Magand – olivier.magand@univ-grenoble-alpes.fr , Dr A. Dommergue – aurelien.dommergue@univ-grenoble-alpes.fr ) as well as other contacts possibly related to the data used (see in contact section in the selected dataset) in the framework of the use of the data and preparation of a research paper (see data policy and contact section).