Workshop 3: The model for wild boar data collection

How to improve the Data Model?

  • People and organisation are the same compared to the first section. Instead, the data model will have multiple contact names and emails
  • The geographical Coverage should include a country dropdown choice (incl. Europe-wide)
  • Is there a shape-file describing boundary? And where is it?
  • Methods: use a dropdown choice (e.g. sighting, hunted, RTA (killed on road data), lab testing, and include multiple (both RTA and sighting)
  • Using hunt data for occurrence (field defining type of occurrence data: sighting, shot animal, road kill, etc.)
  • Occurrence data – start and finish date for camera records
  • Verification per record (e.g. as in National Biodiversity Network of UK – correct, probably correct, etc.)
  • Facilitate data collection by simplifying the table (or using an App)
  • Create automatic/closed fields – able to choose among listed (not open questions)
  • Define a minimum number of mandatory fields: might need a field that determines data type. Certain fields would be mandatory conditional on the type of data
  • Add in cell of table – not available vs missing vs zero
  • Give data providers freedom to define areas from which they can provide data, stressing the need for the highest possible spatial resolution, and not make them fit to NUT/LAU system
  • No need for 2 separate databases for occurrence and harvest/abundance data; just add a field at the start directing to either occurrence or harvest/density database
  • Metadata has also be filled by the data providers
  • Same fields as for the GBIF data file but we will use a selection of fields for our database model
  • The DarwinCore is the basis of selection of fields
  • Select fields that are mandatory
  • Dataset (Resource) should be completely selected, but mandatory:
    • Title, creator; add ownership
  • Temporal and geographical coverage of data is mandatory
  • Methods should be completely selected, but mandatory
    • methodStep, qualityControl
  • Intellectual property rights is mandatory (additional field regarding display of data or mapping on the EFSA website might be useful)
  • Citation might be useful, but not mandatory
  • Metadata elements should remain and be the same for all the data type generated (occurrence, density and harvest). Workshop participants provided a minimum metadata set

How to improve the Data Model?

  • People and organisation are the same compared to the first section. Instead, the data model will have multiple contact names and emails
  • The geographical Coverage should include a country dropdown choice (incl. Europe-wide)
  • Is there a shape-file describing boundary? And where is it?
  • Methods: use a dropdown choice (e.g. sighting, hunted, RTA (killed on road data), lab testing, and include multiple (both RTA and sighting)
  • Using hunt data for occurrence (field defining type of occurrence data: sighting, shot animal, road kill, etc.)
  • Occurrence data – start and finish date for camera records
  • Verification per record (e.g. as in National Biodiversity Network of UK – correct, probably correct, etc.)
  • Facilitate data collection by simplifying the table (or using an App)
  • Create automatic/closed fields – able to choose among listed (not open questions)
  • Define a minimum number of mandatory fields: might need a field that determines data type. Certain fields would be mandatory conditional on the type of data
  • Add in cell of table – not available vs missing vs zero
  • Give data providers freedom to define areas from which they can provide data, stressing the need for the highest possible spatial resolution, and not make them fit to NUT/LAU system
  • No need for 2 separate databases for occurrence and harvest/abundance data; just add a field at the start directing to either occurrence or harvest/density database
  • Metadata has also be filled by the data providers
  • Same fields as for the GBIF data file but we will use a selection of fields for our database model
  • The DarwinCore is the basis of selection of fields
  • Select fields that are mandatory
  • Dataset (Resource) should be completely selected, but mandatory:
    • Title, creator; add ownership
  • Temporal and geographical coverage of data is mandatory
  • Methods should be completely selected, but mandatory
    • methodStep, qualityControl
  • Intellectual property rights is mandatory (additional field regarding display of data or mapping on the EFSA website might be useful)
  • Citation might be useful, but not mandatory
  • Metadata elements should remain and be the same for all the data type generated (occurrence, density and harvest). Workshop participants provided a minimum metadata set

Share

Facebook
Twitter

COLLABORATE IN WILD BOAR DATA CALL.

Become in a ENETWILD collaborator!

Click to access to Wild Boar Data Model and download the instructions and excel files to fill in with your data.

Recent Posts

Twitter