Pipedream or pipeline: delivering regular, reliable, up-to-date information on biodiversity through repeatable workflows

6f6914b1cdb438695ec1aaabba7463bb?s=47 Peter Desmet
October 02, 2017

Pipedream or pipeline: delivering regular, reliable, up-to-date information on biodiversity through repeatable workflows

Talk at the TDWG 2017 annual conference in Ottawa, Canada - October 6, 2017.

The current paradigm for studies on biodiversity change are single studies, of finite duration, and a single published output. Yet the results of a such a workflow become out-of-date quickly, particularly as the speed of environmental change increases. If new environmental policies are implemented it is important to monitor their effects, which implies having results from before and after the policy implementation. Furthermore, given the difficulty of influencing policy, the results of analysis need to be reliable, have clearly communicated uncertainty, and should be open to scrutiny. The timely provision of such information could be possible by using open data, shared standards, and automation.

The TrIAS (Tracking Invasive Alien Species) project in Belgium is attempting to build a workflow from raw biodiversity data to policy advice, specifically to provide useful information on alien species and their associated risks. We are developing scripts (R, Python) to simplify the repeated Darwin Core standardization of species checklists and observations from a wide-range of sources and their publication to the Global Biodiversity Information Facility. We also aim to propose controlled vocabularies for alien species related Darwin Core terms where these data are needed for downstream analysis. Challenges include entrenched non-standard working methods, heterogeneity of data availability, and the sheer complexity of the biosphere itself. We will discuss our plans, the obstacles and potential solutions. Furthermore, we look to the future for what we might be able to achieve if we are successful.

https://doi.org/10.3897/tdwgproceedings.1.20749

6f6914b1cdb438695ec1aaabba7463bb?s=128

Peter Desmet

October 02, 2017
Tweet

Transcript

  1. Pipedream or pipeline Delivering regular, reliable, up-to-date information on biodiversity

    through repeatable workflows 2 October 2017, Ottawa TDWG 2017
  2. Quentin Groom Botanic Garden Meise Tim Adriaens INBO Diederik Strubbe

    Ghent University Sonia Vanderhoeven Belgian Biodiversity Platform Peter Desmet INBO Authors
  3. We want to study biodiversity change to help guide environmental

    policy
  4. None
  5. None
  6. This gets outdated:

  7. How to set up this So we can update this

  8. None
  9. We want to study invasive alien species to help guide

    environmental policy
  10. • What species? • Where are they? • How are

    they getting here? • What is their impact? • Future distributions? • Future impact? IAS in Belgium
  11. Standards

  12. What species are alien in Belgium?

  13. Let’s check the Alien species checklist for Belgium

  14. We don’t have one! And certainly not one that is

    unified, verified & standardized
  15. We do have A number of authoritative checklist with a

    more specialized scope
  16. Let’s develop a standard so we can combine these

  17. Let’s develop a standard so we can combine these

  18. Let’s use an existing standard so we can combine these

  19. Darwin Core

  20. Controlled vocabularies

  21. Proposal for terms & vocabularies

  22. Proposal for terms & vocabularies

  23. Data management

  24. Make a unified IAS checklist by combining the authoritative checklists

  25. Checklists authors should now manage the unified checklist

  26. Checklists authors should now manage the unified checklist

  27. Checklists authors should manage their checklist and get the tools

    to publish it
  28. Checklist management

  29. Repeatable publication

  30. Published checklist on GBIF

  31. Open data

  32. Where and when are the alien species in Belgium?

  33. • Citizen science projects • Monitoring programs • Research projects

    • Collections • Ad hoc information • GBIF Sources of IAS distribution data
  34. Let’s create a database to combine these

  35. Let’s create a database to combine these

  36. Let’s create a database to combine these

  37. Let’s publish as open data on GBIF

  38. Citizen science dataset

  39. Infrastructure & software

  40. Checklist & occurrence data on GBIF

  41. Checklists Occurrence sources

  42. Checklists Occurrence sources Unified checklist

  43. Checklists Occurrence sources

  44. Checklists Occurrence sources

  45. Checklists Occurrence sources

  46. Checklists Occurrence sources

  47. Feeds into further research pipeline

  48. Conclusions

  49. Repeatable workflows are feasible!

  50. Checklists Occurrence sources

  51. Checklists Occurrence sources

  52. Checklists Occurrence sources

  53. Checklists Occurrence sources

  54. Checklists Occurrence sources

  55. None
  56. trias-project.be @trias_project Thank you! Abstract: https://doi.org/10.3897/tdwgproceedings.1.20749 Presentation: bit.ly/2x9XSKF