Biodiversity and Climate Change Virtual Laboratory. Presentation by Willow Hallgren at ESA conference workshop 1 October 2014

1. Workshop: Advances in data
management practices and
technologies for ecosystem science
Where Ecology Meets Big Data
By Dr Willow Hallgren, Professor Brendan Mackey and Mr Andrew Bowness
Presented by Willow Hallgren

2. Aim of Presentation
 Learn what a Virtual Laboratory is, and what it can do
 Introduce you to, and give you a quick tour, of the
Biodiversity and Climate Change Virtual Laboratory
(BCCVL)
 Hopefully get you to start thinking how Big Data is
influencing research, and how you might apply the
BCCVL in your own research, to utilise your own
datasets and expand the scope of the research
questions you ask in the future.

3. Outline of Presentation
 Current capabilities in terms of the concept of Virtual
Laboratories
 BCCVL specific details and video
 How to use the BCCVL
 Example experiment
 Future Directions for the BCCVL.

4. Current capabilities of Virtual Laboratories
Virtual Laboratories:
 Are formed around engaged research communities
 Build on existing research capabilities
 Support research workflows
They are designed to:
 Allow much easier access to data archives and modelling tools, thereby reducing the
technical barriers to using state of the art tools,
 Facilitate the sharing of data, experiments and results
 Reduce the time to conduct scientific research studies
 Provide a platform for collaboration and contributions by the Australian research
community.

5. Currently Existing Virtual Labs
 Biodiversity and Climate Change – Griffith University
 Understanding Biodiversity response to climate change through integrated observation
and modelling
 Genomics Virtual Laboratory – University of Queensland
 Collaborative genomics workflows in the cloud
 Virtual Geophysics Laboratory – CSIRO
 Workflows and access to geophysical tools, data and resources
 Marine Virtual Laboratory (MARVL) – University of Tasmania
 Virtual environments to unify marine modelling and observation
 Climate and Weather Science Laboratory – BoM
 Integrated environment for climate and weather science modelling and data
 Industrial Ecology Virtual Laboratory – Sydney University
 Supporting comprehensive environmental footprinting and sustainability assessments

6. Currently Existing Virtual Labs
 All Sky Virtual Observatory – ANU
 Federating astronomy data: SkyMapper and the Theoretical Astronomical
Observatory
 Humanities Networked Infrastructure – Deakin University
 Unlocking and uniting Australia’s cultural data
 Human Communications Science – University of Western Sydney
 Above and beyond Speech, Language and Music: A national collaborative
environment for HCS research.
 Endocrine Genomics VL – University of Melbourne
 Integrating –omics capabilities to realise personalised e-Health in a translational
researcher-oriented environment
 Characterisation Virtual Laboratory – Monash University
 Research environments for exploring inner space

7. www.bccvl.org.au

8. BCCVL specific details and show
 The Biodiversity & Climate Change Virtual Laboratory (BCCVL)
provides an easy-to-use, web-based platform for modelling the
potential responses of Australia’s biodiversity to climate change.
 The BCCVL gives researchers access to a suite of modelling tools,
data collections, access portals and visualisation capabilities to
facilitate and accelerate research on the impact of climate change
on biodiversity
 Modelling can be at multiple scales down to 250 m resolution
 The BCCVL allows you to upload your own data, share it with other
BCCVL users and conduct modelling experiments on it.

9. BCCVL specific details and show
The BCCVL helps researchers transform biodiversity and climate change
research by:
 Enabling existing research questions to be investigated far more efficiently
and effectively;
 Providing the means for Australian researchers to address new important
questions.
Quantitative modelling is one of our most transparent and science-based
tools for the prediction of how (e.g.) species ranges will shift as a result of
climate change
The Information generated is needed to inform adaptation management
strategies for species and ecosystems.

10. How to use the BCCVL
STEP 1: Go to www.bccvl.org.au (Beta release)
STEP 2: Log in with your Australian Federated Password
(your Institutional user name and password)
(1)
(2)
(3)

11. The BCCVL is organised under three
Datasets
• Species location
and trait data,
• Current and future
climate data, and
• Other
environmental
data, (e.g., soil,
geological and
vegetation type)
main headings:
Experiments
• Access a suite of
statistical
modelling tools.
Knowledge Base
• Documentation
and relevant
literature.

12. BCCVL Datasets
STEP 3: Find the datasets you want to use:
• Many datasets already uploaded into BCCVL
• Species data (distribution, traits), climate data, environmental • 3 icons: view map, download, metadata
Search among the
datasets listed
• Facility to search the web for desired data
• Currently restricted to Atlas of Living Australia
• Search, download into BCCVL, and view
Discover a dataset
on the web
• Can upload species occurrence, abundance or trait data, or other
environmental datasets
• View, download, metadata, share with other BCCVL users
Upload your own
dataset

13. BCCVL Experiments
STEP 4: What do you want to do with the data?
You can run 5 types of experiments:
•Find current potential distribution of a species
Species Distribution Modelling
Experiment
•Project a species distribution into the future based on a climate projection,
for one or more emission scenarios Climate Change Experiment
•Calculate biodiversity statistics (species richness and endemism) based on
SDM results Biodiverse Experiment
•Find current or future distribution of a particular species trait e.g. Leaf Area
Index ***
Species Trait Modelling
Experiment
•Use multiple models- either SDMs or climate models- or emission scenarios,
to mitigate the uncertainty in the single-model/scenario approach Ensemble Modelling

14. BCCVL Knowledge Base
STEP 5: Learn more, before you start, or as you go.
Knowledge base
• Species Distribution Modelling Experiment
• Find current potential distribution of a species
• Reference datasets available for use in the BCCV
• Environmental (geoscientific, elevation, vegetation)
• Climate (current and future climate, water availability, PET, etc.)
• Biological (Atlas of Living Australia, eBird, GBIF)
• Functional response (i.e. species trait) modelling
• Glossary…
• etc. (still very much a work in progress!)

15. Example experiment
(1) Species Distribution Modelling Experiment:
 Predicting the current and future distribution of the Snow Gum
(Eucalyptus Pauciflora).

19. Example Experiment
• (plan to add more slides or videos to show
how to conduct an experiment in the BCCVL)

20. Future Directions
 Utilisation of more, and different types of
datasets (e.g., palaeodata)
 Possibility of incorporating more algorithms
and different types of modelling tools
 Expand the BCCVL internationally

21. Contact Details
Dr Willow Hallgren, Research Fellow
Griffith Climate Change and Adaptation Program
Griffith University, Southport Campus,
Gold Coast
Email: w.hallgren@griffith.edu.au

22. Relevant Links and References
Links:
• www.bccvl.org.au
References
• Peters, D. P. C., K. M. Havstad, J. Cushing, C. Tweedie, O.
Fuentes, and N. Villanueva-Rosales. 2014. Harnessing the
power of big data: infusing the scientific method with
machine learning to transform ecology. Ecosphere 5(6):67.
http://dx.doi.org/10.1890/ES13-00359.1