BCCVL: Opening Up a New Field of Enquiry

BCCVL: Opening Up a New Field of Enquiry
June 15, 2016 Karen Mecoles

“Essentially the BCCVL has enabled us to ask questions that we couldn’t ask before.”

— Professor Brendan Mackey

The Biodiversity and Climate Change Virtual Laboratory (BCCVL) integrates modelling tools and datasets with high performance computers and major data storage facilities, to enable efficient investigation of biological systems. Directed by leading biodiversity and climate change researchers from 16 universities, the BCCVL was established to simplify biodiversity-climate change modelling.

Climate change in its simplest form is an alteration in the statistical distribution of weather patterns. Because of these statistical alterations, researchers are now entering uncharted territory, where future climates can no longer be extrapolated from past climate patterns, and new models must be used for predicting future changes.

These global climate models (GCMs) are similar to those used for daily weather prediction, however, being based on the laws of physics, the use of these models can be particularly prohibitive for a number of researchers – particularly those with limited programming skills.
To overcome this problem, the BCCVL was established in 2013 as a “one stop modelling shop” to simplify the process of biodiversity-climate change modelling.

Through the direction of leading biodiversity and climate change researchers from 16 universities, the goal of the BCCVL is to integrate modelling tools and datasets with high-performance computers and major data storage facilities, to enable more efficient investigation of biological systems.

“We were very excited when we heard about the Nectar Labs,” Professor Brendan Mackey, Director of the Griffith Climate Change Response Program said. “Having worked in this area of species distribution modelling for many years, as ecologists and ecological modellers, we were all very experienced in putting together the data, the compute, and the work flows associated with doing our research.

“But with such heavy start up costs and all of the other development costs associated with doing that, we saw virtual laboratories as a way of enabling us to not only save an enormous amount of time, but to also allow collaboration with those ecologists who weren’t so skilled at programming.”

Previously, the lines of inquiry into biodiversity and climate change impacts have been stymied due to the inability of researchers to access a standardised set of tools for analysis and computational limitations.

Supported by Nectar, lead partners Griffith University and James Cook University, applied for funding to establish the BCCVL, which has allowed them to set up a series of online modelling tools that can be shared among over twenty academic partners across Australia, saving researchers time and precious resources.

“Firstly, when you’re starting up a new project that involved species distribution modelling, you spend 40% of your time just getting all of the data and work flows in place,” Professor Mackey said.

“Secondly, there are a lot of ecologists who would like to incorporate this kind of modelling in their work, but can’t because the level of programming expertise needed is too high.

“So we wanted to lower the bar in terms of accessibility, but at the same time raise the bar in terms of the quality of the modelling – ultimately providing a more rigorous approach.”

In addition to improving accessibility and rigour, Professor Mackey said that the BCCVL will also allow researchers to focus on more than one statistical modelling technique at a time.
“Prior to the BCCVL you’d have to pick one future climate change scenario out of 17 or 18 different statistical modelling processes, 40 global climate changes and at least five future scenarios, then narrow all of that down to a specific time period, which could be any month, from now to 2100,” he said.

“The BCCVL enables us to now run intermodal comparisons, comparing different modelling techniques such as different climate change models, different scenarios, or a combination of both, which makes for much more rigorous modelling in terms of quantifying the uncertainties.

“You can also do ensemble models, where you can take ten of the most reliable climate change models in Australia and run a project based on the statistical integration on all of their outputs.

“It’s early days and there’s a lot of experimenting still to be done, but essentially the BCCVL has enabled us to ask questions that we couldn’t ask before – questions we may have wanted to ask but couldn’t logistically hope to answer, so it’s opened up a whole new field of enquiry. From 2015 the Griffith Climate Change Response Program, will use these new rigorous modelling techniques to focus on integrating the university’s research expertise in the fields of regional and urban planning, coastal management, engineering, architecture, governance, environmental sciences, ecology, information technology, environmental modelling and governance.

It will also apply these to its focal topic of “Adapting to a rapidly changing climate in coastal regions.”

For more information about BCCVL see www.bccvl.org.au