Conference summary

Links:

The conference was organised by a group of 18 EU-funded projects (see About the Conference) with the aim of structuring the biodiversity informatics community at the European level and beyond. Specifically we wanted the community to prepare for the release of funding calls for H2020 in a spirit of cooperation rather than competition.

The primary outcome was that the recognition of a goal, sine qua non, to deliver predictive modelling of the biosphere. We recognise that this is a greater challenge than modelling the global climate, but the climate modellers have provided a template against which we can work. A key difference between the global biosphere and the global climate is that the climate can be assumed to be in equilibrium at a global scale and when perturbed will return to equilibrium. Biodiversity, on the contrary, is prone to extinctions and its components evolve. It cannot therefore return to a pre-disturbance state. The goal, therefore, will be decades in the realisation.

To begin down such a path, it is necessary to have a clear vision of both the overall goal and the contribution any particular effort will make to that goal. In such a way consortia within the community can bid for funding and can communicate the benefits of the work in a clear, straightforward manner.

Various groups within the biodiversity sciences already produce good, powerful tools, but we need to ensure that they are easily interoperable. Ideally, one should be able to conceive an analysis, then design a workflow that will link data through a variety of such tools to reach a conclusion without needing to re-format the products of the component tools along the way.

There is little doubt that we have the technological toolkit to begin this work, but we lack the necessary social cohesion. We need those contributing to this effort to identify themselves as belonging to a specific community in a manner simmilar to particle physicists, astrophysicists and climate modellers. At present we have a strong tendency to emphasise our differences without the balance of recognising those factors that link us together.

Biodiversity informatics has already amassed a large body of coherent data. It is time that we begin to use these data to ask questions. Doubtless it will emerge that the data are insufficient for some questions, but that will provide a powerful motivation to generate and mobilise certain specific data sets, rather than the apparently random aggregation that we have employed thus far. Museum collections, for example, are likely to provide one of the few sources of historical data against which potential models can be calibrated. New data will have to be global, of course, and will need to include those areas that, for various reasons, lack data compatible with those from European and North American, especially Russia, China and Brazil (see "Panel Discussion 1: International Cooperation"). Europe in particular is well placed to offer leadership in this area and should, in addition to serving the needs of scientists, should also seek to serve the needs of decision-makers, specifically the Convention on Biodiversity Diversity (CBD) and the Intergovernmental Panel on Biodiversity and Ecosystem Services (IPBES).

Finally, the goal is beyond the abilities of the biodiversity informatics community alone. We have to forge links with the other disciplines engaged in understanding the biosphere: ecology; microbiology; agriculture; socioeconomics; remote sensing; taxonomy; molecular biology; and, of course, computing. As potential models emerge and are explored, we will also determine the roles these disciplines can play in reaching our ultimate goal.

Scratchpads developed and conceived by (alphabetical): Ed Baker, Katherine Bouton Alice Heaton Dimitris Koureas, Laurence Livermore, Dave Roberts, Simon Rycroft, Ben Scott, Vince Smith