Tammseminarium: LIFEPLAN – Combining Big Ecological Data with Novel Statistical Methods

Press release: Life on Earth to be mapped in huge EU project

Tomas Roslin about the project

Biodiversity underlies ecosystem functioning. To achieve the basis for a sustainable management of natural resources under current environmental change, we need a unified theory of the forces structuring vast sets of ecosystems and taxonomical groups.

For the first time, such a synthesis is now within reach, based on a recent revolution in sampling methodology, globally relevant ecological data, and advances in statistical methods for linking immense data to community ecological theory. LIFEPLAN is a six-year project funded by an ERC Synergy Grant, bringing together the key expertise needed to generate and interpret Big Ecological Data: community ecology, methods for automated species recognition, and Bayesian statistics for immense data. Our objectives are to generate a new understanding of biodiversity patterns and dynamics by developing new methods for big data statistics.

To this end, we will generate fully standardized, global big data on a range of species groups, thus allowing quantification of variation in ecological communities at spatial scales covering six orders of magnitude (from 0.1 km to 10,000 km), across tens of thousands of species. The resulting data motivate the development of transformative big data statistics, in particular highly scalable algorithms for spatio-temporal data, as well as methods for automated species identification from DNA, audio and image samples. As a key deliverable, we will develop global joint species distribution models describing the spatiotemporal structure of communities around the globe.

In this talk, I will argue that there is substance to these pompous claims, and specify what the project is about.