Improving biodiversity monitoring and conservation practice with statistical modelling

Written by Ann Kinzer, Impact Officer, University of Kent

The University of Kent’s Statistical Ecology Group (SE@K) is a group of statisticians that aims to develop new statistical tools for monitoring wildlife populations. Its members focus on various different types of ecological data and associated questions. Both in the UK and abroad, the group engages with stakeholders, such as ecologists, practitioners, charities, and government organisations.

Many stakeholders, including the Bumblebee Conservation Trust and Butterfly Conservation are working with the wider population for their actual data gathering. The Bumblebee’s Conservation Trust’s BeeWalk, for instance, currently involves more than 800 volunteers. Members of SE@K, in turn, develop models to help and analyse such citizen science data. These statistical models are complex, because they have to account for various factors potentially influencing data collection in wildlife populations. In addition to this, models have to constantly evolve as new data collection methods, such as acoustics, drones and environmental DNA are being utilised for data gathering.

The work of SE@K is motivated by and directly guided by the needs of their stakeholders, who are the ones that encounter and pose interesting ecological questions.

‘Our work is problem driven and therefore it is absolutely essential for us to have a wide network of well-established collaborations from different fields […]. If we didn’t have those collaborations, we would have no motivation to do what we do. Developing a statistical model that has no application whatsoever in the world of ecology, would be a waste of everyone’s time.’ (Eleni Matcheou).

The group has built their networks over decades through conferences, meetings and workshops. In 2008, Prof. Byron Morgan – founding member of SE@K and of the National Centre for Statistical Ecology– together with Professor Steve Buckland, from St Andrews, initiated the UK’s first International Statistical Ecology Conference, which was designed to bring statisticians and practicing ecologists together. The conference has grown significantly in both size and importance ever since, nowadays taking place bi-annually in different parts of the world.

Stakeholders, in turn, are not only able to better monitor wildlife populations, thanks to the models developed at Kent, but also to answer important ecological questions and to better prioritise conservation efforts. Work undertaken by SE@K has generated new approaches to biodiversity surveys, assessment and indicators, which impacts conservation policies and allows to more easily identify species in need of protection. The group is also in the process of developing a knowledge-transfer partnership and benefits from student placements with key stakeholders.