A review of modelling approaches for predicting bird population responses to land-use change
1 Centre for Agri-Environmental Research, University of Reading, Reading RG6 6AR, UK
2 Central Science Laboratory, Sand Hutton, York YO41 1LZ, UK
3 British Trust for Ornithology, Thetford, Norfolk IP24 2PU, UK
4 School of Conservation Sciences, University of Bournemouth, Poole BH12 5BB, UK
5 School of Biology, University of Newcastle, Newcastle NE1 7RU, UK
* Email: s.j.butler@reading.ac.uk
To date, most farmland bird research has been retrospective, exploring the status of current bird populations, causes of declines and potential remedial actions in the context of past agricultural intensification and land-use change. As farming adapts to an increasingly dynamic trading environment there is clearly also a need to look forwards, to develop models for predicting how populations will respond to future land-use changes be they policy-driven, for socio-economic reasons or in response to large-scale drivers like climate change. This paper reviews the modelling approaches available for making such predictions. We assess the relative merits and limitations of mechanistic, demographic, habitat association and niche-based models and discuss the potential benefits of adopting Bayesian modelling approaches to account for uncertainties and improve model accuracy. In particular, we assess the suitability of each approach for answering the “How much is enough?” question, asking to what extent they can be used to predict the required levels of resource provision across agricultural landscapes that will be required to deliver population recovery.
Simon Butler is currently a post-doctoral research fellow at the Centre for Agri-Environmental Research, University of Reading. He undertook his PhD in the Farmland Bird Group at the Edward Grey Institute for Field Ornithology in Oxford and then worked as a research biologist for the RSPB before taking up his current position in 2006.
