Dr Tim Newbold
My current work is focused on understanding and predicting human impacts on the biosphere. This work is centred around two major projects: one building statistical models relating human activities to changes in biodiversity (the PREDICTS project) and one attempting to build a process-based model of the world’s terrestrial and marine ecosystems at a global scale (the Madingley Model). Until recently, my work on these two projects was separate. I am now working on a project to integrate the two approaches, focusing on impacts of human activities on the structure and function of African ecosystems. I am also interested in how ecological traits of species determine their responses to land-use change, and what this means for changes in community structure and overall biodiversity. I was involved in the recent Global Biodiversity Outlook 4 report and in the associated work projecting whether we are on track to meet the biodiversity targets for 2020.
Dynamics of African Ecosystems Project
I have recently started work on a new project, seeking to understand how human activities are changing the structure and function of African ecosystems. To do this we are improving the Madingley Model so that it better captures land-use effects on organisms and ecosystems. To inform these improvements we are using the results from the PREDICTS models. This work is in collaboration with Ben Collen and Elizabeth Boakes at University College London. This work is in a very early stage - more information will follow.
The PREDICTS Project
The PREDICTS project assembled a very large and representative (of geographic regions and of species groups) database of species abundances in different habitats, and then used these data to construct statistical models of local responses of species, communities and biodiversity to human activities – such as land-use, land-use intensity and climate change, at a global scale. So far, the PREDICTS work has shown that responses of species to land use vary among taxonomic groups, has quantified past and possible future local biodiversity losses under land-use change, has documented the turnover in the composition of ecological assemblages among land uses, and has shown how the traits of bee species affect how they respond to land use. The structure of the database was described in a paper in Ecology & Evolution. The first phase of this project, led by Andy Purvis at the Natural History Museum, Rob Ewers at Imperial College London, Jorn Scharlemann at Sussex University, Drew Purves, until recently at Microsoft Research and Georgina Mace at University College London, has now finished. A second phase is underway also led by Andy Purvis, with Adriana De Palma, also at the Natural History Museum.
The Madingley Model
The Madingley Model attempts to simulate the fate of, and interactions between, all photosynthetic and heterotrophic organisms on Earth, in order to simulate the dynamics of whole ecosystems and to simulate the effects of human impacts on these ecosystems. We have published papers describing the philosophy of our approach, which was summarized excellently in the Financial Times. We have also published a paper describing the mathematical formulation of the model, and are now working on applications of the model to simulate human impacts. The Madingley Model is a collaborative effort withDerek Tittensor, Mike Harfoot, Chris McOwen and Phil Underwood at UNEP-WCMC, Drew Purves, Stephen Emmott, Lucas Joppa and Matthew Smith at Microsoft Research Cambridge, Mike Bithell at Cambridge University, and Ben Collen, Elizabeth Boakes, Georgina Mace, Tania Barychka at University College London.
Ecological Traits and Responses to Land-Use Change
This project has focused on quantifying communities in terms of the functional traits of the organisms they contain, understanding how functional traits affect the way that species respond to land use (work that was summarized nicely by a colleague in a blog post), and finally projecting what trait-mediated responses of species mean for the functional composition of whole ecological communities. This work has been a collaboration with Jorn Scharlemann, now at Sussex University, Rob Alkemade at the Netherlands Environmental Assessment Agency, Stu Butchart at Birdlife International, Cagan Sekercioglu at Utah State University, and Lucas Joppa at Microsoft Research.