TREE Lab develops practical solutions to conservation challenges by bridging ecological science with on-the-ground action.Â
Across conservation, restoration and environmental management, decision-makers are often asked to act despite considerable uncertainty. Ecological systems are complex, context-dependent and constantly changing. Actions that succeed in one place may fail in another, and interventions can create unintended consequences.
To address these challenges, research is increasingly informed by ideas from causal inference: a collection of approaches for understanding why ecological patterns emerge, predicting the consequences of interventions, and making more reliable decisions under uncertainty. We are interested in how causal thinking can improve ecological prediction, evidence synthesis and environmental management.
We are learning! The application of causal methods in ecology is not yet widespread, and many exciting developments are happening now. We see this as an opportunity to contribute to an ongoing "causal revolution" in ecology and conservation science while remaining grounded in practical real-world problems.
Our work spans three interconnected research areas.
Every forest, farm and restoration site is different. The outcomes of management actions depend on local climate, soils, species communities, landscape context and management history. As a result, interventions that work well in one location may have very different outcomes elsewhere.
We combine ecological understanding with advances in causal inference and machine learning to investigate ecological processes and predict the consequences of interventions. Drawing inspiration from developments in fields such as medicine and economics, we aim to move beyond "one-size-fits-all" recommendations towards approaches that recognise ecological complexity and local context.
Develop methods for predicting site-level responses to conservation interventions
Investigate the causes of ecological variation and context dependence
Build models that identify where management actions are most likely to succeed
Develop tools that support more targeted and transparent decision-making
Advance methods for causal inference in ecological systems
By understanding both cause and context, conservation actions can be better targeted, improving outcomes while making more efficient use of limited resources.
TREE: TRansferablE Ecology for a Changing World (NERC Exploring the Frontiers)
PRECISE: Precision Ecology for a Changing World (NERC PUshing the Frontiers)
Spake, R., Jackson, E.E., Bullock, J.M., Gardner, E., Tipton, E., Grainger, M. & Doncaster, C.P. (2025) Precision ecology for targeted conservation action. Nature Ecology & Evolution.
Jackson, E.E., Snäll, T., Gardner, E., Bullock, J.M. & Spake, R. (2025) Towards causal predictions of site-level treatment effects for applied ecology. EcoEvoRxiv.
Deer and grey squirrels can have substantial impacts on woodlands and farmland throughout the UK, creating conflicts between conservation, forestry and agricultural objectives. These species move freely across landscapes, meaning that management decisions in one location can influence ecological outcomes elsewhere.
Understanding and managing these impacts requires a landscape-scale perspective.
We study how mammals move through and use landscapes, evaluate the effectiveness of management interventions and investigate the causes of variation in impacts across space and time. We then work with practitioners to develop practical tools that support more effective management.
As elsewhere in our research, understanding cause and effect is central. We seek not only to describe patterns, but also to understand the processes that generate them and the likely consequences of alternative management actions.
Trial and evaluate impact mitigation methods
Model habitat selection and movement patterns at landscape scales
Monitor impacts on woodlands and farmland
Co-develop decision-support tools with land managers and conservation practitioners
Nurse planting field trial in the New Forest
Grey squirrel bark stripping: understanding the causes of bark damage to inform more targeted woodland management
Research prioritisation for deer management: identifying the most important unanswered questions for policy and practice
iDeer: an integrated deer-management platform that maps deer impact risk under alternative planting and management scenarios (UKRI Future of UK Treescapes project, NERC)
Our mammal research is highly collaborative. We work with woodland managers, conservation organisations and public forestry bodies. We recognise that people value deer in different ways. Some view them primarily as a source of damage, while others value them for cultural, recreational or economic reasons. Effective management requires approaches that recognise and work within these diverse perspectives.
Spake, R., Bellamy, C., Gill, R., Watts, K., Wilson, T., Ditchburn, B. & Eigenbrod, F. (2020) Forest damage by deer depends on cross-scale interactions between climate, deer density and landscape structure. Journal of Applied Ecology.
Spake, R., Gresham, A. & Jam, D. iDeer: A tool to support woodland creation alongside deer populations. Future of UK Treescapes Impact and Insights Note.
Decision-makers must decide how and where to invest limited conservation resources. However, ecological evidence is often fragmented, context-dependent and difficult to apply directly to management decisions.
Traditional evidence syntheses commonly focus on average effects, which may not transfer well to new settings or answer the questions practitioners actually face.
We synthesise evidence, identify knowledge gaps and work directly with practitioners to understand what information is needed for decision-making. We are particularly interested in how ecological evidence can be used to support causal understanding and improve predictions across different contexts.
We also develop new methods for evidence synthesis, recognising that the challenge is not only to summarise findings, but also to understand when and where they are likely to apply.
Conduct targeted evidence syntheses that identify critical knowledge gaps
Prioritise research questions that matter most for policy and practice
Co-design research with experts across science, policy and conservation
Develop frameworks for translating ecological findings across contexts
Advance quantitative and qualitative synthesis methods
Improve communication of ecological evidence and uncertainty
Better decisions depend on better evidence. By improving how ecological knowledge is generated, synthesised and communicated, we help ensure that conservation resources are directed towards actions with the greatest potential benefit.
Spake, R., O'Dea, R.E., Nakagawa, S., Doncaster, C.P., Ryo, M., Callaghan, C.T. & Bullock, J.M. (2022) Improving quantitative synthesis to achieve generality in ecology. Nature Ecology & Evolution.
Spake, R., Mori, A.S., Beckmann, M., Martin, P.A., Christie, A.P., Duguid, M. & Doncaster, C.P. (2021) Implications of scale dependence for cross-study syntheses of biodiversity differences. Ecology Letters.