In this Leverhulme funded project, the WrEN team joined forces with Prof. Nick Hanley, Dr Mary Nthambi and Dr Katherine Simpson at Glasgow and Dr Tom Finch at RSPB. We employed a combination of ecological science with economics to quantify the biodiversity consequences and economic costs of competing landscape-scale conservation strategies.
There is currently much debate on how to conserve biodiversity in agricultural landscapes, but understanding how to achieve this effectively is undermined by a lack of information on three types of spill-over effects: spatial (effects of adjacent agriculture on wildlife in adjacent semi-natural habitats), temporal (arising the effects of past land use on current biodiversity) and behavioural (e.g. landowner decisions).
There is currently much debate on how to conserve biodiversity in agricultural landscapes, but understanding how to achieve this effectively is undermined by a lack of information on three types of spill-over effects: spatial (effects of adjacent agriculture on wildlife in adjacent semi-natural habitats), temporal (arising the effects of past land use on current biodiversity) and behavioural (e.g. landowner decisions).
The potential influence of spill-overs on farmland biodiversity |
We used a variety of data sources on biodiversity in agricultural landscapes (including from the WrEN project), maps of agricultural intensity land-use, national farm data and historic land-use maps.
Our objectives were to:
|
|
Summary of project outcomes 1. Spillovers and legacies of land management on temperate woodland biodiversity First we investigated how both current habitat configuration and historical land-use legacy shape species distributions, using Bayesian hierarchical occupancy models fitted to data on 373 species across four taxa (ground beetles, birds, vascular plants, and small terrestrial mammals) from 134 woodlands ranging from 10 to over 250 years old in UK agricultural landscapes (Bradfer-Lawrence et al. 2025). |
|
Key findings:
The conclusion is that both spatial and historical context are important when designing biodiversity-focused land management, as neither factor alone is sufficient to predict species distributions reliably.
- Both spillovers and legacies matter — Spatial spillovers and temporal legacies both influenced species richness and community composition, and critically, these two factors interacted with one another, producing different biodiversity outcomes than either considered in isolation.
- Taxon-specific responses — Woodland patches in landscapes with more old woodland and lower historical woodland loss tended to support more bird and plant species, but fewer beetle species.
- Ignoring legacies distorts predictions — Failing to account for legacy effects gave a misleading picture of habitat suitability, meaning the same management actions could produce unexpectedly different outcomes depending on landscape context.
The conclusion is that both spatial and historical context are important when designing biodiversity-focused land management, as neither factor alone is sufficient to predict species distributions reliably.
2. Economic incentives for woodland creation on farmland: Modelling the impacts on biodiversity
Next, we modelled the effects of economic incentives for woodland planting on UK farmland and their impacts on three bird species. An agent-based economic model simulated farmer decisions by comparing returns from current agricultural land use against woodland planting incentives, while an ecological model (from Bradfer-Lawrence et al. 2025) predicted effects of both parcel-level and landscape-level woodland cover on species distributions. Two case study areas were compared, varying in the spatial relationship between opportunity costs and ecological potential (Nthambi et al. 2024).
Key findings:
The core message is that the same economic instrument can produce markedly different biodiversity outcomes depending on the spatial alignment between where farming is least profitable and where woodland creation would be most ecologically beneficial.
Next, we modelled the effects of economic incentives for woodland planting on UK farmland and their impacts on three bird species. An agent-based economic model simulated farmer decisions by comparing returns from current agricultural land use against woodland planting incentives, while an ecological model (from Bradfer-Lawrence et al. 2025) predicted effects of both parcel-level and landscape-level woodland cover on species distributions. Two case study areas were compared, varying in the spatial relationship between opportunity costs and ecological potential (Nthambi et al. 2024).
Key findings:
- Subsidies increase biodiversity indicators — As the per-hectare subsidy value increased, biodiversity indicators improved, but the rate of improvement varied by location and species.
- Cost-effectiveness depends on spatial context — The effectiveness of financial incentives was strongly influenced by whether opportunity costs and ecological potential were positively or negatively correlated spatially across the landscape.
The core message is that the same economic instrument can produce markedly different biodiversity outcomes depending on the spatial alignment between where farming is least profitable and where woodland creation would be most ecologically beneficial.
3. Combining occupancy and dispersal models to predict the conservation benefits of land‑use change
Finally we modelled how different woodland creation scenarios might affect populations of ten woodland-affiliated bird species in a real-world landscape (a 25×25km area of primarily arable land in England). An economic agent-based model from objective 2 (Nthambi et al. 2024) simulated landowner decisions under financial incentives for woodland creation, generating landscapes with varying spatial configurations of new woodland. An individual-based model was then used to simulate bird dispersal and settlement across these landscapes (Dobson et al. 2025).
Key findings:
The main conclusion is that cost-effectiveness of woodland creation for biodiversity is strongly determined by spatial configuration. Short-term targets for woodland birds are unlikely to be met through woodland creation alone given the time needed for trees to mature, and current creation rates may only produce modest population changes. Interim targets are recommended to manage expectations realistically.
Finally we modelled how different woodland creation scenarios might affect populations of ten woodland-affiliated bird species in a real-world landscape (a 25×25km area of primarily arable land in England). An economic agent-based model from objective 2 (Nthambi et al. 2024) simulated landowner decisions under financial incentives for woodland creation, generating landscapes with varying spatial configurations of new woodland. An individual-based model was then used to simulate bird dispersal and settlement across these landscapes (Dobson et al. 2025).
Key findings:
- Spatial configuration matters — Population increases were greater when woodland creation plots were clumped together rather than randomly sited. Planting adjacent to existing woodland produced comparable benefits to clumped arrangements, and was more beneficial at lower planting densities.
- Benefits are slow to materialise — In all scenarios, population increases were gradual, typically taking at least 30 years to be fully realised.
The main conclusion is that cost-effectiveness of woodland creation for biodiversity is strongly determined by spatial configuration. Short-term targets for woodland birds are unlikely to be met through woodland creation alone given the time needed for trees to mature, and current creation rates may only produce modest population changes. Interim targets are recommended to manage expectations realistically.