Modelling economic impact and strategies to increase resilience against tree disease outbreaks

Lead Research Organisation: University of Cambridge
Department Name: Plant Sciences


Forests in the United Kingdom are facing increasing pressures from pests and pathogens. An increase in the volume and global scale of trade, coupled with the evolution and adaptation of pests in the context of global change, has led to new diseases and pests appearing in the UK at an increasing rate. Thus, the issue is not so much whether a new epidemic will emerge in the next years, but when and how it will happen. Tree diseases result in economic costs in terms both of losses in timber values and environmental costs such as impacts on forest landscapes. However, measures to reduce the likelihood of occurrence or speed and extent of spreading the UK are also costly to forest managers. There is thus an important balance to be struck. Understanding the costs and benefits of management options to reduce disease risk requires the researcher to jointly consider both environmental and economic drivers. The characteristics of the physical and biological environment affect the spread of plant diseases, but the response of forest managers also determines this spread. Moreover, the outcomes for a given forest manager of their own actions depends on the actions of others in the landscape, since it is the overall spatial pattern of land management which determines disease spread.

Forests in the UK provide a range of important ecosystem services. Forests are used for the production of timber, but also have recreational and aesthetic values and play a key role in the carbon cycle. Land management policies which maximize the non-monetary (e.g. conservation) or monetary (e.g. timber) value of output do not necessarily lead to a forest landscape structure, tree species composition or forestry practices that are best for disease risk minimization. There is a potential conflict between enhancing the carbon sequestration and provisioning and cultural ecosystem services of forests on one hand, and increasing the resistance and resilience of forest ecosystems with regard to outbreaks of pests and pathogens on the other.

Aims and objectives
In this project, an interdisciplinary team of mathematicians, forest ecologists and economists will work together to build a series of models to allow us to study the ways in which different management options can reduce risks and expected damages from a range of forest diseases. Besides considering the effects of disease management options on the spread of diseases, the models will generate information on the effects of these options on the supply of other ecosystem services from forests, such as carbon sequestration and storage, and recreation opportunities, as well as on an indicator of forest biodiversity. Combined with a "Choice Experiment" implemented with members of the general public, these models will allow us to measure the benefits and costs of these management options, and to study which policy options offer best value for public money. An important aspect of the modelling work will be to show the spatial interactions between forests which are key to understanding disease spread. We will also study the implications of uncertainty on the part of forest managers over the effects of their actions on the best choice of option, using a real options approach.

Potential applications and benefits
The project will provide results that will be of use to forest managers and to government agencies and departments which are concerned with tree and plant diseases and forest management. A risk ranking of forest management options, and a cost-benefit ranking of management options will help improve the quality of forest disease management and policy design. A wide range of stakeholders will be included in the project's knowledge exchange activities to ensure the widest application of these insights. The project will also advance scientific methods in the study of invasive diseases which are linked to land management.

Technical Summary

Our proposal aims to construct a new generation of models built upon an integration of epidemiological, ecological and economic factors that together determine the resilience of forest systems to incursions by infectious agents, and the supply of forest ecosystem services. It will allow the economic costs and benefits of disease response strategies to be evaluated from the perspectives of both private landowners and public policy choice. A key underlying idea behind the modelling is to characterise forest management options in terms of resistance and resilience to disease. Our starting point is a combination of an optimal control model and a form of discounted net benefit tailored towards multiple outcomes of forestry (timber, C sequestration and biodiversity). The model will be extended to include multiple forest types and multiple outcomes. A decision maker will have a choice of different forest types to be established in different locations. Subsequently we will develop a model incorporating multiple forest owners, where each agent is a land manager who controls a number of patches. A combination of multiple objectives and management strategies with spatial interactions across a potentially heterogeneous landscape results in an analytically intractable model and we will therefore use an agent-based model. We will also develop a real-options approach to explore how the inclusion of uncertainty and options changes the management strategies emerging from the optimal control and agent-based models described above. In order to fully parameterise the model particularly for non-market environmental costs and benefits, we will undertake a choice modelling exercise to measure the public's willingness to pay for disease mitigation and the value of avoided damages. Finally, we will use the modelling framework developed in the project to rank the policy options available to both the individual manager and to the government organisations.

Planned Impact

Who will benefit?
We anticipate that the following groups will benefit from our work: (i) The private forestry sector, including both large international firms such as UPM Tilhill, and smaller scale forest owners and managers, along with the tree nursery sector and private estates; (ii) The voluntary forestry sector, including the Woodland Trust and forest-owning charities such as RSPB and National Trust, and county Wildlife Trusts; (iii) Government policy makers such as DEFRA, FERA, the Scottish Government and the Welsh Government; (iv) The Forestry Commission, Natural Resources Wales, Scottish Natural Heritage, and Forest Research; (v) The scientific and economics research communities internationally; (vi) Members of the public who care about woodlands and forests.

How will they benefit?
The high current profile of tree diseases in the UK and internationally has created an unprecedented level of attention from a wide range of stakeholders in government, agencies and the private sector. We will early disseminate the message that inter-disciplinary ecological and economic research has the potential to provide a significant advance in the evidence required to improve the management of forest systems for disease prevention, control and mitigation. We will use key contacts as well as more open media to actively engage with these sectors at the start of the project and invite broad participation at the project stakeholder workshops.
For the private forestry sector, the main benefits will be in terms of new knowledge about the effectiveness of management options at the level of individual forests and the landscape. The voluntary forestry sector also faces similar problems with regard to disease risk and the need for spatially-coordinated management. Provision of a ranking of management options according to their impact on resistance and resilience will also be of use to private forest managers in terms of which kinds of forest they prioritise for investments.
Governments will potentially benefit from new evidence of the spatial- and cost-effectiveness of policy options, along with information on how the public prioritises? and values government interventions in disease control and mitigation. Government can use the ranking of management options in deciding details of advisory and funding schemes for changes in forestry practice.
The Forestry Commission, Natural Resources Wales, Scottish Natural Heritage, and Forest Research will benefit through the generation of insights into the social benefits and costs of alternative management interventions and from improved knowledge of new research and modelling methods acquired through their close participation in the project work. The spatial modelling work will also assist these bodies in understanding how economic incentives combine with environmental and epidemiological factors to determine the spread of disease. The generic modelling approach will have impact on understanding of diseases in other land-use sectors.
The international research community will benefit from our fundamental research, development of models, and provision of novel data. We will share our data primarily via EIDC. We will engage with the international research community via publications in high profile journals and contributions at international conferences and workshops; we will organise an international conference to communicate the results to a wide interdisciplinary audience.
Serious forest dieback, and control through tree felling will have a direct impact on wider society through reduced provision of ecosystem services and some control measures may necessitate severe restrictions on public access to forest landscapes. Members of the public will therefore gain in that our research will facilitate more effective interventions for tree diseases, which could reduce the significant losses in ecosystem services, and restrict the geographical areas that need to be subjected to drastic control measures.
Description The uncertainty in the future spread of an infection means that it is optimal to wait in order to learn more rather than adopting control measures immediately.
How long a forest manager should wait before adopting control measures depends both on the objectives of the manager and the extent to which the infection reduces the different benefits provided by the forest.
At the landscape scale this leads to variability in the timing of control adoption across multiple forests with different owners.
Subsidy schemes can be used to align the timing of disease control measures for managers with different objectives.
We have produced a summary review that outlines the effect of forest management options on forest resilient pathogens. This review is of broad use to policymakers and to scientists involved in the selection and planning of resources for developing future forests and is particularly timely given the commitment by the UK government to increase significantly the area of land planted to trees.
Exploitation Route This work could be extended to incorporate strategic interactions between two forest managers with differing objectives, so as to investigate how a managers decision of whether or not to adopt control impacts the timing of control adoption for a neighbouring forest manager with different objectives.
Sectors Agriculture, Food and Drink,Environment

Description One of the major challenges in disease control for major epidemics at the landscape scale involves a decision of when, if at all, to apply the measure of control. Too soon risks wasting resource if the epidemic is likely to fade out; too late leads to a failure in control because the epidemic has by then moved beyond the zone of feasible economic control. The decision is complicated by the inherent uncertainty of epidemics. Our methods, derived and adapted from the fields of stochastic differential equations and financial mathematics provide a tool to aid policy makers and regulators make more informed decisions based on the available information for an epidemic.
First Year Of Impact 2016
Sector Environment
Impact Types Economic

Description The work developed in the project has led to advice on disease outbreaks in the UK, US and multiple countries in Africa
Geographic Reach Multiple continents/international 
Policy Influence Type Participation in a national consultation
Impact Provided UK Government with outputs for predictive models to assess when, where and how it is appropriate to deploy resources to manage emerging epidemics of plant disease and plant pests in the natural and semi-natural environments. Work has informed Government policy. Also provided warning systems for spread of wheat rust diseases in East African countries and risks to Middle East and South Asia
Description Optimal timing of disease control 
Organisation University of Warwick
Department Warwick Medical School
Country United Kingdom 
Sector Academic/University 
PI Contribution We have provided expertise on epidemiological modelling and also led on the construction of the model and the development of the computer simulations. We also led on the analysis of the results as well as the writing up of the results for publication.
Collaborator Contribution Our partner provided expert knowledge on real options theory as well as the numerical and analytical methods used to solve such problems. The collaboration has been very important to ensure the accuracy of our model and simulation methods. Our partner has also been involved in the analysis and writing up of results for publication.
Impact Publication What a difference a stochastic process makes: epidemiological-based real options models of optimal treatment of disease (accepted March 2017, Environmental and Resource Economics Journal)
Start Year 2015
Description Podcase 
Form Of Engagement Activity A broadcast e.g. TV/radio/film/podcast (other than news/press)
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact We set up and run a podcast series (Greedy Planet, which investigates issues surrounding Food Security and Plant Biosecurity through interviews with academics. The purpose was to engage with the general public on a wide range of food security and plant biosecurity issues and publicise current research.
Year(s) Of Engagement Activity 2015,2016,2017
Description Science Festival stall 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Public/other audiences
Results and Impact We ran a stall in the Plant Sciences marquee where we provided demonstration of webidemics (an interactive tool developed by members of the research group) as well as displaying posters related to the work of the project. Throughout the day we showed members of the general public how modelling can be used to help control tree disease which sparked their interest in the importance and challenges of controlling tree disease.
Year(s) Of Engagement Activity 2015,2016
Description Tree Health Modelling Workshop 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact We co-organised a one day interdisciplinary workshop (attendees included academics, policy makers and relevant stakeholders) to present findings of the project, hear about results from relevant projects and discuss components of tree health modelling that would be relevant for stakeholders. The outcome was that we identified key questions that were of interest to policy makers and stakeholders as well as communicating our results to relevant members of the forestry community.
Year(s) Of Engagement Activity 2016
Description Twitter account 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact We created and help maintain a twitter account for the project which we use to publicise information on workshops, publications and key project findings. This has helped us to engage with relevant stakeholders such as Sylva Foundation and Royal Forestry Society.
Year(s) Of Engagement Activity 2015,2016,2017