Earth observation for the improved management of tree health in urban and peri-urban settings

Trees and forests are a key provider of multiple ecosystem services to the UK, as well as having commercial and cultural value. However, UK tree and forest health is under pressure both from climate change and from introduced pests and diseases. While commercial forestry operations are fairly well mapped, smaller woodland blocks and individual trees in agricultural, private, urban and peri-urban settings are less well recorded. In order to understand the sensitivity and resilience of such trees and woodlands to potential pressures on tree health it is important to understand their location and composition. A recent report by the Tree Council into the national response to ash dieback highlighted the problem of not having a national tree inventory, with estimates of the UK non-woodland ash tree population ranging between 27 and 60 million trees.
The overall objective of this project is to develop a scalable tree species distribution map for urban and peri-urban environments to support policy decisions related to managing tree health. Such a tree map would be of use to multiple stakeholders who benefit from ecosystem services derived. These range from national and local government, to conservation charities and academia, to other managers of estates (in the broadest sense which includes business centres, education facilities, NHS etc.) and utilities that have to manage the tree stock. This project aims to produce these maps by analysis of high resolution satellite imagery from the Copernicus Programme Sentinel satellites trained on very high resolution imagery from unmanned aerial vehicles (UAVs) and ground based surveys. The second half of the project will focus on the delivery of these maps using web technologies, which would include contextual mapping of ecosystem service provision, cultural value and sensitivity to pest, disease and climate pressures.
The commercial product that will be developed by this project is a mapping service. Ultimately we would see this developing into a national woodland map. However, the output for this project will be a proof of concept map for a single council area. This will be delivered as a licenced web mapping service which can be consumed by bespoke applications or off-the-shelf GIS software. On top of this we will build web and native OS applications that display not only the tree species distribution maps, but also allow visualisation of socio-economic values for amenity woodland and interaction with the woodland mapping dataset.
The key stakeholders for tree distribution mapping service are Defra, the Animal and Plant Health Agency, local government agencies, charities and community groups concerned with trees and tree health. If future responses to invasive pests and diseases follow the same model as the response to ash die back, then the management of infected/infested trees will be the responsibility of local authorities and action groups. Tree distribution maps based on the methods and technologies delivered by this project would provide local estimates of numbers and locations of different tree species, allowing for more effective short and long term planning to maintain tree health, a reduction in survey costs and a reduction in the time to implement management actions in response to pest or disease outbreaks.

For those organisations that have to manage tree stocks, currently there is a poor understanding of the composition and distribution of trees in the UK outside the larger well managed forestry or woodland blocks, and this is particularly the case for those smaller pockets of woodland in urban areas and on the urban fringe. At a local level there is significant variability in the knowledge about the tree stock which local authorities and councils are responsible for. However, if there is an outbreak or significant damage to trees from environmental causes (storms, drought, etc.) there is a significant financial cost to authorities in dealing with the affected trees. For example in Kent, one estimate made for the average cost of removing a tree that poses a risk to UK plant biosecurity or public health is £800, but this cost can be much higher if the tree is located in a difficult location (e.g. next to a main road). Further, the National Ecosystem Assessment Follow on work identified that trees close to people provide the greatest benefit to society. Thus we know the least about the potentially most valuable trees.
The development of a method to map trees at the city to county scale will help local authorities better manage their tree stocks. Having an amenity tree map which includes mapping of the potential values for individual trees or woodland will help local councils and other land managers to undertake long-term management to improve the resilience to climate change or pests and diseases. In the event of an incident (e.g. pest outbreak or storm), a tree distribution map will reduce response times and improve the efficacy of management programmes by aiding in targeting individual management actions. By continuing to add new local tree maps to the national collection, central government can begin to better understand the likely national impact on amenity trees from climate change or the introduction of new pests and diseases.
Urban and peri-urban trees have many more values associated with them than the timber and carbon prices by which the economic value of forestry is often measured. Amenity trees are often the trees that most people interact with on a daily basis, being the trees in their parks, on their streets and outside of their places of work. The value of these trees is significantly related to the local population who interact with them, and so by being able to plot the location and composition of tree species against local population we gain a greater understanding of variations in the importance of amenity trees. In a financial situation where resources to manage trees have to be prioritised, understanding the value placed upon a particular group to trees, along with the sensitivity and resilience of the trees to pressures will allow organisations to assign different management actions depending whether the best option is some form of preventative management or treating the problem once it occurs.
In the term of this project, impact will be realised through interaction with the end-user group and through the two-way exchange of knowledge between this project and the twinned Defra-funded project being delivered by Fera. However, the beneficiaries of the ultimate endpoint of this work will be the public, as the objective of the project is to produce products that aid in management that improves public safety, maintains and enhances the resilience of ecosystem goods obtained from the UK tree inventory, and helps to manage trees before they affect supply of services. The main route for these public goods is going to come from the use by government, utilities and infrastructure organisations to more efficiently manage tree stocks, leading to reduced management costs and minimising disruption.