New approaches for the early detection of tree health pests and pathogens

Lead Research Organisation: STFC - Laboratories
Department Name: RAL Space


The UK's forests, woods and trees are under threat from a growing number of pests and diseases. Many of these threats are alien; historically not present in the UK and having been introduced from overseas. Some of these threats may reach the UK naturally i.e. as wind-borne spores from continental Europe; potentially one pathway for introduction of the disease ash die-back. The alternative and probably more common pathway of introduction is via human activity, especially trade; for example moving infected plants (another pathway identified for ash die-back) or through the shipping of goods associated with infested timber (as was the case with the recent introduction of the Asian long-horn beetle into Kent in packaging crates for stone). These cases clearly demonstrate that we need to do more to improve our nation's biosecurity and protect our plants and trees; both cultivated and in the wider environment.
In order to do this we need better methods for detecting these pests and diseases that allow us to find them earlier and with greater efficiency. By detecting these threats earlier you can minimize the damage they cause, by either preventing an outbreak occurring in the first place or by finding it early and then stopping it from establishing and spreading further. At present we rely on trained inspectors to find these alien pests and pathogens, mainly via visual inspections of imported plants and plant-based products e.g. timber. However, given the volume of inspections required, the finite amount of resource available and the huge practical challenges associated with these inspections, this task is extremely difficult and the efficiency of detection is low.
This project is designed to change that situation by providing better methods for detecting tree pests and pathogens; both moving in trade and in the environment. It will look at new technologies for the detecting changes in infected plants; using either 'sniffer' technology to identify differences in the volatile chemicals given off by diseased and healthy plants or imaging techniques that can detect changes beyond the range of human vision. It will also look at developing and designing novel traps for capturing insects and DNA-based detection approaches that look for air- and water-borne pathogens. This will include better approaches for trapping spores and then applying high-throughput sequencing methods that will allow the identification of not only known pathogens but also new ones too.
However, developing these new technologies is only part of the challenge. It is also necessary to make sure these new methods are fit-for-purpose and that they work in a way that meets the needs of those enforcing tree health regulations (e.g. government), those upon who those regulation impact (e.g. woodland owners and industry) and the end-users who would be expected to use these new tools (e.g. inspectors in the field). We will also examine what type of end-users could be involved; this could be trained government inspectors (the traditional approach) or alternatives such as those working in the industry, volunteers or even the general public. So looking to see if a so-called 'citizen science' approach could be used for any of these new approaches.
It is also important to ensure that these new approaches can be deployed effectively, for example at locations that pose the greatest risk, and in a way that offers the best cost-benefit (i.e. the best balance between cost of using the technology and the improvements it can offer in terms of better pest and disease detection). In order to do this, we will take an interdisciplinary approach; getting experts from many different fields e.g. biology, mathematics, chemistry, engineering, physics, economics and social science, to work together to come up with the best overall solution that works technically, economically and socially.

Technical Summary

This project has 6 work packages (WP), each based around a different combination of skills and expertise. For WP2-6 there will be a focus on a particular detection technology, while WP1 will provide the technical oversight needed for effective deployment of these different technologies, as summarised:
WP1:a participatory interdisciplinary approach will be used to evaluate the needs of stakeholders and to ensure that the technologies meet these. It will also focus on the requirements of effective technology deployment, using mathematical modeling to develop sampling strategies, to create network-based risk maps and economic assessments of cost-effectiveness. Further aspects of deployment will be analysed using social science approaches including end-user acceptability and the potential for using citizen science.
WP2:analytical chemistry approaches will be used to identify diagnostic volatile organic compounds produced by pests, pathogens and diseased hosts and to translate these onto commercial-available portable platforms for use by inspectors in the field.
WP3:multispectral imaging will be used to identify markers for the early detection of biotic/abiotic stress in plants. A prototype bioimaging camera will be constructed that can be used to validate this approach in the field.
WP4:will develop mathematical models of spore movement and investigate metagenomics for broad-spectrum surveillance utilizing existing monitoring networks e.g. pollen traps. In addition, a novel integrated cyclone-based trapping and molecular detection system will be developed and evaluated.
WP5:novel semiochemical attractants will be identified for a range of wood-boring beetle pests, incorporated into traps designed for efficient detection and then deployed in a risk-based network.
WP6:methods for sampling and rapid screening water for Phytopthora spp., including 'unknowns' will be developed and validated. This will combine high-throughput sequencing with a rapid bioinformatic.

Planned Impact

The interdisciplinary design of this proposal will ensure maximum ongoing impact. Central to this is stakeholder engagement and our proposal has adopted a novel approach to facilliate this. Traditional approaches to developing new detection or diagnostic technologies have assumed the 'build it and they will come' approach; where the focus is on the technical aspects of the novel methodology, rather than the needs of end-users and the specifics of how it will be effectively deployed. This proposal reverses that by taking an inclusive view of what is required to achieve a successful outcome i.e. the deployment of a new technology that improves our biosecurity, and then co-designs technologies which fit that purpose. It achieves this by embracing an interdisciplinary approach and through establishing early engagement with stakeholders and end-users. Critical to this is the creation of a Learning Platform (Work package 1) which sits at the core of the project and cuts across the other technology-driven work packages (WPs 2-6). This platform will create communication channels, facilitate collaboration and knowledge sharing across work packages and stakeholder groups, actively disseminating project outcomes and enabling the pathways to impact. This will be delivered as a series of workshops; both cross-cutting (looking at the broader issues associated with detection and its successful deployment) and more focused (looking at specific issues associated with a particular technology and the contexts for its use). In addition to interacting with stakeholders (e.g. policy-makers, inspectors, NGOs, industry), this approach will use the breadth of expertise established within the consortium and assembled from across a wide-range of disciplines. This brings together 'technology-owners' (natural and physical scientists) with 'technology-evaluators' (mathematical and social sciences) to ensure that the best technological approaches are married with suitable sampling and risk-based deployment strategies, that they have stakeholder acceptability and offer genuine cost-efficiency benefits to public and private stakeholders alike.

In addition to the novel approach built into the project design, the effective delivery of impact will also benefit from a consortium which has an extremely strong track record of delivering translation science, to policy and industry alike. As government science agencies, the major remit for both Fera and Forest Research is to take science and technology and to translate it into policy-focused tools and evidence. This is a role they provide routinely for Defra and Forestry Commission, and their associated inspectors on the frontline in the field (e.g. Fera PHSI and FC Inspectors). In terms of delivery of technologies to end-users including industry, there is also a strong track record across the consortium in a whole range of contexts e.g. Worcester (horticulture industry diagnostics), JHI (potato industry diagnostics), Greenwich (pest trap deployment) and Fera (field diagnostics deployment). The integration of a number of SMEs within the consortium is another pathway to impact; providing a route for new technologies to be made freely available beyond the end of the project. Finally as plant and tree health sits within a European regulatory framework, the ability to engage with international partners and stakeholders is important. The consortium has a wealth of experience and contacts in this area, in particular through its central role in a range of related EU-funded projects e.g. Q-Detect (Fera-led), ISEFOR (Aberdeen-led) and PERMIT (FR-led). It will also build upon existing systems for knowledge exchange within our region, especially through the use of the European Plant Protection Organization (EPPO). By working with EPPO, using activities such as its workshops and conferences, we will be able to reach out to tree health practitioners across Europe; in many cases the real frontline for UK biosecurity.


10 25 50
Description We have developed a hyperspectral imaging camera system for the spectral analysis and the generation of a spectral library of disease dothistroma in Scots pine. We have developed a data set of key spectral features in artificially inoculated and naturally infected samples to determine stressor associated due to biotic or abiotic factors. These have been combined into a complete handheld system for in field analysis
Exploitation Route From this reserch we have been successful in two further research projects - ST/N006801/1 with Newcastle university and ST/P007066/1
Sectors Aerospace, Defence and Marine,Agriculture, Food and Drink,Environment,Manufacturing, including Industrial Biotechology,Culture, Heritage, Museums and Collections,Security and Diplomacy

Title hyperspectral imaging camera 
Description direct spectral imaging for the analysis of biotic and abiotic stresses in plants and leaf samples 
Type Of Material Biological samples 
Provided To Others? No  
Impact The technology is under development 
Title Controlled growth study 
Description Collection of spectral data from needles collected from Pine saplings grown in a controlled environment. Some of the saplings were infected with Dothistroma, while others were left uninfected as controls. Data set can be used to see the changes with time as the disease progresses. The disease status of the needles was confirmed with PCR genetic testing. 
Type Of Material Database/Collection of data 
Year Produced 2016 
Provided To Others? No  
Impact We have used this dataset to simulate hyperspectral data and to build models to predict the disease status of a Pine sapling. Also to get an idea of when spectral methods can be used to detect Dothistroma in a sample. 
Description Detector manufacturer Ximea 
Organisation XMOS
Country United Kingdom 
Sector Private 
PI Contribution We are jointly development of the control software and analysis of data from the detectors that we are using
Collaborator Contribution Iterative software and firmware upgrades based on our input to their detectors
Impact Knowledge exchange
Start Year 2015
Description Development of Statistical Analysis Techniques 
Organisation University of Oxford
Department Department of Statistics
Country United Kingdom 
Sector Academic/University 
PI Contribution We have been able to provide spectral and hyperspectral data from healthy and diseased trees under controlled conditions. As well as time to carry out statistical analyses to refine and develop robust statistical models
Collaborator Contribution Dr Dan Lund from Department of Statistics, University of Oxford has been able to provide his expert opinion on the best way to analyse our data to develop the robust models to predict tree health
Impact development of statistical models for the detection of tree health. Multi-disciplinary collaboration between statistics and chemistry/physics
Start Year 2015
Title Hyperspectral imaging software 
Description collecting and processing hyperspectral images from Ximea cameras using Labview 
Type Of Technology Software 
Year Produced 2017 
Impact we are able to collect and process the images from our Ximea hyperspectral cameras without having to use proprietary software 
Description Learning Labs 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Industry/Business
Results and Impact Visiting potential end stakeholders for which our research with hyperspectral cameras could be useful, gave us opportunities to discuss their problems and our research techniques
Year(s) Of Engagement Activity 2015
Description User engagement meeting was set up and attended by the water companies that would use the final instrument 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Industry/Business
Results and Impact A meeting was set up between the project partners and water industries.
Year(s) Of Engagement Activity 2014,2015