IAP-24-001

Developing drone and satellite remote sensing to support catchment scale monitoring and management of invasive plant species.

Biological invasions and the impacts of invasive alien species are known to pose a significant threat to global biodiversity and the recent Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services assessment has highlighted that freshwater environments are particularly susceptible to invasions. The impacts of invasive species are linked to the declining conservation status of threatened species, and their accelerated spread into protected and biodiverse habitats are a major global concern. Riparian habitats are one of the richest and most complex habitats globally and the widespread establishment of invasive alien species pose additional challenges to their unique flora and fauna, as well as ecosystem function. Management measures for widespread well established plant species include physical removal which is resource intensive or the use of herbicides which is problematic near watercourses so novel control methods are needed. Challenges of monitoring these less accessible and structurally complex systems are also considerable.

Himalayan balsam, Impatiens glandulifera, is an annual plant native to the western Himalayas, prolific along waterways. It is listed under Schedule 9 of the Wildlife and Countryside Act 1981, therefore it is an offence to plant or otherwise cause this species to grow in the wild. Himalayan balsam, by forming tall, dense colonies, shades out and results in the extirpation of native vegetation. This species is notoriously difficult to manage, particularly if an adjacent water way provides a constant input of viable seeds. In 2014, a biological control agent, rust fungus Puccinia komarovii var. glanduliferae var. nov, was approved for release in the EU. The rust fungus was deployed at various sites from 2015 to better manage I. glandulifera in the UK by the Centre for Agriculture and Bioscience International (CABI).

CABI found initial levels of infection from inoculated populations of I. glandulifera to be variable, an a greater understanding of the environmental conditions that lead to successful establishment overwintering are needed. Where the rust fungus may spread naturally from the initial release site, little is known of the timing and efficacy of natural spread. The temporal and spatial scales over which spread occurs make ground survey time consuming and expensive, even where areas are accessible. Where establishment of the rust fungus does occur, novel modelling and cost-effective monitoring are required to assess the spatial extent of the control method. New techniques to such as UAV and satellite based remote sensing have the potential to provide information on spatial structure of invasions, needed for monitoring and control.

The aim of this project is to assess the in-situ drivers of rust fungus infection success of I. glandulifera. It will develop a remote sensing based monitoring method using UAVs and satellite data to map spread and control, combining this with a modelling framework to determine the natural dispersal efficacy of the rust fungus to other invaded sites.
Specifically, the project objectives are to:
1) Determine the key environmental variables that impact the efficacy of rust fungus infection and survival in-situ.
2) Collect remotely sensed datasets with differing spectral, spatial and temporal resolutions, to develop a cost-effective monitoring method for I. glandulifera at multiple scales.
3) Determine the efficacy of local inoculations using UAV imagery in assessing reduction in population size of infected I. glandulifera populations and determining infection status of I. glandulifera populations,
4) Assess the natural dispersal capabilities of the rust fungus using a modelling approach

The project will be carried out in collaboration with CABI and Tweed Forum who are conducting rust fungus release trials.
1) The student will undertake a full site assessment, mapping key topographical features of the invaded riparian sites and catalogue flora present at the site using botanical survey methodology.
2) Environmental conditions (monitored with data loggers) conducive to the rust fungus growth and infection rate will be modelled over time.
3) Initial ground surveys will be replicated using varied UAV flights to establish the most appropriate contributions of multi-spectral and lidar drone platforms.
4) The student will complete fieldwork to monitor rust fungus efficacy via infection rate, documenting I. glandulifera population size, using UAV and/or satellite imagery. These images will also be used to develop an approach for identifying I. glandulifera populations that have been infected by the rust fungus, using ground truthing surveys.
6) To assess the natural dispersal potential of the rust fungus to other invaded riparian sites, the student will develop a spread model, considering fluvial connectivity of the catchment area and populations of I. glandulifera in the catchment.

Year 1

Literature review as a meta-analysis to assess biological control efficacy globally
(months1-4); Satellite imagery downloads and drone flight course (months 5-6); prepping for fieldwork, mapping site features; spring community and success of infection survey (month 7-9); summer community and environmental conditions survey (months 11-12)

Year 2

Data management and modelling method assessment (months 13-16); image processing and protocol development (months 17-18); spring
community assessment (month 19); UAV flights and data acquisition
(months 20-22); summer community and environmental conditions, ground truthing surveys (months 23-24)

Year 3

Analyses of UAV imagery (Months 25); Data analyses, model development, writing up
results (months 26-30); Writing thesis chapters, attendance at an international
conference (months 31-36)

Year 3.5

Writing publications and thesis submission (months 37-42)

The student will receive training from an interdisciplinary supervisory team,
particularly in some of the key NERC most wanted skills:
*Fieldwork: the large fieldwork element of this PhD means the student will be exposed to a variety of sampling and experimental techniques in the field
*Taxonomic Identification: to create an inventory of fauna and flora the student will
receive species ID training, particularly botanical.
*Data management and modelling: the student will be part of the Modelling, Evidence and Policy Research group at Newcastle and as such will have the opportunity to learn an array of modelling methods best suited to their data.
*Translating research into practice: As the output of this PhD will be integral to management of invasive species, the student will receive training in science communication to multiple audiences (e.g. policy makers, non-governmental organisations (NGOs) and the wider public).
The student will also participate in IAPETUS training and events. A training budget is included for any external training required by the student