NERC E4 What is required to declare an eradication? Using simulation models to inform current attempts to eradicate invasive Australian Acacias from South Africa

Interested individuals must follow the "how to apply" link on the Geosciences E4 Doctoral Training Partnership web page: http://www.ed.ac.uk/e4-dtp/how-to-apply

Summary

The aim of this project is to inform attempts to eradicate populations of nine species of Australian Acacias from South Africa by simulating both acacia populations and an observer with limited knowledge of the system.

Project background

Invasive alien species that have been transported beyond their native range by humans are amongst the main threats to biodiversity and cost billions of dollars each year. Australian Acacia species are amongst the most damaging invasive species affecting biodiversity in one of the most important florally megadiverse countries in the world—South Africa.

Although prevention is the most efficient approach against biological invasions, once a population of invasive species has become established, eradication is one of the few interventions in conservation that can permanently improve biodiversity.

In practice, it is difficult to know when eradication has been successful, particularly for plant populations that possess significant seed banks that can persist in the ground for a long time, and play an important role for population dynamics.

Current tools for planning the management of Australian acacias rely on guess estimates of invaded area, combined with practical but non-optimal monitoring approaches. Due to the large uncertainty in the true extent of invasions in real systems, estimating the efficacy of such tools is complex.

This PhD project will explore how spatially explicit, simulation population models for Australian acacia species that account for the seed bank can be used to assess and improve the efficacy of existing management strategies. While focussing on a specific system, this project offers an opportunity to address fundamental ecological questions concerning the dynamics and monitoring of biological invasions, with direct real-world application to a pressing conservation problem.  

The PhD will be based at the University of Edinburgh with potential field work in South Africa, to collaborate with those working on the ground and to assess the feasibility of management recommendations.

Research questions

This project will aim at improving management practices for invasive plant species eradication (using Australian acacias in South Africa as a case study) by addressing the following research questions:

• What is an optimum eradication strategy for invasive plant species with abundant seed banks under perfect knowledge and different constraints?
• How does imperfect knowledge about invasive plant population distribution and abundance impact our ability to eradicate them?
• Are existing heuristic tools for the management of invasive plant species appropriate to guide knowledge acquisition and optimise eradication?

Methodology

A “virtual ecologist” approach will be used, in which both the system of interest (Acacia populations) and the observer’s acquisition of information about the system are modelled. In such a model, virtual data is generated by simulating (a) a virtual ecological model which includes key processes of the ecological system, (b) a virtual sampling model mimicking the observation procedure, and (c) the methodological tools used to analyse the ‘virtually’ observed data. Results are then evaluated against ‘true’ simulated data.

The virtual ecologist approach will allow us to account explicitly for the effect of imperfect knowledge and to assess the gains obtained from improving monitoring of biological invasions, an aspect that is often neglected in simulation modelling approaches.

The model will make use of data gathered over the past decade, and inform how data should be collected in future. There is potential for the candidate to join in field work that uses fake plants to assess how well human observers can detect specific plants - providing data that should improve the realism of the virtual ecologist model.

Year 1: 

• Literature review on invasive species and their management. You will become familiar with the concept of invasive alien species, how they are introduced and spread into novel environments, their potential impacts on biodiversity and human livelihood, and how they can be managed.
• Implement a population model of Acacia incorporating seed bank dynamics and explore management approaches.
• One scientific paper presenting eradication strategy for invasive Acacia under perfect knowledge and different constraints.

Year 2: 

• International experience visiting co-supervisor and research partners in South Africa. Being a modelling project, it will be possible to do it without going to South Africa if the pandemic situation prevents it. However, since it would be very beneficial for you to better know the modelled system, and to meet and collaborate directly with local researchers and managers, we hope this will be possible!
• Define a modelling protocol to simulate a virtual ecologist in the population model and account for imperfect knowledge about the system.
• Explore model sensitivity to imperfect knowledge.
• One scientific paper in an international journal presenting these results.

Year 3: 

• Implement existing heuristic tools and adaptive management approaches for the management of Acacia in the population model. Based on the literature, you will explore how existing approaches to acquire knowledge about invasive species population perform to improve their management.
• You will explore the possibility to improve such tools and approaches.
• One scientific paper in an international journal presenting these results.

Training

The recruited student will gain skills in population ecology, invasion science, population management, data analysis, statistical modelling, programming, and training in science best practice including FAIR data principles and science communication. The PhD programme will include some international collaboration and the recruited student will have the opportunity to develop their own targeted research questions. A comprehensive training programme will be provided comprising both specialist scientific training and generic transferable and professional skills.

Requirements

We seek am enthusiastic and motivated student with a background in ecology and quantitative analyses. Applicants should have very good quantitative and mathematical abilities. Good programming skills are essential, particularly in R, but other programming languages are also appropriate. Good statistical skills are recommended. And knowledge of or a willingness to learn version control in GitHub and open science best practice is encouraged.