Study level

  • PhD
  • Master of Philosophy
  • Honours
  • Vacation research experience scheme

Faculty/School

Topic status

We're looking for students to study this topic.

Research centre

Centre for Data Science

Supervisors

Dr Matthew Adams
Position
Research Fellow (DECRA)
Division / Faculty
Faculty of Science
Professor Chris Drovandi
Position
ARC Future Fellow
Division / Faculty
Faculty of Science
Dr Robert Salomone
Position
Senior Research Fellow in Data Science
Division / Faculty
Faculty of Science

Overview

This project seeks to use a combination of matrix algebra and advanced Monte Carlo methods for rare-event simulation to identify efficient algorithms for ensemble ecosystem modelling in large ecosystems. This project has the potential to significantly change the methods that researchers useto investigate complex ecosystems for the purposes of environmental management and future predictions.

Ensemble ecosystem modelling (EEM) is a novel and increasingly popular method for generating predictions of future species abundance in complex ecosystems represented by generalisations of the Lotka-Volterra equations. The method involves simulation of thousands of individual ecosystems with different random parameters. However, the parameters of the individual ecosystems must first be drawn according to a probability distribution that is conditional on the system being feasible and stable.

For a small number of species (<15), this procedure is straightforward: random parameters are generated repeatedly, and only tose which satisfy the constraint are used. In higher dimensions, the probability of obtaining a sample that satisfies the constraint becomes very small and EEM is thus computational infeasible. However, the criteria for ecosystem feasibility involve relatively simple matrix algebraic expressions which have yet to be exploited to improve sampling efficiency.

Research activities

  • Develop new statistical methods.
  • Code up and compare the methods.
  • Apply the methods to a case study (e.g. ecosystem models in MATLAB).

Outcomes

  • New statistical methods.
  • Development of expertise in multidisciplinary research, cutting across mathematics, ecology and statistics.
  • Write a paper to be published in the scientific literature.

There is also opportunity to collaborate with and/or present results to quantitative ecology researchers from external institutions.

Skills and experience

Excellent skills in coding, matrix algebra, statistical inference, and/or ordinary differential equations, an interest in using a combination of mathematics and statistics to address applied ecology problems, and a potential interest in working/communicating with a multidisciplinary team of researchers.

Scholarships

You may be eligible to apply for a research scholarship.

Explore our research scholarships

Keywords

Contact

Contact the supervisor for more information.