Project outline

Integrating advanced climate projections with machine learning techniques represents a transformative approach to addressing the challenges posed by climate change to global food security. This innovative methodology promises to enhance our ability to predict, understand, and mitigate food security risks, ultimately contributing to the development of resilient agricultural systems and the sustainability of food supplies in the face of a changing climate in Australia.

The primary objectives of this research are:

• To develop predictive models that integrate climate change projections with machine learning algorithms to assess future food security risks
• To identify the most vulnerable regions and crops by analysing model outputs
• To propose data-driven adaptive strategies and policies that enhance the resilience of food systems.

Research Questions

• How can machine learning algorithms be integrated with climate change projections to predict food security risks?
• What are the predicted impacts of climate change on agricultural productivity and food security for different regions and crops?
• Which adaptive strategies are most effective based on predictive model outcomes?

Hypotheses

• Integrating climate change projections with machine learning models enhances the accuracy of food security risk predictions.
• The impacts of climate change on food security will vary significantly across regions and crops, with some areas and crops being more vulnerable.
• Data-driven adaptive strategies derived from predictive models will improve the resilience of food systems to climate change.

Methodology

 1. Data Collection

• Climate projections: Data from global climate models (e.g., CMIP6) under different emission scenarios.
• Agricultural data: Historical crop yield, soil quality, and land use data from agricultural databases.
• Socioeconomic data: Information on population growth, economic conditions, and food distribution networks.

2. Machine Learning Models
Data preprocessing: Cleaning, normalizing, and integrating datasets.
Model selection: Evaluating different machine learning algorithms (e.g., random forests, neural networks, support vector machines) for predictive accuracy.
Training and validation: Using historical data to train models and validate their performance.

3. Scenario Analysis
Developing future scenarios based on different climate projections and socioeconomic pathways.
Running predictive models to assess food security risks under each scenario.

4. Case Studies
Selecting regions with diverse climatic conditions and agricultural practices.
Conducting in-depth analyses to validate model predictions and explore local adaptation strategies.
 

The student will learn:

• development of robust predictive models that integrate climate change projections with machine learning techniques, offering detailed insights into future food security risks in Australia
• comprehensive identification of regions and crops most vulnerable to climate change, facilitating targeted adaptation efforts
• data-driven policy recommendations that support the development of resilient food systems in Australia, informed by predictive modelling outcomes.

To register an expression of interest, click here. You will need to outline why you have selected the research project and how your skills, experience and/or knowledge meet the project requirements.