Undergraduate opportunities at UNSW
The ARC Centre of Excellence for the Weather of the 21st Century explores how Australia’s weather is being reshaped by climate change. We offer highly competitive scholarships intended to provide undergraduate students from Australian universities with an introduction to cutting-edge climate science and weather change research.
Students should be in their second, third, or post-honours year and interested in pursuing honours or a postgraduate degree in climate or weather change science. At UNSW, scholarship projects may either run on a full-time basis over the summer or other mid-semester/trimester breaks, or part-time for the equivalent of six weeks fulltime work throughout the academic year. The scholarships are valued at $3,800.
If you have any questions about our undergraduate research scholarships, please contact the Centre’s Associate Director Leadership and Training Melissa Hart.
To apply for an undergraduate research project, please complete this form, which is also available at the bottom of the page.
Machine learning for climate
Supervisor(s): Professor Steven Sherwood, Dr Abhnil Prasad, Dr David Fuchs
Description: Most of our understanding of changes in atmospheric temperature and wind come from reanalysis products, but these are problematic for looking at small changes over long time periods. The student will examine a new homogenised global radiosonde dataset for climate-change signals including trends in upper tropospheric temperatures and winds.
Experience required: Familiarity with Python will be required for this project
Processing urban morphology for computer vision applications in city’s weather and climate
Supervisor(s): Dr Negin Nazarian, Dr Jiachen Lu, Dr Sanaa Hobeichi
Description: Complex flow patterns within urban environments are significantly influenced by the diversity of urban layouts but have only been studied from generalizations based on conventional urban geometrical parameters in climate models. However, the inner- and ultra-variability of cities’ layouts including the street orientations, building shapes, and building height distributions challenge the generalization validity. Considering the scope of the study is for global cities, the validation work is better assisted by computer vision techniques that require a strong database of urban morphology. Based on the recent progress in satellite data processing (e.g., OpenStreetMap (OSM) and Microsoft Building Footprints) and building height estimation (World Settlement Footprint (WSF)), the high-resolution urban morphology is ready for this purpose. In this project, the selected student will learn and apply image pre-processing techniques for computer vision applications in weather and climate. The data produced will contribute to enhancing the understanding of urban heterogeneities’ impact on climate models. In this project, the student will be coding and adapting existing scripts.
Experience required: The applicant needs to have programming experience in Python to be successful.
Validating and preparing an Australian drought inventory for machine learning training
Supervisor(s): Dr Sanaa Hobeichi, Dr Elisabeth Vogel
Description: This project aims to validate a comprehensive drought inventory and prepare it for Machine Learning training. The inventory is compiled from over a hundred drought reports and climate statements and includes detailed information on the locations, times, and impacts of past droughts on Australian communities and ecosystems. Examples of documented impacts include statements such as “crops are cut for hay and silage”, “water supplies in major population centres have been affected”, and “inadequate water availability in the main storage dam”. The selected student will use various climate observations, such as streamflow and precipitation data from monitoring stations, crop yield datasets, and satellite-derived vegetation indices to validate the reported drought impacts. This validation process is essential to identify any erroneous information and ensure the accuracy of the drought database. The validated database will be a valuable resource for advancing drought research and developing accurate drought models using Machine Learning.
Experience required: Students need to have experience in Python or R programming to be considered for this project.
Investigating Extreme Rainfall in Townsville using Reanalysis Data
Supervisor(s): Dr. Leena Khadke and Prof. Jason Evans
Description: Townsville experienced record-breaking rainfall in February and March 2025, with 1,033 mm falling in just the first eight days of February. On March 18–19, the city recorded 301.4 mm in 24 hours which is the heaviest rainfall in 27 years, caused widespread flooding across north Queensland. In the present study, the student will explore the atmospheric and oceanic conditions that contributed to these extreme rainfall events. Using reanalysis data, the student will compare the behaviour of key atmospheric and oceanic variables such as sea surface temperature, outgoing longwave radiation, and atmospheric temperature against its climatology (1995 to 2024). The student will also examine wind patterns and vertical motion to understand how they influenced moisture transport and instability in the atmosphere. By analyzing these events, the study can identify the large-scale weather patterns responsible for the heavy rainfall. The study will help student to understand the mechanism and processes behind the occurrence of such extreme rainfall events.
Experience required: Students need to have experience in Python or MATLAB programming to be considered for this project.