Project Overview: This PhD project at the University of Southampton aims to advance our understanding of the Earth's magnetosphere-ionosphere (M-I) system by leveraging machine learning techniques. The M-I system is driven by its interaction with the solar wind through a process known as 'reconnection.' The project will use deep learning and data science methods to analyze extensive ionospheric radar datasets, with the goal of elucidating how solar wind conditions control the size and rate of this coupling process. Research Context: Understanding the M-I system is crucial for space weather research, which has significant implications for both space- and ground-based technologies. Ionospheric radar observations are uniquely suited to infer the global extent and rate of reconnection, providing key insights into how the M-I system responds to solar wind driving. Despite recent advances, fundamental dependencies—such as how upstream solar wind conditions control the spatial extent of reconnection—remain poorly understood. This project builds on recent work and aims to address these questions through large-scale statistical studies enabled by advanced data science techniques. Methodology: The student will develop automated algorithms to identify reconnection events in ionospheric radar data, enabling transformative statistical studies of M-I driving. The project will involve training and validating deep learning models to detect subtle spatial-temporal patterns in noisy data, followed by statistical analysis to determine the response of the M-I system to different solar wind conditions. The research will be conducted in collaboration with the Space Environment Physics group (School of Physics & Astronomy) and the Vision, Learning and Control group (School of Electronics and Computer Science). Training and Development: The IGNITE programme offers comprehensive personal and professional development, including subject-specific training in solar-terrestrial physics, deep learning frameworks, and high-performance computing. Students will participate in group meetings, journal clubs, and present findings at national and international conferences. Additional opportunities include postgraduate lecture series, research seminars, and the chance to work as a lab demonstrator. Programming skills development is integral to the project. Funding: The IGNITE Doctoral Landscape Award provides full funding for 3.5 years, including a tax-free stipend at the UKRI standard rate (£20,780 for 2025/26), Home tuition fees, and a Research Training Support Grant (£2,200 per year). The University waives the difference between Home and International tuition fees, making the award accessible to both UK and international students (with a cap on international places). Eligibility: Applicants should have a UK bachelor’s degree with upper second-class honours or higher in a relevant subject (international equivalents accepted). English language proficiency is required (IELTS 6.5 overall, minimum 6.0 in all components, or equivalent). Strong programming skills and an interest in machine learning, data science, and space physics are highly desirable. Application Process: Applications must be submitted by 11:59pm on 8 January 2026. Prospective students are encouraged to contact the lead supervisor to discuss suitability before applying. Applicants may apply for up to two projects but are advised to tailor their application to one. The IGNITE programme supports diversity and offers a guaranteed interview scheme for qualifying UK applicants from racially minoritised backgrounds. For more information, visit the project and group pages: Space Environment Physics Group Vision, Learning and Control Group Project Listing