PhD Studentship: Detecting Ecological Change in Intertidal Habitats Using Repeat Photography and Machine Learning

Intertidal habitats provide important ecosystem functions and services, supporting biodiversity, fisheries productivity, coastal resilience and cultural values. However, these dynamic environments are increasingly exposed to acute and chronic climate-driven pressures. Conventional ecological monitoring is constrained by cost, spatial coverage and sampling frequency and scalable approaches that can reliably detect ecological change are urgently needed.

This project will investigate whether repeat photography, combined with citizen science and machine learning, can provide robust monitoring of ecological change in intertidal habitats. Smartphone imagery can generate large-scale observations, but variation in viewpoint, lighting, tidal state, season and image quality presents a major challenge.

The project will address two questions:

1. How should repeat-photography surveys be designed to reliably detect ecological change in intertidal habitats?
2. Can machine-learning approaches account for environmental and observational variability, whilst robustly detecting and quantifying ecological change from repeat photographs?

The Doctoral Researcher will combine field data collection, statistical analysis and image-based AI methods, testing methodology across locations, seasons, and environmental conditions. Methods will be validated against the MarClim 25-year time series, comprising annual surveys of rocky intertidal species across 100 UK sites. The project offers flexibility to shape its direction, with scope to place greater emphasis on environmental monitoring, image analysis, or machine learning as the project develops and the Doctoral Researcher's interests evolve.

The project brings together expertise from four leading marine research and higher education institutions in the South West, providing access to complementary expertise, facilities and networks. The Doctoral Researcher will gain training in fieldwork, experimental design, image analysis, machine learning and ecological monitoring, alongside statistics, scientific writing and communication. They will also benefit from stakeholder engagement, established monitoring programmes, and GW4+ and University of Exeter researcher-development opportunities. The project will integrate marine ecology, statistics and AI to develop innovative tools for addressing real-world ecological challenges.