Dr Amy Hinsley, Dr Cait Newport, and Dr Lydia France have joined the Eric and Wendy Schmidt AI in Science Postdoctoral Fellowship. A programme of Schmidt Sciences, the interdisciplinary programme aims to accelerate the next scientific revolution by supporting talented postdoctoral researchers to apply AI techniques across the natural sciences, engineering, and mathematical sciences.
Predicting trends in the illegal and unsustainable wildlife trade
The illegal and unsustainable wildlife trade is complex and trends are difficult to predict. Trends change, and poor-quality and incomplete data make it difficult to anticipate where and when shifts in species, products, or trade methods may appear. Amy’s research will focus on filling this knowledge gap with ‘nowcasting’.
Like forecasting but for real-time predictions, nowcasting is useful when lag-times in data availability hinder effective decision-making. Official wildlife trade statistics are published with 1–2-year delays, so threats are often identified after they happen. However, nowcasting has not been used for global wildlife markets until now. Amy will develop an early-warning system to aid more effective, proactive approaches to reducing threats to wild species.
Through applying machine learning to this challenge, the project will provide insights into emerging wildlife trade threats and allow them to be tackled in real time.
Investigating how fish navigate their environment
Understanding how animals move in their natural environments is key to understanding many of their behaviours. Cait’s research will focus on how wild fish navigate visually in 3D through reefs, including how water clarity impacts behaviour.
With video photogrammetry animals can be tracked, and their environments mapped, in 3D. While relatively easy on land, in underwater environments it needs specialist hardware and/or bespoke software; this has prevented researchers from exploring questions around social, homing, predatory, and foraging behaviours underwater.
Cait will develop new open-source computer vision tools for 3D reconstruction of aquatic animal movements and the underwater environments in which they move, using low-cost consumer-grade hardware – transforming how biologists measure and monitor the movements and behaviour of wild aquatic animals.
Exploring the underpinnings of bird flight
When flying, birds can react to unpredictable forces and obstacles, control extreme manoeuvres, and move efficiently when faced with physical constraints. Lydia’s research will focus on gaining insights from wing and tail movement in flight.
Complex dynamic motion is difficult to research. Wings moving in space create complex 3D paths, and each wing flap differs from the last – each a subtly different twisting path. While previous research has relied on simplifying data, Lydia will take a new, data-driven AI approach.
The project will find signatures in the motion of wings in flight and apply machine learning algorithms, which can uncover hidden information and find underlying physics in complex dynamical systems. The research will enable the development of methods that can be applied to other birds, to inform bioinspired designs and to study natural motion in general.
