Data Science | Biology
I am a scientist at The Open University interested in the intersection of data science and environmental biology. I like difficult challenges that need a mixture of high performance computing, statistics and domain knowledge of plant/pest science. My PhD was in Environmental Science at Lancaster University where I first became interested in analysis of image-type data. I then joined Professor Sarah Bridle at the Jodrell Bank Centre for Astrophysics to develop new techniques on the interface between Astrophysics and Remote Sensing. In 2020 I started work with Dr Kadmiel Maseyk at The Open University to apply machine learning techniques to huge hyperspectral datasets for understanding physiology and ecology of trees in urban environments.
My current research aims are to improve tree trait measurement and disease monitoring in the UK. We are working closely with an aerial survey provider, and other partners, to develop robust data processing pipelines for monitoring trees across the UK using different data types and resolutions.
I recently finished working on a project with CABI and Professor Rene Breton to map an invasive weed (Parthenium hysteropheros) across Pakistan. The project collected extensive ground data at sites across the country and produced one of the most comprehensive studies of the biogeography of this plant. We then used a combination of supervised machine learning and Sentinel-2 imagery to produce maps at 10m resolution. Papers are due for release by the end of the year but more info available here
I collaborated with Dr Jon Murray to develop fast algorithms to recover tree health indicators from terrestrial LiDAR data. The work is published here
I worked with Professor Bruce Grieve to develop a method for classifying Bemisia tabaci biotypes. We used a new machine learning approach for segmenting and classifying hyperspectral imagery of the frozen insects. The work is published here
During my PhD I worked on using the light in polytunnels to disrupt flying pests, such as aphid and thrip. In 2019 I published an up-to-date review of pest responses to colour. In 2020 I published some exciting work demonstrating how the Green Peach aphid uses vision to find safe feeding sites including full analysis code and data published on GitHub (here).
If you're new to working in Python and would like a longer introduction with details on setting up your Python installation, more examples and advanced techniques, you might find this repository useful. If you're interested in learning some basics of analysis of time domain (a radioastronomy pulsar example) and spatial domain (an optical imagery example) data, check out this one instead.
I've developed/modified various small code repositories for specific tasks and projects:
I enjoyed developing resources for public engagement on climate change and food with the GGDOT team. The outreach work and various methods used are described fully here. We developed a 'Top Trumps'-style game where the players try to get the lowest impact food cards. These are available for download along with other resources from GGDOT/Take A Bite:
I currently work for The Open University and I am always keen to collaborate with exciting people in academia and industry. I am very happy to discuss a variety of ways in which we can work together including consultancy, collaboration, and through sponsored research studentships. Send me a message to begin the conversation.
Areas of expertise