Gail Robertson

Rosheen Luo has written the following article as part of our series of Academic Interviews; featuring Gail Robertson!

Gail Robertson is a statistical consultant for the Statistical Consultancy Unit based at the University of Edinburgh. As an experienced applied statistician, with a quantitative ecology and epidemiology background, she has worked with various types of data, providing advice to group members and students, and analysing large-scale datasets using various statistical methods.

Early experiences in academia

Gail first studied at the University of Glasgow for a bachelor's degree in zoology. At that time, she was interested in ecological relationships between predators and prey animals, a mathematical and statistical problem. However, after graduation, her interest turned from this population ecology to conservation science, due to the call of the climate change and biodiversity crisis. So, she moved on to an MSc at the Imperial College and completed a questionnaire-based study on 'people's perceptions of avian predators' for her MSc thesis. The purpose of this study was to gauge public opinion on how predator populations were declining due to people shooting or illegally trapping them. Afterwards, she achieved a PhD in ecology from the University of Glasgow, analysing seabird tracking data to identify important foraging areas at sea. To complete her work, Gail lived on an island for three months per year to collect data on kittiwakes and tern movement, using GPS tags. After doing several postdocs, she now works as a consultant statistician, applying statistical solutions to real-world problems in industry and academia.

From ecology to statistics

Gail claimed her interest in statistics came gradually. When doing her PhD and MSc, she always found that she was good at analysing multivariate data while her colleagues were struggling. As her research progressed, she repeatedly found that her interests included areas in statistical ecology, such as analysing animal movement data and spatial modelling. She also confirmed that once she had stayed on an island a few times, she didn't know if she wanted to keep doing so; as living on an island for long time periods can be challenging!

For the statistical side of her work, Gail said she loves the puzzles that are inherent to it. However, she confirmed that she‘s more interested in its applications, rather than the theoretical side. She explained that there were always some problems involving birds that she had to try and think of a way of solving, using a statistical method. Gail then gave an example: if you are working on puffins and want to predict how many there are in a given colony, you can do so by taking lots of photographs over time. But, puffins live in holes, which will introduce errors to the collected data. If you want to estimate the population and other ecological questions, you need complex statistical tools to solve them.

Stand out memory – living on an Island

When completing her PhD, Gail thought she'd be doing work modelling data. However, shortly after starting her PhD she was given the opportunity to collect seabird data from colonies on a small island in the North Sea. The island is off the north coast of England and is tiny, no people live on it and there’s only an automatic lighthouse. It is run by the RSPB, the Royal Society for the protection of birds, to try to support all the different birds that live there, mainly terns. These are small sea birds with sharp beaks that dip into the sea to collect tiny fish. 

In her first year, Gail was investigating the feeding rates of terns and was comparing them to different species; this involved a lot of field observations. In her second year, she put some GPS tag devices on kittiwakes, a small seabird living on the cliffs, and to catch them, she used a long pole with a plastic noose at the end. After catching the kittiwake, she would then put the GPS devices on its back, tying it on with some tape, before letting it go to fly and forage. After several days, she would repeat the process with the same bird, taking the tag off this time, and then downloading the data collected via GPS. Although the living conditions were sometimes challenging, Gail said it was a lovely place to work, with some very supportive and friendly colleagues collaborating to achieve the same goal.

Call for math students to join in ecology

Gail suggested that math students should think about how they would apply their knowledge in a real-world context. There is so much data coming in from different sources, such as animal movements, animal populations or climate data, that there is an urgent need for people who have a mathematical and statistical brain to work on those ecological and environmental problems. She suspects that a maths student joining a different field would help many people who are keen to work with statistical and mathematically minded people.