The first-hand experience of the devastating storm that hit the North West of England in December 2015 inspired Environmental Science student Duncan Chapman to investigate the storm for his undergraduate research project and landed him a coveted prize.
“I went down to the river the night of Storm Desmond and it was completely mad how fast the flow was,” said Duncan, who has just graduated with a First Class Degree from the Lancaster Environment Centre (LEC) at Lancaster.
“Afterwards there were a lot of news stories about what happened, with the Environment Agency classing the event as unprecedented while being in line with climate change predictions.
“I wondered was it really so extreme, was what the newspapers were saying about it in line with the science?”
This question came back to Duncan when he was searching for a dissertation topic. He’d really enjoyed the field trips he’d been on during his degree, particularly a hydrological field course in Slapton Ley, Devon.
“It was really good to see the stuff we’d discussed in lectures, like hydraulic conductivity, and apply it to real life.”
So, with encouragement from his supervisor, Dr Andy Jarvis, Duncan decided to follow his curiosity.
His resulting research project was awarded the Keith Beven prize for the best undergraduate hydrology dissertation, presented in honour of Lancaster Professor Keith Beven FRS, one of the world’s top hydrologists. It also landed Duncan a job with a leading flood and water management consultancy.
Duncan analysed 75 years of data from a gauge at Caton, at the bottom of the River Lune near Lancaster, which showed the greatest river flow ever recorded in the UK during Storm Desmond. He also had access to 124 years of data from a flood stone at Caton Mill.
Duncan set out to discover whether Storm Desmond was outside of previous trends. So he analysed the data using a standard single site flood frequency method. He fitted the Gumbel distribution to 75 years of annual maximum data (AMAX). He looked at two different scenarios, one including Storm Desmond and one excluding it.
“The results showed that in the scenario when Storm Desmond wasn’t included, the Gumbel distribution fitted the data really well. But when Storm Desmond was included it didn’t fit. Storm Desmond was on the edge of the 95% confidence limit, suggesting that it was coming from a new different trend.”
While it might seem likely that what has changed is the climate, Duncan argued there were too many other factors influencing Storm Desmond to conclude climate change was the only driver. He believes more research needs to be done.
Duncan also estimated the return period - predicting the probability of Storm Desmond to occur in the future.
“My conclusion was it was highly uncertain, with scenarios between more than 1300 years, and 500 years.
“The analysis assumed all the data is stationary, i.e climate change isn’t happening. There is so much uncertainty, to try and estimate it is not that useful.”
But despite the limitations of his findings, Duncan found the experience of doing the research very satisfying.
“It was really fascinating to go from an idea and be able to carry it through. The biggest challenge was just trying to get the statistics to work. I made a lot of use of the textbook, Hydrology in Practice, which one of my lecturers Nick Chappell had written along with others,” said Duncan, who was studying abroad at the University of Wageningen in the Netherlands while working on his dissertation.
Duncan is now working as an assistant analyst with a leading environmental company, JBA Consulting, which collaborates on research and training with LEC.
“I’d seen posters about the work of JBA Consulting around the Environment Centre and decided to apply through their graduate scheme. It definitely helped that I had done this dissertation.”