On Thursday 14 March 2019, the RSS Lancashire and Cumbria local group held a meeting at Lancaster University. Dr Claire Miller of Glasgow Unversity spoke about global lake water quality and the statistical challenges associated with modelling ecological systems.
Claire began her talk by describing the GloboLakes Project, a five-year research programme investigating the state of lakes across the world and their response to climate change. The programme used long-term satellite observations in order to estimate the water quality of lakes and look for spatial similarities.
The data consisted of 1,000 lakes, including Lake Balaton in Hungary, the largest lake in central Europe. Claire showed us visualisations of both the satellite data and the water samples over time and highlighted the difficulties of matching up the in-situ samples with and remote sensing data.
Some of the challenges that Claire discussed, such as missing data and dimensionality, were perhaps not a surprise to the audience. However, one unusual big challenge that Claire did face was a missing sensor. Unfortunately, one of the satellites providing remote sensing data to the GloboLakes project, Envisat, stopped transmitting data early on in the project. This meant that the team had to resort to an alternative data source.
Claire also discussed other recent projects such as modelling river networks and the hydroscape project. It was great to hear that most of Claire's projects resulted in R packages, making this research more accessible for other researchers.
The event concluded with a stimulating discussion of the concepts raised in Claire’s talk, involving a number of audience members from a variety of backgrounds.