Event Details

The seminar will be in-person.

Title: Modelling spatial extreme temperature events on the Greenland ice sheet

Abstract: The Greenland ice sheet has experienced significant melt over the past six decades. Rises in global temperatures caused by climate change have contributed to spatial extreme temperature events that cause large areas of the ice sheet to melt simultaneously. Identifying and modelling extreme temperatures on the ice sheet is challenging due to a soft upper limit in temperatures around 0◦C caused by the melt process of ice. This limit causes the distributions of temperatures to vary greatly across the ice sheet and have poorly defined upper tails for extreme value analysis. To account for this, we use a Gaussian mixture model fit to MODIS satellite data to provide a more contextual definition of an extreme event. The mixture model allows melt to be estimated as a probability rather than using a fixed melt threshold to calculate melt empirically from the data. The model is then used as a marginal model for a Spatial Conditional extremes model. This spatial model allows us to simulate spatial melt events and understand their structure across different locations and scales.