Identifying the origins of extreme rainfall using storm track classification

Andrew Paul Barnes,Marcus Suassuna Santos,Carlos Garijo,Luis Mediero,Ilaria Prosdocimi,Nick McCullen,Thomas Rodding Kjeldsen
DOI: https://doi.org/10.2166/hydro.2019.164
IF: 3.058
2019-10-23
Journal of Hydroinformatics
Abstract:Abstract Identifying patterns in data relating to extreme rainfall is important for classifying and estimating rainfall and flood frequency distributions routinely used in civil engineering design and flood management. This study demonstrates the novel use of several self-organising map (SOM) models to extract the key moisture pathways for extreme rainfall events applied to example data in northern Spain. These models are trained using various subsets of a backwards trajectory data set generated for extreme rainfall events between 1967 and 2016. The results of our analysis show 69.2% of summer rainfall extremes rely on recirculatory moisture pathways concentrated on the Iberian Peninsula, whereas 57% of winter extremes rely on deep-Atlantic pathways to bring moisture from the ocean. These moisture pathways have also shown differences in rainfall magnitude, such as in the summer where peninsular pathways are 8% more likely to deliver the higher magnitude extremes than their Atlantic counterparts.
environmental sciences,computer science, interdisciplinary applications,engineering, civil,water resources
What problem does this paper attempt to address?