Analysing the temporal clustering and interaction of storm-tide hazards around the coast of the UK

Abstract

Coastal flooding is one of the most destructive natural hazards affecting the United Kingdom (UK). The temporal clustering of storm-tide hazards present consecutive threats that can compound impacts to the natural and human elements of the coastal zone through a lack of recovery and repair time. It has been established that coastal flooding events in the UK are dominated by the tide, and that the interaction between the storm surge and tidal elements of sea level can have a significant impact on the magnitude, and timing, of flooding hazards. Despite this, there is still a lack of understanding of the temporal clustering and interaction of storm-tide hazards. This thesis aims to quantify and characterise the temporal clustering and interaction of storm-tide hazards around the UK under present and simulated conditions without consideration of future changes in climate. Objective 1 of this thesis characterises the levels of temporal clustering of storm-tide hazards around the coast of the UK for the first time and finds clustering to be inherent in the UK record. Clustering occurs on both interannual and intra-annual timescales, with quick succession events constituting a high proportion of the total number of events – between ~35-44% (~15-22%) of sea level, storm surge and wave height exceedances occur within 50 days of the last at a given location. Still sea levels cluster the most. A quarter of 1 in 1-year sea level exceedances occurred within 2 days of the last. In objective 2, to determine the levels of temporal clustering one may expect to occur naturally, a near 500-year storm surge model forced without anthropogenic climate change was analysed to provide the best estimate of the levels of temporal clustering under natural variability. For the whole timeseries, clustering statistics evaluate to near their statistical expectancies. However, when applying the calculations over a discrete rolling window that represents the average length of a tide gauge in the UK National Tide Gauge Network, clustering is prominent. For the 50-year periods, at a given location, the percentage of years with a sea level or storm surge exceedance can vary by up to ~34%, ~24% and ~18% at the 1 in 1-, 5-, and 10-year return level, respectively. The mean number of days between consecutive sea level or storm surge exceedances can vary by ~231, ~14,780, and ~17,793 days at the same return levels. Lastly, in objective 3, to assess the interaction of storm-tide hazards, the interactions of skew surge and tidal high water, non-tidal residual and tidal phase, and non-tidal residual and tidal level were determined for the observational record, the near-500 year model and in a modelling exercise with an artificially time-adjusted storm arrival time. Interaction is relatively insensitive to event threshold and declustering window size. Low levels of interaction occur between skew surge and tidal high water, but high levels of interaction occur between the non-tidal residual and the astronomical tide – extreme non-tidal residuals occur favourably between 1-5 hours before tidal high water and at tidal levels that are at, or below, the average tidal level. Storm arrival time does not significantly affect the levels of interaction seen. The temporal clustering and interaction of storm-tide hazards were consistently shown to be underrepresented in modelled data when compared to observed records. This thesis quantifies and characterises the temporal clustering and interaction of storm-tide hazards around the UK, providing invaluable information for coastal practitioners looking to forecast flooding hazards and manage the coastline.

More information

Identifiers

ORCID for Luke Joe Jenkins: orcid.org/0000-0002-7206-7242

ORCID for Ivan Haigh: orcid.org/0000-0002-9722-3061

ORCID for Hachem Kassem: orcid.org/0000-0002-5936-6037

Catalogue record

Export record