Skip to main content


Title: Extreme sea levels for section 105 surveys : summary report
Author: Environment Agency South West Region
Document Type: Monograph
Annotation: EA additional title info: summary report - July 1998
Extreme sea levels have been calculated at 10 sites around the North West coast of the UK using three different methods. These methods include; (i) the Spatial Revised Joint Probability Method (SRJPM) described by Dixon and Tawn (1997), (ii) a General Extreme Value (GEV) distribution fitted to the annual maximum data by the method of maximum likelihood (GEV-MLE), and (iii) a General Extreme Value distribution fitted by probability weighted moments (GEV-PWM). A comparison of these 3 methods, and those provided by Graff (1981) and Coles and Tawn (1990), indicates that the SRJPM overestimates extreme sea levels at most locations around the north-west coast Dixon and Tawn (1997) suggest using local data to revise the I-year level based on 6 or preferably 12 months of hourly tidal data. However, it is considered that this would influence the origin of the relationship between return period and sea level and not the slope of the relationship, which is shown to be inaccurate at many locations, and thus little improvement would be gained. To effect an improvement in the SRJPM, the slope of the relationship was revised by recalculating the return period adjustment factors at relevant nodes based on the GEV-MLE estimates described above. However, comparison of the GEV with the original and revised SRJPM estimates indicates that the SRJPM does not accurately predict extreme sea levels. It is therefore considered that the GEV-MLE distribution provides the best fit to the data.
Publisher: Environment Agency
Publication Date: 1998
Publication Place: Exeter
Subject Keywords: Sea levelWeatherWeatherCoastsTidesFloodingForecasting
Geographic Keywords: LiverpoolWorkingtonBarrowHeyshamLlandudno
Extent: 2
Total file downloads: 248

Download PDF    Display PDF in separate tab