Abstract
This paper describes methods of obtaining improved estimates of long-term sea level trends for the British Isles. This is achieved by lengthening the sea level records where possible, then removing known sources of variability, and then further adjusting for datum errors that are revealed by the previous processes after verification using metadata from archived sources. Local sea level variability is accounted for using a tide and surge model. Far field variability is accounted for using a “common mode”. This combination reduces the residual variability seen at tide gauges around the coast of the British Isles to the point that a number of previously unrecognised steps in individual records become apparent, permitting a higher level of quality control to be applied. A comprehensive data archaeology exercise was carried out which showed that these step-like errors are mostly coincident with recorded site-specific changes in instrumentation, and that in many cases the periodic tide gauge calibration records can be used to quantify these steps. A smaller number of steps are confirmed by “buddy-checking” against neighbouring tide gauges. After accounting for the observed steps, using levelling information where possible and an empirical fit otherwise, the records become significantly more consistent. The steps are not found to make a large difference to the trend and acceleration observed in UK sea level overall, but their correction results in much more consistent estimates of first order (Sea Level Rise) and second order (Sea Level Acceleration) trends over this 60-year period. We find a mean rate of sea level rise of 2.39 ± 0.27 mm yr−1, and an acceleration of 0.058 ± 0.030 mm yr−2 between Jan. 1958 and Dec. 2018. The cleaner dataset also permits us to show more clearly that the variability other than that derived from local meteorology is indeed consistent around the UK, and relates to sea level changes along the eastern boundary of the North Atlantic.
Highlights
Our overall aim in this paper is to extend and improve the British Isles monthly Mean Sea Level (MSL) dataset, to begin to understand the sources of the observed variability in the improved dataset, and to quantify sea level trends and accelerations.This paper significantly improves the sea level records: (1) by using results of a data archaeology exercise to extend the sea level data set where possible; (2) by making use of a barotropic model to remove much of the variability due to local meteorology; (3) by deriving and subtracting a common mode, representing variability from more distant sources
A data archaeology exercise was carried out using historical documents archived at the National Oceanography Centre (NOC) (Liverpool), UK Hydrographic Office (UKHO) archives (Taunton) and older editions of large-scale Ordnance Survey (OS) maps
These comprised OS levelling records, tide gauge calibration records, annual Admiralty Tide Tables, Institute of Oceanographic Sciences (IOS) and National Tidal and Sea Level Facility (NTSLF) reports, paper records of tide gauge history (e.g. Tide Gauge Inspectorate (TGI) reports) and large amounts of correspondence between the UKHO and the Permanent Service for Mean Sea Level (PSMSL). This resulted in additional information being recovered for each tide gauge site, such as older local bench mark elevations, semi-annual or annual tide gauge zero check sheets, Ordnance Survey tide gauge zero levelling history, and elevation changes to the local port or chart datum
Summary
Our overall aim in this paper is to extend and improve the British Isles monthly Mean Sea Level (MSL) dataset, to begin to understand the sources of the observed variability in the improved dataset, and to quantify sea level trends and accelerations. The tide gauge measurements around the coast of the British Isles are affected by a range of different physical processes These include: responses to local atmospheric pressure (Doodson, 1924; Ponte, 2006); wind stress (Thompson 1980; 1981) including tide and storm surges (Frederikse et al, 2016a; Frederikse and Gerkema, 2018); a response to more distant ocean variability (Wakelin et al, 2003; Calafat et al, 2012; Frederikse et al, 2016b; Volkov et al, 2019) which modulates and includes global mean sea level changes; local vertical land motion (VLM) (Wöppelmann and Marcos, 2016) due to present day processes (such as localised subsidence due to groundwater extraction, or potentially for North East England, coal mining) (Rossiter and Gray, 1972); and both land and gravitational responses to past glaciations, known as Glacial Isostatic Adjustment (GIA) (Bradley et al, 2011).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
