Continuity and Discontinuity in British Storm Surge Science, 1919–1959

By Anna Carlsson-Hyslop

University of Manchester


The effect of wind and barometric pressure on sea level can cause dramatic changes in sea level; what is today called a storm surge. In 1953 over 300 people died on the East Coast of England when an extra-tropical storm surge caused severe flooding. Today such events are forecast using computer models developed at the National Oceanography Centre (NOC) in Liverpool.

Work on forecasting storm surges had been undertaken in Liverpool since NOC’s predecessor, the Liverpool Tidal Institute (TI), was established in 1919. A key continuity in this work was the use of multiple regression statistics. Having worked with the statistician Karl Pearson during WW1, Arthur Doodson (1890-1968) introduced this particular mathematical practice, which was then a relatively new technology of calculation, into the study of storm surges. While not unquestioned, the theoretical/mathematical practice of multiple regression remained in continuous use in TI’s surge science until at least the 1950s.

At the same time the patronage of the work was quite discontinuous. Initially TI’s work on forecasting surges was funded primarily by the local shipping industry and aimed at improving the accuracy of the periodic tidal predictions these patrons used. After a flood event in 1928, when 14 died in central London, the work become about preventing flooding. From then their work was given patronage first by local government actors, then increasingly by the navy during WW2, and then, after the 1953 event, by central government. Alongside these changes of patronage went changes in disciplinary identity, from mathematics towards oceanography.