Tide Gauges and Bidston Observatory

Philip L. Woodworth, 4 August 2016.

Everyone knows that the level of the sea goes up and down. Most of these changes in level are due to the ocean tide (at Liverpool the level changes due to the tide by more than 8 metres at ‘spring tides’), but changes of several metres can also occur due to ‘storm surges’ that occur during bad weather, while slow changes in level can take place due to climate change and because of the geology of the adjacent land.

Changes in sea level are measured by devices called ‘tide gauges’: the more suitable name of ‘sea level recorders’ has never been widely adopted in the UK although Americans often call them ‘water level recorders’. There are as many types of tide gauge such as:

Vertical scales fixed to a jetty or dock entrance.

These were simple ‘rulers’ (sometimes called ‘tide poles’ or ‘tide boards’), by means of which the sea level could be measured by eye. An example is shown in Figure 1.

Figure 1. A simple ‘tide pole’ or ‘tide board’ installed vertically in the water by means of which the water level can be estimated by eye.
Figure 1. A simple ‘tide pole’ or ‘tide board’ installed vertically in the water by means of which the water level can be estimated by eye.
Float and stilling well gauges.

This way of measuring sea level was first proposed by Sir Robert Moray in the mid-17th century. However, over a century went by before the first practical systems were introduced at locations in the Thames during the 1830s. They quickly become the standard way of measuring sea level and by the end of the 19th century they had spread to major ports around the world.

A stilling well is a vertical tube with a hole at its base through which sea water can flow. The level inside will be, in principle, the same as that of the open sea outside, but energetic wave motion will be damped (or ‘stilled‘) inside due to the hole acting as a ‘mechanical filter’. In the well is a float which rises and falls with the water level, and is attached via a wire over pulleys to a chart recorder driven by an accurate clock. The rise and fall of the water level is thereby recorded as a line traced by a pen on paper charts that are regularly replaced, the charts finding their way to a laboratory such as that at Bidston Observatory, where an operator ‘digitises’ the pen trace and so provides the measurements of sea level.

Figure 2(a) demonstrates how the level of the float is recorded on the paper chart, while Figure 2(b) is a photograph of the tide gauge station at Holyhead where there are two exceptionally large stilling wells.

Figure 2a. An example of a float and stilling well tide gauge. In modern gauges of this type, the recording drum and the paper charts are replaced by digital shaft encoders and electronic data loggers.
Figure 2a. An example of a float and stilling well tide gauge. In modern gauges of this type, the recording drum and the paper charts are replaced by digital shaft encoders and electronic data loggers.
Figure 2b. Two large stilling wells at Holyhead in North Wales.
Figure 2b. Two large stilling wells at Holyhead in North Wales.

This type of gauge is of historical importance as they were used for almost two centuries (although with modern improvements such as replacing the paper charts with modern electronic data loggers) and so data from them make up the data sets of sea level change that are nowadays archived at the Permanent Service for Mean Sea Level (PSMSL) in Liverpool and used for studies into long-term climate change. During the 19th century, most of these gauges were operated in the UK by the major ports, and even by the railway companies which operated ferries. Bidston Observatory operated one at Alfred Dock in Birkenhead for many years. A number of countries still operate float and stilling well gauges although most in the UK have been replaced with other types.

Pressure gauges.

These gauges measure sea level by recording water pressure with the use of a pressure sensor that is installed well below the lowest likely level of the water. The recorded pressure will be the sum of two forces pressing on the sensor: the pressure due to the water above it (which will be the sea level times the water density and acceleration due to gravity) and the pressure of the atmosphere pressing down on the sea surface. In practice, the latter can be removed from the pressure measurement using what is called a ‘differential’ sensor, thereby, after some calculation, providing a measurement of the sea level.

We mention two types of pressure sensor below, which were both developed at Bidston. One type (the bubbler pressure gauge) has been used at 45 locations around the UK for several decades and remains the main technology for sea level measurements in this country. Until recently (mid-2016), this large network was operated for the Environment Agency by a group at Bidston called the Tide Gauge Inspectorate, and then, following relocation, at the National Oceanography Centre in Liverpool.

Ranging tide gauges.

These devices consist of a transducer that is installed over the sea so that it can transmit a pulse down to the water, where the pulse is reflected back and recorded by the transducer, so measuring the time taken to travel down and back. If one knows what the speed of the pulse is, then one can readily compute the height of the transducer above the sea, and so measure sea level. The transmitted pulse can be either an acoustic one (sound), or electromagnetic (radar) or optical (light). During the last decades of the 20th century, acoustic systems became very popular and replaced float gauges, and even replaced pressure gauges in some countries. However, they have since been largely replaced in their turn by radar gauges for several reasons. One simple reason is relative cost. However, radar gauges are potentially more accurate than acoustic systems owing to the speed of a radar pulse, unlike sound, being independent of air temperature. Optical ranging gauges use lasers to transit the pulses but, to my knowledge, are used in only two countries (Canada and South Korea).

Bidston Observatory had expertise in all of these types of tide gauge, but three can be mentioned in which Bidston scientists took a special lead.

Bubbler pressure gauges.

In the late 1970s, the Institute of Oceanographic Sciences (IOS, as Bidston Observatory was then known) was encouraged by the government to see if the new types of tide gauge then becoming available would be suitable for replacing the float and stilling well gauges then standard in the UK. This led to a programme of research by David Pugh and others into the use of different types of pressure gauge, including the bubbler gauge, and the curiously-named ‘non-bubbling bubbler gauge’ which we shall not explain.

Bubbler gauges were not invented at Bidston but they were developed there into practical instruments. They offered advantages over other pressure sensor systems in which the sensors themselves are installed in the water. In a bubbler system, the only equipment in the water is a tube through which gas flows at a rate sufficient to keep the tube free of water, such that the pressure in the tube is the same as that of the water head above the ‘pressure point’ at the end of the tube (Figure 3). The pressure sensor itself is located safely at the ‘dry land’ end of the tube, so there are no expensive electronic components that could be damaged in the water. If the tube is damaged it is simple and cheap to replace. The only drawback is that a diver is needed to install the tube, although the same need for a diver applies to all other pressure systems.

Figure 3. A outline of the bubbler pressure gauge system. (From Pugh and Woodworth, 2014).
Figure 3. A outline of the bubbler pressure gauge system. (From Pugh and Woodworth, 2014).

Comparisons of the old (float and stilling well) and new (bubbler) gauges were made at various locations, including at the important tide gauge station at Newlyn in Cornwall. In addition, the way that they measure sea level was thoroughly understood from both theoretical and experimental perspectives. The conclusion of the research was that bubbler pressure gauges could be reliably installed across the network. Bubblers are now standard in the UK and Ireland although they have since been replaced in countries such as the USA by other systems.

‘B’ gauges (where B stands for Bidston).

These gauges were developed in the 1990s by Bob Spencer, Peter Foden, Dave Smith, Ian Vassie and Phil Woodworth for the measurement of sea level at locations in the South Atlantic. They are rather complicated to explain in this short note, but the gist of the technique is that it uses three pressure sensors to measure sea water pressure (as in a bubbler gauge) and also maintain the datum (measurement stability) of the data in the record. ‘B gauges’ are probably the most accurate and stable types of tide gauge ever invented, but they are expensive (because of the requirement for three sensors) and were never developed commercially. Nevertheless, the principle of the ‘B technique’ was eventually incorporated into the way the bubblers were operated in the UK network, which remains the situation today.

Radar tide gauges.

Bidston Observatory cannot claim to have invented radar tide gauges; these radar transducers were developed first for the measurement of liquids and solids in giant industrial tanks, and were then applied to the measurement of river levels. However, Bidston can claim to have been one of the first laboratories to have used radar gauges for measuring sea level, a one year comparison of radar and bubbler data from Liverpool having shown that radar was a suitable technique for a tide gauge (Figure 4). Radar gauges have since fallen in price, are even more accurate than they were, can be readily interfaced to any kind of computer, and consume less power (an important feature in remote locations where gauges have to be powered from solar panels). They have become the standard technique for measuring sea level around the world and look like remaining so in the future.

Figure 4. A radar tide gauge at Gladstone Dock in Liverpool. The gold-coloured radar transducer transmits pulses down to the water and so measures sea level. The grey box on the wall is a satellite transmitter that sends the data to the laboratory.
Figure 4. A radar tide gauge at Gladstone Dock in Liverpool. The gold-coloured radar transducer transmits pulses down to the water and so measures sea level. The grey box on the wall is a satellite transmitter that sends the data to the laboratory.

Some References for More Information

  • Bradshaw, E., Woodworth, P.L., Hibbert, A., Bradley, L.J., Pugh, D.T., Fane, C. and Bingley, R.M. 2016. A century of sea level measurements at Newlyn, SW England. Marine Geodesy, 39(2), 115-140, doi:10.1080/01490419.2015.1121175.
  • IOC. 2015. Manual on Sea Level Measurement and Interpretation. Manuals and Guides 14. Intergovernmental Oceanographic Commission. Volumes I-V may be obtained from http://www.psmsl.org/train_and_info/training/manuals/.
  • Pugh, D.T. and Woodworth, P.L. 2014. Sea-level science: Understanding tides, surges, tsunamis and mean sea-level changes. Cambridge: Cambridge University Press. ISBN 9781107028197. 408pp.
  • Woodworth, P.L., Vassie, J.M., Spencer, R. and Smith, D.E. 1996. Precise datum control for pressure tide gauges. Marine Geodesy, 19(1), 1-20.

Bidston Observatory and Its Tide Prediction Machines

This article originally appeared in the newsletter of the Friends of Bidston Hill in February 2016. It is reproduced here with the permission of the author.

The role of Bidston Observatory has changed several times through the years. In its early decades, following the decision in the 1860s by the Mersey Docks and Harbour Board to move the Liverpool Observatory from Waterloo Dock to Bidston Hill, the focus was on astronomical measurements. These were required in order, amongst other things, to determine accurately the latitude and longitude of the site. Famous names involved included John Hartnup and his son (also John) and W.E. Plummer. Other areas of science undertaken by the Observatory included meteorology and seismology. In addition, it provided several local services, such as the calibration of accurate chronometers for port users and precise timing via the “One O’Clock Gun”.

By the 1920s, the Observatory had become ‘moribund’ (to quote from the excellent book by David Cartwright) and, after the death of its then Director Plummer, the decision was made to combine its work with that of the University of Liverpool Tidal Institute, with both to be located at Bidston. The latter had been founded in 1919 on the university campus in Liverpool with Joseph Proudman as Director and Arthur Doodson as Secretary, with funding from several sources including the major Liverpool shipping companies. The formal amalgamation of the Observatory and the Tidal Institute took place in 1929.

Proudman is another famous name, with Bidston Observatory later becoming known as the Proudman Oceanographic Laboratory. However, it is Arthur Doodson who is more relevant to this article. In the first year of the Tidal Institute, Doodson and Proudman began work on the problem of predicting tides, especially in shallow waters. They also undertook an evaluation of the benefits of mechanical tide prediction machines, which had been invented in the late 19th century by Lord Kelvin (William Thomson) and later developed by Edward Roberts. In effect, they were ‘analogue computers’. By 1924 Doodson had taken delivery of a brand new tide machine, the so-called ‘Bidston Kelvin machine’ thanks to the generosity of Liverpool ship-owners. Then in 1929, with all staff now installed at Bidston, he acquired and refurbished the so-called ‘Roberts machine’ which had been constructed by Roberts in 1906. The Roberts family had used this machine as part of a business of providing tidal predictions to the government but, due to the death of Roberts’ son, were no longer able to continue.

The Bidston Kelvin Machine and (inset) Arthur Doodson (from Parker, 2011)
The Bidston Kelvin Machine and (inset) Arthur Doodson (from Parker, 2011)

The Roberts machine was in many ways superior to the Kelvin machine, being capable of predicting 40 ‘constituents’ of the tide instead of 29. Such machines can only have a decent stab at simulating the tide at all thanks to the fact that the tide is capable of being described as the sum of individual harmonic constituents. Constituents can be thought of as cosines with particular frequencies (or periods) that are known from astronomy. So, for example, two of the most important constituents are called M2 and S2. These come from the Moon and Sun respectively with periods of 12 hours 25 minutes for M2 and 12 hours exactly for S2. These two terms are responsible for the regular twice-daily tide we have at Liverpool. However, many more constituents than these two are required to do a decent job of simulating the real tide to the accuracy required, and a machine with as many constituents as possible is highly desirable.

The Roberts machine at an exhibition in Paris in 1908. This machine is now on display at the National Oceanography Centre in Liverpool.
The Roberts machine at an exhibition in Paris in 1908. This machine is now on display at the National Oceanography Centre in Liverpool.

These two machines were responsible for many important achievements in the Observatory’s history. Bidston had become the undoubted centre of excellence in tidal research, both from theoretical perspectives (primarily Proudman) and on more practical bases such as the provision of tidal predictions worldwide using these machines (primarily Doodson). Doodson was excellent at devising techniques for handling numbers within complicated scientific calculations that nowadays would be undertaken by digital computers. He also became an expert in the technical design and construction of the tide prediction machines.

Although important individual machines were constructed in Germany and the USA, the majority of the 33 ever made (24 machines) were designed and manufactured in the UK, in either London, Glasgow or Liverpool. The UK was the only country to export machines to other countries. The construction of the majority of the machines made after 1920 was supervised, one way or another, by Arthur Doodson. These included a series of machines made after World War II, of which one (called locally the “Doodson-Légé machine”) was to be found in the lobby of the main POL building for many years until the move of the laboratory to the Liverpool campus in 2004.

The Doodson-Légé machine in the 1990s in the reception area of the Proudman Oceanographic Laboratory. The machine is now on display at the National Oceanography Centre in Liverpool.
The Doodson-Légé machine in the 1990s in the reception area of the Proudman Oceanographic Laboratory. The machine is now on display at the National Oceanography Centre in Liverpool.

Two of the three machines at Bidston have an importance in a notable period in the Observatory’s history, in providing tidal predictions during World War II and, in particular, for the D-Day landings and in other military operations around the world. These were the Kelvin and Roberts machines, which were located in separate buildings at the Observatory during the 1940s in case of bomb damage. The Kelvin machine, Doodson’s first, is now to be found in good condition at the headquarters of the French Hydrographic Service in Brest. Its disposal by Bidston after the war was a financial requirement in order to obtain funding for the Doodson-Légé machine.

The Roberts and Doodson-Légé machines are still located in Liverpool and are now owned by the Liverpool Museum. Recently, they have both been refurbished excellently and are capable of working as well as they can in order to show how things were done at Bidston, before the advent of digital computers in the 1960s saw their demise as the Observatory’s main technical assets.

Both machines are now on long-term loan from the Museum to the National Oceanography Centre in Brownlow Street on the Liverpool University campus, NOC being the successor to POL and therefore the ‘spiritual home’ of the machines. They are available for viewing by the public but arrangements must be made beforehand with the NOC Administration.

For anyone interested in Bidston Observatory and these machines, there is more to read. For an excellent introduction to tidal science, see Cartwight (1999), while histories of the Observatory and the people who worked there are given by LOTI (1945), Jones (1999) and Scoffield (2006). Aspects of Doodson’s career have been described by Carlsson-Hislop (2015). An ‘inventory’ (or overview) of tide prediction machines can be obtained from me, while the story of the use of the Kelvin and Roberts machines in World War II is given by Parker (2011).

Philip L. Woodworth
National Oceanography Centre,
6 Brownlow Street,
Liverpool L3 5DA
December 2015

References

  • Carlsson-Hislop, A. 2015. Human computing practices and patronage: anti-aircraft ballistics and tidal calcuations in First World War Britain. Information and Culture: A Journal of History, 50, 70-109, doi:10.1353/lac.2015.0004.
  • Cartwright, D.E. Tides: a scientific history. Cambridge University Press: Cambridge, 1999. 292pp.
  • LOTI. 1945. Liverpool Observatory and Tidal Institute. Centenary Report and Annual Reports for 1944-5. Available from P.L. Woodworth.
  • Jones, J.E. (original date 1999) From astronomy to oceanography: a brief history of Bidston Observatory. http://noc.ac.uk/f/content/downloads/2011/proudman-history.pdf.
  • Parker, B. 2011. The tide predictions for D-Day. Physics Today, 64(9), 35-40, doi:10.1063/PT.3.1257. Available from http://scitation.aip.org/content/aip/magazine/physicstoday/article/64/9/10.1063/PT.3.1257.
  • Scoffield, J. 2006. Bidston Observatory: The place and the people. Merseyside: Countyvise Ltd. 344pp.