How Does Anybody Know How High Anything Is? A Brass Plug in a Wall

Height has no absolute meaning without a reference, and measuring it requires comparing one place with another, which in practice means carrying a known elevation from somewhere to the point of interest. That is done by levelling, in which a surveyor sets up an instrument between two staffs, reads the difference between them, and repeats along a route, accumulating the differences. The process is slow and errors accumulate, so it is impractical to start from the sea every time. Instead a network of permanent marks with established elevations is placed across a country, and any local survey starts from the nearest one. Those marks are the physical embodiment of the height system.

Marks vary by country and by purpose:

• •The cut mark, an arrow with a horizontal bar chiselled into stonework
• •Brass or bronze plugs set into rock, walls or concrete
• •Flush brackets set into the walls of substantial buildings
• •Fundamental marks, deep concrete pillars founded in bedrock
• •Triangulation pillars, which mark position rather than height primarily
• •Satellite reference stations, which are the modern equivalent

Choosing a zero is harder than it sounds, because the sea moves. Tides raise and lower it twice daily, weather changes it, and its average differs measurably between one coast and another because of currents, salinity and temperature. Countries therefore chose a particular tide gauge and averaged its readings over a defined period, and the resulting figure became the national zero. Britain uses the average at Newlyn in Cornwall between 1915 and 1921. Other countries chose other ports and other periods, which is why heights quoted in neighbouring countries differ by a few tens of centimetres for the same physical point, and why international projects must state which reference they use.

A network of fixed marks measured repeatedly over a century turns out to record how the land itself moves, which was not the intention. Comparing modern measurements against nineteenth and twentieth century ones shows systematic change across whole regions, and in Britain that reveals the land in Scotland rising and the south east sinking, which is the continuing response to the removal of the last ice sheet as the crust slowly readjusts. Coalfields show subsidence over worked seams. Areas of groundwater extraction sink measurably. Tectonically active regions record uplift and displacement across faults. Those long records are valuable precisely because the marks were placed for another purpose and were measured consistently for decades.

Satellite positioning gives heights directly and did not make the marks obsolete, for a reason that surprises people. Satellites report height relative to a mathematical ellipsoid approximating the shape of the Earth, which is not the same as height relative to sea level, because gravity varies across the planet and the surface that water would settle into is lumpy rather than smooth. Converting between the two requires a model of that gravity surface, which is itself derived partly from ground measurements. National mapping agencies have therefore retained a reduced network of physical marks, tied to permanent satellite stations, and many countries have abandoned maintaining the older dense networks, so a great many marks remain in walls with no organisation checking them.

Height is meaningless without a reference, and carrying one from the sea to each site is impractical, so a network of permanent marks with established elevations covers a country and local work starts from the nearest. Britain's zero is the average tide at Newlyn between 1915 and 1921, and other countries chose other ports, so heights differ across borders. Satellites report height against a mathematical shape rather than sea level, which still requires a gravity model.