The Mapping Of Great Britain By Chris Knights

Many navigators, (as well as any other folk that may be interested in maps,) may like to know the story of how our Country came to be surveyed and mapped, ending with the Ordnance Survey Maps without which, for instance, the sport of rallying would be a rather difficult task for the organisers! Let’s start with a little bit of history…

“THE HISTORY OF BRITISH MAPPING”

The first set of maps of Great Britain was produced by the great printer Saxton between 1574 and 1579. Other than this, all mapping was a local activity funded by local subscriptions, but these maps were usually of poor quality and inconsistent.

By 1750 Britain was falling behind other European countries in the extent and quality of its mapping, notably France, and The Society of Arts tried to persuade the Government to organise a full national survey. This failed so in 1759 the Society itself offered a prize of £100 to anyone prepared to undertake an accurate mapping of their own county at a scale of one inch per mile.

This scale continued to be the standard scale of British mapping until the Ordnance Survey announced that it was changing to metric scales in 1965. The first OS 1:50,000 metric map was produced in 1974.

The Society’s prizes were taken up with enthusiasm and a large number of individual county maps were produced over the next thirty years; but it soon became clear that these maps were unsuitable for military purposes which was the Government’s only concern. The idea of a full scientific national survey must be credited to William Roy (1726-1790) who was a civilian draughtsman with the army’s Board of Ordnance at its Edinburgh Office.

Following the Jacobite uprising which culminated in the battle of Culloden in 1746, George II realised that accurate maps of Scotland were necessary to pursue his aim of pacifying the Highlands of Scotland and he authorised a new military survey to be undertaken under the guidance of William Roy who was given a military rank in the Corps of Engineers. The survey of Scotland proceeded well but in 1755 the Seven Years war against France and Spain intervened and Roy and his mapmakers were sent to France to prepare maps of war.

In 1763, the war now over, Roy realised how vulnerable Britain was to invasion and proposed a full national survey for defence purposes, but the Government declined on the grounds of cost. Roy was disappointed but continued to experiment with the technical aspects of surveying. For instance he developed a new barometer, which he used as an altimeter for assessing heights. In 1775 he measured the height of Snowdon as 3568 feet (only 8 feet too high!) and his mathematician friend, Dr Charles Hutton, who was doing some of the calculations for him, invented the use of contour lines.

In 1784 Roy became aware that any future national survey had to be based on ‘triangulation’ as this was the only method, which would provide an accurate indication of the relative positions of the country’s features. Triangulation allows distances between features to be calculated without having to measure them since measurement of distances on the ground is very time-consuming and almost impossible on other than totally flat terrain. (See part Two for a detailed explanation of Triangulation.)
But Roy also realised that even triangulation required a precisely measured base line a few miles long as a starting point and he chose the flat lands of Hounslow Heath, the site of Heathrow Airport today, as his starting point.

Roy tried and rejected several measuring methods such as rods of wood and iron and eventually settled on the use of glass tubes 20’ long and 1” in diameter which were unaffected by temperature and humidity. This was the first precisely measured base line in Britain.

By this time the new King, George III, was anxious to cooperate with the French, who were disputing the exact position (ie latitude) of the Greenwich Observatory whilst promoting their own surveys based on the Paris Observatory. The King agreed to fund a survey of the south east corner of England which could match up to the surveys already completed in northern France by triangulation across the Channel.

Roy therefore ordered at the king’s expense a new super-accurate theodolite that would be needed for the triangulation, from Jesse Ramsden the famed instrument maker. Whilst this instrument was being manufactured Roy roughly surveyed the countryside between Hounslow Heath and the south coast and pinpointed and cleared all the hilltops and other high points that he proposed to use for his triangulation. Ramsden’s theodolite was delivered in July 1787 and the first stage of the survey began.

Roy died in 1790 aged sixty four just as the King’s survey of the south east was completed.

Charles Lennox, 3rd Duke of Richmond had been appointed Master General of the Board of Ordnance in 1782. The role of the Board was to deal with the army’s fortifications, small arms and munitions. Richmond had always had a personal interest in mapmaking and had long been impressed with Roy’s professional approach to the task and his use of scientific methods. In 1791, soon after Roy’s death, Richmond set up ‘The Trigonometrical Survey of the Board of Ordnance’ and appointed Major Edward Williams as its first Director.

The Ordnance Survey was born.

The First OS Maps.

The data collected during the triangulation of the south east coast was expanded by detailed ‘topological’ survey work on the ground and the results were passed on to the mapmakers at the Survey’s headquarters in the Tower of London. In 1801 the first OS map (Number 1) covering the county of Kent was published.

The surveying continued, first along the south coast because the maps were still essentially for military defence purposes, and then gradually northwards. The process was the same throughout: first the primary triangulation with triangle ‘legs’ about 30 miles long, then the secondary triangulation with approximately 5 mile legs inside the primary triangles, then the detailed topographical survey on the ground prepared a survey at a scale of 6 inches to one mile. All this data was passed to the engravers and printers who produced the one-inch maps.

Copyright was always a contentious issue and there were many infringement cases against commercial mapmakers who merely copied the OS maps. For most of the time the OS maps were sold to the public though few could afford them at the equivalent of £50 per map at current prices.

This first national survey was completed in 1841.

Over the next hundred years the initial survey was revised continually and was extended to include Ireland (which was of course part of the Kingdom at the time). Map production was advancing all the time and by the start of the 20th century the look of the one-inch maps would be familiar to those who use the latest versions today. Early experiments with aerial mapping were tried but without much success.

By the 1930s, however, massive changes to the landscape, especially in urban areas, made it clear that a complete new survey was needed, and a new national primary triangulation was started under the auspices of Captain Martin Hotine in 1935.

The New Survey.

Hotine realised that war was looming again so he pushed the survey ahead quickly. First the triangulation sites were recce’d, often using the same sites as the original triangulation 100 years before. The hilltop sites were cleared of undergrowth and new sites established where the original could not be used. Next the pillars were constructed on the sites – these are the still familiar concrete ‘triangulation pillars’ better known as ‘trig points’ so beloved by Navscat organisers.

The Second World War intervened but the national re-triangulation was completed by 1957. A new base line for the re-triangulation was measured along the Ridge Way in Wiltshire using invar tape strung on catenaries with precise weights taking up the slack. At the conclusion of the survey a base line was measured in Caithness on the north coast of Scotland to check that the triangulation calculations were accurate. This 15-mile line was found to be just 20 metres shorter than the calculated distance based on the base line 800 miles to the south.

An embarrassing mistake.

Throughout this time, secondary triangulation and on-the-ground surveys followed the main observations. This revealed an interesting mistake made a century earlier. The Greenwich Meridian (0° longitude) had been defined by Bradley’s transit instrument in the Greenwich Observatory up to 1850 and this position had been used by the OS as the definitive 0°. The new Astronomer Royal, Sir George Airy, built a new transit instrument in 1850 that he positioned at Greenwich in the room next to Bradley’s earlier instrument. Airy’s instrument was accepted by the Washington Conference in 1884 as being the precise location of 0° longitude for international geographical and navigation purposes (much to the disgust of the French who wanted the meridian to run through the Paris Observatory). Unfortunately the Survey never reacted to the Washington agreement and ever since 1850 the position of 0° longitude marked on all the OS maps has actually been 5.79 metres west of the true meridian!

Up to date.

Aerial photography suffered in its early days from lack of sufficiently good equipment, but advances in techniques have ensured that the days of the ground survey were soon to be over. It is interesting to note that the aerial surveys still relied on the trig points to locate precisely the overall positioning controls for the aerial maps. In fact, trig points were repainted in bright white just so that they could be easily seen on the aerial photographs. It didn’t matter that many of the trig points could no longer be seen from adjacent triangulation points so long as they could be spotted from the air!

Measuring methods also advanced from the 1950s. The Swedish Geodimeter issued a pulsed light beam that could be reflected back to the observer and this was tested by the US Army Mapping Service on the Ordnance Survey’s Ridge Way base line in Wiltshire, as was the South African tellurometer which worked similarly but used microwave pulses. The Ridge Way was used because it is considered to be the most precisely measured base line in the world. In fact, as a result of these tests, the speed of light was brought into question and after independent proof by the National Physical Laboratory the new figure for the speed of light was internationally accepted. What a tribute to the OS men with their measuring rods!

Since the 1980s Global Positioning Systems (GPS) using satellites have been widely used and since 1990 the two hundred year old triangulation methods have no longer been used for the primary survey. For the future the mapping scientists are looking to intergalactic bodies such as pulsars and quasars to enable distances such as the complete length of Great Britain to be measured to within a single millimetre.

“TRIANGULATION & TRIG POINTS”

Imagine that we are trying to fix certain locations with precise relationship to other places. A, B, C, D, E and F are all high points with lines-of-sight to each other (figure 1). These are the points chosen by the surveyors for the triangulation exercise and on these points they will eventually build their “triangulation pillars” – better known to navigators as ‘trig points’. These points are where the surveyors set up their theodolites, which are really just telescopes fixed to very accurate brass plates on which were engraved degrees, minutes and seconds. The cross hairs on the telescope had of course to be extremely fine and were originally made from the threads from a spider’s web.

The target high point was usually made easy to see even at a distance of several miles by lighting a phosphorous flame in a copper bowl. This equipment could measure the angle between two target locations with an accuracy of three or four seconds of arc. In fact the great theodolite made by Jesse Ramsden in 1787 had an accuracy of better than two seconds of arc. Compare this with the accuracy with which you can measure the same angles on a modern OS map with a simple protractor – probably no more accurate than half a degree, ie about 2000 seconds of arc!
The whole point of the triangulation method was that distances between fixed points on the ground could be calculated with very great accuracy simply by measuring the angles between points and then applying relatively simple trigonometry. Naturally you needed to start off with just one distance that had been accurately measured on the ground – the ‘base line’. Once that had been done then by chosing a high point visible from the two ends of the known base line and carefully measuring the angle from each end of the line to the high point (to form a triangle) then the distance of the other two legs of the triangle could be calculated, rather than measured. The base line was chosen so that it was flat and therefore easy to measure on the ground.
Once the first ‘triangle’ was thus calculated, each of the three sides could be used as a new base line for further triangles extending ever outwards across the country. Each ‘leg’ had to end at high points which were visible from each other and typically each leg was 7 or 8 miles apart though this varied according to the terrain.

As the ‘triangulation’ gradually proceeded from the south of England to cover the whole country, some accuracy was inevitably lost but when the triangulation eventually reached the northeast of Scotland, near John O’ Groats, which is generally flat, the final leg of the mainland triangles was remeasured on the ground. This leg was about 20 miles long and the difference between the actual measurement and the distance calculated by triangulation was found to be less than 20 metres in 15 miles and this was 800 miles north of the original measured base line!

This is a real tribute to the surveyors of 200 years ago. One of the reasons for inaccuracy was that of ‘spheroidal excess’. This phenomemon is caused by the curvature of the earth and means that if you measure the three included-angles at each corner of a triangle whose legs are some miles long, the sum of the angles is slightly more than 180°. The surveyors had access to tables which allowed this error to be taken into account.

But to return to our local illustration. Imagine that the leg from Ackleton to Upper Farmcote (line AB on figure 1) has already been calculated as a result of earlier triangulation and is 6.5 kilometres long. We now need to calculate the distances from the high point at Apley Terrace to four other high points: Upper Farmcote (CB), The Down (CD), Upton Cressett (CE), Shirlett (CF).

The surveyors now trudge up to Apley Terrace (C) and set up their theodolite equipment whilst other surveyors set up equipment at the five points (A, B, D, E and F) where they can all be seen from Apley Terrace. The surveyor at Apley measures the angles between A and B (55½°), between B and D (67½°), between D and E (22½°) and between E and F (30°). Then the surveyor at Upper Farmcote measures the angle between A and C (37½°) and between C and D (61°). Finally the surveyor at Upton Cresset measures the angle between C and D (84½°) and between C and F (61°).

Now, back in the office, they can begin their trigonometrical calculations. First they calculate any missing angle in triangles where they have only measured two of the angles and make allowances for spheroidal excess.
Then by using the sine rule they can calculate:

These figures would have to adjusted of course to allow for the fact that all the observation points are at different levels above sea level, but the theodolites used in the original survey allowed for vertical as well as horizontal angles to be measured, so a little Pythogoras would soon take care of that.
Later ‘levelling’ as it was called, would be done rather more accurately with levelling staffs but the principal was similar. See ‘What about the Verticals?….later.

Trig Points Redundant?

Where these measurements on the local high points were made, the surveyors in the resurvey which started in 1935 usually built a permanent ‘triangulation pillar’ generally made of concrete and containing brass inserts to which their equipment was fixed and if you visit these points on Figure 1 you will still see them in place. A few years ago the Ordnance Survey accepted that they no longer needed all these trig points and did not want to incur the costs of maintaining them now that most surveying is done by satellite GPS systems. They only wanted to keep a few key points as checks on their GPS measurements so they were faced with a difficult decision. Shoud they remove the unwanted trig points or leave them to deteriorate gradually? Several groups such as The Ramblers Association lobbied them to keep trig points in good order as they were such useful features to their members, particularly in conditions of fog in otherwise featureless countryside where a trig point looming up in the mist was the only way in which the walkers could pinpoint their position with complete accuracy. The Ordnance Survey decided on a typically British compromise: they put up most of their trig points for “adoption” whereby members of the public could agree to adopt a particular trig point which would then be recorded as ‘belonging’ to that individual. The individual would agree to visit ‘their’ trig point each year and carry out any basic maintenance such as repainting in white and to report to the OS any major damage or deterioration in condition. The OS issue an impressive certificate of adoption, too.

I have adopted the trig point on Wart Hill (ref: 137/401847) pictured here after being repainted, which I have named ‘Trevor Trig Point’. Blimey, that’s sad isn’t it? Visit it sometime. There’s a super view over the whole of the south Shropshire Hills from the top.

“WHAT ABOUT THE VERTICALS & WHERE’S SEA LEVEL?”

Triangulation was essentially concerned with calculating horizontal distances but vertical distances ie heights were also important and not just because the height differences had to be taken into account when doing the triangulation calculations. The measurements of heights was originally done using the same theodolite on a vertical plane as was used for the horizontal measurements of angles in the triangulation survey but this proved to be not very accurate especially over long distances as the measurements were affected by refraction. Later measurements of heights were carried out (originally in the 1830s) by means of ‘levelling’ where two ‘levelling staffs’ – or long vertical poles – were placed a few hundred yards apart with a telescope kept horizontal by a spirit level being set up between the staffs. The staffs were marked with heights in inches and the telescope was first trained on one staff and the height shown on the staff was noted, then the telescope was turned round and trained on the other staff and the height again noted. The difference in heights indicated the increase or decrease in the height between the two staff positions.

But height above what? Sea level was the obvious choice, Britain being an island. It was initially thought that sea level was the same everywhere but after some experimentation it was soon discovered that sea level varied from place to place. Originally Liverpool was chosen, then Felixstowe, then Dunbar in Scotland but eventually experiments showed that Newlyn in Cornwall had the most constant mean sea level of all these. A new tide guage was established at the end of the stone pier at Newlyn harbour in 1915 and records of sea level were made hourly for the next six years. From this data ‘mean sea level’ was calculated and Newlyn became the ‘datum’ for all OS maps showing heights above sea level. The Newlyn datum is now accepted as the most accurate datum for sea level anywhere in the world, and it is noted on all OS maps (see the bottom right corner of your Landranger map legend) as ‘Vertical Datum Mean Sea Level Newlyn’ or‘ODN’ – Ordnance Datum Newlyn.

Having established the datum, a more accurate levelling could take place and several did over the next 50 years. The latest levelling survey of the whole country took place in the 1950s and 1960s during which a hundred or so ‘Fundamental Benchmarks’ were levelled from Newlyn. Once these were established and marked, secondary levelling could take place based on the main benchmarks. Secondary levels were marked (diagram above) on any convenient and reasonably permanent feature such as walls, buildings, churches and so on.

The 100 or so fundamental benchmarks were meant to be completely secure and permanent and were usually chosen because the geology was very stable. The levellers dug down to the bedrock and marked the precise level in the bedrock itself, often with a brass bolt. The pit was then filled with gravel and sealed. Above ground the site was simply marked with a granite block about 2’ tall rather like a small trig point. These sites are not marked on the usual maps but I can reveal that the nearest fundamental benchmark to us is near Church Stretton at 137/467927. All the fundamental benchmarks are still used by the Ordnance Survey as ‘controls’ for modern GPS surveying.

Secondary or lesser benchmarks, measured from the fundamental benchmarks, are usually marked with the standard OS benchmark sign – a horizontal line with an vertical arrow underneath – and they are to be found all over the place; all trig points have them and they are also carved onto buildings, walls etc. There are three in Pattingham and the one pictured left is about two feet up the wall on the south east corner of St Chad’s church over the road from The Pigot. The OS database indicates that St Chad’s benchmark is 118.81 metres above mean sea level at Newlyn.