Why GPS?

Scotland has magnificent but potentially dangerous country, and GPS can be a valuable aid to map reading under adverse conditions.

Unsure about using GPS - See write-up

Can you please assist in helping to make a new catalogue of the Munros, listing the location of each Munro by on-the-ground measurements, using GPS measurements. I would also like to include the location of Tops and deleted summits (as listed in Robin Campbell's Variorum in his Munroist's Companion), Corbetts, Grahams, Donalds and Other hills included in SMC lists (collectively referred to as summits). I am also interested in routes - but which hills have a unique, standard route? Nevertheless, there is often a 'key' point on a route, a footbridge, col, sorry bealach, path junction, access point to ridge or more importantly the start of a safe route down. I do not want to generate routes that can be followed 'blind', without reference to a map, but rather to provide safety information as back up. When GPS is used as an aid to navigation it can be a valuable aid to safety by answering the question "exactly where am I?"

Why a new list; surely the SMC Munro tables represent a definitive list? And there are other lists if that isn't enough.

One problem is that current lists are derived from maps, not from the ground. The location of the summit cairn has to be deduced from close reading of a map. Certainly the current maps derived by photogrametry are very precise, but are still an interpretation of the ground from an air photograph.

The second problem is that the 6-figure grid references traditionally given for the location of summits is in some cases too imprecise to identify the point in question. At best, the location is specified to 100 m in the east-west and north-south directions. Different interpretation (see later) of the grid reference can increase this uncertainty to 150 m, even for a well defined summit with a trig point or mapped cairn.

In many cases, the 6-figure grid references is sufficient to define the summit that is intended. However, there is a case where two listed summits have the same grid reference (Bhasteir Tooth and Sgurr a' Fionn Choire, tops of different Munros, in the SMC tables of 1997), plus many cases where there are several knolls and peaks in a grid square so that it is ambiguous as to which high point is the one intended by the keeper of the tables. By way of example, the west top of Beinn a' Chroin is a delightful broad ridge, 200 m × 1000 m, abounding with lochans and rocky knolls. Delightful, unless you are intent on finding the highest point, as there are several cairns, but none at the highest point. This ridge is listed as the West Top of Beinn a' Chroin, but there is some evidence that the highest point is actually higher than the east top, the Munro, so it is quite significant (a) to identify the highest point and (b) to give a more precise location than is possible with a 6-figure grid reference, to enable walkers to locate the spot even in conditions of poor visibility. Robin Campbell has reported that he has had difficulty in identifying what top the cataloguer had in mind to list in "confusing places like the Saddle, the An Teallach ridge, or the awful Mullach na Dhearagain ridge with many small tops."

With care, it is possible to derive 8-figure grid references from maps, particularly the 1:25 000 OS maps or 1:40 000 Harveys maps. An 8-figure grid reference locates the summit to a 10×10 m square, accurate enough for most purposes, if the summit in question can be identified on the map. However, for both Slioch and The Saddle the summit is allocated to slightly higher ground, identifiable on the ground but inspection of the map does not allow the ring contour with the higher centre to be identified unabmiguously without having visited the site.

GPS units work with 10-figure grid references, precise to 1 m, and probably reproducible to about 5 m with current hand-held equipment. Such precision should remove any ambiguity as to which rock has been identified as the summit. The problem is to get hold of GPS readings for all of the summits. Ideally, multiple readings taken by different walkers on different days should be combined, eliminating dubious readings and averaging the others. I am therefore appealing to walkers to make a point of recording the location of summits that you visit. Mark the summit as a waypoint on your GPS unit and pass the relevant data on to me, please. The procedure required to get consistent data is summarised as follows:

There are two distinct data sets I propose: (a) Munros and Tops to 5m (rather than the 100 m of standard OS grid references) (plus Corbetts, Grahams and Donalds) and (b) key waypoints on the route up and (probably more important) the safety route down. I was particularly taken by the fact that altitude readings actually have a meaning.

Munros and Tops:
At the summit, get a time averaged reading of the summit location, with altitude which may have to be recorded manually (now altitude readings are consistent, but is there an inherent displacement?). Put the GPS on the summit cairn, mark the point averaging the data for a few minutes, until the FOM (Figure of Merit) has dropped to less than 5 m. The FOM figure and the altitude given by the GPS should be recorded as well as the name of the Munro/Top.  If the summit has a trig point, please note whether it coincides with the summit cairn, and if different record the location of both.  For a number of other Munros, the trig point is known not to be at the summit. I have identified Spidean Coire nan Clach, Beinn a'Ghlo - both Carn nan Gabhar and Carn Liath, Ladhar Bheinn, Slioch, The Saddle and Ben Vorlich (Lomond).  For the Loch Earn Ben Vorlich, the trig point is, I believe, at the summit, whereas the summit cairn (100 m away) is apparently lower.

GPS waypoints for deleted Munros and Tops, Corbetts, Grahams and Donalds also accepted!

Routes:
GPS data on routes should be seen as back-up to navigation with a map and not a substitute. The first priority is to provide data to lead walkers to a safe route from the summit, i.e. the route to the top of the descent(s) from the summit plateau. It is also useful to identify where the summit path leaves the valley path, often identified by a (missable) cairn. Typically, routes would give the car park (or the point the route leave the public road) so that I can see where you are starting from, with key junctions, the top of the main ascent to the summit plateau, the summit itself (with time averaged GPS data as above) plus tops and the point the route leaves the summit plateau. The data should allow forward and reverse routes (hence the point at the top of the ascent for straight up and down outings rather than traverses of the mountain). These should be specified as waypoints, not just points on routes (which give truncated waypoint data) or tracking data (difficult to identify key points, but useful for checking the distance walked - and interesting if you select a fixed time interval of, say, 4 min, when you look at a graph plot). A written description would be useful.

Please e-mail me, Henry Marston, with attached text files of waypoint and routes and relevant information (background to route e.g. as described in SMC Munros book, altitude of summits as recorded by GPS if not included in downloaded data file), GPS unit (e.g. Garmin 12), and download software package (e.g. PCX5). Make sure you have set Ordnance Survey GB as the map datum and the British OS Grid as the Grid option.

The plan is initially to issue these Munro/Top data (averaged between contributions) and the routes on links from this web page. Currently I have about 250 summits, some supplied by other walkers, in a waypoint file

Please send contributions to Henry Marston  
(given in plain text in case links above fail -  and include a Subject and Body text - unexplained attachments are treated as spam).
Link to GPS waypoint download area

Consider a summit with a 6-figure grid reference quoted as NN 244 456, and assume you have a very large scale map so that you can impose a detailed grid. Within the 100 km grid square NN, you identify the matrix point at an Easting or 244 and a Northing of 456, the centre of the pale blue square in the accompanying diagram. (Note: this diagram is fictitious, and does not correlate with anywhere on the Black Mount) Around this intersection there are four summits, marked as solid dots. Following the instructions with 1:50 000 maps, estimating tenths eastwards and northwards implies that summits are given the grid reference of the nearest matrix point, that is the three dot summits in the pale blue square would be given the grid reference NN 244 456.

However, one logical aspect of the grid system is that as the grid reference of a point is increased in precision, the part already identified remains intact. Thus the square drawn appears in the 10 km grid square (2-figure grid reference) NN 24. The 4-figure grid reference NN 2445 includes all points in this square from NN 240 450 to NN 249 459. Extending this principle to 6-figure grid references, only the two summits in the orange square should be given the grid reference NN 244 456, one at about NN 2442 4563, the other at about NN 2447 4568. The dot adjacent to but southeast of the matrix intersection at NN 244 456 should strictly be listed as NN 244 455, or NN 2441 4559 at higher precision. But what keeper of the tables would not list it as NN 244 456?

Where does this leave the user of the tables? Unfortunately different editors seem to have taken different approaches, increasing the ambiguity of the 6-figure grid references. Unless you know the criterion used by the particular editor of tables, a traditional 6-figure grid reference implies at best an uncertainty of 150 m east-west and north-south, or by Pythagoras a range of 225 m, even with a well-defined summit. For an unidentified contour summit, the uncertainty can be even greater.

If the keeper of the tables worked on the basis of the light blue square, i.e. the grid reference corresponds to the nearest measurement, averaging summit locations would require addition of 00 to convert 6-figure grid references to 10-figure grid references, i.e. eee00 nnn00 would take you to the centre of the most probable square for the summit; it should be within 50 m in any grid direction (eee is the 6-figure grid reference easting, nnn the northing). However, if the keeper of the tables used the orange square criterion, the best derived 10-figure grid references, would be eee50 nnn50. On the basis of 52 summit locations I have checked GPS, I estimate the best conversion to 10- figure grid references involves adding 19 to 6- figure grid references for the 1981/1990 SMC tables (eee19 nnn19) and 42 m for the 1997 tables (eee42 nnn42). This suggests that the earlier tables were based on a 'nearest point' system (but with summits on the boundary being assigned to the SW-wards matrix point) whereas the later tables were based on the (orange) grid square, unless just over the southern or western boundary. Nevertheless, there are some significant discrepancies, and several summits have moved by 200 m to 300 m, plus height changes of almost 30 m.

Note:

OSTN02

The grid transformations used in GPS Utility is based on the standard Transverse Mercator projection. This agrees with the transformation produced by most GPS receivers. However, this is not the same as the OSTN02 grid transformation defined by the UK Ordnance Survey. The difference in grid positions can be up to 15 metres and is most noticeable in north Scotland. 

The OSTN02 grid is defined by a particular set of distortions of the Terrestrial Frame of Reference (TRF). This is incorporated into an "OSTN02 definitive transformation". For more details see the website http://www.gps.gov.uk/ and the page http://www.gps.gov.uk/guide6.asp.

If you wish to improve the accuracy of your coordinates by utilising OSTN02, then you can do this over a local area by using a corrected BNG grid definition. The file 'ostn02grids.txt contains a number of corrected grids. These offer 1 or 2 metre accuracy over a 100km square. This is achieved by correcting the False Easting and False Northing of the standard grid. 

It appears expedient to use the imprecise standard transformation generally provided by GPS receivers, which will be used in the field. 

Back to Help