Belfry Bulletin

Search Our Site

Article Index


Photographic Cave Surveying

A 'Chatty version' – according to the author, John Letheren of M.N.R.C. of a new technique which will be dealt with in greater detail in a future issue of 'Cave Notes'.

The general idea is to survey caves and mines quickly but with reasonable accuracy - much more accurately than 'sketch from memory' and much more quickly than with conventional surveying equipment.

The data required, from one point to the next, is range, bearing and elevation.  Range cannot be measured accurately with a photographic rangefinder as the angle subtended beyond a few yards is so small as to be quite un-measurable - so an alternative method is used.

The one hit on is to photograph something of known size, and measure the size of the image on the negative which, allowing for a small correction for focussing is inversely proportional to the distance of the object from the camera.  To take an extreme case, you could not measure the distance of the moon with a photographic rangefinder to better than 'between 30ft and infinity' but if you were to photograph the moon and measure the size of the image you could, knowing its diameter, find its distance to the same degree of accuracy as you could measure the diameter of the image, say to 1%. This then is the method used to measure range - although it is not necessary to know the dimensions of the object itself, only to compare it with an image taken at a known distance.

The device used consists of a horizontal wooden cross about ten inches each way with a vertical arm fixed to the centre of the cross also about ten inches high.  Each of these five arms (four horizontal and one vertical) has a small filament bulb at its end, and the measurements must be equal from bulb filament to bulb filament.  In addition there is an extra bulb on one arm to denote south.  This instrument, although we have built and used it, has as yet no name, but it is usually called the 'chandelier' for obvious reasons.  It is also fitted with a compass (used only to align the instrument - not for readings) and a spirit level tube on two of the horizontal arms for levelling.  The whole thing is mounted on a short tripod with a ball-joint head, and is connected by a short cable to a 6v. battery which stands on the ground near the tripod.  The camera is normally used with a 200rmn telephoto lens and is mounted on a longer tripod with a ball-joint head.  A 135mn lens would be adequate, but this is about the limit.

The survey team consists of one photographer and one chandelier setter.  The latter sets the chandelier level and north, and the former makes an exposure at 125th of a second at f8, giving no problems with either depth of focus or long exposure.  This records all the information needed to compute the range, bearing and elevation of the device relative to the camera or vice-versa. The camera is then moved to be ahead of the chandelier and the next exposure made, and so on.  The chandelier must be situated so that ALL the lamps are visible (quite easy in practice, although one must always consider the following shot as well as the one first being made) and in addition, the data is only valid if the camera is looking DOWN on the chandelier.  Reverse shots (i.e. those where the camera is looking back along the direction of surveying) are readily distinguished from forward shots by including a caving lamp in the corner of the picture for reverse shots. You then leapfrog wherever possible, keeping the camera always above the chandelier.  At the end of each traverse, one frame is marked with a close-up of the nearest caving lamp for identification purposes, and the survey then continues for the next section or traverse.

Providing one has sufficient cheap black-and-white film, that is all you have to do underground and it takes approximately one minute per leg.  The rest is done at home.  The film is developed and the negative set up (mount each frame in a cardboard frame which is numbered) and each picture (which is, of course, a series of black dots - it being a negative) is projected, square on, to a sheet of paper and the dots marked on the paper.  The first slide must be a reference shot of the chandelier taken from a measured distance.  This allows for the magnification of the projector etc. and is used as the basis for calculating range.  It need hardly be said that the shots are numbered on the sheets of paper also and that reverse shots are distinguished from forward shots.  The spacing between the North-South and East-West lamps are then measured (the units are irrelevant) and also the height from the centre to the top lamp.  Finally, all the data is shovelled into a handy computer (or calculator, which is quite feasible, but tedious) and out pops a survey.  In the computer programme I have written, each section is entered as either' 'open' or 'closed' and if closed, the computer recalculates the points to adjust out the closing error.

I will not deal with the mathematics here, as that is destined for a separate publication in the future. Nevertheless, the system is working, although a lot more experience is required.  Even so, it really does take only one minute per shot.

In the meantime, anyone thinking of trying out the method is welcome to the very simple details of the instrument and the mathematics.  It is quite feasible to use an electronic calculator if a computer is not available, but the latter is much quicker. I would like to close by thanking John Richardson (M.N.R.C.) for a good many trials (in both senses of the word) in the Coombe Down Mines to get the system working.

Editor's Note:     I have seen a closed traverse done by this method, the actual traverse being DRAWN by the computer as well as being calculated by it.  At present, this method is not as accurate as conventional accurate surveying methods, but its interest lies in its automated nature. With hand calculators of computer like complexity becoming available, one might be able to replace the camera with a TV camera and feed the dots straight in as electrical signals; store all the information in the hand computer; feed it, into a computer with a graphics attachment once outside the cave, and just wait for it to draw the survey! Perhaps we should run a 'Tomorrow's World' programme in the B.B.!