ioflogo.gif - 2333 Bytes
Instructor's kit for O-mapping courses
Fieldwork and Survey
2kezfel.gif - 998 Bytes

Planning the survey

Time for survey

monalisa.gif - 1428 BytesThe amount of work involved in survey depends on the following factors:

Planning of survey should be done as soon as possible. Survey is the most time consuming part of mapmaking, and because it is tiring both mentally and physically it is not possible to work too long a day. Do not normally plan to work more than six hours a day. Always remember that you have to draw the survey draft immediately after survey.

The time for survey can vary from 10 hours/km2 to 60 hours/km2. Taking into account the factors above the planning should be done in total man hours, broken down into a number of man days. To this total amount of days should be added a certain number of extra days (at least 20 %) for bad weather conditions, unexpected difficulties and checking.

Example:
   Area to be surveyed: 10 km2
   Time for 1 km2: 25 hrs
   Time for 1 day: 6 hrs
   Total hours (10×25÷6): 42 man days
   Contingency: 8 days

An allowance of 50 days should be made for completing the survey.


Organization of survey

If the survey for a map is to be done by several mappers, it is impossible to get a consistent map if they work independently. It is very important to follow the following guidelines:

Co-ordination: The whole survey has to be co-ordinated by one person (experienced mapper, club-leader, etc.). It may not be necessary for this person to have a lot of mapping experience, but he can prevent conflicts and can be contacted when problems arise.

Agreement of classification: Before survey starts, all the mappers should visit the area together and come to an agreement as to what level of detail should be included and what should be omitted. If there is the opportunity, a sample area mapped by one of the mappers may serve as an excellent demonstration.
Examples of mapping comparison from Norway, Germany and Switzerland

monalisa.gif - 1428 BytesDivision of the area: Each mapper is given an area to map. The division must be based on clear boundaries (streams, roads or open land). It is very useful to have several copies of the base map, so the sections can be cut with overlap. Arrangements must be made for collaboration between the mappers surveying adjacent areas.

Checking process: The work should be checked regularly by the co-ordinator. Contact between the different mappers is very important. All must report on their progress regularly.

Consistency: As each section of the map is completed, the consistency of the survey should be checked. This work could be done by the co-ordinator or by exchanging sections between the mappers. A final visit together could be useful.

Survey draft: The production of the survey draft has to be done very carefully, taking care to use the same symbols and colours. The work along the dividing lines must tie in exactly.


Survey Equipment

Basic survey equipment

monalisa.gif - 1428 Bytes


Digital Rangefinder

A new measuring instrument manufactured by Leica, called Geovid has come onto the market which resembles a pair of field glasses but contains a digital rangefinder and electronic compass. Sighting on the object to be measured and pressing a button activates a laser rangefinder which displays the distance and bearing to the object. A more sophisticated model also records the vertical angle to the horizontal and can store readings in memory for downloading to a computer.

Global Positioning System (GPS)

The Global Positioning System was developed by the United States military to provide military personnel and weapons systems with real time positioning and navigation systems for use anywhere in the world. The enormous potential of this system for use in areas such as surveying, navigation systems (marine, aircraft and in-car), zero visibility landing systems for aircraft, has pressured the US government to make it publicly available. The Russians are developing a rival system known as GLONASS.

Satellite navigation - The GPS system consists of a constellation of 21 operational satellites plus 3 spares each orbiting the earth twice a day at very high altitude (approximately 17700km). Because of their high altitude orbit there are no atmospheric drag effects and so it is very easy to predict their exact location at any instant in time. To ensure complete accuracy the satellites are monitored regularly for any minor ephemeris errors, these are errors in position, altitude and velocity caused by things like gravitational pull from the moon and sun. These ephemuris errors are transmitted to the satellites. The satellites continuosly transmit a timing code and a data message containing it’s orbital location, the ephemeris errors and the systems health, this timing code and data message are read by GPS receivers.

monalisa.gif - 1428 Bytes How far is a satellite? - This is based on the simple velocty x time = distance formula, the velocity of electromagentic waves in a vacuum is a constant (300000 km/s) and so all that we need is the time of travel of the satellite signal. To measure time very accurate clocks are required since the time of travel of a radio signal from a satellite overhead is less than 0.06 seconds. The satellites have atomic clocks, the most accurate clocks in the world on board while most receivers have clocks that can measure time with nanosecond (10-9) accuracy.

Satellite ranging - The basic principle of GPS is measuring our distance from a group of satellites of known position. This is similar to triangulation we use in the forest but with known distances instead of angles and is in 3 dimensions.

monalisa.gif - 1428 BytesIf a satellite’s position and distance from the observer is known then the observer knows he lies somewhere on an imaginary sphere centred on that satellite with a radius of the distance to that satellite. If a second satellite is observed, a second imaginary sphere is created. These two imaginary spheres will intersect, the intersection being an imaginary circle, the observer now knows he lies somewhere on that imaginary circle. A third satellite observation will will produce a third imaginary sphere which will cut the imaginary circle formed by the intersection of the first two imaginary spheres at two points. The observer now knows he is positioned at one of two points.
Because the observer knows he is positioned on the surface of the earth, only one of these will generally be acceptable, the other being a ridiculous location in space! If a fourth observation is taken this will eliminate the rediculous answer, this observation is also required to remove any inherent clock error in the receiver.

Each satellite transmits a timing code along with the data message, the timing code consists of a series of random 0's and 1's which repeat themselves every second, this code is known as psuedo random code and is different for each satellite. The receiver also produces these same psuedo random codes, and can identify which code it is receiving. If the codes are generated by the receiver and satellite at exactly the same time then by comparing the time offset between the receivers code and the delayed satellite code we know how long the signal took to reach the receiver and hence the distance to the satellite.

Perfect timing? - In order to get accurate distances we need accurate timing. An error of 1 millisecond (10-3) will create an error of some 300km. We know each satellite has atomic clocks and so can assume they all have exactly the same time. The clock in the reciever we stated earlier can measure time to the nearest nanosecond but we can be pretty sure that its time is not synchronized with the satellites’.

Assuming the error in the time is constant (ie. the only error is in the receiver’s clock), then taking a fourth satellite reading will remove any timing error. There are four unknowns, X, Y, Z position of receiver and dT timing error, and four simultaneous equations which can be solved to determine the receivers position.

Errors - Apart from errors mentioned earlier such as ephemeris and clock errors. Other errors include:

The ultimate accuracy of GPS is determined by the sum of these errors. With a receiver and a good GDOP accuracy can be expected to be in the order of 20-35m with a worst case of about 60m except when S/A is implemented when it would be about 100m.

monalisa.gif - 1428 BytesDifferential GPS - This level of accuracy is not suitable for orienteering map making and other applications were high accuracy is required such as, land survey, in-car navigation and zero visibilty landing. By introducing a static receiver at a known location and comparing the computed and actual locations means that the positional error can be calculated. As the satellites are so high up mobile receivers in the vicinity can assume to have the same errors at a given time and so a corrected position can be calculated. This method of using one fixed receiver and a mobile receiver is known as ‘Differential GPS’ or ‘DGPS’ and can eliminate virtually all errors.

Land surveyors using DGPS can acheive accuracy to the nearest centimetre by recording fifteen minutes of readings and using sophisticated computer software to calculate position. Wheras everyday users of DGPS should achieve accuracy to between 2 and 5 metres. To record instantaneous position the errors calculated by the static receiver need to be sent to the mobile users using telemetry. In some parts of the world DGPS corrections are now transmitted on a sub carrier wave of local FM wavebands.

GPS receivers - The GPS receiver is the instrument that tracks the satellite signals and computes its location. There are two broad groups of receivers, those that can track four or more satellites simultaneously and those that sequence between satellites. The cost of a receivers can range between a few hundred dollars to many thousand of dollars.

GPS for orienteering map making - There are two areas in map making where GPS could be a useful tool: for establishing control for photogrammetry where no large scale mapping exists and during field work.

With GPS accuracy in the order of 2-5 metres, this would be acceptable for map making. A 5 metre error would introduce 0.33 mm error on a 1:15000 map, less than the size of any of the point symbols and would be hardly noticable.

Trials using GPS have been carried out in various countries with varying degrees of success. Generally the accuracy of position in open spaces have been good but once under tree cover in the forest results have been variable. Successful mapping using GPS has been carried out in Finland whereas in middle Europe the measurements of position were too erratic. Reasons for these variations may be the type of receiver used or density of forest cover under which readings were attempted.

GPS may not yet be an everyday tool for O-mapping, but with the continued development of receivers it may not be long before it is.


Pen Computer

Land surveyors are starting to use pen computers in the field for updating existing maps. Hans Steinegger has developed OCAD so that it can be used with a pen computer.

For a pen computer to be used in the field it will need to have the following qualities: lightweight, rugged, waterproof, long battery life and be cheap!


Fieldwork

Preparation for fieldwork

The original base map should not be cut and taken out in the field. A copy (with north lines) should be made on polyester film and used for fieldwork. The copy may be drawn by hand or produced by one of the commercially available processes. This base is stable even under varying conditions of humidity. If the base map is on paper it must be covered with a polyester film (self adhesive or not) after drawing on the north lines.
Do not use paper copies without checking exactly for possible distorsion!

To ensure accuracy one of the following options is recommended for copying the base map:


Stages of survey

Exploration: The whole area is explored at the outset in order to decide the level of detail to be included on the map. This should be done by all the mappers together, if the area is to be surveyed by several mappers. This first step is very important; for the map to be consistent a size of detail which is significant to an orienteer must be determined and applied all over the map.
Features smaller than the minimum sizes given in the drawing specifications must not be included, but in certain areas larger minimum sizes may need to be used. It is a waste of time to measure and plot a large number of small features on the map, if most of these have to be omitted afterwards to ensure legibility in very detailed areas.

Checking the accuracy of the base map: Before starting the survey, it is necessary to spend some time checking the accuracy of the base map. Even if it is a new photogrammetric plot, it is very useful to know which features are correct and can be used as check points.

Subdivide the area: Roads, paths, streams and vegetation boundaries are suitable boundaries for dividing an area into smaller working areas. These smaller areas should be no larger than can be surveyed in one days work. If there are too few details on the base map, paths, tracks or other line features have to be surveyed before the division can be made. In areas with very few linear features, temporary survey points can be marked using visible coloured streamers as references for fixing other detail.

Progress of survey: There are many different methods of working within the subdivided area though a methodical approach is most important. The individual technique varies from one type of terrain to another. In steep areas, forests with dense vegetation and undergrowth or nearly flat regions, the method of survey could vary very much. The following steps can be recommended in any type of terrain: monalisa.gif - 1428 Bytes

  1. Work round the perimeter of the small area plotting the position and directions of lines and points, classifying paths etc.
  2. Complete the survey along linear features within this area, including vegetation boundaries and the lines of spurs and reentrants and significant points on contour features.
  3. Fill in point features by crossing systematically the whole area.
  4. Classify vegetation.
  5. Complete the contours.
Contour features are plotted together with the line or the point features, because they must be correct in relation to other map detail.

Experienced mappers will soon recognize that this recommended method is very time consuming, and that, with more experience, overlap between the different steps will be possible without loosing accuracy.


Survey Techniques

General techniques

The details to be drawn on an O-map can be divided into:

monalisa.gif - 1428 BytesCompass bearings and pace counting are the basic techniques for fixing positions of lines and points. Areas can be treated in a similar way, their outline being plotted as lines. Spurs and re-entrants are plotted as lines, as a guide for the contours which will be drawn later.

monalisa.gif - 1428 Bytes monalisa.gif - 1428 BytesIt is therefore very important to know one’s own pace length. It should be calculated over known distances, on the flat, uphill and downhill. Usually double paces at a comfortable walking speed are counted. A double pace scale can be fixed on the compass.

Several factors will affect the number of paces per hundred metres:

Experience is needed to reduce the possible error in measurement to a minimum. Other measuring instruments (tapes, wheels, etc.) are generally not very efficient.

monalisa.gif - 1428 BytesIn dense forest distances greater than 100 metres generally cannot be measured accurately enough by compass and pacing, any large distance measurements must be checked from other fixed points. Remember the effect that measured horizontal distances in steep areas get smaller with increasing steepness.


Measuring point features

monalisa.gif - 1428 BytesThe simplest way of fixing an unknown point is to take bearings from two or more known points, this is known as triangulation.
In theory only two measurements are required though as the sighting compass is not the most accurate instrument a third check bearing is taken.

As the sighting compass is only accurate to +/-1 degree bearings should not be taken to objects more 150m.

monalisa.gif - 1428 BytesFor shorter distances where the error will not be so large two bearings or a single bearing and pacing the distance to the object will fix the point. If at all possible these points should be verified from other known points.

monalisa.gif - 1428 BytesIf it is not possible to sight from an unknown point to a fixed point it may be possible to take a reading to an intermediate point or line and first fix that point from a known point.


Measuring line features

monalisa.gif - 1428 BytesTo plot a line (path, stream, boundary), the line feature is broken up to short (less than 100m) segments, which can be measured by compass and pace counting. This method should be used between two known points. If the end point is not known, it should always be carefully checked before it is taken as fixed point for further measurements. It is also always necessary to check newly surveyed features in relation to the contours (direction of slope, ascending/descending).

monalisa.gif - 1428 BytesIf you are using a very detailed base map, you often will not need to measure many lines. But the relation between contour details and line features must be correct. The impression from the competitor's view is always more important than precise distance measurements!


The survey of landforms

monalisa.gif - 1428 Bytes monalisa.gif - 1428 Bytes One of the most important things in fieldwork is to check, correct and complete the contour lines. They should show the competitor using the map the shape, height and steepness of the ground. Contours provide the basic information for the competitor to interpret the terrain and should be as accurate as possible.

monalisa.gif - 1428 Bytes monalisa.gif - 1428 Bytes monalisa.gif - 1428 Bytes monalisa.gif - 1428 Bytes For O-maps it is not necessary to keep strictly to the contour interval. If a land form feature can be shown more clearly by raising or lowering a contour, this is allowable provided that the contour is not moved by more than 25% of the contour interval. To the competitor, this deviation should not be obvious. Additional land form detail can be shown by inserting form lines between contours, but care should be taken as an overuse of form lines reduces the legibility of a map drastically.
Note: It is important that time is set aside for practising. The survey of land forms is the most difficult part of survey!


Survey of land forms in contour detailed areas

Methods of fieldwork in areas with detailed contours are basically the same as in other terrain. But the main work will be to correct contour lines, to add small hills, knolls and depressions; in doing this, it is very important to keep the map readable and to omit details that are too small (even in brown, but especially in black and green). To give the competitor a true and clear picture of the area, the brown information must be prominent. It may be necessary to show the shape of the terrain with the aid of form lines, these can be used to show dimensions of features, the relative height of hills, the change in steepness, etc.

In this detailed terrain the mapper has to find a good compromise between information on the shape of land forms and details like boulders, knolls, cliffs and vegetation boundaries. It cannot be the aim of fieldwork to obscure all the contour information by black and green. Survey in this sort of terrain therefore requires much experience.


Vegetation and Runnability

Mapping vegetation correctly is important to the orienteer as it is one of the considerations that determines route choice.

monalisa.gif - 1428 BytesOpen areas have two basic classifications: open land and rough open land, the orienteer would expect rough open land to be slower than open land. Where these areas have occassional trees they as shown with different screens and classified as having scattered trees, but this should not affect the runnability. Where the runnability is slowed by undergrowth the appropriate undergrowth symbol is superimposed to indicate the lower runnability.

monalisa.gif - 1428 BytesForest and woodland has four classifications of runnability: easy running, slow running, difficult to run, difficult to walk. These indicate the density of the forest and the ease with which an orienteer can pass through ignoring other terrain conditions such as topography, marshes and rock detail. Where the forest is easy running but undergrowth prohibits that rate of progress the appropriate undergrowth symbol may be used.

monalisa.gif - 1428 Bytes monalisa.gif - 1428 BytesMappers often underestimate the ease with which orienteers can pass through even fairly dense forest, especially if they only view it from the path. A simple test can be employed to assess runnability of differing densities and types of vegetation.


Sequence of fieldwork

For beginners it is recommended that they split up their survey into line and point features, and land form survey. More experienced mappers avoid this method, because of the need to cover the terrain twice which increases time of survey and the risk of missing important information relating a feature to the land form. Each part of the map can be surveyed in one step. It is best to concentrate on one part in a small area, which can be surveyed in a few hours:


Classification and generalization

monalisa.gif - 1428 Bytes Before starting fieldwork, the standards for classification and generalization of all the different parts of the area have to be assessed by a short walk around to familiarize oneself with the terrain. In some kinds of terrain, it will not be possible to fix minimum dimensions for the whole area, they may have to be changed in some cases to increase the legibility during survey.
Note: For comments on the slide series on the interpretation of the drawing specification please refer to that manual. We remind you, that it will always be more meaningful to show this in the actual terrain and to use the slides only to show other possibilities and other kinds of terrain.


2kezbal.gif - 965 Bytes 2kezfel.gif - 998 Bytes 2kezjobb.gif - 969 Bytes