Geospatial data stores information about the location, shape and attributes of real objects.
So what? you might say; paper maps have been doing this for centuries. But it is the capturing of this information in digital form that makes it much easier to store and reproduce. It also enables the power of computers to be used in manipulating, updating and analysing the information in many different ways.
Scanning from paper maps
You are probably familiar with scanning technology already – many home and office PCs come with a desktop scanner. The scanner will take any printed image and take a picture of it. By capturing the image in digital form it can be stored on the computer and displayed on screen.
Scanning a map is a straightforward process and generally fast, but it does not provide for the capture of attribute information for features, such as the address of a building. Raster data uses up a lot of disk space, so rasterisation of maps by scanning is not always the most efficient method. However, it is very good for storing the cartographic style of the map.
Digitising from paper maps
Digitising requires the use of special equipment. The source map is laid flat on a table (tablet) and an electronic cursor is passed over the features of the map. In this way, each of the coordinate points which make up the different shapes can be identified. By clicking the cursor when it is held over a point, digitising captures map data in vector form.
For digitising to work, the tablet must have a magnetic field embedded in the flat surface, so as the cursor is moved around the map, its location can be identified.
Digitising can be very time consuming because every single point or vertex must be captured individually. Ordnance Survey's National Topographic Database currently contains more than 230 million features. You can imagine what a time consuming task it was to digitise it originally. Fortunately, the database is maintained by surveying methods that generate digital data directly.
When a cartographer is capturing information by digitising, it is possible to attach attribute information to features. Often, the digitising tablet has some kind of menu of feature types. Once a particular feature is digitised the resulting data contains information about its type and shape.
Some very specialised computer systems are able to convert raster data to vector data by recognising patterns in the image. For instance, it can guess that a sequence of coloured pixels which seem to form a line across the image are showing a linear feature of some kind. If the system knows the extent of the real position of the image, it can convert these shapes into vector information. This vectorisation from raster data can be a fast method of capture because it can be automated, but is usually less accurate than manual digitising.
Early surveying techniques
In simple terms, the job of the surveyor is to measure the size, shape and relative location of physical objects in the outside world. Size and distance are fairly easy – you can use physical measuring tools of stable and constant length to record these dimensions. Some of the earliest long measurements were made using glass rods end to end, to fix a distance between two points on the ground. Such rudimentary methods are still in use today.
From the earliest days of surveying, surveyors have exploited the rules of trigonometry to deduce distances between points on the ground without actually having to measure them directly. Once you have accurately recorded the distance between two points, you can then identify the distance to any third point by simply measuring the angles between all three. This process is called triangulation and was the basis for Ordnance Survey's original creation of detailed mapping for the whole of Britain. The theodolite was the traditional optical tool used to survey in this way, and more recently electrical devices were developed to conduct this kind of ground measurement.
Photogrammetry – remote sensing
Photogrammetry is the science of measuring objects from photographs. Historically, this meant using aerial photographs to capture topographic information. The first photogrammetric surveys were conducted more than 100 years ago. Now satellite pictures are also used to record the location and geometry of features on the ground. Remote sensing is another term describing the use of aerial and space imagery to record geographical information. It includes the interpretation of other phenomena such as vegetation type or land use shown in the Earth's reflectiveness to different wavelengths of electromagnetic radiation.
Initially, maps were created from aerial photographs by various kinds of tracing mechanism. Sophisticated devices were engineered to allow an operator to view and trace a pointer around the visible features on the photograph. Using a system of wheels and pulleys, this motion was mechanically reproduced by a drawing arm. Such machines used stereoscopic viewing to survey in 3-D (three dimensions). As technology has advanced, the techniques of digitising and scanning have become important aspects of photogrammetry. Ordnance Survey has captured a significant amount of its detailed mapping by digitising from aerial photographs. Remote sensing by satellite is now widely used for data capture and, as the accuracy increases, this method could replace ground survey and aerial photos.
The Global Positioning System (GPS)
GPS enables positioning of objects on or above the earth's surface in an absolute sense, not just in relation to other nearby features (as in the use of photogrammetry described previously, in which locations are defined relative to the known position of certain features in the image).
GPS can be used almost anywhere in the world, 24 hours a day, in all weathers. A constellation of 24 satellites orbit the earth and send signals that can be picked up by GPS receivers. GPS measurements are taken by computing the distance between the receiver and the satellite. If a receiver picks up signals from four or more satellites, a 3-dimensional position can be calculated. Certain methods can be used to increase the accuracy of the position to the 1cm level, either in real time or afterwards during post-processing.
GPS measurements are obtained in the GPS coordinate system: World Geodetic System 1984 (WGS84). Users should be aware that this position usually needs to be transformed into the local coordinate system for the region, OSGB36 in Great Britain, enabling GPS to be used alongside the local mapping. GIS data collectors can make use of the free GPS service provided to locate map objects and features directly in the field.
Ordnance Survey uses GPS to locate map detail. However, there are many other uses for GPS, such as navigating boats, planes or cars, monitoring the stability of structures, and providing location information for recreational users.
Whatever the method used for measuring the shape and location objects, modern surveyors rarely record information by hand drawing detail on a master survey document (see surveyor on left of picture with large board). Instead, they use hand-held pen computers equipped with flat, touch-sensitive screens (see surveyor on right of picture). These computers allow the surveyor to draw and click directly onto the screen to update map information while out in the field. Importantly, this means that new map features can be inputted directly as digital data.
Ordnance Survey surveyors use such a system known as PRISM – standing for Portable Revision and Integrated Survey Module. It enables the coordinates of new objects to be added in reference to the existing features. Features that have been demolished can be deleted while text names can be added using a freehand character recognition facility. So even if the geographical objects have been measured with the most rudimentary and time-honoured techniques – a tape measure for instance – the information will still be recorded in electronic form out in the field.