Generalisation
As maps represent reality at a very reduced scale, no map, even at large scales, can show everything. It would be totally illegible and very chaotic if we did. This is because there is a reduced amount of space available on a page/screen to represent features compared to the real world and this space reduces as the scale decreases.
Before jumping into generalisation, it is important to think carefully about the purpose of the map, who the user is and how the map will be used, as this will help determine the scale of your map and what features will be required to meet the requirements of the user. The smaller the scale of your map, the less information you’ll be able to show. For example, on a map of the world, you won’t be able to show every building outline or every road. A larger scale map of the same area will require more space, which may not be possible. It’s a balancing act but the general rule is that if the feature is not critical to the purpose or understanding of the map, it should be removed.
Once you’ve decided what features you need on the map, you will likely need to generalise them (reduce their complexity) in order to maintain their legibility when moving to a smaller scale. There are many techniques which can be deployed to achieve this aim. Here are a few of the techniques cartographers use to generalise a map:
Generalisation Techniques
Elimination: Removing certain features which are not necessary at the new scale such as removing a small building from a built-up area or removing individual bus stops. It might also include removing certain classes of features such as removing all minor roads and keeping just the major ones – this may not work for every use case as a minor road in a rural area may be more important than in an urban area – selective elimination, at the discretion of the cartographer, may therefore be required.
Amalgamation (or combination): The joining together of adjacent objects of the same type, such as merging individual buildings into a single built-up area, or multiple patches of woodland into a single wooded area.
Simplification: Makes features legible by removing some of the variation such as smoothing out meanders in a river or removing some of the smaller tributaries of a river network. The key is to maintain the general characteristics of the feature you are simplifying.
Exaggeration: Small or narrow features can be increased in size in order to maintain their legibility whilst retaining a defining feature. For example, maintaining a visible opening to a bay or showing roads wider on the map than their true scale to make them visible.
Displacement: Moving features apart to ensure clarity, such as moving a road and river apart to prevent them overlapping and create visible separation between the features when the scale is reduced.
Collapse: Similar to amalgamation but may result in the change of dimension of a feature such as a polygon to a point symbol. For example, a city may be shown as a polygon at large scales but as a single point at smaller scales. Similarly, dual carriageways shown with a central reservation may be reduced to just a single line road feature.
But how do we decide how to generalise things?
Still to this day, the method of generalisation and level of generalisation is down to the discretion of cartographers. It is a cartographer’s job to understand what the user requirements are and the purpose of the map and then select and symbolise the data appropriately. It is often quite subjective and varies from place to place (and cartographer to cartographer). For example, a small building in an urban area may be eliminated due to its size, however the same sized building in a very rural area, where it is the only building for miles and thus a key feature of the landscape, would be retained. Features are also interrelated, making generalisation complex. For example, if you simplified a river network by ‘smoothing’ out a few of the meanders, the contours would have to be simplified in a way that they match the river network such that rivers appear at the lowest point in the topography and flow downhill.
Successful generalisation ensures that the graphic legibility of features are retained, the general characteristics of the features are retained, and the relative importance of features and the spatial relationships between them are considered.
Computer generalisation
More recently, computer algorithms have been developed which allow automatic generalisation, based on a set of pre-programmed rules. However, the success of these is still quite varied and often human judgement gives more pleasing results.
Last updated
