Depicting a map is not just a matter of personal taste. If our purpose is essentially to convey certain information in the clearest way possible, we must not forget that the format and final use of the map has to be taken into account when choosing the colour palette. An online site of Penn University will assist you with your choices. The site is http://colorbrewer2.org/
Firstly, you will not load your map in the site. The displayed map is only a fixed support on which will appear the different choices that you will make in order to see and judge them.
The choices ColorBrewer will offer you are the following:
- choice of colours according to the »Possible use of your map (screen, print,photocopy, …)
- choice of pallets according to the type of data (continuous, classified, …)
- choice of colours according to the number of colours (classes) to be visualized
Depending on the choices you make fort the number of classes, the type of data and the palette, you will have symbols indicating which formats are suitable and which are not. You have a symbol bar:
From left to right the symbols indicate the suitability of the chosen pallet with:
- the aesthetics of visualization
- the map photocopy
- LCD display
- colour printing
If the symbol is highlighted, it indicates that the choice of colours is in line with this use.
If the symbol is crossed by an X, it indicates that the choice of colours is not adequate for this use.
If the symbol is crossed out by a? it indicates that the choice of colours may not be adequate for this use.
At a glance you have a summary of the advantages and disadvantages of the chosen palette
Selecting the number of classes
Selecting the number of data classes is very important when designing a map. Increasing the number of data classes will result in a map with plenty of information at the expense of the data generalization. However, too many classes of data can overwhelm the reader and distract him from perceiving the general trends in the distribution of the phenomenon. In addition, a large number of classes can compromise readability because it requires more colours that become more and more difficult to distinguish.
Many cartographers will advise you to use between five and seven classes for a choropleth map. Maps of isolines or choropleths with very regular spatial distributions can safely use more classes of data because similar colours are seen side by side, making them easier to distinguish.
Choosing the nature of your data
1. Sequential patterns are adapted to organized data that progress from the lowest to the highest. Brightness changes dominate the style of these schemes, with light colours for low data values to dark colours for high data values.
2. The divergent patterns emphasize just as much in the middle of the range of values as in the extremes of the data range. The critical class in the middle of the legend is highlighted with light colours and the lower and upper limits are highlighted with dark colours that have contrasting hues.
The divergent patterns are well adapted when the class in the middle of the sequence is important in relation to the mapped data, for example when it refers to the average, median, or zero. The colours will be darker to represent the distance of the data from this pivot value.
3. Qualitative patterns do not reflect differences in magnitude among the classes of the legend. Hues are used to create visual differences among primary classes. The qualitative schemes are best suited to the representation of nominal or organized in categories data.
Most of the qualitative schemes are based on differences in hue with very little difference in brightness among colours. You can choose a subset of colours from a palette with more classes than you actually need.
Compared to what has just been said, there are two exceptions to the rule:
Matched classes: This diagram shows a series of brightness pairs for each colour (for example, light green and dark green). Use this option when you have categories that should be visually related, even if they are not explicitly ordered. For example,”forest” and “wood” would be adequately represented with dark green and light green accent scheme
Highlighted diagram: It is used to highlight certain classes that are of particular interest. While all classes are represented by similar brightness tones, these classes are represented by dark tones. This makes them ” obvious “. Warning! Beware of highlighting classes that have no particular interest.
If you have a well-defined final support, such as printing, you can check the corresponding box in “Only show:”. The pallets proposed will necessarily be those which are adequate for this type of product.
In “Context “you can check Roads to bring up a line layer and Cities to reveal a layer of points. You can give a different colour to each of these layers to see the suitability of the colour combination. It goes without saying that Roads can represent a stream layer or any other linear type layer.
The drop-down menu allows you to choose the desired colour code type:hexadecimal, RGB or CMYK If you click EXPORT you have a series of options for exporting the palette:
It’s simple but effective.