types of visualisation

This post will be rather technical. While reading the book  “Data Visualization: a successful design process’” by Andy Kirk, I started questioning whether we can apply his categorisation of visualisation types to scientific visualisations.

The book summarises three main functions of visualisation:

  1. Convey an explanatory portrayal of data to a reader

  2. Provide an interface to data in order to facilitate visual exploration

  3. Use data as an exhibition of self-expression

Most of the time, scientific results follow an explanatory approach – their function is to explain. It is about conveying information to a reader. In this example, the decline of sea ice extent:

 

Example of explanatory visualisation. Taken from the ‘Climate Energy Balance of the Earth’ by Alan Betts (http://alanbetts.com/workspace/uploads/climate-energy-balance-1366392908.pdf)
Example of explanatory visualisation. Taken from the ‘Climate Energy Balance of the Earth’ by Alan Betts (http://alanbetts.com/workspace/uploads/climate-energy-balance-1366392908.pdf)

 

In contrast to the explanatory-based function, the exploratory data visualisation seeks to provide us with a visual analysis instead of just a visual presentation of data. This function is used in science too but more often in the analysis process than for the data presentation. It is clearly a function for scientists to find patterns before publication. These findings require then the use of an explanatory-based visual evidence – which is often a hard step and not always successfully done. However, choosing the exploration form of visualisation means that the scientist needs to come up with a clear portrayal of interesting findings.

Here are two nice examples of exploratory data visualisations of scientific results:

The Global Carbon Atlas is a platform to explore and visualize the most up-to-date data on carbon fluxes resulting from human activities and natural processes:

CarbonAtlas_exp

The global forest change analysed and visualised by the University of Maryland:

forestGoogleMaryland_exp

The third function, to exhibit data as a means of self-expression, is very rare in science. It is sometimes referred to as ‘data art’. It aims rather for an aesthetic reaction than for a clear exploration or explanatory portrayal. I would consider the wind and ocean currents by Cameron Beccario as an example of convincing exhibitions-type design of scientific data:

Picture from Cameron Beccario, for more information see: http://twitter.com/cambecc or http://earth.nullschool.net

 

a visualisation by Cameron Beccario of global wind conditions forecast by supercomputers updated every three hours updated every five days.  He continuously adds other climate parameters, i.e. relative humidity, air temperature or total cloud water:

or this hybrid form – an explorative-exhibitation type – by the Guardian showing which fossil fuel companies are most responsible for climate change:

guardian_co2

Leave a Reply

Your email address will not be published. Required fields are marked *