Visualisation in the modern world is essential and foundational for communication. It is easy to overload the readers senses with too much information, yet with some time and effort we can convey and impart complex ideas and structures that otherwise may be difficult to explain through the written word alone. Good visualisation is the art of simplifying, creating context, and imparting meaning to data, to tell a coherent story.

The MinFuture project seeks to provide methods and guidelines for structuring MFA data, enabling a more comprehensive understanding of mineral and material systems. Visualisation is one of the key components described in the MinFuture Pyramid, while the MinFuture Core Dimensions describe four dimensions which are key to MFA studies: stages, trade, linkages, time. We add uncertainty and stocks to this list to create six core dimensions for which visualisation might be required.

Good visualisation

Visualisation matters because it is an essential part of the way we communicate information to others. Tufte (2006) comments that ‘the world we seek to understand is profoundly multivariate’ and therefore visualisations by association must also be multivariate. The aim of the visual designer is to draw out clarity from this complexity. Two types of visualisation tools are required for telling data driven stories. Elicitation tools are used for extraction and interrogation of the MFA data, to ensure credibility and extract clear narratives. Communication tools are used to convey the data structure and narrative to the reader. The process of telling a visual story is analogous to the Google Mapping approach which takes traditional maps, creates journey options and keeps standout routes. Similarly, for visual story-telling, MinFuture sees an interactive data environment where tables of MFA data (maps) are structured, allowing the creation of data stories (journeys) and communication of these stories to decision makers (standout routes).

Review of Visualisation Theory and Principles

There is a long history of visual theorists and designers defining principles of analytical design’ with key contributors including Edward Tufte, Jacque Bertin, Stephen Few, and Jock Mackinlay. Effective visualisation is found in simplicity, data visualised in its most naked or pure form, void of frivolous additions. Tufte (2006) presents six foundational principles of analytical design for communicating the essential information in visualisations:

• Principle 1: Show comparisons, contrasts, differences.
• Principle 2: Show causality, mechanism, explanation, systematic structure.
• Principle 3: Show multivariate data.
• Principle 4: Integrate words, numbers, images, diagrams.
• Principle 5: Describe the evidence.
• Principle 6: Content must be relevant, have integrity and be of significant quality.

Bertin in his book ‘Semiology of Graphics’ (1983) describes seven foundational variables of graphical perception, which relate to the way we perceive information through sight. The illustration below shows these position and retinal variables against their strengths and weakness for displaying information as a point, area and line.

Use of Visualisation in MFA

As a part of the MinFuture project we undertook of the types and forms of visualisation used across 48 MFA studies, sourced from research publications and online interactive models. The charts below show the frequency with which the MinFuture Core Dimensions (plus Uncertainty and Stocks) and Bertin’s retinal variables, were included in these studies. Clearly stages, time and stocks are important dimensions in MFA studies, while size is used almost exclusively for displaying quantitative data. Yet, apart from size, there was little consistency across the MFA studies in the use of retinal variables to display information. Many of the visuals reviewed were judged to be overly complicated and difficult to interpret. This finding suggests that designers of visualisations in the MFA community are mostly ignorant of the long history of visualisation theory and design principles. MinFuture seeks to redress this problem by providing a Best Practice Guide for Visualising MFA data.

Best Practice Guide for Visualising MFA

Sankey Diagrams are judged as the preferred primary diagram for visualising MFA data, as they convey both the material system structure and the quantitative values of material flows ins a clear manner. However, to communicate all of the MinFuture Core Dimensions, plus uncertainty and stocks, requires the use of secondary visuals (the most important are shown below.) These diagrams support the data and invite the viewer to see deeper insights. The use interactive visual platforms, allows ‘pop-up’ windows to be simply accessed with a ‘click’. A full catalogue of best visual options approaches for communicating the core dimensions is provided in this document.

Creating good visualisation

The following list of questions is useful creating good visuals from complex MFA data:

• What is it for?
• What is the best way to represent the information?
• How should I display multivariate information?
• What does the data look like? Sketch a wire frame
• Have I included titles and captions?
• Does it support the literature, is it consistent with other visualisations?
• Does the visualisation tell a narrative?
• Can the visualisation be made interactive?

Our final word is a plea that more time be given to creating visuals in MFA research. When a researcher undertakes a typical MFA study, the data collection takes many months, the writing up takes several, yet we are lucky to spend more than a few days on the visuals. However, the reader under time pressure looks first at the title, followed by the abstract, and then dwells on the figures. Good visualisation takes time and requires multiple iterations to perfect. Allocating more time to visualisation, not only helps us communicate our message well, but it also gives us the skills to create better visuals in the future. Practice makes perfect!