Adjacent versus coincident representations of geospatial uncertainty: Which promote better decisions?

Abstract

International audience3D geological models commonly built to manage natural resources are much affected by uncertainty because most of the subsurface is inaccessible to direct observation. Appropriate ways to intuitively visualize uncertainties are therefore critical to draw appropriate decisions. However, empirical assessments of uncertainty visualization for decision making are currently limited to two-dimensional map data, while most geological entities are either surfaces embedded in a 3D space or volumes. This paper first reviews a typical example of decision making under uncertainty, where uncertainty visualization methods can actually make a difference. This issue is illustrated on a real Middle East oil and gas reservoir, looking for the optimal location of a new appraisal well. In a second step, we propose a user study that goes beyond traditional 2D map data, using 2.5D pressure data for the purposes of well design. Our experiments study the quality of adjacent versus coincident representations of spatial uncertainty as compared to the presentation of data without uncertainty; the representations quality is assessed in terms of decision accuracy. Our study was conducted within a group of 123 graduate students specialized in geology

How to Assess Visual Communication of Uncertainty? A Systematic Review of Geospatial Uncertainty Visualisation User Studies

“…While most studies assess coincident approaches, there are a number of studies that involve a direct comparison between adjacent and coincident views (Aerts et al, 2003;Edwards and Nelson, 2001;Evans, 1997;Gerharz and Pebesma, 2009;Kardos, 2003;Kardos, 2007;Kubíček and Šašinka, 2011;Kunz et al, 2011;MacEachren et al, 1998;Retchless, 2012;Senaratne et al, 2012;Viard et al, 2011).…”

“…'which entity is more uncertain?') (Alberti, 2013;Blenkinsop et al, 2000;Evans, 1997;Kinkeldey et al, 2014;MacEachren et al, 1998;MacEachren et al, 2012;Schweizer and Goodchild, 1992;Viard et al, 2011). Some studies involve comparisons for which subjects have to aggregate uncertainty over an area or several entities first.…”

“…extremely high or low values. Several studies incorporate such tasks, for example, the search for the highest data value (Viard et al, 2011), the lowest and/or the highest uncertainty (Sanyal et al, 2009;Wray, 2007), or the identification of patterns such as clusters in the data (Edwards and Nelson, 2001;MacEachren et al, 1998) or in uncertainty only (Drecki, 2002;Edwards and Nelson, 2001;Sanyal et al, 2009). Sanyal et al (2009) extend this task type further to multiple entities using a counting task for data values and uncertainty meaning that the number of clusters in data and uncertainty has to be determined.…”

“…Aerts et al, 2003;Edwards and Nelson, 2001;Leitner and Buttenfield, 2000;MacEachren et al, 1998;Nadav-Greenberg et al, 2008;Retchless, 2012;Schweizer and Goodchild, 1992;Slocum et al, 2003), transparency (e.g. Drecki, 2002;Newman and Lee, 2004;Slocum et al, 2003;Viard et al, 2011), or colour saturation that is used in a number of studies (e.g. Drecki, 2002;Kubíček and Šašinka, 2011;Kunz et al, 2011;Leitner and Buttenfield, 2000;Retchless, 2012;Sanyal et al, 2009).…”

“…Owing to requirements of statistical analysis, the number is much higher in studies using quantitative approaches. Lab studies, the most common evaluation type, range from 9 (Zhang et al, 2008) to 123 (Viard et al, 2011) participants -the median is 31. Compared to studies carried out in a laboratory, web-based studies have the advantage that high numbers of subjects can easily be recruited; the participant numbers in webbased studies reviewed reflect this as they range from 32 (Kinkeldey et al, 2014) to 274 (Retchless, 2012) with a median of 82.…”