Abstract: This paper explains why it is vital to account for uncertainty when utilising socioeco-nomic data in a GIS, focusing on a novel and intuitive method to visually represent the uncertainty. In common with other data, it is not possible to know exactly how far from the truth socioeconomic data are. Therefore, when such data are used in a decision-making environment an approximate measure given for correctness of data is an essential component. This is illustrated, using choropleth mapping techniques on census data as an example. Both attribute and spatial uncertainty are considered, with Monte Carlo statistical simulations being used to model attribute uncertainty. An appropriate visualisation technique to manage certain choropleth issues and uncer-tainty in census type data is introduced, catering for attribute and spatial uncertainty simultaneously. This is done using the output from hierarchical spatial data structures, in particular the region quadtree and the HoR (Hexagon or Rhombus) quadtree. The variable cell size of these structures expresses uncertainty, with larger cell size indicating large uncertainty, and vice versa. This technique is illustrated using the New Zealand 2001 census data, and the TRUST (The Representation of Uncertainty using Scale-unspecific Tessellations) software suite, designed to show spatial and attribute uncertainty whilst simultaneously displaying the original data.

Abstract: This study explores the effect of spatial resolution on outcomes of dasymetric mapping using remotely sensed data, particularly the 30 m multispectral (MS) and 15 m panchromatic (PAN) spatial resolutions of Landsat Enhanced Thematic Mapper (ETM), 15 m of ASTER, and 1 m of airborne color infrared (CIR) imagery for Black Hawk County, Iowa. These images were classified to produce urban land cover maps. Then, dasymetric maps for population density were developed using residential areas derived from remotely sensed data and various GIS layers. The effect of spatial resolution on dasymetric maps was evaluated using traditional choropleth maps and a differentiation matrix. The accuracy of dasymetric maps with different spatial resolutions was assessed using highly detailed population density maps generated from the residential building footprints. The analysis demonstrates that dasymetric maps with higher spatial resolution tend to exhibit higher levels of accuracy. However, with higher spatial resolution data, ...

Abstract: Choropleth maps can be used to compare the patterns exhibited by different spatial variables In this paper, we develop an evolutionary algorithm that can be used to generate classifications that allow a user to explore the spatial patterns of multiple choropleth maps in terms of their visual correlation and the equality of area contained in each class Synthetic and census data are used to demonstrate the effectiveness of our approach

Abstract: Common Spatial Data Models (SDMs) such the vector, raster, and quadtree have well understood and widely practiced conventions of storage and visualization This paper explores what happens when the conventions of visualization are not strictly adhered to, and the SDMs are used in an atypical fashion A framework based on a quasi similarity measure is presented, which quantifies (in terms of "distance") the relationship between the storage format and the visualization output, following an accepted protocol This research used a transformation process (Tp) to define this distance Then, the atypical use of the quadtree SDM to represent choropleth spatial boundary uncertainty and attribute uncertainty was quantified using the same framework This research shows that if a SDM is used outside of its original context, then the distance between the storage format and its visual output can alter; in our case, the distance decreased This result was interpreted as evidence for the creation of a new spatial data st

Abstract: Traditionally, choropleth maps are used to examine spatial patterns present in a data distribution rather than to examine specific data values within that distribution. The transformation of initial numeric data values into graphic values for visual display differs for classed and unclassed maps. Classed maps first group data into classification intervals and each interval is then assigned a graphic value usually using the principle of maximum contrast. Unclassed maps skip the classification step and directly assign graphic value based on the original numeric data value. The paper examines the differences in defining algebraic-to-graphic transformation lines for both classed and unclassed classed maps and especially the role of the maximum contrast principle in constructing classed choropleth maps. Results show that the maximum contrast principle increases the visual complexity of spatial patterns in a classed map over the level present in a corresponding unclassed map.

Abstract: Cette these definit des differentes techniques de representations visuelles de l'incertitude dans les systemes d'information geographiques Elle permet de choisir des techniques efficaces de visualisation de l'incertitutde des attributs, l"incertitude des frontieres spatiales et l'incertitutde temporelle de Choropleth L"etude porte sur le recensement de 2001 en Nouvelle Zelande, appuyee par des enquetes Internet, des cartes explicatives et la realisation d'un logiciel Trust et propose de developper et de tester de nouvelles methodes de representation de l'incertitutde

Abstract: This is a revised version of the package published in The Stata Journal 4(4):361-378 (2004) for carrying out simple thematic mapping. This new release should be considered as a beta version: comments and problem reports to the author are welcome. After the final revision, the resulting version will be submitted for publication to The Stata Journal.

Abstract: Multivariate choropleth maps are often used to compare patterns of different spatial variables. This approach can be implemented by simultaneously drawing a series of choropleth maps, with each representing a particular variable. In this paper, we develop an evolutionary algorithm that can be used to generate a set of classifications that allow a user to explore the spatial patterns of multiple choropleth maps in terms of their visual correlation. Synthetic and census data are used to demonstrate the effectiveness of our approach. We also discuss the role of our approach in an interactive mapping environment and its implication for spatial data mining.

Abstract: The article describes a recently developed template for multivariate data analysis called conditioned choropleth maps (CCmaps). This template is a two-way layout of maps designed to facilitate comparisons. The template can show the association between a dependent variable, as represented in a classed choropleth map, and two potential explanatory variables. The data-analytic objective is to promote better-directed hypothesis generation about the variation of a dependent variable. The CCmap approach does this by partitioning the data into subsets to control the variation in the dependent variable that is associated with two conditioning variables. The interactive implementation of CCmaps introduced here provides dynamically updated map panels and statistics that help in comparing the distributions of conditioned subsets. Patterns evident across subsets indicate the association of conditioning variables with the dependent variable. The patterns lead to hypothesis generation about scientific relation...