No abstract

Airborne laser scanning (ALS) is becoming an increasingly popular data capture technique for a variety of applications in urban surface modeling. Raw ALS data are captured and supplied as a 3D point cloud. Many applications require that these data are interpolated onto a regular grid in order that they may be processed. In this article, we identify and analyze the magnitudes and spatial patterning of residuals from ALS models of urban surfaces, at a range of different scales. Previous research has demonstrated the effects of interpolation method and scale upon the nature of error in digital surface models (DSMs), but the size and spatial patterning of such errors have not hitherto been investigated for urban surfaces. The contribution of this analysis is thus to investigate the ways in which different methods may introduce error, and to understand the uncertainty that characterizes urban surface models that are devised for a wide range of applications. The importance of the research is shown using examples of how the different methods may introduce different amounts of error and how the uncertainty information may benefit users of ALS height models. Our analysis uses a range of validation techniques, including split-sample, cross-validation, and jackknifing, to estimate the error created in DSMs of urban areas. IntroductionDigital elevation models (DEMs) are used in many analytical operations such as slope and aspect calculations, and processes such as image segmentation and filtering in the creation of bare-earth digital terrain models (DTMs). When combined in applications, these operations can be highly sensitive to the quality of the elevation data used (Wood 1994;Florinsky 2002). However, many end users of height models in geographical information systems are unaware of the issues surrounding the quality of the underlying height data, partly because of the lack of diagnostic tools for dealing with quality information in many commercial software packages (Desmet 1997). Even where users are aware of quality issues with the DEM, there is a lack of methods available to accommodate them in subsequent spatial analysis (Goodchild and Gopal 1989). The sum consequence is a rather limited understanding of the uncertainties inherent in applications using DEMs based on airborne laser scanning (ALS) data.Previous research on error in DEMs has been restricted to analyzing the nature of the errors within bare-earth surfaces in predominantly natural environments (e.g., MacEachren and Davidson 1987;Xie et al. 2003).Very few studies have begun to consider the problem of understanding error in urban surface models. Assessment of the quality of urban digital surface models (DSMs) poses very different challenges to those of evaluating bare-earth models, because of the inherent complexities of the urban surface-including the frequent discontinuities, the variety of textures and shapes, and the rate of surface change. This paucity of error investigations for urban DEMs arises in spite of the fact that high spatial resolution, accu...

Abstract. Formal models of urban systems have the potential to reveal a lot about the form and functioning of urban settlements, yet much of this potential has still to be realised. In this paper we focus on the extent to which this has reflected the dearth of digital data that arc rich* relevant, and disaggregate. Geodcmogrnphic classifications have made important and enduring contributions to small-area analysis. Yet, on the one hand, reliance upon census data makes them outdated and irrelevant and, on the other, fragmentation and diversification of social areas in cities has made the 'mosaic metaphor' of small-area analysis untenable. As part of the quest for a new perspective on data modelling, we investigate in this paper the potential of 'lifestyles' data sets for creating richer, more relevant digital models of human activity patterns in cities.

At the core of geographic information science (GIScience) lie enduring organising principles and concepts that have been developed using research methods that are robust, transparent and scientifically reproducible. Yet it is also a science of real-world problem solving, which has come to prominence at a time in which the nature and volume of geographic information (and the human activities that generate it) are profoundly changing. This article assesses and interprets these changes in the context of Goodchild's assessment of challenges to GIScience, using examples from 'geodemographics' -the analysis of people according the places where they live. The conclusions have repercussions not only for the way that we think about neighbourhood profiling, but also for the practice of GIScience itself, specifically with regard to reconciling new sources of 'big' spatial data and understanding the inherent vagaries of citizen science; linkage of conventional social, economic and demographic geographies to patterns of virtual interactions at fine levels of spatial granularity; and improving understanding of the ways in which 'open' geodemographics are specified, estimated and tested.