Showing papers on "Spatial analysis published in 1981"

Spatial Processes Models and Applications

Abstract: Describing the various ways the degree of spatial autocorrelation in a set of variate values can be assessed and to which the pattern formed by the location of objects treatable as points can be examined.

Spatial variability of flow parameters in a stratified sand

Abstract: The spatial variability in porosity, hydraulic conductivity, compressibility, and various grain size fractions is analyzed for several sets of samples from the Quadra Sand. This unit is a well-sorted, medium grained, horizontally stratified sand with relatively few silt or gravel interbeds. Both random and uniformly spaced sample plans are used. The heterogeneity of the flow parameters is characterized by frequency histograms and their estimated moments, by their sample autocorrelation functions, and the estimated power spectra. Emphasis is placed on the nature of the spatial dependence between neighboring values of the flow parameters. A nearest neighbor stochastic process model is fit to the data to consider its adequacy in describing the spatial dependence within the porosity and hydraulic conductivity sequences. Even though the Quadra Sand is relatively uniform, a fairly complex spatial structure is observed. A simple monotonically decaying autocorrelation function may not adequately represent the spatial continuity. Statistical anisotropy is observed in both the extent of the spatial autocorrelation and in its functional form. Results show the importance of scale in constructing a probability model to describe the spatial variability.

Destination Choice and Processing Spatial Information: Some Empirical Tests with Alternative Constructs

Abstract: A number of researchers in geography [5; 6; 7; 10; 11; 12; 13; 24; 28] have carried out studies which incorporate various disaggregate spatial choice models (integration theory and functional measurement; multiattribute attitudinal, markovian, MDS and other scaling models; portfolio theory). The findings are inconclusive in the absence of commonly accepted theories of choice behavior, and the models themselves vary substantially in their ability to predict actual patterns of spatial preference and spatial choice. Details of these approaches are, however, well-documented [4; 9; 14]. A noticeable gap in behavioral research has been the limited number of empirical comparisons of alternative modeling approaches. Certainly, there are exceptions [7; 25], but most geographic studies merely highlight conceptual differences [1; 18; 27]. This has hampered necessary advances in behavioral geography. Central to the development of a theory of choice is a procedure (i.e., a decision rule) that specifies how information is to be processed in order to arrive at a selection. The functional form of the decision rule has been a basic concern of geographers. However, a point of controversy which has not been satisfactorily resolved is: to what extent is compensation (or non-compensation)' a modeling assumption, and to what extent is it a substantive finding? The issue is complicated by the fact that empirical tests demonstrate that both compensatory and noncompensatory decision rules can exhibit a reasonable degree of convergent validity under experimental conditions [7; 13; 23]. This paper examines the predictive efficacy of three different modes of processing subjective information in spatial choice situations. The research is innovative in two distinct ways. First, the results are validated by using data collected from the same sample population to predict how subjects make preference and patronage judgments among spatial opportunities. Second, in terms of methodology, two compensatory models (additive conjoint measurement-a decomposition model; and multinomial logit-a composition model)2 are compared di-

The spatial autocorrelation matrix

Abstract: Among the numerous approaches proposed for the analysis of spatial processes, the autocorrelation index (Geary, 1954) introduced by Cliff & Ord (1973) possesses remarkable properties. For any given variable (presence/absence, cover, quantitative measures of abundance) and random distribution of sampling points in space, this index tests the null hypothesis of absence of correlation for values recorded at two neighbouring points, using a non parametric model for equiprobability of N ! attributions of N numerical values at N sampling points.

Texture Features in Remote Sensing Imagery

Abstract: The remote-sensing main automatic approach is nowadays based on pixel radiometry and so necessarily incomplete as it takes no account of its environment. Our aim is to improve the real efficiency of spatial information (texture), the principle being that information derived from one pixel should be regarded as a “spectral-spatial” information pair. It seems indeed obvious that texture is going to assume an ever increasing importance as spatial resolution of satellite imagery becomes more and more precise.

Stochastic models of cover class dynamics

Abstract: Investigations related to satellite remote sensing of vegetation have been concerned with questions of signature identification and extension, cover inventory accuracy, and change detection and monitoring. Attention is given to models of ecological succession, present directions in successional modeling and analysis, nondynamic spatial models, issues in the analysis of spatial data, and aspects of spatial modeling. Issues in time-series analysis are considered along with dynamic spatial models, and problems of model specification and identification.

Ordinal econometrics in regional and urban modelling

Abstract: Geographical research is often suffering from inaccurate and unreliable data This paper deals with the treatment of soft (ordinal) data in spatial statistics and econometrics After a brief discussion of soft spatial data, a set of recently developed techniques is introduced aiming at drawing quantitative (cardinal) inferences from a soft data input These techniques are inter alia multidimensional scaling, rank order statistics, logit analysis, interdependence analysis, discriminant analysis and canonical correlation A key stone of this paper is formed by combining Theil's approach to logit analysis with Kendall's rank correlation method based on pairwise comparisons The paper offers several new perspectives for the treatment of soft data in multivariate techniques (such as multiple regression and clustering analysis) and in decision problems

Simulation and mapping of space-dependent data in geology

Abstract: This paper shows how it can be possible to generate the values of two-dimensional variables (top of stratigraphic horizon, thickness or grade of a vein, trace-element content of stream-sediment samples, etc.) on a very fine grid around existing control points. The simulated values show the same probability distribution and spatial autocorrelation as existing data. Also, the similarity between the control data and the neighbour simulated values is in accordance with the spatial autocorrelation. It is shown that the advantage of the maps of simulated values over maps of interpolated values is the ability to restore the natural variability of the data and thus lead to more-accurate estimates of the proportion of data above a given threshold in any region.

The role of remotely sensed and other spatial data for predictive modeling: the Umatilla, Oregon example

Abstract: Developpement et experimentation de techniques appliquees a des modeles de prevision pour evaluer les besoins d'irrigation dans le bassin de l'Umatilla