Showing papers on "Spatial analysis published in 1996"

The modifiable areal unit problem and implications for landscape ecology

Abstract: Landscape ecologists often deal with aggregated data and multiscaled spatial phenomena. Recognizing the sensitivity of the results of spatial analyses to the definition of units for which data are collected is critical to characterizing landscapes with minimal bias and avoidance of spurious relationships. We introduce and examine the effect of data aggregation on analysis of landscape structure as exemplified through what has become known, in the statistical and geographical literature, as the Modifiable Areal Unit Problem (MAUP). The MAUP applies to two separate, but interrelated, problems with spatial data analysis. The first is the “scale problem”, where the same set of areal data is aggregated into several sets of larger areal units, with each combination leading to different data values and inferences. The second aspect of the MAUP is the “zoning problem”, where a given set of areal units is recombined into zones that are of the same size but located differently, again resulting in variation in data values and, consequently, different conclusions. We conduct a series of spatial autocorrelation analyses based on NDVI (Normalized Difference Vegetation Index) to demonstrate how the MAUP may affect the results of landscape analysis. We conclude with a discussion of the broaderscale implications for the MAUP in landscape ecology and suggest approaches for dealing with this issue.

An autologistic model for the spatial distribution of wildlife

Abstract: 1. A new method for estimating the geographical distribution of plant and animal species from incomplete field survey data is developed. 2. Wildlife surveys are often conducted by dividing a study region into a regular grid and collecting data on abundance or on presence/absence from some or all of the squares in the grid. Generalized linear models (GLMs) can be used to model the spatial distribution of a species within such a grid by relating the response variable (abundance or presence/absence) to spatially referenced covariates. 3. Such models ignore or at best indirectly model dependence on unmeasured covariates, and the intrinsic spatial autocorrelation arising for example in gregarious populations. 4. We describe a procedure for use with presence/absence data in which spatial autocorrelation is modelled explicitly. We achieve this by extending a logistic model to include an extra covariate which is derived from the responses at neighbouring squares. The extended model is known as an autologistic model. 5. To allow fitting of the autologistic model when only a random sample of squares is surveyed, we use the Gibbs sampler to predict presence/absence at unsurveyed squares. 6. We compare the autologistic model with the ordinary logistic model using red deer census data. Both models are fitted to a subsample of 20% of the data and results are compared with the 'true' abundance and spatial distribution indicated by the full census. We conclude that the autologistic model is superior for estimating the spatial distribution of the deer, whereas the ordinary logistic model yields more precise estimates of the overall number of squares occupied by deer at the time of the survey.

Variowin: Software for Spatial Data Analysis in 2D

Abstract: This pack provides Windows software for analyzing spatial data and graphical tools for variogram surfaces, directional variograms, h-scatterplots and variogram clouds. By making use of several measures of spatial continuity, it provides an environment for geostatistical estimation and simulation. The four components to the software enable researchers in spatial statistics and geostatistics to interactively analyze and model the spatial continuity of spatial data, produce variogram surfaces and directional variograms in any direction, interactively mask data pairs thought to adversely affect measures of spatial continuity, and produce maps of experimental and modelled variogram surfaces.

Effects of changing spatial resolution on the results of landscape pattern analysis using spatial autocorrelation indices

Abstract: Understanding the relationship between pattern and scale is a central issue in landscape ecology. Pattern analysis is necessarily a critical step to achieve this understanding. Pattern and scale are inseparable in theory and in reality. Pattern occurs on different scales, and scale affects pattern to be observed. The objective of our study is to investigate how changing scale might affect the results of landscape pattern analysis using three commonly adopted spatial autocorrelation indices,i.e., Moran Coefficient, Geary Ratio, and Cliff-Ord statistic. The data sets used in this study are spatially referenced digital data sets of topography and biomass in 1972 of Peninsular Malaysia. Our results show that all three autocorrelation indices were scale-dependent. In other words, the degree of spatial autocorrelation measured by these indices vary with the spatial scale on which analysis was performed. While all the data sets show a positive spatial autocorrelation across a range of scales, Moran coefficient and Cliff-Ord statistic decrease and Geary Ratio increases with increasing grain size, indicating an overall decline in the degree of spatial autocorrelation with scale. The effect of changing scale varies in their magnitude and rate of change when different types of landscape data are used. We have also explored why this could happen by examining the formulation of the Moran coefficient. The pattern of change in spatial autocorrelation with scale exhibits threshold behavior,i.e., scale effects fade away after certain spatial scales are reached (for elevation). We recommend that multiple methods be used for pattern analysis whenever feasible, and that scale effects must be taken into account in all spatial analysis.

Measurement of genetic structure within populations using moran's spatial autocorrelation statistics

Abstract: Spatial structure of genetic variation within populations, an important interacting influence on evolutionary and ecological processes, can be analyzed in detail by using spatial autocorrelation statistics. This paper characterizes the statistical properties of spatial autocorrelation statistics in this context and develops estimators of gene dispersal based on data on standing patterns of genetic variation. Large numbers of Monte Carlo simulations and a wide variety of sampling strategies are utilized. The results show that spatial autocorrelation statistics are highly predictable and informative. Thus, strong hypothesis tests for neutral theory can be formulated. Most strikingly, robust estimators of gene dispersal can be obtained with practical sample sizes. Details about optimal sampling strategies are also described.

The spatial autocorrelation coefficient Moran's I under heteroscedasticity.

Abstract: The spatial distribution of rates used in epidemiology often raises questions concerning the randomness of the observed pattern. In order to provide a first answer to this kind of question, the well-known spatial autocorrelation coefficient Moran's I is frequently used. Unfortunately, under heteroscedasticity, that is, unequal variances of the rates due to different population sizes, the moments of the test distribution of Moran's I under H(0) differ from the usually used moments. To obtain a less biased test, it is proposed in this paper and validated by simulation results, to approximate the moments of Moran's I by means of incorporating population size into the covariance matrix of the rates.

Multifractal modeling and spatial statistics

Abstract: In general, the multifractal model provides more information about measurements on spatial objects than a fractal model. It also results in mathematical equations for the covariance function and semivariogram in spatial statistics which are determined primarily by the second-order mass exponent. However, these equations can be approximated by power-law relations which are comparable directly to equations based on fractal modeling. The multifractal approach is used to describe the underlying spatial structure of De Wijs 's example of zinc values from a sphalerite-bearing quartz vein near Pulacayo, Bolivia. It is shown that these data are multifractal instead of fractal, and that the second-order mass exponent (=0.979±0.011 for the example) can be used in spatial statistical analysis.

Integrated Analysis of Spatial Data from Multiple Sources: Using Evidential Reasoning and Artificial Neural Network Techniques for Geological Mapping

Abstract: shows different slopes with an ordinal scale such as flat, As the availability of digital spatial data, other than from re- middle, and steep. Remote sensing images, on the other mote sensing, increases, it becomes increasingly important to hand, record radiance of surface targets with a ratio scale. develop algorithms to handle both remote sensing and other From a user's point of view, it is difficult to use all these spatial data. F~~ purposes, common~y used re- digital data together because they have different levels of re

Multifactor spatial analysis for landfill siting

Abstract: Siting a landfill typically requires processing a significant amount of spatial data with respect to various siting rules, regulations, factors, and constraints. Manually performing such a spatial analysis with drawing tools is generally tedious. A modern geographical information system (GIS), although capable of manipulating spatial data to facilitate the analysis, lacks the ability to locate an optimal site when compactness and other factors are simultaneously considered. An appropriate siting model was therefore explored for use with a raster-based GIS. A mixed-integer programming model was developed to obtain a site with optimal compactness. A comparison was made between the model and two other previously proposed models in terms of their applicability and simplicity for raster-based data. The compactness model was further extended to include multiple siting factors with weights determined using map layer analysis functions provided by a GIS. This multifactor model was applied to analyze the effects of varied weights and factors on making a siting decision.

Database design for a multi-scale spatial information system

Abstract: Growth in the available quantities of digital geographical data has led to major problems in maintaining and integrating data from multiple sources, required by users at differing levels of generalization. Existing GIS and associated database management systems provide few facilities specifically intended for handling spatial data at multiple scales and require time consuming manual intervention to control update and retain consistency between representations. In this paper the GEODYSSEY conceptual design for a multi-scale, multiple representation spatial database is presented and the results of experimental implementation of several aspects of the design are described. Object-oriented, deductive and procedural programming techniques have been applied in several contexts: automated update software, using probabilistic reasoning; deductive query processing using explicit stored semantic and spatial relations combined with geometric data; multiresolution spatial data access methods combining poini,...

Spatial pattern and process in plant-pathogen interactions

Abstract: An individual host's likelihood of acquiring an infectious disease depends, in large part, on the location of the host relative to sources of infection, proximity to other hosts, and occupation of specific microhabitats that confer increased susceptibility. Spatial organization of the host and pathogen populations then may be critical in determining patterns of disease occurrence and dynamics. We discuss three fundamental problems as- sociated with understanding the interactions of plants, pathogens, and spatial structure: (1) how one characterizes spatial patterns, (2) how we determine spatial dynamic processes from a given spatial pattern, and (3) how we then simulate the spatial dynamics presumed to be dominating a given host-pathogen interaction. To demonstrate methods for the characterization of spatial pattern, we analyzed spatial maps of a Silene latifolia population infected with the anther smut Microbotryum violaceum, using join-count statistics, continuous spatial autocorrelation, and Mantel's test. Different statistical techniques provided different interpretations of the same data set, indicating the value of using multiple methods. We described spatial structure in the plant population, the pathogen population (factoring out the plant population structure), and spatial cross correlation between two variables (plant gender and disease status). Each of these tests provides information on how the disease may be spreading in the population. We are not very optimistic about the prospect of determining underlying disease-spread processes purely from analysis of spatial pattern. Most spatial patterns can potentially be generated by a number of biological processes (e.g., "true" vs. "apparent" contagion), and it is not possible to distinguish between hypotheses without additional information about the system. A difficulty in modeling spatial processes is simulation of populations with given spatial patterns. Simulated spatially structured populations are useful for predicting disease dy- namics in a spatial context. We present a number of techniques for creating these spatially structured populations, including a method we developed for modeling the effect of vector behavior on the spread of a plant virus.

Spatial analysis of the distribution of tsetse flies in the Lambwe Valley, Kenya, using Landsat TM satellite imagery and GIS

Abstract: Satellite imagery, geographic information systems (GIS) and spatial statistics provide tools for studies of population dynamics of disease vectors in association with habitat features on multiple spatial scales. Tsetse flies were collected during 1988-90 in biconical traps located along transects in Ruma National Park in the Lambwe Valley, western Kenya. Fine spatial resolution data collected by Landsat Thematic Mapper (TM) satellite and reference ground environmental data were integrated in a GIS to identify factors associated with local variations of fly density. Statistical methods of spatial autocorrelation and spatial filtering were applied to determine spatial components of these associations. Strong positive spatial associations among traps occurred within transects and within the two ends of the park. From satellite data, TM band 7, which is associated with moisture content of soil and vegetation, emerged as being consistently highly correlated with fly density. Using several spectral bands in a multiple regression, as much as 87% of the variance in fly catch values could be explained. When spatial filtering was applied, a large component of the association between fly density and spectral data was shown to be the result of other determinants underlying the spatial distributions of both fly density and spectral values. Further field studies are needed to identify these determinants. The incorporation of remotely sensed data imagery into a GIS with ground data on fly density and environnmental conditions can be used to predict favourable fly habitats in inaccessible sites, and to determine number and location of fly suppression traps in a local control programme.

As the Crow Flies: Bias in Consumers' Map-Based Distance Judgments

Abstract: Consumers make distance judgments when they decide which store to visit or which route to take. However, these judgments may be prone to various spatial perception biases. While there is a rich literature on spatial perceptions in urban planning and environmental and cognitive psychology, there is little in the field of consumer behavior. In this article we introduce the topic of spatial perceptions as an area of research in marketing. We extend the literature on spatial perceptions by proposing that consumers use the direct distance between the endpoints of a path, or the distance “as the crow flies,” as a source of information while making distance judgments—the shorter the direct distance, the shorter the distance estimate. We study two spatial features that affect direct distance—path angularity (i.e., the size of the angle between path segments) and path direction (i.e., whether the path retraces back or not). We further propose and demonstrate that the direct-distance bias is due to the perceptual salience of direct distance and is used by consumers in an automatic manner. Theoretical implications for the manner in which consumers process spatial information and the use of cognitive heuristics while making spatial judgments are discussed.

Efficient algorithms for constructing proper higher order spatial lag operators

Abstract: . This paper extends the work of Blommestein and Koper (1992)–BK–on the construction of higher-order spatial lag operators without redundant and circular paths. For the case most relevant in spatial econometrics and spatial statistics, i.e., when contiguity between two observations (locations) is defined in a simple binary fashion, some deficiencies of the BK algorithms are outlined, corrected and an improvement suggested. In addition, three new algorithms are introduced and compared in terms of performance for a number of empirical contiguity structures. Particular attention is paid to a graph theoretic perspective on spatial lag operators and to the most efficient data structures for the storage and manipulation of spatial lags. The new forward iterative algorithm which uses a list form rather than a matrix to store the spatial lag information is shown to be several orders of magnitude faster than the BK solution. This allows the computation of proper higher-order spatial lags “on the fly” for even moderately large data sets such as 3,111 contiguous U. S. counties, which is not practical with the other algorithms.

Topological queries in spatial databases

Abstract: Handling spatial information is required by many database applications, and each poses different requirements on query languages. In many cases the precise size of the regions is important, while in other applications we may only be interested in the TOPOLOGICAL relations- hips between regions — intuitively, those that pertain to adjacency and connectivity properties of the regions, and are therefore invariant under homeomorphisms. Such differences in scope and emphasis are crucial, as they affect the data model, the query language, and performance. This talk focuses on queries targeted towards topological information for two- dimensional spatial databases, where regions are specified by polynomial inequalities with integer coeficients. We focus on two main aspects: (i) languages for expressing topological queries, and (ii) the representation of topological information. In regard to (i), we study several languages geared towards topological queries, building upon well-known topologi- cal relationships between pairs of planar regions proposed by Egenhofer. In regard to (ii), we show that the topological information in a spatial database can be precisely summarized by a finite relational database which can be viewed as a topological annotation to the raw spatial data. All topological queries can be answered using this annotation, called to- pological invariant. This yields a potentially more economical evaluation strategy for such queries, since the topological invariant is generally much smaller than the raw data. We examine in detail the problem of transla- ting topological queries against the spatial database into queries against the topological invariant. The languages considered are first-order on the spatial database side, and fixpoint and first-order on the topological in- variant side. In particular, it is shown that fixpoint expresses precisely the PTIME queries on topological invariants. This suggests that topolo- gical invariants are particularly well-behaved with respect to descriptive complexity. (Based on joint work with C.H.Papadimitriou, D. Suciu and L. Segoufin.)

Spatial Statistics and GIS: Software Tools to Quantify Spatial Patterns

Abstract: With the increasing use of GIS by planners, statistics routines are needed that quantify relationships taking into account spatial location. Spatial statistics is a branch of statistics that includes measures of spatial distribution, spatial autocorrelation, and spatial association. Five statistics packages are described here that calculate various spatial indices useful for planners. In the future, planners will use these methods along with GIS programs to draw more rigorous and quantifiable deductions from their data.

Spatial autocorrelation and optimal spatial resolution of optical remote sensing data in boreal forest environment

Abstract: The aim of the study was to examine the spatial autocorrelation and the optimal spatial resolution of optical remote sensing images in a forested landscape. Also the effect of tree species and forest age on this optimum was investigated. Two different methodologies were applied. Semivariograms were used to measure the autocorrelation of pixels while local variance curves were used to define the spatial resolution that maximizes the variance between neighbouring pixels. The range of the semivariograms was 5 m. for Scots pine and 7 m for Norway spruce. Range is also weakly dependent on forest age class. Thereafter sill was found to be strongly dependent on tree species and forest age class. The local variance maximum in infrared and green band was obtained with a spatial resolution of 3 m and in red channel with a resolution of 2 m. Like the semivariance the local variance was at a higher level in spruce and old forests.

Fields as a framework for integrating GIS and environmental process models. Part 1: Representing spatial continuity

Abstract: Linking a GIS to a spatially distributed, physically-based environmental model offers many advantages. However, the implementation of such linkages is generally problematic. Many problems arise because the relationship between the reality being represented by the mathematical model and the data model used to organize the spatial data in the GIS has not been rigorously defined. In particular, while many environmental models are based on theories that assume continuity and incorporate physical fields as independent variables, current GISs can only represent continuous phenomena in a variety of discrete data models. This paper outlines a strategy in which field variables are used to enable modellers to work directly with the spatial data as spatially continuous phenomena. This allows the manner in which the spatial data has been discretized and the ways in which it can be manipulated to be treated independently from the conceptual modelling of physical processes. Modellers can express their spatial data needs as representations of reality, rather than as elements of a GIS database, and a GIS-independent language for model development may result. By providing a formal linkage between the various models of spatial phenomena, a mechanism is created for the explicit expression of transformation rules between the different spatial data models stored and manipulated by a GIS.

External-Memory Algorithms for Processing Line Segments in Geographic Information Systems

Abstract: In the design of algorithms for large-scale applications it is essential to consider the problem of minimizing I/O communication. Geographical information systems (GIS) are good examples of such large-scale applications as they frequently handle huge amounts of spatial data. In this paper we develop efficient new external-memory algorithms for a number of important problems involving line segments in the plane, including trapezoid decomposition, batched planar point location, triangulation, red-blue line segment intersection reporting, and general line segment intersection reporting. In GIS systems, the first three problems are useful for rendering and modeling, and the latter two are frequently used for overlaying maps and extracting informationfrom them.