Previous topic Next topic 
  

  Translate this page (Google, opens new window/tab):  

Copyright and licensing information ii

Contents, figures and tables v

Foreword to the Third Edition xix

Acknowledgements xxi

1      Introduction and terminology 23

1.1       Motivation and Media. 23

1.1.1        Guide overview. 23

1.1.2        Spatial analysis, GIS and software tools 25

1.1.3        Intended audience and scope. 29

1.2       Software tools and Companion Materials 31

1.2.1        GIS and related software tools 31

1.2.1.1     Sample software products 32

1.2.1.2     Software performance. 32

1.2.2        Suggested reading. 33

1.3       Terminology and Abbreviations 36

1.3.1        Definitions 36

1.4       Common Measures and Notation. 43

1.4.1        Notation. 43

1.4.2        Statistical measures and related formulas 45

1.4.2.1     Counts and specific values 45

1.4.2.2     Measures of centrality 46

1.4.2.3     Measures of spread. 47

1.4.2.4     Measures of distribution shape. 50

1.4.2.5     Measures of complexity and dimensionality 50

1.4.2.6     Common distributions 51

1.4.2.7     Data transforms and back transforms 52

1.4.2.8     Selected functions 53

1.4.2.9     Matrix expressions 54

2      Conceptual Frameworks for Spatial Analysis 57

2.1       The geospatial perspective. 57

2.2       Basic Primitives 58

2.2.1        Place. 58

2.2.2        Attributes 59

2.2.3        Objects 61

2.2.4        Maps 62

2.2.5        Multiple properties of places 63

2.2.6        Fields 64

2.2.7        Networks 65

2.2.8        Density estimation. 65

2.2.9        Detail, resolution, and scale. 65

2.2.10       Topology 66

2.3       Spatial Relationships 68

2.3.1        Co-location. 68

2.3.2        Distance, direction and spatial weights matrices 68

2.3.3        Multidimensional scaling. 70

2.3.4        Spatial context 70

2.3.5        Neighborhood. 70

2.3.6        Spatial heterogeneity 71

2.3.7        Spatial dependence. 71

2.3.8        Spatial sampling. 72

2.3.9        Spatial interpolation. 72

2.3.10       Smoothing and sharpening. 73

2.3.11       First- and second-order processes 73

2.4       Spatial Statistics 75

2.4.1        Spatial probability 75

2.4.2        Probability density 75

2.4.3        Uncertainty 75

2.4.4        Statistical inference. 76

2.5       Spatial Data Infrastructure. 78

2.5.1        Geoportals 78

2.5.2        Metadata. 79

2.5.3        Interoperability 79

2.5.4        Conclusion. 79

3      Methodological Context 81

3.1       Spatial analysis as a process 81

3.2       Analytical methodologies 83

3.3       Spatial analysis and the PPDAC model 87

3.3.1        Problem: Framing the question. 88

3.3.2        Plan: Formulating the approach. 90

3.3.3        Data: Data acquisition. 91

3.3.4        Analysis: Analytical methods and tools 93

3.3.5        Conclusions: Delivering the results 95

3.4       Geospatial analysis and model building. 96

3.5       The changing context of GIScience. 102

4      Building Blocks of Spatial Analysis 105

4.1       Spatial Data Models and Methods 105

4.2       Geometric and Related Operations 107

4.2.1        Length and area for vector data. 107

4.2.2        Length and area for raster datasets 109

4.2.3        Surface area. 111

4.2.3.1     Projected surfaces 111

4.2.3.2     Terrestrial (unprojected) surface area. 113

4.2.4        Line Smoothing and point-weeding. 114

4.2.5        Centroids and centers 116

4.2.5.1     Polygon centroids and centers 116

4.2.5.2     Point sets 119

4.2.5.3     Lines 121

4.2.6        Point (object) in polygon (PIP) 121

4.2.7        Polygon decomposition. 123

4.2.8        Shape. 124

4.2.9        Overlay and combination operations 125

4.2.10       Areal interpolation. 128

4.2.11       Districting and re-districting. 130

4.2.12       Classification and clustering. 135

4.2.12.1    Univariate classification schemes 135

4.2.12.2    Multivariate classification and clustering. 138

4.2.12.3    Multi-band image classification. 140

4.2.12.4    Uncertainty and image processing. 145

4.2.12.5    Hyperspectral image classification. 146

4.2.13       Boundaries and zone membership. 149

4.2.13.1    Convex hulls 149

4.2.13.2    Non-convex hulls 150

4.2.13.3    Minimum Bounding Rectangles (MBRs) 152

4.2.13.4    Fuzzy boundaries 153

4.2.13.5    Breaklines and natural boundaries 156

4.2.14       Tessellations and triangulations 157

4.2.14.1    Delaunay Triangulation. 157

4.2.14.2    TINs — Triangulated irregular networks 158

4.2.14.3    Voronoi/Thiessen polygons 159

4.3       Queries, Computations and Density. 163

4.3.1        Spatial selection and spatial queries 163

4.3.2        Simple calculations 163

4.3.3        Ratios, indices, normalization, standardization and rate smoothing. 166

4.3.4        Density, kernels and occupancy 171

4.3.4.1     Point density 171

4.3.4.2     Kernel density for networks 178

4.3.4.3     Line and intersection densities 179

4.3.4.4     Cartograms 179

4.4       Distance Operations 184

4.4.1        Metrics 186

4.4.1.1     Introduction. 186

4.4.1.2     Terrestrial distances 187

4.4.1.3     Extended Euclidean and Lp-metric distances 188

4.4.2        Cost distance. 190

4.4.2.1     Accumulated cost surfaces and least cost paths 191

4.4.2.2     Distance transforms 196

4.4.3        Network distance. 202

4.4.4        Buffering. 203

4.4.4.1     Vector buffering. 203

4.4.4.2     Raster buffering. 205

4.4.4.3     Hybrid buffering. 205

4.4.4.4     Network buffering. 205

4.4.5        Distance decay models 205

4.5       Directional Operations 210

4.5.1        Directional analysis — overview. 210

4.5.2        Directional analysis of linear datasets 210

4.5.3        Directional analysis of point datasets 215

4.5.4        Directional analysis of surfaces 217

4.6       Grid Operations and Map Algebra. 219

4.6.1        Operations on single and multiple grids 219

4.6.2        Linear spatial filtering. 220

4.6.3        Non-linear spatial filtering. 223

4.6.4        Erosion and dilation. 224

5      Data Exploration and Spatial Statistics 225

5.1       Statistical Methods and Spatial Data. 225

5.1.1        Descriptive statistics 227

5.1.2        Spatial sampling. 228

5.1.2.1     Sampling frameworks 230

5.1.2.2     Declustering. 234

5.2       Exploratory Spatial Data Analysis 236

5.2.1        EDA, ESDA and ESTDA. 236

5.2.2        Outlier detection. 238

5.2.2.1     Mapped histograms 238

5.2.2.2     Box plots 239

5.2.3        Cross tabulations and conditional choropleth plots 241

5.2.4        ESDA and mapped point data. 243

5.2.5        Trend analysis of continuous data. 244

5.2.6        Cluster hunting and scan statistics 244

5.3       Grid-based Statistics 247

5.3.1        Overview of grid-based statistics 247

5.3.2        Crosstabulated grid data, the Kappa Index and Cramer’s V statistic 248

5.3.3        Quadrat analysis of grid datasets 250

5.3.4        Landscape Metrics 252

5.3.4.1     Non-spatial landscape metrics 254

5.3.4.2     Spatial landscape metrics 255

5.4       Point Sets and Distance Statistics 259

5.4.1        Basic distance-derived statistics 259

5.4.2        Nearest neighbor methods 260

5.4.3        Pairwise distances 263

5.4.4        Hot spot and cluster analysis 267

5.4.4.1     Hierarchical nearest neighbor clustering. 268

5.4.4.2     K-means clustering. 269

5.4.4.3     Kernel density clustering. 269

5.4.4.4     Spatio-temporal clustering. 270

5.4.5        Proximity matrix comparisons 272

5.5       Spatial Autocorrelation. 274

5.5.1        Autocorrelation, time series and spatial analysis 274

5.5.2        Global spatial autocorrelation. 276

5.5.2.1     Join counts and the analysis of nominal-valued spatial data. 276

5.5.2.2     Moran I and Geary C. 282

5.5.2.3     Weighting models and lags 289

5.5.3        Local indicators of spatial association (LISA) 290

5.5.4        Significance tests for autocorrelation indices 292

5.6       Spatial Regression. 294

5.6.1        Regression overview. 294

5.6.2        Simple regression and trend surface modeling. 299

5.6.3        Geographically Weighted Regression (GWR) 301

5.6.4        Spatial autoregressive and Bayesian modeling. 305

5.6.4.1     Spatial autoregressive modeling. 305

5.6.4.2     Conditional autoregressive and Bayesian modeling. 308

5.6.5        Spatial filtering models 311

6      Surface and Field Analysis 313

6.1       Modeling Surfaces 313

6.1.1        Test datasets 313

6.1.2        Surfaces and fields 314

6.1.3        Raster models 315

6.1.4        Vector models 318

6.1.5        Mathematical models 319

6.1.6        Statistical and fractal models 320

6.2       Surface Geometry. 323

6.2.1        Gradient, slope and aspect 323

6.2.1.1     Slope. 323

6.2.1.2     Aspect 325

6.2.2        Profiles and curvature. 328

6.2.2.1     Profiles and cross-sections 328

6.2.2.2     Curvature and morphometric analysis 328

6.2.2.3     Profile curvature. 331

6.2.2.4     Plan curvature. 331

6.2.2.5     Tangential curvature. 332

6.2.2.6     Longitudinal and cross-sectional curvature. 332

6.2.2.7     Mean, maximum and minimum curvature. 332

6.2.3        Directional derivatives 332

6.2.4        Paths on surfaces 333

6.2.5        Surface smoothing. 334

6.2.6        Pit filling. 335

6.2.7        Volumetric analysis 336

6.3       Visibility. 337

6.3.1        Viewsheds and RF propagation. 337

6.3.2        Line of sight 340

6.3.3        Isovist analysis and space syntax. 341

6.3.3.1     Isovists 341

6.3.3.2     Space syntax. 343

6.4       Watersheds and Drainage. 345

6.4.1        Overview of watersheds and drainage. 345

6.4.2        Drainage modeling. 345

6.4.3        D-infinity model 346

6.4.4        Drainage modeling case study 347

6.4.4.1     Flow accumulation. 347

6.4.4.2     Stream network construction. 347

6.4.4.3     Stream basin construction. 348

6.5       Gridding, Interpolation and Contouring. 349

6.5.1        Overview of gridding and interpolation. 349

6.5.2        Gridding and interpolation methods 350

6.5.3        Contouring. 356

6.6       Deterministic Interpolation Methods 358

6.6.1        Inverse distance weighting (IDW) 359

6.6.2        Natural neighbor 361

6.6.3        Nearest-neighbor 363

6.6.4        Radial basis and spline functions 363

6.6.5        Modified Shepard. 364

6.6.6        Triangulation with linear interpolation. 365

6.6.7        Triangulation with spline-like interpolation. 365

6.6.8        Rectangular or bi-linear interpolation. 366

6.6.9        Profiling. 366

6.6.10       Polynomial regression. 366

6.6.11       Minimum curvature. 367

6.6.12       Moving average. 367

6.6.13       Local polynomial 367

6.6.14       Topogrid/Topo to raster 368

6.7       Geostatistical Interpolation Methods 369

6.7.1        Core concepts 369

6.7.1.1     Geostatistics 369

6.7.1.2     Geostatistical references 370

6.7.1.3     Semivariance. 370

6.7.1.4     Sample size. 371

6.7.1.5     Support 371

6.7.1.6     Declustering. 371

6.7.1.7     Variogram. 372

6.7.1.8     Stationarity 372

6.7.1.9     Sill, range and nugget 372

6.7.1.10    Transformation. 373

6.7.1.11    Anisotropy 374

6.7.1.12    Indicator semivariance. 375

6.7.1.13    Cross-semivariance. 375

6.7.1.14    Comments on geostatistical software packages 376

6.7.1.15    Semivariance modeling. 377

6.7.1.16    Fractal analysis 381

6.7.1.17    Madograms and Rodograms 381

6.7.1.18    Periodograms and Fourier analysis 381

6.7.2        Kriging interpolation. 382

6.7.2.1     Core process 382

6.7.2.2     Goodness of fit 384

6.7.2.3     Simple Kriging. 384

6.7.2.4     Ordinary Kriging. 385

6.7.2.5     Universal Kriging. 386

6.7.2.6     Median-Polishing and Kriging. 386

6.7.2.7     Indicator Kriging. 386

6.7.2.8     Probability Kriging. 386

6.7.2.9     Disjunctive Kriging. 386

6.7.2.10    Non-stationary Modeling and Stratified Kriging. 387

6.7.2.11    Co-Kriging. 387

6.7.2.12    Factorial Kriging. 387

6.7.2.13    Conditional simulation. 387

7      Network and Location Analysis 391

7.1       Introduction to Network and Location Analysis 391

7.1.1        Overview of network and location analysis 391

7.1.2        Terminology 391

7.1.3        Source data. 393

7.1.4        Algorithms and computational complexity theory 395

7.2       Key Problems in Network and Location Analysis 397

7.2.1        Overview — network analysis 397

7.2.1.1     Key problems in network analysis 398

7.2.1.2     Network analysis software. 401

7.2.1.3     Key problems in location analysis 404

7.2.2        Heuristic and meta-heuristic algorithms 406

7.2.2.1     Greedy heuristics and local search. 407

7.2.2.2     Interchange heuristics 408

7.2.2.3     Metaheuristics 409

7.2.2.4     Tabu search. 409

7.2.2.5     Cross-entropy (CE) methods 410

7.2.2.6     Simulated annealing. 410

7.2.2.7     Lagrangian multipliers and Lagrangian relaxation. 411

7.2.2.8     Ant systems and ant colony optimization (ACO) 414

7.3       Network Construction, Optimal Routes and Optimal Tours 416

7.3.1        Minimum spanning tree. 416

7.3.2        Gabriel network. 417

7.3.3        Steiner trees 419

7.3.4        Shortest (network) path problems 420

7.3.4.1     Overview of shortest path problems 420

7.3.4.2     Dantzig algorithm. 421

7.3.4.3     Dijkstra algorithm. 421

7.3.4.4     A* algorithm. 422

7.3.4.5     GIS implementations of SPAs 422

7.3.4.6     Further SPAs applications 424

7.3.5        Tours, travelling salesman problems and vehicle routing. 425

7.3.5.1     Capacitated vehicle routing. 428

7.4       Location and Service Area Problems 430

7.4.1        Location problems 430

7.4.2        Larger p-median and p-center problems 433

7.4.2.1     Simple heuristics 433

7.4.2.2     Lagrangian relaxation. 433

7.4.2.3     Comparison of alternative p-median heuristics 436

7.4.3        Service areas 438

7.4.3.1     Travel time zones 439

7.5       Arc Routing. 441

7.5.1        Network traversal problems 441

8      Geocomputational methods and modeling 445

8.1       Introduction to Geocomputation. 445

8.1.1        Geocomputational methods 445

8.1.2        Modeling dynamic processes within GIS. 446

8.1.2.1     Representing time and change within GIS. 447

8.1.2.2     Linkage/coupling versus integration/embedding. 449

8.2       Geosimulation. 452

8.2.1        Introduction to geosimulation. 452

8.2.2        Cellular automata (CA) 452

8.2.3        Agents and agent-based models 456

8.2.3.1     Agent-based models 456

8.2.3.2     Agents 457

8.2.4        Applications of agent-based models 459

8.2.5        Advantages of agent-based models 462

8.2.6        Limitations of agent-based models 463

8.2.7        Explanation or prediction? 464

8.2.8        Developing an agent-based model 466

8.2.9        Types of simulation/modeling (s/m) systems for agent-based modeling. 467

8.2.10       Guidelines for choosing a simulation/modeling (s/m) system. 469

8.2.11       Simulation/modeling (s/m) systems for agent-based modeling. 470

8.2.12       Verification and calibration of agent-based models 482

8.2.13       Validation and analysis of agent-based model outputs 484

8.3       Artificial Neural Networks (ANN) 486

8.3.1        Introduction to artificial neural networks 486

8.3.1.1     Multi-level perceptrons (MLP) 487

8.3.1.2     Learning and back-propagation for MLPs 489

8.3.1.3     MLP Example 1: Function approximation. 492

8.3.1.4     MLP Example 2: Landcover change modeling (LCM) 494

8.3.1.5     MLP Example 3: Spatial interaction modeling. 497

8.3.2        Radial basis function networks 499

8.3.3        Self organizing networks 501

8.3.3.1     Self Organizing Maps (SOMs) 501

8.3.3.2     SOM unsupervised classification of hyper-spectral image data. 502

8.3.3.3     TSP optimization using SOM concepts 507

8.4       Genetic Algorithms and Evolutionary Computing. 509

8.4.1        Genetic algorithms — introduction. 509

8.4.2        Genetic algorithm components 510

8.4.2.1     Encoding or representation. 510

8.4.2.2     Fitness function. 511

8.4.2.3     Population initialization. 512

8.4.2.4     Selection. 512

8.4.2.5     Reproduction. 513

8.4.2.6     Crossover 513

8.4.2.7     Mutation. 514

8.4.2.8     Local search. 514

8.4.2.9     Termination. 514

8.4.3        Example GA applications 514

8.4.3.1     GA Example 1: TSP. 514

8.4.3.2     GA Example 2: Clustering. 514

8.4.3.3     GA Example 3: Map labeling. 515

8.4.3.4     GA Example 4: Optimum location. 517

8.4.4        Evolutionary computing and genetic programming. 518

Afterword 519

References 521

CATMOG Guides 538

R-Project spatial statistics software packages 540

Fragstats landscape metrics 543

Web links 545

Associations and academic bodies 545

Online technical dictionaries/definitions 546

Spatial data, test data and spatial information sources 546

Selected national and international data and information sources 546

Network analysis test datasets 547

Statistics and Spatial Statistics links 547

Other GIS web sites 547

Index 549

 

List of Figures

Figure 1‑1 3D Physical GIS models 28

Figure 2‑1 Attribute tables – spatial datasets 59

Figure 2‑2 Cyclic attribute data — Wind direction, single location. 61

Figure 2‑3 An example map showing points, lines, and areas appropriately symbolized. 62

Figure 2‑4 Layers and overlay 63

Figure 2‑5 Noise level raster 64

Figure 2‑6 Filled contour view of field data. 64

Figure 2‑7 Topological relationships 67

Figure 2‑8 Spatial weights computation. 69

Figure 2‑9 Three alternative ways of defining neighborhood, using simple GIS functions 70

Figure 2‑10 Simple interpolation modeling. 73

Figure 2‑11 Four distinct patterns of twelve points in a study area. 74

Figure 2‑12 The process of statistical inference. 76

Figure 2‑13 Pupil performance and a school catchment area in the East Riding of Yorkshire, UK. 79

Figure 3‑1 Analytical process — Mitchell 83

Figure 3‑2 Analytical process — Draper 84

Figure 3‑3 PPDAC as an iterative process 86

Figure 3‑4 Noise map, Augsburg. 87

Figure 3‑5 Simple GIS graphical model (ESRI ArcGIS) 96

Figure 3‑6 Dynamic residential growth model (Idrisi) 98

Figure 3‑7 Modeling wildfire risks, Arizona, USA. 99

Figure 4‑1 Area calculation using Simpson’s rule. 107

Figure 4‑2 3x3 grid neighborhood. 110

Figure 4‑3 5x5 grid neighborhood. 110

Figure 4‑4 Planimetric and surface area of a 3D triangle. 111

Figure 4‑5 DEM surface area. 112

Figure 4‑6 Surface model of DEM. 113

Figure 4‑7 Smoothing techniques 115

Figure 4‑8 Triangle centroid. 117

Figure 4‑9 Polygon centroid (M2) and alternative polygon centers 117

Figure 4‑10 Center and centroid positioning. 118

Figure 4‑11 Polygon center selection. 119

Figure 4‑12 Point set centers 120

Figure 4‑13 Point in polygon — tests and special cases 122

Figure 4‑14 Skeletonised convex polygon. 123

Figure 4‑15 GRASS overlay operations, v.overlay 126

Figure 4‑16 Areal interpolation from census areas to a single grid cell 129

Figure 4‑17 Proportionally assigned population values 129

Figure 4‑18 Grouping data — Zone arrangement effects on voting results 132

Figure 4‑19 Creating postcode polygons 134

Figure 4‑20 Automated Zone Procedure (AZP) 134

Figure 4‑21 AZP applied to part of Manchester, UK. 135

Figure 4‑22 Jenks Natural Breaks algorithm. 138

Figure 4‑23 SPOT Band 1 image histogram — distinct peaks highlighted (CLUSTER) 145

Figure 4‑24 2D map of Cuprite mining district, Western Nevada, USA. 147

Figure 4‑25 3D hypercube visualization of Cuprite mining district, Western Nevada, USA. 147

Figure 4‑26 Single class assignment from spectral angle analysis 148

Figure 4‑27 Convex hull of sample point set 149

Figure 4‑28 Alpha hulls 151

Figure 4‑29 Interpolation within “centroid” MBR. 153

Figure 4‑30 Point locations inside and outside bounding polygon. 153

Figure 4‑31 Sigmoidal fuzzy membership functions 155

Figure 4‑32 Delaunay triangulation of spot height locations 158

Figure 4‑33 Voronoi regions generated in ArcGIS and MATLab. 160

Figure 4‑34 Voronoi cells for a homogeneous grid using a 3x3 distance transform. 161

Figure 4‑35 Network-based Voronoi regions — Shibuya district, Tokyo. 162

Figure 4‑36 Cell-by-cell or Local operations 165

Figure 4‑37 Map algebra: Index creation. 166

Figure 4‑38 Normalization within ArcGIS. 169

Figure 4‑39 Quantile map of normalized SIDS data. 169

Figure 4‑40 Excess risk rate map for SIDS data. 170

Figure 4‑41 Point data. 172

Figure 4‑42 Simple linear (box or uniform) kernel smoothing. 172

Figure 4‑43 Univariate Normal kernel smoothing and cumulative densities 173

Figure 4‑44 Alternative univariate kernel density functions 173

Figure 4‑45 2D Normal kernel 174

Figure 4‑46 Kernel density map, Lung Case data, 3D visualization. 175

Figure 4‑47 Univariate kernel density functions, unit bandwidth. 177

Figure 4‑48 Cartogram creation using basic Dorling algorithm. 180

Figure 4‑49 Cartogram creation using Dougenik, Chrisman and Niemeyer algorithm. 181

Figure 4‑50 World Population as a Cartogram. 181

Figure 4‑51 Cartograms of births data, 1974. 182

Figure 4‑52 Hexagonal cartogram showing UK mortality data, age group 20-24. 183

Figure 4‑53 Alternative measures of terrain distance. 184

Figure 4‑54 Glasgow's Clockwork Orange Underground. 186

Figure 4‑55 Great circle and constant bearing paths, Boston to Bristol 187

Figure 4‑56 p-metric circles 189

Figure 4‑57 Cost distance model 191

Figure 4‑58 Cost surface as grid. 191

Figure 4‑59 Grid resolution and cost distance. 192

Figure 4‑60 Accumulated cost surface and least cost paths 193

Figure 4‑61 Alternative route selection by ACS. 194

Figure 4‑62 Steepest path vs tracked path. 195

Figure 4‑63 3x3 Distance transformation – scan elements 196

Figure 4‑64 5x5 DT mask. 198

Figure 4‑65 Distance transform, single point 198

Figure 4‑66 Urban traffic modeling. 199

Figure 4‑67 Notting Hill Carnival routes 199

Figure 4‑68 Alternative routes selected by gradient constrained DT. 200

Figure 4‑69 Hellisheiği power plant pipeline route selection. 201

Figure 4‑70 Shortest and least time paths 202

Figure 4‑71 Simple buffering. 203

Figure 4‑72 Manifold: Buffer operations 204

Figure 4‑73 Manifold: Buffering options 204

Figure 4‑74 Inverse distance decay models, α/dβ 208

Figure 4‑75 Exponential distance decay models, αe‑βd 208

Figure 4‑76 Directional analysis of streams 212

Figure 4‑77 Two-variable wind rose. 214

Figure 4‑78 Standard distance circle and ellipses 215

Figure 4‑79 Correlated Random Walk simulation. 216

Figure 4‑80 Slope and aspect plot, Mt St Helens data, USA. 217

Figure 4‑81 Wind flow grid simulation using WindNinja. 218

Figure 4‑82 Dilation and erosion operations 224

Figure 5‑1 Point-based sampling schemes 229

Figure 5‑2 Grid generation examples 231

Figure 5‑3 Grid sampling examples within hexagonal grid, 1 hectare area. 231

Figure 5‑4 Random point generation examples — ArcGIS. 233

Figure 5‑5 Random point samples on a network. 233

Figure 5‑6 Brushing and linking, GeoDa. 236

Figure 5‑7 Parallel coordinate plot 237

Figure 5‑8 Star plot 237

Figure 5‑9 Histogram linkage. 239

Figure 5‑10 Simple box plot 239

Figure 5‑11 Mapped box plot, GeoDa. 240

Figure 5‑12 Conditional Choropleth mapping. 242

Figure 5‑13 Exploratory analysis of radioactivity data. 243

Figure 5‑14 Trend analysis of radioactivity dataset 244

Figure 5‑15 Quadrat counts 250

Figure 5‑16 Texture analysis — variability 253

Figure 5‑17 Nearest Neighbor distribution. 260

Figure 5‑18 Ripley’s K function computation. 263

Figure 5‑19 Ripley K function, shown as transformed L function plot 265

Figure 5‑20 Thomas Poisson Cluster Process (20 clusters, SD=0.03, mean=5) 265

Figure 5‑21 Lung cancer incidence data. 267

Figure 5‑22 Lung cancer NNh clusters 269

Figure 5‑23 KDE cancer incidence mapping. 270

Figure 5‑24 Time series of stock price and volume data. 274

Figure 5‑25 Join count patterns 277

Figure 5‑26 Join count computation. 278

Figure 5‑27 Homogeneous and non-homogenous probability images 279

Figure 5‑28 Grouping and size effects 281

Figure 5‑29 Irregular lattice dataset 282

Figure 5‑30 Adjacency matrix, W.. 283

Figure 5‑31 Moran's I computation. 285

Figure 5‑32 Revised source data. 285

Figure 5‑33 Sample dataset and Moran I analysis 287

Figure 5‑34 Moran I (co)variance cloud, lag 1. 288

Figure 5‑35 Local Moran I computation. 291

Figure 5‑36 LISA map, Moran I 291

Figure 5‑37 Significance tests for revised sample dataset 293

Figure 5‑38 Georgia educational attainment: GWR residuals map, Gaussian adaptive kernel 303

Figure 6‑1 East Sussex test surface, OS TQ81NE. 313

Figure 6‑2 Pentland Hills test surface. 314

Figure 6‑3 Linear regression surface fit to NT04 spot heights 315

Figure 6‑4 Raster file neighborhoods 316

Figure 6‑5 Vector models of TQ81NE. 319

Figure 6‑6 First, second and third order mathematical surfaces 320

Figure 6‑7 Random and fractal grids 321

Figure 6‑8 Pseudo-random surfaces 322

Figure 6‑9 8-triangle slope computation. 324

Figure 6‑10 Gradient and sampling resolution. 325

Figure 6‑11 Slope computation output 325

Figure 6‑12 Frequency distribution of aspect values 326

Figure 6‑13 Aspect computation output 327

Figure 6‑14 Profile of NS transect, TQ81NE. 328

Figure 6‑15 Multiple profile computation. 328

Figure 6‑16 Surface morphology 329

Figure 6‑17 Path smoothing — vertical profile. 334

Figure 6‑18 Grid smoothing. 335

Figure 6‑19 Viewshed computation. 338

Figure 6‑20 3D Urban radio wave propagation modeling using Cellular Expert and ArcGIS. 339

Figure 6‑21 Radio frequency viewshed. 340

Figure 6‑22 Line of sight analysis 341

Figure 6‑23 Viewsheds and lines of sight on a synthetic (Gaussian) surface. 341

Figure 6‑24 Isovist analysis, Street network, central London. 342

Figure 6‑25 Axial lines and connectivity 343

Figure 6‑26 Depthmap — Gallery space visibility map. 344

Figure 6‑27 D-Infinity flow assignment 346

Figure 6‑28 Flow direction and accumulation. 347

Figure 6‑29 Stream identification. 348

Figure 6‑30 Watersheds and basins 348

Figure 6‑31 Contour plots for alternative interpolation methods — generated with Surfer 8. 353

Figure 6‑32 Linear interpolation of contours 356

Figure 6‑33 Contour computation output 357

Figure 6‑34 IDW as surface plot 359

Figure 6‑35 Contour plots for alternative IDW methods, OS NT04. 360

Figure 6‑36 Natural Neighbor interpolation ― computation of weights 362

Figure 6‑37 Clough-Tocher TIN interpolation. 366

Figure 6‑38 Regression fitting to test dataset OS NT04. 367

Figure 6‑39 Sample variogram. 371

Figure 6‑40 Sill, range and nugget 373

Figure 6‑41 Data transformation for Normality 374

Figure 6‑42 Anisotropy 2D map, zinc data. 375

Figure 6‑43 Indicator variograms 375

Figure 6‑44 Variogram models — graphs 380

Figure 6‑45 Fractal analysis of TQ81NE. 381

Figure 6‑46 Ordinary Kriging of zinc dataset 385

Figure 6‑47 Conditional simulation of untransformed zinc test dataset 389

Figure 7‑1 Network topologies 393

Figure 7‑2 Visualization of lane/movement simulation (Dynameq) 403

Figure 7‑3 LP Solution graphs 413

Figure 7‑4 Minimum Spanning Tree. 416

Figure 7‑5 Gabriel network construction. 417

Figure 7‑6 Relative neighborhood network and related constructions 418

Figure 7‑7 Steiner MST construction. 419

Figure 7‑8 Dantzig shortest path algorithm. 421

Figure 7‑9 Salt Lake City — Sample networking problems and solutions 423

Figure 7‑10 Shortest obstacle-avoiding path. 424

Figure 7‑11 MST, TSP and related problems 427

Figure 7‑12 Heuristic solution and dual circuit TSP examples 428

Figure 7‑13 Tanker delivery tours 429

Figure 7‑14 Optimum facility location on a network — LOLA solution. 432

Figure 7‑15 Comparison of heuristic p-median solutions, Tripolis, Greece. 437

Figure 7‑16 Facility location in Tripolis, Greece, planar model 438

Figure 7‑17 Service area definition. 439

Figure 7‑18 Travel-time or drive-time zones 440

Figure 7‑19 Routing directions 441

Figure 7‑20 Arc routing. 442

Figure 8‑1 Game of Life Model 453

Figure 8‑2 Heatbugs Model 453

Figure 8‑3 Moore and von Neumann neighborhoods 455

Figure 8‑4 Schelling segregation model 460

Figure 8‑5 Pedestrian movement simulation — Subway hall model 461

Figure 8‑6 Model development balance. 470

Figure 8‑7 Geometric and Locational Features of the Notting Hill Carnival Swarm model 472

Figure 8‑8 RepastS point and click modeling and runtime environments 474

Figure 8‑9 RepastCity — importing GIS network data into Repast Simphony 475

Figure 8‑10 Repast Simphony — agent-based model visualized using NASA’s Worldwind. 475

Figure 8‑11 StarLogo TNG drag and drop programming interface and 3D view of a simulation. 478

Figure 8‑12 Outputs from OBEUS: Schelling residential dynamics model 479

Figure 8‑13 AgentSheets: The Boulder Mountain Biking Advisor 481

Figure 8‑14 An urban and transport dynamics model developed in AnyLogic (2006) 482

Figure 8‑15 Simple 3-5-2 feedforward artificial neural network. 486

Figure 8‑16 MLP 3-5-2 with bias nodes 487

Figure 8‑17 ANN hidden node structure. 487

Figure 8‑18 Sample activation functions 488

Figure 8‑19 MLP: Test data and fitted model 493

Figure 8‑20 MLP: RMSE curves 493

Figure 8‑21 Land cover, 1986, Chiquitania. 495

Figure 8‑22 Distance raster (meters), anthropogenic disturbance, Chiquitania. 495

Figure 8‑23 MLP Classifier — Idrisi 496

Figure 8‑24 Transition potential map. 497

Figure 8‑25 MLP trip distribution model 1. 498

Figure 8‑26 MLP trip distribution model 2. 499

Figure 8‑27 Radial basis function NN model 500

Figure 8‑28 SOM grids 502

Figure 8‑29 SOM classification of remotely-sensed hyperspectral data. 504

Figure 8‑30 Self Organizing Map (SOM) classification — Idrisi 506

Figure 8‑31 SOM classified 3-band image. 507

Figure 8‑32 Rank score transform. 512

 

List of Tables

Table 1‑1 Selected terminology 36

Table 1‑2 Notation and symbology 43

Table 1‑3 Common formulas and statistical measures 45

Table 4‑1 Geographic data models 105

Table 4‑2 OGC OpenGIS Simple Features Specification — Principal Methods 106

Table 4‑3 Spatial overlay methods, Manifold GIS. 127

Table 4‑4 Regional employment data — grouping affects 131

Table 4‑5 Selected univariate classification schemes 136

Table 4‑6 Image classification facilities — Selected classifiers 142

Table 4‑7 Selected MATLab/GRASS planar geometric analysis functions 159

Table 4‑8 Widely used univariate kernel density functions 176

Table 4‑9 Interpretation of p-values 190

Table 4‑10 3x3 Chamfer metrics 197

Table 4‑11 Linear spatial filters 222

Table 5‑1 Implications of Data Models 226

Table 5‑2 Description of methods for analysis of spatial data in ecology 227

Table 5‑3 Voronoi-based ESDA. 244

Table 5‑4 Sample statistical tools for grid data — Idrisi 247

Table 5‑5 Simple 2-way contingency table. 248

Table 5‑6 Simple Chi-square frequency table computation. 251

Table 5‑7 NN Statistics and study area size. 262

Table 5‑8 Join count analysis results 278

Table 5‑9 Join count mean and variance formulas 280

Table 5‑10 Tabulated lattice data. 283

Table 5‑11 Selected regression analysis terminology 298

Table 5‑12 Georgia dataset — global regression estimates and diagnostics 303

Table 5‑13 Georgia dataset — comparative regression estimates and diagnostics 307

Table 6‑1 Morphometric features — a simplified classification. 331

Table 6‑2 Gridding and interpolation methods 351

Table 6‑3 Variogram models (univariate, isotropic) 379

Table 7‑1 Network analysis terminology 392

Table 7‑2 Some key optimization problems in network analysis 399

Table 7‑3 Sample network analysis problem parameters 400

Table 7‑4 Routing functionality in selected logistics software packages 402

Table 7‑5 Taxonomy of location analysis problems 404

Table 8‑1 Agent-based modeling and GIS coupling. 450

Table 8‑2 Agents and environments 467

Table 8‑3 Comparison of open source simulation/modeling toolkits 471

Table 8‑4 Comparison of shareware/freeware simulation/modeling systems 476

Table 8‑5 Comparison of proprietary simulation/modeling systems 480

Table 8‑6 W weights matrix, Chiquitania MLP model 497

Table 8‑7 SOM neighborhood and learning rate functions 505

 

  Back to Top    Back to Home  Previous topic Next topic