Transcript Chapter 2
Chapter 6 Part B: Surface analysis – gridding and interpolation www.spatialanalysisonline.com Surface analysis - Gridding & Interpolation Gridding (a form of systematic interpolation): to convert a sample of data points to a complete coverage (set of values) for the study region to convert from one level of data resolution or orientation to another (resampling) to convert from one representation of a continuous surface to another, e.g. TIN to grid or contour to grid Quality of process: Input data quality/accuracy Data density Data distribution Spatial variability of data Model applied – typically weighted average: n zj z i i i 1 3rd edition www.spatialanalysisonline.com 2 Surface analysis - Gridding & Interpolation Some other key issues: Validity of continuity assumption/handling of breaklines/stratifications and missing data Validity of weighted average model How to choose weights? How many points to include in summation? Shape and size of neighbourhood region Selection of model and parameters Evaluation of quality Availability of related datasets 3rd edition www.spatialanalysisonline.com 3 Surface analysis - Gridding & Interpolation Comparison of interpolation methods Method Speed Type Inverse distance weighting (IDW) Fast Exact, unless smoothing factor specified Natural neighbour Fast Exact Nearest-neighbour Fast Exact Kriging -Geostatistical models (stochastic) Slow/Medium Exact if no nugget (assumed measurement error) Conditional simulation Slow Exact Radial basis Slow/Medium Exact if no smoothing value specified Modified Shepard Fast Exact, unless smoothing factor specified Triangulation with linear interpolation Fast Exact Triangulation with spline interpolation Fast Exact Profiling Fast Exact Minimum curvature Medium Exact/Smoothing Spline functions Fast Exact (smoothing possible) Local polynomial Fast Smoothing Polynomial regression Fast Smoothing Moving average Fast Smoothing Topogrid/Topo to Raster Slow/Medium Not specified 3rd edition www.spatialanalysisonline.com 4 Surface analysis - Gridding & Interpolation Contour generation from grids Simple linear interpolation Smoothing – splines or iterative thresholding 3rd edition www.spatialanalysisonline.com 5 Surface analysis - Gridding & Interpolation Deterministic interpolation 3rd edition www.spatialanalysisonline.com 6 Surface analysis - Gridding & Interpolation Deterministic interpolation: IDW General model: IDW version: n zj n z , i i i 1 zj i 1 i 1 1/ dij z , 1/d i i i ij i i The values are computed for all or selected i (e.g. 12) The value for alpha is typically 1 or 2 The division by the sum of weighted distances ensures that the weights add up to 1 3rd edition www.spatialanalysisonline.com 7 Surface analysis - Gridding & Interpolation Deterministic interpolation: IDW Simple IDW calculations for point 5,5 % each data point contributes Grid point (5,5) Calculation of weights using alpha=1 and alpha=2 3rd edition www.spatialanalysisonline.com 8 Surface analysis - Gridding & Interpolation Sample data for main examples (x,y,z) 316467 316139 319451 317374 319128 316684 314144 315284 313741 314441 310382 310987 311728 311608 310713 309731 3rd edition 650940 652115 653028 653494 655535 654997 655537 659375 658288 657566 651060 654741 656272 657781 658370 652685 294 237 287 279 298 285 535 479 517 537 358 534 464 478 562 448 309724 309012 308632 307917 306469 307272 306512 305480 306083 303423 301975 301388 318247 319469 318800 319382 653873 655066 651036 652199 651249 653817 653133 653039 654457 655030 650487 651527 649157 648372 647423 645231 465 518 326 407 445 434 448 415 417 336 297 296 357 479 445 476 318555 313677 313344 312471 315503 318573 317466 316285 313444 312553 310398 308987 307987 306795 307082 306179 644731 646485 647439 645750 644081 643128 640234 642153 641290 641625 643270 646909 645958 646096 648950 648702 www.spatialanalysisonline.com 465 309 310 307 376 395 456 365 536 557 347 438 516 388 400 297 304905 303036 300638 300158 302636 308279 309067 308526 306512 305435 305290 304370 302488 302066 647876 648380 649904 648596 645845 640150 641210 642825 641175 640415 641442 642830 642215 641652 293 248 294 294 264 355 344 426 303 359 284 308 307 294 9 Surface analysis - Gridding & Interpolation Deterministic interpolation: IDW Source data 3rd edition IDW grid: alpha=2, no smoothing www.spatialanalysisonline.com 10 Surface analysis - Gridding & Interpolation IDW: All grid points in rectangle are assigned a value Sample points are exactly fitted Surface looks rather peaky with odd ‘hollows’/bull’s eye’s affected by point locations, alpha and grid resolution How do we know what values to choose - number of neighbours, alpha, grid res., break lines? Can we judge the ‘goodness of fit’ of the resultant surface? Generally no, but could test the fit by (a) additional sampling; (b) comparison with other sources of data (e.g. aerial photographs); (c) computational procedures 3rd edition www.spatialanalysisonline.com 11 Surface analysis - Gridding & Interpolation Deterministic interpolation: Nearest neighbour Simple assignment on nearest neighbour basis Uniform within zones, stepped between zones 3rd edition www.spatialanalysisonline.com 12 Surface analysis - Gridding & Interpolation Deterministic interpolation: Natural neighbour Assumes each existing point, i, has a “natural” sphere of influence – its Voronoi polygon with area Ai Assumes a newly added point, p, (e.g. grid point) captures part of these natural areas for itself. These captured parts have areas Aip Weights are based on initial and captured Voronoi regions Aip i k A ip i 1 3rd edition www.spatialanalysisonline.com 13 Surface analysis - Gridding & Interpolation Deterministic interpolation: Natural neighbour Source data 3rd edition Nat N grid: extent limited to convex hull www.spatialanalysisonline.com 14 Surface analysis - Gridding & Interpolation Deterministic interpolation: Radial basis/splines A large family of interpolation functions Exact (or smoothing) Exhibit well-defined characteristics – e.g. properties of smoothness and fit Thin plate splines widely used in Earth Sciences Core model is a form of weighted transformed distances: ‘bias’ or offset value n zp z-value at point p 3rd edition w (r ) m i i i 1 Weight for ith point Radial basis function, for radius ri www.spatialanalysisonline.com 15 Surface analysis - Gridding & Interpolation Deterministic interpolation: Radial basis/splines Computation: Compute matrix D of all interpoint distances Apply radial basis function to all elements of D For grid point p, compute distance to all data points, as a vector r Apply radial basis function to all elements of r Form augmented matrix equation shown and solve by inversion Φ 1 λ φ 1' 0 m 1 3rd edition www.spatialanalysisonline.com 16 Surface analysis - Gridding & Interpolation Deterministic interpolation: Thin Plate Spline (r) (c 2 r 2 )ln(c 2 r 2 ) Using 12 point neighbourhoods Using all 62 data points 3rd edition www.spatialanalysisonline.com 17 Surface analysis - Gridding & Interpolation Deterministic interpolation: Modified Shepard Hybrid IDW-like system Typically uses local IDW or quadratic fit plus longer range IDW with separate parameter Example using 8 local points for quadratic fit and 16 for longer range IDW 3rd edition www.spatialanalysisonline.com 18 Surface analysis - Gridding & Interpolation Deterministic interpolation: Triangulation with linear interpolation (i.e. 2D linear model) Source data 3rd edition Linear grid: extent limited to convex hull www.spatialanalysisonline.com 19 Surface analysis - Gridding & Interpolation Deterministic interpolation: Minimum curvature Seeks smoothed ‘elastic membrane’ fit to surface Fit to residuals after linear regression of original dataset Linear component then added back 3rd edition www.spatialanalysisonline.com 20 Surface analysis - Gridding & Interpolation Deterministic interpolation: Local polynomial Fits a local polynomial (e.g. quadratic) to each grid point, using window (e.g. circle) of given size or number of points Smoothing Example: local quadratic distanceweighted OLS, with 12 points per fit 3rd edition www.spatialanalysisonline.com 21 Surface analysis - Gridding & Interpolation Deterministic interpolation – some other methods: Triangulation with non-linear interpolation - Clough-Tocher: bicubic patches fitted to every triangle, such that edge and vertex joins are smooth Bi-linear: suitable for dense uniform datasets, e.g. grids with some missing cell values – simple weighted averaging of directly or indirectly neighbouring cells/points Profiling: input dataset is contour vectors. Typically interpolation is linear between closest pair of contours (8-point) Moving average – uses a moving circular or elliptical window (similar to time series analysis) to obtain simple local average Topogrid – conversion of contour vectors to hydrologically consistent DEM. Includes enforcement of local drainage, lake boundaries, ridges and stream lines 3rd edition www.spatialanalysisonline.com 22 Surface analysis - Gridding & Interpolation Geostatistical interpolation 3rd edition www.spatialanalysisonline.com 23 Surface analysis - Gridding & Interpolation Geostatistical interpolation: Core concepts Geostatistics definition: “…models and methods for data observed at a discrete set of locations, such that the observed value, zi, is either a direct measurement of, or is statistically related to, the value of an underlying continuous spatial phenomenon, F(x,y), at the corresponding sampled location (xi,yi) within some spatial region A.” (Diggle et al) 3rd edition www.spatialanalysisonline.com 24 Surface analysis - Gridding & Interpolation Geostatistical interpolation: Core concepts Addresses questions such as: how many points are needed to compute the local average? what size, orientation and shape of neighbourhood should be chosen? what model and weights should be used to compute the local average? what errors (uncertainties) are associated with interpolated values? 3rd edition www.spatialanalysisonline.com 25 Surface analysis - Gridding & Interpolation Geostatistical interpolation: Variograms Analyse the observed variation in data values by distance bands using a spatial autocorrelationlike measure, Typically, analyse all pairs of observations that lie within specific distance bands Compute the average values of for each band Plot these averages against distance Fit an experimental curve to the observed pattern 3rd edition www.spatialanalysisonline.com 26 Surface analysis - Gridding & Interpolation Geostatistical interpolation: Variograms Analyse the observed variation in data values by distance bands using a spatial autocorrelationlike measure, : Semivariance measure is most often used: dij h /2 ˆ(h) 1 2N(h) d (zi z j )2 ij h /2 Bands have width . N(h) is the number of pairs in the band with mid-point distance h 3rd edition www.spatialanalysisonline.com 27 Surface analysis - Gridding & Interpolation Geostatistical interpolation: Zinc data (x,y,z) Burrough & McDonnell (1998) App.3 98 soil samples (z=zinc ppm) x 1637 1894 2110 2356 1755 1503 1665 1373 1226 1500 1961 1851 1087 1692 1249 1302 1507 1004 1192 1810 1683 1042 1862 926 1573 3rd edition y 2651 2630 2630 2618 2599 2597 2597 2555 2517 2513 2493 2475 2461 2457 2442 2413 2401 2359 2335 2331 2314 2255 2247 2236 2223 z 812 298 321 213 746 1548 832 1839 1528 1571 167 176 933 746 703 550 262 1190 432 258 464 253 142 907 210 x 1179 1682 1562 1878 895 1044 1429 1670 1747 1867 889 1599 924 1535 1719 814 1851 1117 959 846 1562 734 1317 874 655 y 2219 2161 2137 2122 2070 2030 2020 2007 2000 1994 1933 1891 1865 1852 1840 1794 1773 1741 1723 1722 1687 1666 1625 1604 1564 z 139 365 210 119 761 203 198 282 152 133 659 375 232 222 176 643 117 166 317 191 136 801 141 241 784 x 1131 1222 1467 611 1601 1728 1057 870 1391 476 1676 540 1092 381 1174 1281 422 528 465 1514 1252 693 312 1380 862 www.spatialanalysisonline.com y 1545 1542 1485 1475 1460 1458 1450 1425 1369 1305 1263 1255 1233 1224 1221 1212 1173 1167 1164 1103 1101 1097 1079 1073 1066 z 128 158 143 765 126 113 140 1060 129 1161 130 703 119 1383 221 240 545 676 793 192 778 186 1136 203 226 x 510 353 407 580 210 432 495 520 805 434 101 1182 275 458 1018 485 5 606 429 866 1721 1551 203 y 1058 1042 1027 1010 966 945 936 878 867 861 857 840 816 810 758 733 706 698 694 681 666 653 649 z 593 505 685 198 560 549 680 206 180 451 577 157 420 539 199 296 553 187 400 162 722 1672 332 28 Surface analysis - Gridding & Interpolation Geostatistical interpolation: Variograms Smallest observed separation Fitted curve sill Average semivariance for band 4 C1=C0+C (structural variance) Range, A0 model 3rd edition C0 “Nugget” Lag (distance band) www.spatialanalysisonline.com 29 Surface analysis - Gridding & Interpolation Geostatistical interpolation: Variogram models Model Nugget effect Formula Notes (0) C 0 Simple constant. May be added to all models. Models with a nugget will not be exact Linear (h) C1 h Exponential Exp() (h) C 1 e Spherical Sph() (h) C1 3rd edition 1 No sill. Often used in combination with other functions. May be used as a ramp, with a constant sill value set at a range, a kh k is a constant, often k=1 or k=3. Useful when there is a larger nugget and slow rise to the sill Useful when the nugget effect is important 3h 1 3 h , h 1 but small. Given as the default model in 2 2 some packages. (h) C1, h 1 www.spatialanalysisonline.com 30 Surface analysis - Gridding & Interpolation Geostatistical interpolation: Core concepts Geostatistical methods assume a general form of model which is a mix of: (a) a deterministic model m(x,y) (b) a regionalised statistical variation from m(x,y) (c) a random noise (Normal error) component (actually two components which we cannot generally separate) (b) is chosen by analysing the semivariance, 3rd edition www.spatialanalysisonline.com 31 Surface analysis - Gridding & Interpolation Geostatistical interpolation: Further key concepts Sample size Support Declustering Thresholding Stationarity Transformation Anisotropy Madograms and Rodograms Periodograms and Fourier analysis 3rd edition www.spatialanalysisonline.com 32 Surface analysis - Gridding & Interpolation Geostatistical interpolation: Kriging Examine the data (z values) for spatial trends and Normality and transform variables if necessary, e.g. loge(z) or loge(z+1) Compute the experimental variogram and fit a suitable model IFF the pattern warrants this (i.e. the data is not just noise) Check the model by cross validation and examine size of mean squared deviation Decide on method to use – Kriging or Conditional simulation Generate a grid from the selected model and plot Compare the results to (a) the input data points and (b) other sources of data/information 3rd edition www.spatialanalysisonline.com 33 Surface analysis - Gridding & Interpolation Geostatistical interpolation: Kriging Ordinary Kriging – basic procedure is exactly the same as for radial basis functions, but with function applied being the modelled variogram Φ 1 λ φ 1' 0 m 1 Computation: Compute matrix D of all interpoint distances Apply variogram function to all elements of D For grid point p, compute distance to all data points, as a vector r Apply variogram function to all elements of r Form augmented matrix equation shown and solve by inversion 3rd edition www.spatialanalysisonline.com 34 Surface analysis - Gridding & Interpolation Geostatistical interpolation: Ordinary Kriging (OK) n Estimated value at p is then: zp Estimated variance at p is: ˆ p2 z i i i 1 n c i i m i 1 where m is the Lagrangian multiplier and the c i are the modelled semivariance values (the values) 3rd edition www.spatialanalysisonline.com 35 Surface analysis - Gridding & Interpolation Geostatistical interpolation: Ordinary Kriging Zinc data: Predicted values 3rd edition Zinc data: Estimated standard deviation www.spatialanalysisonline.com 36 Surface analysis - Gridding & Interpolation Geostatistical interpolation: Kriging Goodness of fit methods: Residual plots, standard deviation plots Examination for artefacts Simple cross-validation Jack-knifing Re-sampling Examination of independent datasets Stratification 3rd edition www.spatialanalysisonline.com 37 Surface analysis - Gridding & Interpolation Geostatistical interpolation: Kriging – other models Universal Kriging – Kriging with a trend Indicator Kriging – thresholding Stratified Kriging – regionalised approach Co-Kriging – use of associated data 3rd edition www.spatialanalysisonline.com 38 Surface analysis - Gridding & Interpolation Geostatistical interpolation: Conditional simulation (Gaussian sequential method) Similar to OK approach. 3 core steps: 1. Analyse the data/transform as necessary, fit model variogram, define grid to use (possibly multi-level) 2. Randomly select a grid node to visit (e.g. apply a random walk process) as execute step 3 3. Apply OK estimation at node based on local observed data points. Take estimated value and apply Normal random process with mean as estimate and variance as estimated variance. Return to step 2 This process is then repeated until all nodes have been visited. Then iterated M times 3rd edition www.spatialanalysisonline.com 39 Surface analysis - Gridding & Interpolation Geostatistical interpolation: Conditional simulation, M=100 iterations Zinc data: Predicted values 3rd edition Zinc data: Estimated standard deviation www.spatialanalysisonline.com 40