1 GEOG4650/5650 – Fall 2007 Spatial Interpolation Triangulation Inverse-distance Kriging (optimal interpolation)

Slides:



Advertisements
Similar presentations
Spatial point patterns and Geostatistics an introduction
Advertisements

School of Geography FACULTY OF ENVIRONMENT Point-surface interpolation.
Basic geostatistics Austin Troy.
University of Wisconsin-Milwaukee Geographic Information Science Geography 625 Intermediate Geographic Information Science Instructor: Changshan Wu Department.
WFM 6202: Remote Sensing and GIS in Water Management © Dr. Akm Saiful IslamDr. Akm Saiful Islam WFM 6202: Remote Sensing and GIS in Water Management Akm.
19 th Advanced Summer School in Regional Science Combining Vectors and Rasters in ArcGIS.
ANALYSIS 3 - RASTER What kinds of analysis can we do with GIS? 1.Measurements 2.Layer statistics 3.Queries 4.Buffering (vector); Proximity (raster) 5.Filtering.
Z – Surface Interpolation…. INTERPOLATION Procedure to predict values of attributes at unsampled points Why? Can’t measure all locations: Time Money Impossible.
Introduction This project deals with conversion of Vector based Probable Maximum Precipitation (PMP) data into Raster based PMP data using different interpolation.
Spatial Analysis Longley et al., Ch 14,15. Transformations Buffering (Point, Line, Area) Point-in-polygon Polygon Overlay Spatial Interpolation –Theissen.
Spatial Interpolation
Concept Course on Spatial Dr. A.K.M. Saiful Islam Developing ground water level map for Dinajpur district, Bangladesh using geo-statistical analyst.
Week 17GEOG2750 – Earth Observation and GIS of the Physical Environment1 Lecture 14 Interpolating environmental datasets Outline – creating surfaces from.
Spatial Interpolation, geostatistics and sampling
Topic 6: Spatial Interpolation
Lecture 4. Interpolating environmental datasets
Lecture 07: Terrain Analysis Geography 128 Analytical and Computer Cartography Spring 2007 Department of Geography University of California, Santa Barbara.
Applications in GIS (Kriging Interpolation)
Introduction to ArcGIS for Environmental Scientists Module 2 – GIS Fundamentals Lecture 5 – Coordinate Systems and Map Projections.
Slope and Aspect Calculated from a grid of elevations (a digital elevation model) Slope and aspect are calculated at each point in the grid, by comparing.
Spatial Analyst Toolbox Lecture 17. Spatial Analyst Tool Sets  Conditional  Density  Distance  Generalization  Ground Water  Interpolation  Conditional.
ESRM 250 & CFR 520: Introduction to GIS © Phil Hurvitz, KEEP THIS TEXT BOX this slide includes some ESRI fonts. when you save this presentation,
Using ESRI ArcGIS 9.3 Spatial Analyst
$88.65 $ $22.05/A profit increase Improving Wheat Profits Eakly, OK Irrigated, Behind Cotton.
Basic geostatistics Austin Troy.
Interpolation.
Spatial Analysis.
Intro. To GIS Lecture 9 Terrain Analysis April 24 th, 2013.
Interpolation Tools. Lesson 5 overview  Concepts  Sampling methods  Creating continuous surfaces  Interpolation  Density surfaces in GIS  Interpolators.
Geographic Information Science
Model Construction: interpolation techniques 1392.
GEOSTATISICAL ANALYSIS Course: Special Topics in Remote Sensing & GIS Mirza Muhammad Waqar Contact: EXT:2257.
Chapter 8 – Geographic Information Analysis O’Sullivan and Unwin “ Describing and Analyzing Fields” By: Scott Clobes.
Interpolation Content Point data Interpolation Review Simple Interpolation Geostatistical Analyst in ArcGIS IDW in Geostatistical Analyst Semivariograms.
1 Peter Fox GIS for Science ERTH 4750 (98271) Week 10, Friday, April 6, 2012 Lab:
Ripley K – Fisher et al.. Ripley K - Issues Assumes the process is homogeneous (stationary random field). Ripley K was is very sensitive to study area.
Spatial Interpolation Chapter 13. Introduction Land surface in Chapter 13 Land surface in Chapter 13 Also a non-existing surface, but visualized as a.
Geographic Information Systems
Interpolation of Surfaces Spatial Data Analysis. Spatial Interpolation Spatial interpolation is the prediction of exact values of attributes at un-sampled.
NR 143 Study Overview: part 1 By Austin Troy University of Vermont Using GIS-- Introduction to GIS.
Chapter 16 - Spatial Interpolation
What’s the Point? Working with 0-D Spatial Data in ArcGIS
1 Overview Importing data from generic raster files Creating surfaces from point samples Mapping contours Calculating summary attributes for polygon features.
Grid-based Map Analysis Techniques and Modeling Workshop
Statistical Surfaces Any geographic entity that can be thought of as containing a Z value for each X,Y location –topographic elevation being the most obvious.
L15 – Spatial Interpolation – Part 1 Chapter 12. INTERPOLATION Procedure to predict values of attributes at unsampled points Why? Can’t measure all locations:
Lecture 6: Point Interpolation
Ran TAO Missing Spatial Data. Examples Places cannot be reached E.g. Mountainous area Sample points E.g. Air pollution Damage of data E.g.
Interpolation and evaluation of probable Maximum Precipitation (PMP) patterns using different methods by: tarun gill.
Machine Learning 5. Parametric Methods.
Geo479/579: Geostatistics Ch12. Ordinary Kriging (2)
INTERPOLATION Procedure to predict values of attributes at unsampled points within the region sampled Why?Examples: -Can not measure all locations: - temperature.
Interpolation Local Interpolation Methods –IDW – Inverse Distance Weighting –Natural Neighbor –Spline – Radial Basis Functions –Kriging – Geostatistical.
URBDP 422 URBAN AND REGIONAL GEO-SPATIAL ANALYSIS
Objective Analysis and Data Assimilation
Synthesis.
Labs Put your name on your labs! Layouts: Site 1 Photos Title Legend
Creating Surfaces Steve Kopp Steve Lynch.
Statistical surfaces: DEM’s
Chapter 10 Problems Even problems are at end of text. 19. What is a kernel in a moving window operation? Does the kernel size or shape change for different.
Problems with Vector Overlay Analysis (esp. Polygon)
Interpolation of Surfaces
Review- vector analyses
Spatial Analysis Longley et al..
Inference for Geostatistical Data: Kriging for Spatial Interpolation
Interpolation & Contour Maps
Spatial interpolation
Interpolating Surfaces
Creating Surfaces with 3D Analyst
Concepts and Applications of Kriging
Presentation transcript:

1 GEOG4650/5650 – Fall 2007 Spatial Interpolation Triangulation Inverse-distance Kriging (optimal interpolation)

2 What is “Interpolation”? Predicting the value of attributes at “unsampled” sites from measurements made at point locations within the same area or region Predicting the value outside the area - “extrapolation” Creating continuous surfaces from point data - the main procedures

3 Types of Spatial Interpolation Global or Local Global-use every known points to estimate unknown value. Local – use a sample of known points to estimate unknown value. Exact or inexact interpolation Exact – predict a value at the point location that is the same as its known value. Inexact (approximate) – predicts a value at the point location that differs from its known value. Deterministic or stochastic interpolation Deterministic – provides no assessment of errors with predicted values Stochastic interpolation – offers assessment of prediction errros with estimated variances.

4 Classification of Spatial Interpolation Methods Global DeterministicStochastic Local Deterministic Stochastic Thiessen (exact) Density estimation(inexact) Inverse distance weighted (exact) Splines (exact) Kriging (exact) Regression (inexact) Trend surface (inexact)

5 Global Interpolation use all available data to provide predictions for the whole area of interest, while local interpolations operate within a small zone around the point being interpolated to ensure that estimates are made only with data from locations in the immediate neighborhood. Trend surface and regression methods

6 Trend Surface Analysis Approximate points with known values with a polynomial equation. Math equation – you don’t want to know…. Local polynomial interpolation – uses a sample of known points, such as convert TIN to DEM

7 Local, deterministic methods Define an area around the point to be predicted finding the data points within this neighborhood choosing a math model evaluating the point

8 Thiessen Polygon (nearest neighbor) Any point within a polygon is closer to the polygon’s known point than any other known points. One observation per cell, if the data lie on a regular square grid, then Thiessen polygons are all equal, if irregular then irregular lattice of polygons are formed Delauney triangulation - lines joining the data points (same as TIN - triangular irregular network)

9 Thiessen polygons Delauney triangulation

10 Example data set soil data from Mass near the village of Stein in the south of the Netherlands all point data refer to a support of 10x10 m, the are within which bulked samples were collected using a stratified random sampling scheme Heavy metal concentration measured

11 Exercise: create Thiessen polygon for zinc concentration Create a new project Copy “classfiles\GEOG4650-Li\data\10- 24\Soil_poll.dbf” and add it to the project. After add the table into the project, you need to create an event theme based on this table Go to Tools > Add XY Data and make sure the “Easting” is shown in “X” and “Northing” is in “Y”. (Don’t worry the “Unknown coordinate” Click on OK then the point theme will appear on your project.

12 Exercise 1) Try to plot this point data based on Zinc concentration (Try “Graduate Color” and “Graduate Symbol”)

13 This is what you might see on screen

14 Create a polygon theme The next thing you need to do is provide the Thiessen polygon a boundary so that the computing of irregular polygons can be reasonable Use ArcCatalog to create a new shapefile and name it as “Polygon.shp” Add this layer to your current project. Use “Editor” to create a polygon.

15 Creating Polygon Theme

16 Notes: 1)Remember to stop Edits, otherwise your polygon theme will be under editing mode all the time 2)Remember to remove the “selected” points from the “Soil_poll_data.txt”. If you are done so, your Thiessen polygons will be based on the selected points only.

17 Thiessen Polygon from ArcToolBox In Arctoolbox | Analysis Tools | Proximity | Create Thiessen Polygon Make sure the soil_poll event is the Input Features and output to your own folder. Select “All” for Output Fields.

18 Result from Thiessen polygon

19 Create Thiessen Polygons from Spatial Analyst: Set Extent and Cell Size Go to “Spatial Analyst > Options” and click on tab and use “Polygon” as the “Analysis Mask”. If the Analysis Mask is not set, the output layer will have rectangular shape.

20 Thiessen Polygon from Spatial Analyst Select Spatial Analyst > Distance > Allocation. In “Assign to”, select “soil_poll Event” and click OK to create cell in temporary folder.

21

22 Vector vs Raster Results Polygons from “Analysis Tools” are vector polygons with attributes. Polygons from “Spatial Analyst” are raster polygons with same values inside of each polygon, required to be converted to vector and NO attributes..

23 Zinc Concentration Plot thiessen polygons using zinc concentrations from the attribute table. Before you plot the map, trim thiessen polygons based on polygon.shp.

24 Inverse Distance Weighted the value of an attribute z at some unsampled point is a distance-weighted average of data point occurring within a neighborhood, which compute: =estimated value at an unsampled point n= number of control points used to estimate a grid point k=power to which distance is raised d=distances from each control points to an unsampled point

25 Computing IDW 246X246X Z 1 =40 Z 2 =60 Z 4 =40 Z 3 =50 Do you get 49.5 for the red square?

26 Exercise - generate a Inversion distance weighting surface and contour Spatial Analyst > Interpolate to Raster > Inverse Distance Weighted Make sure you have set the Output cell size to 50.

27 Contouring create a contour based on the surface from IDW Spatial Analyst | Surface Analysis | Contour

28 IDW and Contouring

29 Problem - solution Unsampled point may have a higher data value than all other controlled points but not attainable due to the nature of weighted average: an average of values cannot be lesser or greater than any input values - solution: Fit a trend surface to a set of control points surrounding an unsampled point Insert X and Y coordinates for the unsampled point into the trend surface equation to estimate a value at that point

30 Splines draughtsmen used flexible rulers to trace the curves by eye. The flexible rulers were called “splines” - mathematical equivalents - localized piece-wise polynomial function p(x) is

31 Spline - math functions piece-wise polynomial function p(x) is p(x)=p i (x) x i <x<x i+1 p j (x i )=p j (x i ) j=0,1,,,, i=1,2,,,,,,k-1 i+1 x0x0 xkxk x1x1 x k+1 break points

32 Spline r is used to denote the constraints on the spline (the functions p i (x) are polynomials of degree m or less r = 0 - no constraints on function

33 Exercise: create surface from spline have point data theme activated Spatial Analyst | Interpolate to Raster | Spline Define the output area and other parameters Select “Zn” for Z Value Field and “regularized” as type and “50” for Output cell size.

34 Kriging comes from Daniel Krige, who developed the method for geological mining applications Rather than considering distances to control points independently of one another, kriging considers the spatial autocorrelation in the data

35 Semivariance (  ) Z1Z2Z3Z4Z5Z1Z2Z3Z4Z Zi = values of the attribute at control points h =multiple of the distance between control points n =number of sample points

36 Semivariance h=1,h=2h=3h= (Z 1 -Z 1+h ) 2 (Z 2 -Z 2+h ) 2 (Z 3 -Z 3+h ) 2 (Z 4 -Z 4+h ) 2 sum 2(n-h)

37 semivariance the semivariance increases as h increases : distance increases -> semivariance increases nearby points to be more similar than distant geographical data

38 data no longer similar to nearby values h sill range

39 kriging computations we use 3 points to estimate a grid point again, we use weighted average =w 1 Z 1 + w 2 Z 2 +w 3 Z 3 = estimated value at a grid point Z 1,Z 2 and Z 3 = data values at the control points w 1,w 2, and w 3 = weighs associated with each control point

40 In kriging the weighs (wi) are chosen to minimize the difference between the estimated value at a grid point and the true (or actual) value at that grid point. The solution is achieved by solving for the wi in the following simultaneous equations w 1  ( h 11 ) + w 2  ( h 12 ) + w 3  ( h 13 ) =  ( h 1g ) w 1  ( h 12 ) + w 2  ( h 22 ) + w 3  ( h 23 ) =  ( h 2g ) w 1  ( h 13 ) + w 2  ( h 32 ) + w 3  ( h 33 ) =  ( h 3g )

41 w 1  ( h 11 ) + w 2  ( h 12 ) + w 3  ( h 13 ) =  ( h 1g ) w 1  ( h 12 ) + w 2  ( h 22 ) + w 3  ( h 23 ) =  ( h 2g ) w 1  ( h 13 ) + w 2  ( h 32 ) + w 3  ( h 33 ) =  ( h 3g ) Where  ( h ij )=semivariance associated with distance bet/w control points i and j.  ( h ig ) =the semivariance associated with the distance bet/w ith control point and a grid point. Difference to IDW which only consider distance bet/w the grid point and control points, kriging take into account the variance between control points too.

42 Example Z1(1,4)=50 Z2(2,1)=40 Z3(3,3)=25 Zg(2,2)=? g g123g distance h   =10h w w w =22.4 w w w =10.0 w w w =14.1

43 =0.15(50)+0.55(40) (25) = 37

44 Homework 6 – due next Friday midnight (11/2/07) See website