Cover Page

Grid Analysis User Guide

Version 4.4

March 2008

Copyright 2008 Mentum S.A. All rights reserved.

Notice This document contains confidential and proprietary information of Mentum S.A. and may not be copied, transmitted, stored in a retrieval system, or reproduced in any format or media, in whole or in part, without the prior written consent of Mentum S.A. Information contained in this document supersedes that found in any previous manuals, guides, specifications data sheets, or other information that may have been provided or made available to the user. This document is provided for informational purposes only, and Mentum S.A. does not warrant or guarantee the accuracy, adequacy, quality, validity, completeness or suitability for any purpose the information contained in this document. Mentum S.A. may update, improve, and enhance this document and the products to which it relates at any time without prior notice to the user. MENTUM S.A. MAKES NO WARRANTIES, EXPRESSED OR IMPLIED, INCLUDING, WITHOUT LIMITATION, THOSE OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, WITH RESPECT TO THIS DOCUMENT OR THE INFORMATION CONTAINED HEREIN. Trademark Acknowledgement Mentum, Mentum Planet and Mentum Ellipse are registered trademarks owned by Mentum S.A. MapInfo Professional is a registered trademark of PB MapInfo Corporation. RF-vu is a trademark owned by iBwave. WaveSight is a trademark of Wavecall. This document may contain other trademarks, trade names, or service marks of other organizations, each of which is the property of its respective owner.

Contents

Contents
MENTUM PRODUCTS CONTACTING MENTUM INTRODUCTION List of products Getting technical support Send us your comments Using this documentation Online Help Documentation library Notational conventions Performing analysis Working with grid files Understanding grids What is a grid? The gridding process Grid types Numeric grids Classified grids Grid file architecture Tables Grid resolution Understanding estimation techniques Interpolation techniques Modeling techniques Other grid creation techniques Setting your preferences 2 4 4 6 6 8 9 12 12 13 14 14 14 14 15 16 16 17 17 17 17 18 18

CHAPTER 1 Understanding Grids

Contents Grid Analysis User Guide

CHAPTER 2
Creating Grids Using Interpolation

Understanding interpolation Types of interpolation Choosing an interpolation technique Using the Interpolation Wizard to create a grid To create a grid using the Interpolation Wizard Triangulation with smoothing Inverse Distance Weighting interpolation Natural Neighbor interpolation Rectangular interpolation Kriging interpolation How kriging works Understanding kriging techniques Using the Kriging interpolation technique Generating a semivariogram Tuning the model Custom Point Estimation Suggested reading on interpolation techniques Natural neighbors Kriging Model types Using the Location Profiler model Setting the number of points Using point weighting Using distance decay functions To create a grid using the Location Profiler model Using the Trade Area Analysis models Trade areas To create a grid using a trade area analysis model Suggested reading on trade area analysis

20 20 21 24 24 25 27 28 30 31 32 33 34 34 38 40 40 41 41 44 44 45 46 47 49 50 50 53 55

CHAPTER 3 Creating Grids Using Spatial Models

ii

Contents Grid Analysis User Guide

CHAPTER 4
Creating Grids Using Other Methods

Introduction Creating point density grids Calculating point density using square area To create a square area point density grid Calculating point density using kernel smoothing To create a kernel smoothing point density grid Converting regions to a grid To convert a table of regions to a grid Buffering map objects to create a grid To create a grid using the Grid Buffer tool Preparing data using Poly-to-Point To create a point table using Poly-to-Point Importing grids Grid formats that you can import To import a grid Working with the Grid Manager To display the Grid Manager Using the Grid Manager Info function Using the Grid Info tool To use the Grid Info tool Using the Region Info tool To use the Region Info tool Using the Line Info tool To use the Line Info tool Using color in numeric grids To use the Grid Color tool Using color in classified grids To use the Dictionary Editor Creating and displaying legends To quickly create and display a legend To open the Grid Legend dialog box To customize a legend Modifying information about a grid To modify information about a grid

58 58 59 59 59 60 60 61 61 63 64 64 65 65 68 70 70 72 76 77 77 77 78 78 78 80 82 82 83 84 85 86 87 88

CHAPTER 5 Working with the Grid Manager

iii

Contents Grid Analysis User Guide

Changing grid file projections To change the projection of a grid file To change the projection of a vector-based .tab file Updating the projection file To add the Projection File Updater to the Tools menu To update the projection file Reclassifying grids To reclassify a numeric grid To reclassify a classified grid To reclassify isolated areas of a classified grid Converting grids To convert a numeric grid to a classified grid To convert a classified grid to a numeric grid Trimming grids To trim a grid Splicing grids To splice grids Resizing grids To resize a grid using the Resizer To resize a grid using the Numeric Grid Filter Exporting grids To export a grid CHAPTER 6 Working with Graphs Introduction Customizing graphs To customize a graph Changing graph viewing options To maximize a graph To zoom in and out on a graph To show or hide the legend To show or hide the graph data Printing graphs To print a graph Exporting graphs To export a graph

88 89 90 90 91 91 91 91 92 94 95 95 96 96 97 97 99 100 101 102 103 104 108 108 108 109 109 109 109 109 110 110 110 111

iv

Contents Grid Analysis User Guide

CHAPTER 7
Using Grids for Spatial Analysis

Introduction Using the Grid Calculator To use the Grid Calculator To use the Grid Calculator with saved expressions Understanding grid queries Creating and editing conditional queries To create a conditional query Creating a cross section To create a cross section graph Using the Point Inspection function To perform point inspection Using the Line Inspection function To perform line inspection Using the Region Inspection function To perform region inspection Introduction When to aggregate data Techniques for data aggregation Simple point aggregation To perform simple point aggregation Point aggregation with statistics To perform point aggregation with statistics Forward Stepping Aggregation Cluster density aggregation Square bin aggregation Building a table of standard deviation ellipses Understanding natural neighbors Creating regions from points (Voronoi) To create regions from points Calculating the region area To calculate the region area

114 114 115 116 116 118 118 119 119 119 119 120 121 122 122 124 124 125 125 126 127 127 128 130 131 132 136 137 137 138 138

CHAPTER 8 Aggregating Data

CHAPTER 9
Using Voronoi Diagrams

Contents Grid Analysis User Guide

CHAPTER 10
Creating 3D Views Using GridView

Introduction Creating a 3D scene To create a three-dimensional scene Adjusting scene properties Setting the viewing mode Setting the scene lighting Setting the surface appearance Creating a drape file To create a drape file Adding a drape file to a scene To load a drape file in GridView Saving a GridView scene To save a GridView scene To open a saved scene in Mentum Planet

142 142 142 143 143 143 144 145 145 146 146 146 146 147 149

INDEX

vi

Mentum Products

This chapter contains the following section:

The Mentum Product portfolio provides a range of products for planning and maintaining wireless networks. This section describes the products that are available as part of the portfolio. For additional details about any of these products, see the Mentum web site at http://www.mentum.com.

List of products

Mentum Products Grid Analysis User Guide

List of products
The following table describes wireless network planning and optimization products. The table does not provide details about specific features and tools. For more information, see the introductory chapters in the User Guide for the specific product or visit the Mentum web site at http://www.mentum.com.
Product Mentum Planet Description A Windows-based wireless network planning and analysis tool. You can add technologies and tools to support the planning functions that you require. Depending on the options that you choose, Mentum Planet provides support for the following technologies: TDMA/FDMAGSM (including GPRS and EGPRS), IS-136, AMPS, NAMPS, and iDEN

CDMAW-CDMA (UMTS, including HSPA), cdma2000 (including IS-95, 1xRTT, EV-DO)

Specialized modules Measurement Data Package Universal Model Test mobile and scan receiver functionality that can be added to Mentum Planet so that you can import and analyze measurement data and increase the accuracy of predictions. Propagation model that automatically adapts to all engineering technologies (micro, mini, small and macro cells), to all environments (dense urban, urban, suburban, mountainous, maritime, open), and to all systems (GSM, GPRS, EDGE, UMTS, WIFI, WIMAX) in a frequency range that spans from 400MHz to 5GHz. Indoor/outdoor module that links Mentum Planet with iBwave RF-vu allowing you to view and plan indoor/outdoor networks and manage RFvu projects using the Mentum Planet Data Manager.

Indoor/Outdoor

Optimization applications Mentum Ellipse Renaissance An integrated software solution for the optimal planning and design of point-to-point and point-to-multipoint radio transmission links. Frequency planning tool that uses evolutionary algorithms to find the very best frequency plan that will minimize interference across the network. Optimisation tool that enables engineers to improve upon manual optimisation techniques by allowing them to consider and adjust multiple input parameters simultaneously. The result is a quicker and more costeffective convergence towards a 'best network' configuration.

Capesso

Contacting Mentum
This chapter contains the following sections:

Mentum is committed to providing fast, responsive technical support. This section provides an extensive list of contacts to help you through any issues you may have. We also welcome any comments about our documentation. Customer feedback is an essential element of product development and supports our efforts to provide the best products, services, and support we can.

Getting technical support Send us your comments

Contacting Mentum Grid Analysis User Guide

Getting technical support

You can get technical support by phone or email, or by going to http://www.mentum.com/customercare/customercare.asp. Email is the best way of getting technical support.
North America Phone: +1 866 921-9219 (toll free), +1 819 483-7094 Fax: +1 819 483-7050 Email: [email protected] Hours: 8am 8pm EST/EDT (Monday-Friday, excluding local holidays) Europe, Middle East, and Africa Phone: +33 1 39264642 Fax: +33 1 39264601 Email: [email protected] Hours: 9am 6pm CET/CEST (Monday-Friday, excluding local holidays) Asia Pacific Phone: +852 2824 8874 Fax: +852 2824 8358 Email: [email protected] Hours: 9am 6pm HKT (Monday-Friday, excluding local holidays)

When you call for technical support, ensure that you have your product ID number and know which version of the software you are running. You can obtain this information using the About command from the Help menu. When you request technical support outside of regular business hours, a Product Support Specialist will respond the next working day by telephone or email, depending upon the nature of the request.

Send us your comments

Feedback is important to us. Please take the time to send comments and suggestions on the product you received and on the user documentation shipped with it. Send your comments to: [email protected]

Introduction

This chapter contains the following sections:

This user guide provides you with the necessary information to create mapping solutions for a wide range of applications.

Using this documentation

Introduction Grid Analysis User Guide

Using this documentation

Before using this documentation, you should be familiar with the Windows environment. It is assumed that you are using the standard Windows XP desktop, and that you know how to access ToolTips and shortcut menus, move and copy objects, select multiple objects using the Shift or Ctrl key, resize dialog boxes, expand and collapse folder trees. It is also assumed that you are familiar with the basic functions of MapInfo Professional. MapInfo Professional functions are not documented in this User Guide. For information about MapInfo Professional, see the MapInfo online Help and MapInfo Professional User Guide. You can access additional MapInfo user documentation from the MapInfo website at www.mapinfo.com. All product information is available through the online Help. You access online Help using the Help menu or context-sensitive Help from within a dialog box by pressing the F1 key. If you want to view the online Help for a specific panel or tab, click in a field or list box to activate the panel or tab before you press the F1 key. The following sections describe the structure of the online Help.

Online Help
From the Help menu, you can access online Help for Mentum Planet software and for MapInfo Professional. This section describes the structure of the Mentum Planet online Help. The online Help provides extensive help on all aspects of software use. It provides

help on all dialog boxes procedures for using the software an extensive Mentum Planet documentation library in PDF format User Guides

The following sections provide details about the resources available through the online Help.
Resource Roadmap

When you first use the online Help, start with the Resource Roadmap. It describes the types of resources available in the online Help and explains how best to use them. It includes a step-by-step guide that walks you through the available resources.

Introduction Grid Analysis User Guide

Printing

You have two basic options for printing documents:

If you want a good quality print of a single procedure or section, you can print from the Help window. Click Print in the Help window. If you want a higher quality print of a complete User Guide, use Adobe Reader to print the supplied print-ready PDF file contained in the Mentum Planet documentation library. Open the PDF file and choose File Print.

Library Search

You can perform a full-text search on all PDF files contained in the Mentum Planet documentation library if you are using a version of Adobe Reader that supports full-text searches. The PDF files are located in the Mentum Planet 4\Help folder.
You can also perform a search on all online Help topics by clicking the Search tab in the Help window. Type a keyword, and click List Topics to display all Help topics that contain the keyword. The online Help duplicates the information found in the User Guide PDF files in order to provide more complete results. It does not duplicate the information in the Release Notes, or Glossary. Frequently Asked Questions

The Frequently Asked Questions section provides answers to common questions about Mentum Planet. For easy navigation, the section is divided into categories related to product functionality.
Whats This? Help

Whats This? Help provides detailed explanations of all dialog boxes.

User Guides

All User Guides for Mentum Planet software is easily accessible as part of the online Help.

Introduction Grid Analysis User Guide

Documentation library
Mentum Planet comes with an extensive library of User Guides in PDF format. The following table provides details about the documentation supplied with Mentum Planet. Additional documents, including Application Notes and Technical Notes, are available on the Mentum Web site: http://www.mentum.com.

Document Mentum Planet User Guide Grid Analysis User Guide

Enables you to Plan and analyze simulated wireless communication networks. Perform operations on spatial data that is stored in grids, and display, analyze, and export digital elevation models (DEM) and other grid-based data. Model indoor networks and learn how to view, edit, and manage generic and RF-vu projects from within Mentum Planet. Plan and analyze TDMA/FDMA networks. Plan and analyze W-CDMA (UMTS) and cdma2000 networks. Learn how to use the Data Manager. The Data Manager enables users to work with centralized Mentum Planet data stored in an Oracle or Microsoft SQL Server database. Learn how to install and configure the Data Manager Server on database and file servers in a network environment, and how to manage access to project data. Install Wireless Network Planning software. Search for commonly used technical terms. Learn about new features and known issues with the current release of software.

Indoor Analysis User Guide

TDMA/FDMA User Guide CDMA User Guide Data Manager User Guide

Data Manager Server Administrator Guide

Installation Guide Glossary Release Note

Introduction Grid Analysis User Guide Document Data Manager Server Release Note MapInfo Professional User Guide Enables you to Learn about new features and known issues with the current release of Data Manager Server software. Learn about the many features of MapInfo Professional, as well as basic and advanced mapping concepts.

Notational conventions
This section describes the textual conventions and icons used throughout this documentation.
Textual conventions

Special text formats are used to highlight different types of information. The following table describes the special text conventions used in this document.
bold text Bold text is used in procedure steps to identify a user interface element such as a dialog box, menu item, or button. For example: In the Select Interpolation Method dialog box, choose the Inverse Distance Weighting option, and click Next. Courier text is used in procedures to identify text that you must type. For example: In the File Name box, type Elevation.grd. Bright blue text is used to identify a link to another section of the document. Click the link to view the section. Menu arrows are used in procedures to identify a sequence of menu items that you must follow. For example, if a step reads Choose File Open, you would click File and then click Open. <> Angle brackets are used to identify variables. For example, if a menu item changes depending on the chosen unit of measurement, the menu structure would appear as Display <unit of measurement>.

courier text

bright blue text

Icons

Throughout this documentation, icons are used to identify text that requires special attention.

Introduction Grid Analysis User Guide

This icon identifies a workflow summary, which explains a series of actions that you will need to carry out in the specified order to complete a complex task.

This icon identifies a cautionary statement, which contains information required to avoid potential loss of data, time, or resources. This icon identifies a tip, which contains shortcut information, alternative ways of performing a task, or methods that save time or resources. This icon identifies a note, which highlights important information or provides information that is useful but not essential.

10

1.
This chapter contains the following sections:

Chapter 1: Understanding Grids

Understanding Grids

This chapter covers the concept of grids and explains how they are used in Mentum Planet.

Performing analysis Working with grid files Understanding grids Grid types Grid file architecture Tables Grid resolution Understanding estimation techniques Setting your preferences

11

Chapter 1 Grid Analysis User Guide

Performing analysis
The analysis component of Mentum Planet provides a mapping technique for calculating and displaying the trends of data that vary continuously over geographic space, and provides a mechanism for sophisticated comparison and analysis of multiple map layers. Three main object types are currently used by GIS applications to represent the spatial distribution of data: regions, lines, and points. None of these objects is very well suited to representing data that varies continuously through space such as ground-level air temperature, elevation, distance from a store location, or the distribution of wealth across a city. Values for this type of data must all be collected at discrete locations, but the way they change over space is very significant. Traditional ways of indicating variation are labeling individual sample locations with a known value, creating graduated symbols at each sample site, where the size reflects the samples value, and generating contour lines or regions depicting locations of equal value (Figure 1.1).

Figure 1.1 Three examples of how a traditional vector-based GIS system, such as MapInfo, displays data that varies continuously.

The problem with these methods is that they do not portray how the data changes between known locations. To address this problem, Mentum Planet creates a type of spatial data representation called a grid. Grids enable you to represent data as a continuous coverage. You can see how values change in space and query any location to obtain a meaningful value.

Working with grid files

You should be familiar with the concept of map layers when you work with Mentum Planet. Each unique layer of information exists as a separate file that can be added as a layer in a Map window.

12

Understanding Grids Grid Analysis User Guide

Figure 1.2 Various map layers covering the same geographical area can hold different types of information

Just as each layer can be visualized above or below another layer, layers can be compared using spatial analysis functions. The GIS functionality of Mentum Planet works with layers of map data that are vector-based, i.e., point, line and polyline, or polygon information. Points can represent soil samples or retail store locations; lines and polylines can represent roads; and polygons can represent trade areas, bodies of water, or municipal boundaries. Mentum Planet also offers another level of data representation: grid-based layers. Grid data is the best way to represent phenomena that vary continuously through space. Elevation, signal strength, soil chemistry, and income are excellent examples of properties that are distributed in constantly varying degrees through space and are best represented in grid form as map layers.

Understanding grids
Grids represent the basic structural component for contouring, modeling, and displaying spatial data in Mentum Planet. Grids can be considered the fourth spatial data type after regions, lines, and points. A grid can be used to effectively visualize the trends of geographic information across an area. Grids give you the power to mathematically compare and query layers of information, create new derived grids, or analyze grid layers for such unique properties as visual exposure, proximity, density, or slope.

13

Chapter 1 Grid Analysis User Guide

What is a grid?
A grid is made up of regularly spaced square bins arranged over a given area. Each bin has a node, which is a point located at its center. Each bin can be given a value and a color representing the value. If there are several bins between two known locations, such as two contour lines, the change in color indicates how the values change between the locations.

The gridding process

Typically, the gridding process begins by overlaying an array of grid nodes over the original point file. This can be visualized as a regularly spaced point file arranged in the form of a grid (Figure 1.3). Each grid node is then attributed with an estimated value based upon the values of the surrounding points. The grid is then displayed in a Map window. The display takes the form of a raster image, where the colors reflect the estimated node values. It is because of this display component that grid-based GIS systems are called raster GIS.

Figure 1.3 The gridding process begins by creating an array of grid nodes geographically coincident to the sample file. Each grid node is attributed with an estimated value that is displayed as a raster image in a Map window.

Grid types
Mentum Planet supports two types of grids: numeric grids, which have numeric attribute information, and classified grids, which have character attribute information.

Numeric grids
The most illustrative example of a numeric grid is a digital terrain model where each bin is referenced to a value measured in units of distance above sea level (Figure 1.4). Numeric grids are best used to define continuously varying surfaces of information, such as elevation, in which bin values are

14

Understanding Grids Grid Analysis User Guide

either mathematically estimated from a table of point observations or assigned real numeric values. For example, in Figure 1.4, each bin was calculated (interpolated) from a table of recorded elevation points. In Mentum Planet, numeric grid files are given the extension .grd.

Figure 1.4 An example of a numeric grid showing the continuous variation of elevation across an area.

Classified grids
Classified grids are best used to represent information that is more commonly restricted to a defined boundary. They are used in the same way that a region is used to describe a boundary area, such as a land classification unit or a census district. In this case, the grid file does not represent information that varies continuously over space. In Figure 1.5, a land classification grid displays each bin with a character attribute attached to it that describes the land type underlying it. In Mentum Planet, classified grid files are given the extension .grc.

15

Chapter 1 Grid Analysis User Guide

Figure 1.5 An example of a classified grid representing land use where each bin is referenced to a descriptive attribute.

Grid file architecture

Every Mentum Planet grid file (.grc or .grd) is divided into two sections. The first section, called the file header, contains several pieces of information including the following:

Map name Map size (number of bins in height and width) Bin size Coordinates of first bin Grid Projection Grid value description

The second section, the body of the grid file, contains the attribute data for every bin in the map.

Tables
For GIS tables, such as vector tables and point tables, up to five different files are created (.tab, .map, .dat, .id, and .ind). For grid files, all the information is contained in either a .grc or a .grd file. An associated .tab file is created that points to the .grc or .grd file.

16

Understanding Grids Grid Analysis User Guide

Grid resolution
The resolution of a grid is the size of the bins. Mentum Planet has square bins, so the width and height are identical. The smaller the grid, the higher the resolution (the more detailed the information depicted). For example, what appears as a spike at 1000 m resolution may be clearly discernible as part of a mountain at 1 m resolution. Appropriate resolution depends on the application. For example, for data at the world level, 1 km resolution is relatively high. For wireless planning, however, 1 km resolution is low, and 1 m resolution produces much more accurate results.

Understanding estimation techniques

In order to represent how data changes between known values, some type of estimation must be made. There are several kinds of estimation technique that can be applied to a point file. These techniques fall into two different categories: interpolation techniques and modeling techniques. They approach the creation of a grid from entirely different perspectives, but both are mathematical construction tools designed to build grids that assign values to grid nodes from a geographically coincident point file.

Interpolation techniques
These techniques are used to build grids that are an estimation of the same variable as the underlying points. Each new bin has the same unit of measure as the point value. Mentum Planet currently supports six interpolation techniques:

Triangulation with Smoothing (TIN) Inverse Distance Weighting (IDW) Natural Neighbor (simple and advanced) Rectangular Kriging Custom Point Estimation

Modeling techniques
These techniques create grids of derived values. For example, one of the modeling techniques included with Mentum Planet is trade area analysis. This modeling technique uses store locations and their relative attractiveness to

17

Chapter 1 Grid Analysis User Guide

calculate grid values measuring the percent probability of customer patronage. In this case, the units of the resulting grid are different from the units of the originating point file. Mentum Planet currently supports two modeling techniques. The first, used by the Location Profiler, creates a grid measuring the average distance to point locations from anywhere within a map area. The second, trade area analysis, is based on the Huff model and calculates a grid measuring the probability of customer patronage within a trade area.

Other grid creation techniques

You can also create grids using the following methods:

Convert existing regions to a grid. This is also known as vectorto-raster conversion and is given the term Region to Grid in Grid Analysis on the GIS menu. The process simply converts the existing region file to a grid where every grid node is given the same value as the region it falls in. Because Mentum Planet grids can be attributed with only one piece of information, you are prompted to choose the column in the region file to attribute to the grid. In this way, you choose what type of grid will be created. If a numeric column is chosen, then a numeric grid is created. If a character column is chosen, a classified grid is created. Import grid files from external sources. This technique offers more flexibility with regard to the source of the information used and analyzed. Analyze existing grids. Each tool used to analyze an existing grid creates a new grid with the results of the analysis. There are many ways to analyze existing grids, for example, by performing a viewshed analysis, by using the Grid Calculator, by using the Grid Query tool, and by creating slope and aspect.

Setting your preferences

You can determine the default settings that control grid file management and dialog box usage in the Preferences dialog box.

Choose GIS Grid Analysis Preferences.

For more information on the fields and options in the Preferences dialog box, press the F1 key.

18

2.
This chapter contains the following sections:

Chapter 2: Creating Grids Using Interpolation

Creating Grids Using Interpolation

This chapter describes all of the basic commands associated with the creation of numeric grid files, and explains interpolation techniques.

Understanding interpolation Types of interpolation Choosing an interpolation technique Using the Interpolation Wizard to create a grid Triangulation with smoothing Inverse Distance Weighting interpolation Natural Neighbor interpolation Rectangular interpolation Kriging interpolation Custom Point Estimation Suggested reading on interpolation techniques

19

Chapter 2 Grid Analysis User Guide

Understanding interpolation
Interpolation is the process of estimating grid values using measured observations taken from a point file. New values calculated from the original point observations form a continuous, evenly spaced grid surface that fills in the gaps between the non-continuous points. Many mathematical formulae can be applied to estimating or interpolating grid values from an existing point file. There is no perfect solution, and many techniques are in use. The validity of each method depends entirely upon the type of data being interpolated, and each generates a unique style of interpolation surface.

Types of interpolation
The Interpolation Wizard enables you to create grid files using six different techniques. The Natural Neighbor technique has two variants: Simple and Advanced.
Triangulation with Smoothing Original data points are joined by a network of lines to build a mesh of triangular faces, called a Triangular Irregular Network (TIN). These faces represent the original data surface. New grid values are then estimated according to the slope of the TIN surface at the nearest points. Inverse Distance Weighting Original data points lying within a prescribed radius of a new grid node are weighted according to their distance from the node and then averaged to calculate the new bin value.

Natural Neighbor A network of natural neighbor regions (Voronoi diagram) is built using the original data. This creates an area of influence for each data point that is used to assign new values to overlying bins. Natural neighbor interpolation has two options: Simple and Advanced. Natural Neighbor (Simple) The Simple option offers the first-time user a two-step process for implementing the interpolation technique. Many of the controls have been pre-set to generate the most appropriate surface given the distribution of points.

20

Creating Grids Using Interpolation Grid Analysis User Guide Natural Neighbor (Advanced) The Advanced option gives you access to a variety of controls in the Natural Neighbor interpolation technique that you can use to make subtle adjustments to the grid surface generated from a points table. Rectangular (Bilinear) Original data points are joined by a network of lines to build a rectilinear mesh. New grid values are then estimated using the slopes of the double linear (bilinear) framework formed by the nearest four points. Kriging Kriging is a geostatistical interpolation technique that considers both the distance and the degree of variation between known data points when estimating values in unknown areas. Graphing tools help you understand and model the directional (e.g., north-south, east-west) trends of your data. Custom Point Estimation Bin values are calculated based upon a user-defined mathematical operation and performed using the data points found within a given search radius around each bin. Operations include sum, minimum, maximum, average, count, and median.

Choosing an interpolation technique

The most challenging task in creating a surface through interpolation is choosing the most appropriate technique. All interpolation techniques create gridded surfaces; however, the results may not properly represent how the data behaves through space, i.e., how the values change from one location to the next. For example, if an elevation surface is created from sample points taken in a mountainous area, you need to choose a technique that can simulate the severe elevation changes because this is how this type of data behaves. It is not always easy to understand how data behaves before you start the gridding process and, therefore, it can be difficult to know what technique to use. The answers to the following questions will help you determine the most appropriate technique to use.

21

Chapter 2 Grid Analysis User Guide

What kind of data is it, or, what do the data points represent?

Some interpolation techniques can be automatically applied to certain data types.

Data type Elevation Soil Chemistry Demographic Drive Test Possible interpolation Triangular Irregular Network (TIN), Natural Neighbor (NN) Inverse Distance Weighting (IDW), Kriging NN, IDW, Kriging NN

How accurate is the data?

Some techniques assume that the value at every data point is an exact value and will honor it when interpolating. Other techniques assume that the value is more representative of an area.
Point value accuracy Very Accurate Not Very Accurate Possible interpolation technique NN, TIN, Rectangular IDW, Kriging

What does the distribution of the points look like?

Some interpolation techniques produce more reasonable surfaces when the distribution of points is truly random. Other techniques work better with point data that is regularly distributed.
Point distribution Possible interpolation technique Most interpolation techniques work well with randomly scattered data points. NN, TIN, IDW, Kriging

Highly clustered data presents problems for many interpolation techniques. NN, IDW, Kriging TIN for slightly clustered data points

22

Creating Grids Using Interpolation Grid Analysis User Guide Point distribution Possible interpolation technique Rectangular can only handle data that is distributed in an evenly spaced pattern. Rectangular, NN, Kriging

This type of linear pattern generally occurs when data is collected from aircraft. Samples are taken close together but flight lines are some distance apart. IDW, NN, Kriging

This type of linear pattern generally occurs when samples are taken along roads. NN, Kriging

Is interpolation speed a factor?

Certain factors will influence the speed of interpolation. The smaller the bin and/or the more points in the data, the longer it takes to calculate the surface. However, some interpolation techniques are faster than others.
Interpolation technique TIN IDW Rectangular NN Kriging Speed Fast Fast Very Fast Slow Slow Limiting factors None Search and display radius size Search radius size Point distribution Number of directions analyzed

23

Chapter 2 Grid Analysis User Guide

Is it necessary to overshoot or undershoot the local minimum and maximum values?

Overshooting and undershooting the local minimum and maximum values is generally necessary when interpolating elevation surfaces.
Over/Undershoot Yes No Possible interpolation technique TIN, NN IDW, Rectangular, Kriging

Using the Interpolation Wizard to create a grid

Using the Interpolation Wizard to create grid files streamlines the process of estimating grid values.

To create a grid using the Interpolation Wizard

1 2 3 4 5 6 7 Choose GIS Grid Analysis Create Grid Interpolation. In the Select Interpolation Method dialog box, choose the interpolation method you want to use. Click Next. In the Select Table and Column dialog box, click Open Table to add a table to the Select Table To Grid list. From the Select Table To Grid list, choose the appropriate table of points that contains the data to be gridded. From the Select Column list, choose the column that contains the attribute values. To use an unmapped data file (an x, y, z file) that has not been converted to a vector point table using the Create Points command, do the following:

From the X-Column list, choose the column containing the x-coordinates for each point. From the Y-Column list, choose the column containing the y-coordinates for each point. Click Projection and choose the coordinates system of the location data.

In the Enter Data Description box, type an annotation (maximum 31 characters). This will be carried as a header in the grid file.

24

Creating Grids Using Interpolation Grid Analysis User Guide

To set the unit of measurement for the z-value, do one of the following:

From the Unit Type list, choose the appropriate unit of measurement of the z-value. In the Enter User Defined Type box, type a user-defined unit of measurement.

10 If you want to include only non-zero records, enable the Ignore Records Containing Zero check box. 11 Click Next. A dialog box specific to the type of interpolation opens. Each dialog box is discussed in the following sections.

Triangulation with smoothing

Triangulation is a process of grid generation that is usually applied to data that requires no regional averaging, such as elevation readings. The surface created by triangulation passes through (honors) all of the original data points while generating some degree of overshoot above local high values and undershoot below local low values. Elevation is an example of point values that are best surfaced with a technique that predicts some degree of overand under- estimation. In modeling a topographic surface from scattered elevation readings, it is not reasonable to assume that data points were collected at the absolute top or bottom of each local rise or depression in the land surface.
.

Local maximum value

overshoot > local maximum Interpolated surface

Data point

Local minimum value Figure 2.1 TIN interpolation profile: using the triangulation technique, the interpolated surface passes through the original data points. However, peaks and valleys will extend beyond the local maximum and minimum values.

Triangulation involves a process whereby all the original data points are connected in space by a network of triangular faces, drawn as equilaterally as

25

Chapter 2 Grid Analysis User Guide

possible. This network of triangular faces is referred to as a Triangular Irregular Network (TIN) as shown in Figure 2.2. Points are connected based on the nearest neighbor relationship (the Delaunay criterion) which states that a circumcircle drawn around any triangle will not enclose the vertices of any other triangle.

Figure 2.2 A three dimensional view of a Triangular Irregular Network (TIN).

A smooth grid surface is then fitted to the TIN using a bivariate fifth-order polynomial expression in the x- and y- direction for each triangle face. This method guarantees continuity and smoothness of the surface along the sides of each triangle and smoothness of the surface within each triangle. The slope blending algorithm is designed to calculate new slope values for each of the triangle vertices (i.e., each point of the data) where the influence of adjacent slopes in the blending calculation is weighted according to specified triangle properties. Five properties of data point geometry and value greatly influence the ability of the slope-blending algorithm to control smoothing of the TIN surface:

the triangle centroid location the triangle aspect ratio the triangle area the angle versus slope of the triangle the statistically-derived slope of a triangle vertex

For example, triangles with centroids farther from the vertex being solved have less influence on the slope calculation than triangles whose centroids are closer; similarly, triangles with greater areas have greater influence in the slope calculation than triangles with a smaller area. The end result is a smoothing process that significantly reduces the frequency of angular

26

Creating Grids Using Interpolation Grid Analysis User Guide

artifacts, representing remnants of the original TIN facets in the final gridded surface.

Inverse Distance Weighting interpolation

Inverse Distance Weighting (IDW) interpolation is a moving average interpolation technique that is usually applied to highly variable data. For certain data types, it is possible to return to the collection site and record a new value that is statistically different from the original reading but within the general trend for the area. Examples of this type of data include environmental monitoring data such as soil chemistry and consumer behavior observations. It is not desirable to honor local high and low values but rather to look at a moving average of nearby data points and estimate the local trends.
Local maximum value Interpolated surface Data point

Local minimum value Figure 2.3 IDW interpolation profile: the interpolated surface, estimated by using a moving average technique, is less than the local maximum value and greater than the local minimum value.

The IDW technique calculates a value for each grid node by examining surrounding data points that lie within a user-defined search radius. Some or all of the data points can be used in the interpolation process. The node value is calculated by averaging the weighted sum of all the points. Data points that lie progressively farther from the node influence the computed value far less than those lying closer to the node (Figure 2.4).

27

Chapter 2 Grid Analysis User Guide

Search radius

Grid node Data point Figure 2.4 IDW Calculation: a radius is generated around each grid node from which data points are selected for use in the calculation.

Natural Neighbor interpolation

Natural Neighbor interpolation is a geometric estimation technique that uses natural neighbor regions generated around each point in the data. This technique is particularly effective for dealing with a variety of spatial data themes exhibiting clustered or highly linear distributions. The natural neighbor technique is designed to honor local minimum and maximum values in the point file and can be set to limit overshoots of local high values and undershoots of local low values. This technique thereby enables the creation of accurate surface models from data that is very sparsely distributed or very linear in spatial distribution.
Interpolated surface using overshooting Local maximum value Interpolated surface without overshooting

Local minimum value Figure 2.5 Natural neighbor interpolation profile: the interpolated surface is tightly controlled by the original data points by honoring the value at each point. It also provides the option to over- or under-shoot point values.

28

Creating Grids Using Interpolation Grid Analysis User Guide

Put simply, natural neighbor interpolation makes use of an area-weighting technique to determine a new value for every grid node. As shown in Figure 2.6, a natural neighbor region is first generated for each data point. Then, a new natural neighbor region is generated at every node in the new grid which effectively overlies various portions of the surrounding natural neighbor regions defining each point. The new grid value is calculated as the average of the surrounding point values proportionally weighted according to the intersecting area of each point.

Figure 2.6 A display of the natural neighbor regions around the point file as well as those created around a grid node.

You can choose one of three variations on the natural neighbor technique. Figure 2.7 illustrates the behavior of each variation.

Constant Value Solutioneach grid node takes on the value of the underlying natural neighbor region. Linear Solutionthe grid value is determined by averaging the point values associated with surrounding natural neighbor regions and weighted according to the area that is encompassed by a temporary natural neighbor region generated around the bin (Figure 2.6). Slope-Based Solutionthe grid value is determined by averaging the extrapolated slope of each surrounding natural neighbor region and area weighted as in the linear solution. By examining the adjacent points, a determination is made as to whether that point represents a local maximum or minimum value. If it does, a slope value of zero is assigned to that value

29

Chapter 2 Grid Analysis User Guide

and the surface will honor that point by neither overshooting nor undershooting it.
Slope-based solution Constant solution

Linear solution Figure 2.7 A graph showing the three variations of Natural Neighbor interpolation.

The Interpolation Wizard provides two methods of performing natural neighbor interpolation. The simple option offers the first-time user a two-step process for implementing the interpolation method. Many of the controls have been pre-set to generate the most appropriate surface given the distribution of points. The Advanced option enables you to fine-tune the grid surface.

Rectangular interpolation
Rectangular interpolation is usually applied to data that is regularly and closely spaced, such as points generated from another gridding application. This technique creates an interpolation surface that passes through all points without overshooting the maximum values or undershooting the minimum values. Rectangular interpolation locates the four nearest data points lying within a circular search zone, one from each quadrant, and connects them with a double linear rectangular framework (Figure 2.8). An appropriate value is calculated for each node using the slopes of the connecting sides of the rectangle. However, in the absence of additional smoothing, linear artifacts are often generated across the surface when working with an irregular data point distribution.

30

Creating Grids Using Interpolation Grid Analysis User Guide

Search radius

Grid node Data point Figure 2.8 Rectangular interpolation calculation: a radius is generated around each grid node from which the closest data point in each quadrant is selected to be used in the calculation.

Kriging interpolation
Kriging is a geostatistical interpolation technique that considers both the distance and the degree of variation between known data points when estimating values in unknown areas. A kriged estimate is a weighted linear combination of the known sample values around the point to be estimated. Applied properly, kriging enables you to derive weights that result in optimal and unbiased estimates. It attempts to minimize the error variance and set the mean of the prediction errors to zero so that there are no overestimates or underestimates. Included with the kriging function is the ability to construct a semivariogram of the data, which is used to weight nearby sample points. It also provides a means for you to understand and model the directional (e.g., north-south, east-west) trends of your data. A unique feature of kriging is that it provides an estimation of the error at each interpolated bin, providing a measure of confidence in the modeled surface. The effectiveness of kriging depends on the correct specification of several parameters that describe the semivariogram and the model of the drift (i.e., the mean value does or does not change over distance). Because kriging is a robust interpolation technique, even a naive selection of parameters will provide an estimate comparable to many other grid estimation procedures. The trade-off for estimating the optimal solution for each point by kriging is computation time. Given the additional trial and error time necessary to select appropriate parameters, kriging should be applied where best estimates are required, data quality is good, and error estimates are essential.

31

Chapter 2 Grid Analysis User Guide

Mentum Planet provides three different methods of kriging interpolation: ordinary kriging, simple kriging, and universal kriging.

How kriging works

Kriging is a weighted moving average technique that is similar to Inverse Distance Weighting (IDW) interpolation. With IDW, each grid node is estimated using sample points that fall within a circular radius. The degree of influence each point has on the calculated value is based upon the weighted distance of each point from the grid node being estimated. In other words, points that are closer to the node will have a greater degree of influence on the calculated value than those points farther away. The general relationship between the amount of influence a sample point has with respect to its distance is determined by the IDW Exponent setting, as shown below.
1
Degree of influence Distance decay curves for exponents 1, 2, and 3

2 3
Grid node

A B
Data point Distance from node Search radius

Figure 2.9 Effect of IDW Exponent on decay curves.

The disadvantage of IDW interpolation is that it treats all points that fall within the search radius the same way. For example, if an exponent of 1 is specified, a linear distance decay function is used to determine the weights for all points that lie within the search radius (Figure 2.9). This same function is used for all points regardless of their geographic orientation to the node (north, south, etc.) unless a sectored search is implemented. Kriging, on the other hand, uses different weighting functions depending on the distance and orientation of sample points with respect to the node and the manner in which sample points are clustered.

32

Creating Grids Using Interpolation Grid Analysis User Guide

Search radius

B
Grid node Figure 2.10 This figure illustrates the influence points have, using IDW interpolation, on the calculated value based on the same distance decay function when one of the points is northeast of the grid node (point A) and the other point is southeast (point B). With kriging, the grid node may be calculated using a different weighting function for every point in the search radius.

Before interpolation begins, every possible distance weighting function is calculated by generating an experimental semivariogram and choosing a mathematical model to approximate its shape. The mathematical model provides a smooth, continuous function for determining appropriate weights for increasingly distant data points.

Understanding kriging techniques

Mentum Planet provides three variations of kriging interpolation that you can apply in two forms, although they all operate in a similar way. The three methods are ordinary kriging, simple kriging, and universal kriging, and all three of these techniques can be applied in one of two forms: punctual or block.
Ordinary kriging

This method assumes that the data has a stationary variance and a nonstationary mean value within the search radius. Ordinary kriging is highly reliable and is recommended for most data sets.
Simple kriging

This method assumes that the data has a stationary variance and a stationary mean value and requires you to enter the mean value.
Universal kriging

This method represents a true geostatistical approach to interpolating a trend surface of an area. The surface representing the drift of the data is built first, then the residuals for this surface are calculated. With universal kriging, you

33

Chapter 2 Grid Analysis User Guide

can set the polynomial expression used to represent the drift surface. The following equation shows the most general form of this expression. F(x, y) = a20 * x2 + a11 * xy + a02 * y2 + a10 * x + a01 * y + a00
Equation 2.1 Universal kriging Where

a00 is always present but rarely set to zero in advance of the calculation. However, the other coefficients can also be set to zero. The recommended setting is a first degree polynomial which will avoid unpredictable behavior at the outer margins of the data.
Punctual and block kriging

All three kriging interpolation techniques can be applied in one of two forms: punctual or block. The most commonly used is punctual kriging (the default), which estimates the value at a given point. Block kriging uses the estimate of the average expected value at a given location (such as a block) around a point. Block kriging provides better variance estimation and has the effect of smoothing interpolated results.

Using the Kriging interpolation technique

There are four basic steps in the kriging process:

aggregate the data choose the kriging parameters complete the variogram analysis perform the kriging estimation

If you choose all the system defaults, the kriging type will be ordinary, the experimental semivariogram will be calculated, a model will be automatically fitted to the data, and kriging interpolation will be performed. However, experienced users will always spend some time fitting a model to the semivariogram.

Generating a semivariogram
Kriging uses a different weighting function depending on both the distance and geographic orientation of the sample point to the node being calculated. It is difficult for you to know, at a first glance, precisely how data varies outward from any one location with respect to distance and direction. There are, however, many techniques available to help you determine how data varies. The most popular is a variance analysis.

34

Creating Grids Using Interpolation Grid Analysis User Guide

Data points

Constant data trend Steeply changing data trend

Figure 2.11 Example of data that does not vary crosswise but varies greatly along the y-axis of the data.

Kriging uses a property called semivariance to express the degree of relationship between points on a surface. The semivariance is simply half the variance of the differences between all possible points spaced a constant distance apart. The semivariance at a distance d = 0 will be zero because there are no differences between points that are compared to themselves. However, as points are compared to increasingly distant points, the semivariance increases. At some distance, called the range, the semivariance will become approximately equal to the variance of the whole surface itself. This is the greatest distance over which the value at a point on the surface is related to the value at another point. The range defines the maximum neighborhood over which control points should be selected to estimate a grid node, taking advantage of the statistical correlation among the observations. The calculation of semivariance between sample pairs is performed at different distances until all possible distance combinations have been analyzed. The initial distance used is called the lag distance, which is increased by the same amount for each pass through the data. For example, if the lag distance is 10 meters, the first pass calculates the variance of all sample pairs that are 10 meters apart. The second pass calculates the variance of all sample pairs 20 meters apart, the third at 30 meters and so on until the last two points that are the farthest apart have been examined. Put simply, every point is compared to every other point to determine which points are approximately the first lag distance apart. When points this distance apart are found, the variance between their values and their geographical orientation is determined. Once the first lag distance has been analyzed, the process is repeated using the second lag distance and then the third, and so on

35

Chapter 2 Grid Analysis User Guide

until all distance possibilities are exhausted. When the variance analysis is completed, the information is displayed in a semivariogram. A semivariogram is a graph that plots the semivariance between points on the y-axis and distance at which the semivariance was calculated on the x-axis. An example of a semivariogram is shown in Figure 2.12. The jagged line on the graph is the plot of calculated semivariances, plotted on the y-axis, and their corresponding lag distances on the x-axis. This plot is given the term experimental semivariogram. The jagged nature of the experimental semivariogram makes it unsuitable for use in calculating the kriging weights, so a smooth mathematical function (model) must be fit to the variogram. The model is shown as the smooth line on the graph (Figure 2.12).
Semivariogram view
11000 10000 9000 8000 Variance 7000 6000 5000 4000 3000 2000 1000 0 0 2500 5000 7500 10000 Distance 12500 15000 17500

Model Curve

Experimental Semivariogram

Figure 2.12 An example of an omni-directional semivariogram.

Although the strength of kriging is its ability to account for directional trends of the data, it is possible to analyze variance with respect to distance only and disregard how points are geographically oriented. The above experimental semivariogram is an example of this, called an omni-directional experimental semivariogram. If the geographic orientation is important, then a directional semivariogram should be calculated such as the one shown in Figure 2.13.

36

Creating Grids Using Interpolation Grid Analysis User Guide Semivariogram view
15000 12500 10000 Variance 7500 5000 2500 0

Model Curve

North-South Experimental Semivariogram

East-West Experimental Semivariogram

0 2500 5000 7500 10000 Distance 12500 15000 17500

Figure 2.13 An example of a directional semivariogram. Notice the two experimental semivariograms, one representing points oriented north and south of each other, and the other representing points oriented east and west of each other.

When two or more directions are analyzed, an experimental semivariogram will be generated for each direction. In Figure 2.13, two directions are being investigated and therefore two experimental semivariograms are plotted. Semivariogram experimentation can uncover fundamental information about the data, such as whether the data varies in more than one direction. In more technical terms, the semivariogram experimentation can reveal whether the data is isotropic (the data varies the same in all directions) or anisotropic (the data varies differently in different directions) as demonstrated in Figure 2.13. When investigating these directional trends, you will have to modify parameters such as the directions in which the variances will be calculated. These parameters are discussed in the following section.
Modifying directional parameters

Up to this point the directional calculation of variance has been discussed as being north-south and east-west. In reality, data will not have directional variations that are described in these exact directions. Therefore, you will need to create a model that looks in the direction in which the data is varying. This is done by modifying the number of different directions analyzed, the angle in which they are oriented, and the degree of tolerance that will be afforded to each direction.

37

Chapter 2 Grid Analysis User Guide

Sector A
N W S E

Sector B

Figure 2.14 A two-directional semivariance search showing direction/angle/tolerance.

In Figure 2.14, two directions are analyzed, represented by the dark and light gray pie shapes. It is important to note that although the diagram shows four pies, variance analysis is always performed in opposing directions. When more than one direction is set, the angle to which these sectors will be oriented must be specified. In the above diagram, the angles are zero degrees (Sector A) and 80 degrees (Sector B). It is unlikely to find data pairs at exactly 0 degrees or 80 degrees, thus you will need to define an interval around these exact values for which points will be considered. This interval is known as the tolerance. In the above diagram, the zero degree direction has a tolerance of 45 degrees and the 80 degree direction has a tolerance of 20 degrees.

Tuning the model

After you have generated the experimental semivariogram, you can calculate a model curve that closely fits the semivariogram.
Changing the semivariogram model

The semivariogram models included with Mentum Planet are Spherical, Exponential, Gaussian, Power, Hole Effect, Quadratic, and RQuadratic (Rational Quadratic). By applying one or more of these models to the different directional semivariograms, the model curve can be adjusted to better represent the variance in the data. After any of these models are applied, they can be further modified by changing the sill and range values. The range is the greatest distance over which the value at a point on the surface is related to the value at another point. Variance between points that are farther apart than the range does not increase appreciably. Therefore, the semivariance curve flattens out to a sill. Not all data sets exhibit this behavior. In Figure 2.15, the sill value is at variance of 12 200, and the range value occurs at a distance of 12 000.

38

Creating Grids Using Interpolation Grid Analysis User Guide Semivariogram view
15000 12500 10000 Variance 7500 5000

Sill

Range
2500 0

2500

5000

7500

10000 Distance

12500

15000

17500

Figure 2.15 The sill is a variance value that the model curve ideally approaches but does not cross. The range is the distance value at which the variogram model determines where the sill begins.

Anisotropic modeling

It is quite natural for the behavior of a data set to vary differently in one direction as compared to another. For example, a steeply sloping hill will typically vary in two directions. The first is up and down the hill where it varies sharply from the top to bottom, and the second is across the hill where it varies more gradually. When this occurs in a data set, it is called anisotropy. When performing anistropic modeling, you are essentially guiding the kriging interpolation to use sample data points that will most accurately reflect the behavior of the surface. This is achieved by creating additional models for each direction analyzed. When interpolating points are oriented in a northsouth direction, kriging weights can be influenced to use the parameters of one model while the points oriented in an east-west direction will be weighted using a different model.

39

Chapter 2 Grid Analysis User Guide Semivariogram view

15000 12500 10000 Variance 7500 5000 2500 0

Two Model Curves

2500

5000

7500

10000 Distance

12500

15000

17500

Figure 2.16 A two-directional, two-model semivariogram.

Custom Point Estimation

The Custom Point Estimation technique is similar to the IDW technique, in which grid values are calculated based upon the points found within a predefined search radius. The main difference between these two techniques is that using the Custom Point Estimation technique, you can choose from six different calculations to perform on data points.

Suggested reading on interpolation techniques

For further information that covers interpolation and contouring techniques, including an exhaustive reference list, see Watson, D.F. Contouring: A Guide to the Analysis and Display of Spatial Data. Elsevier Science Inc., Tarrytown, New York, NY, 1992. For information on the fifth-order bivariate interpolation applied to the TIN solution, see Akima, H. A Method for Bivariate Interpolation and Smooth Surface Fitting for Irregularly Distributed Data Points. ACM Transactions on Mathematical Software, v.4, no. 2, 1978: pp. 148-159.

40

Creating Grids Using Interpolation Grid Analysis User Guide

Natural neighbors
Gold, C.M. and Roos, T. Surface Modelling with Guaranteed Consistency An Object-Based Approach in Nievergelt, J., Roos, T., Schek, H.-J., and Widmayer, P. (eds.), IGIS 94, Proceedings of the International Workshop on Advanced Research in Geographic Information Systems, Lecture Notes in Computer Science 884; Springer-Verlag, 1994: pp. 7087.

Kriging
Olea, R., ed. Geostatistical Glossary and Multilingual Dictionary. Oxford University Press, New York, NY, 1991. Deutsch, Clayton V., and Journel, Andre G. GSLIB, Geostatistical Software Library and Users Guide. Oxford University Press, New York, NY, 1992. Isaaks, Edward H., and Srivastava, R.Monah. Applied Geostatistics. Oxford University Press, New York, Oxford, 1989. Goldberger, A. Best Linear Unbiased Prediction in the Generalized Linear Regression Model. JASA, 57, 1962: pp. 369-375.

41

Chapter 2 Grid Analysis User Guide

42

3.
This chapter contains the following sections:

Chapter 3: Creating Grids Using Spatial Models

Creating Grids Using Spatial Models

This chapter describes all of the basic commands associated with the creation of numeric grid files using the Location Profiler and Trade Area Analysis modeling techniques.

Model types Using the Location Profiler model Using the Trade Area Analysis models

43

Chapter 3 Grid Analysis User Guide

Model types
Modeling techniques create grids of derived values. The modeling wizard enables you to create grid files using two basic techniques.
Location Profiler The Location Profiler creates a model where bin values are calculated as the average distance to all points found within a user-defined radius

Trade Area Analysis - Single Site Trade Area Analysis enables you to generate trade areas around stores or other services based on the probability of an individual patronizing a particular location. The trade area for a single site is calculated using the distance customers live from the site and the ability of the site to attract them. Bin values represent the probability of a customer patronizing the site. Trade Area Analysis - Multiple Site Trade areas for multiple sites are calculated using the distance customers live from the site and the ability of the site to attract them. Bin values represent the probability of a customer patronizing any one of the sites in the selection.

Using the Location Profiler model

The Location Profiler computes and averages the distance to a series of points from anywhere within a map area. The algorithm generates a grid, where at each bin a value is calculated that represents the average distance to all point locations surrounding that bin and lying within a defined search radius. You are creating a geographic profile of an area that measures proximity to a series of sites using spatial relationships (spacing, distribution, and density). By identifying those locations on the map that are the shortest average distance to all or some of the sites, the grid file can be used to represent geographic centers of activity. The model can be further refined to take into account weighting factors that specify the relative influence of each site compared with those surrounding it.

44

Creating Grids Using Spatial Models Grid Analysis User Guide

Three settings are critical to making the Location Profiler a truly effective modeling tool:

the number of points to which the distance is computed from each grid node the relative influence or weighting of each point in the averaging calculation the distance decay function of the weighting factor

Setting the number of points

At each bin of the grid, the Location Profiler measures the distance to every point lying within the search area and calculates the average value. However, the pattern of geographic centers generated by the process will vary depending upon the number of points used for each bin calculation. The number of points is determined by defining a search area around each bin to select points, and defining the minimum and maximum number of points to be used for each calculation. Figure 3.1 and Figure 3.2 show the result of varying the number of points used by the Location Profiler for each bin calculation. By default, the algorithm sets the search radius and the maximum number of data points to include 100 percent of the points in the table. Using the default settings thereby creates a grid that defines the most central region of the point location database.

45

Chapter 3 Grid Analysis User Guide Figure 3.1 This illustration shows the geographic profile of a table of point locations (represented by black dots) calculated using 25 percent of the total number of points. Sites lying within the more central contour regions are closer to the surrounding points than sites lying within the outermost regions.

Reducing the number of points involved in each bin calculation, by decreasing both the search radius and the maximum number of data points, tends to create a more complex profile that highlights local zones of greater point density within the overall distribution of points. This may be useful if, for example, you want to highlight local areas of greater site density across a geographic area.

Figure 3.2 This illustration shows the geographic profile of the same table of locations calculated using only one percent of the total points. The highlighted local concentrations of points give an effective visual representation of variations in point density across the map area.

In addition to defining a search area that controls the number of sites used in the distance averaging calculation, you can also define an exclusion radius that is used to create a circular area immediately surrounding each bin within which site locations are excluded from the calculation.

Using point weighting

The Location Profiler enables the use of weighting values attached to points. If all data points are considered to be of equal importance, then their weighting values will all be equal (typically one). If some data points are considered to be more important, or perhaps more reliable, correspondingly higher values may be attributed to them. In real terms, the weighting factor

46

Creating Grids Using Spatial Models Grid Analysis User Guide

may represent a relative confidence value assigned to each point if, for example, each site represented a human observation. The factor may also be a measure of frequency if, for example, each site represented a freight companys customer and showed the number of deliveries per month. In each case, the distance calculation is treated as a multiple of the weighting factor. For example, if the weight represented frequency of deliveries, then each distance calculation would be handled as a multiple measurement based upon the frequency value.

Using distance decay functions

A distance decay function can be applied to all points in the location table assuming that a weighting factor has also been assigned. Distance decay can be used in the Location Profiler model if it is assumed that the data points close to the grid point are more relevant than those lying farther from the grid point. In other words, you must decide if it is reasonable to expect that, as the distance to the data points approaches the outer edge of the search radius, the contribution of these values approaches zero. A decay value of one is suggested for data points close to the grid node, and it falls to zero for data points closer to the edge of the search radius. A distribution of this type typically has an inflection point somewhere along its length. An inflection point is the point along the distribution where the slope either stops increasing in value or stops decreasing in value. For typical distance decay curves, as shown in Figure 3.3, the slope is zero where x = 0 and gradually decreases, becoming more negative, until the inflection point is reached. At this point, the slope gradually increases, becoming more positive and returning to zero again as x approaches one.
Grid Node 1.0 0.8 Decay Factor (Y) 0.6 0.4 0.2 0 0 0.2 0.4 0.6 0.8 Distance Ratio (X) 1.0 Search Radius

Curve Parameters
Cruve No 1 2 3 4 5 6 Inflection Point X .99 .9 .7 .5 .3 .1 Y .01 .2 3 5 6 8 Slope -99 -5 -3 -2 -3 -5

47

Chapter 3 Grid Analysis User Guide Figure 3.3 Shown here are six search radius distance decay curves supported by the Location Profiler. Values along the x-axis represent a ratio obtained by dividing the distance between the grid point and data point by the search radius. Curve 3 represents the most typical search radius decay function that can be applied to the widest variety of modeling parameters.

A similar decay function can be applied to the exclusion zone of the bin but using a curve with a positive slope. In this case, the slope is zero where x = 0 and gradually increases, becoming more positive, until the inflection point is reached and then gradually decreases and returns to zero as x approaches one. Various methods can be used to obtain the distributions shown in Figure 3.3 and Figure 3.4. Most techniques require that you specify the location (x- and y- coordinates) of the inflection point and the slope at this inflection point. Given an appropriate distance decay curve, a decay factor can therefore be determined for any distance value measured between the bin and a weighted point location. When x = 0 (at the bin assuming no exclusion radius setting), then y = 1 (no distance decay factor), and the slope is 0. When x = 1 (at the outer edge of the search radius), then y = 0 (100 percent decay), and again the slope is 0.
Grid Node 1.0 0.8 Decay Factor (Y) 0.6 0.4 0.2 0 0 0.2 0.4 0.6 0.8 Distance Ratio (X) 1.0 Exclusion Radius

Curve Parameters
Cruve No 1 2 3 4 5 6 Inflection Point X .01 .05 .05 1 .3 5 Y .01 .4 .1 2 .4 5 Slope 99 20 10 5 3 2

Figure 3.4 Shown here are six exclusion radius distance decay curves supported by the Location Profiler. Values along the x-axis represent a ratio obtained by dividing the distance between the grid point and data point by the exclusion radius. Curve 3 represents the most typical exclusion radius decay function that can be applied to the widest variety of modeling parameters.

The y-axis indicates the decay factor associated with the data point. This decay factor, which is always between zero and one, is then multiplied by the weighting factor associated with the data point to obtain the weighted value. This weighted value is then multiplied by the measured distance between the

48

Creating Grids Using Spatial Models Grid Analysis User Guide

grid point and data point, to obtain a weighted distance. The weighted average distance for the grid point is obtained by summing all the weighted distances for each data point, then divided by the sum of all the weighted values of the data points.
Grid Node 1.0 0.8 Decay Factor (Y) 0.6 0.4 0.2 0 0 0.2 0.4 0.6 0.8 Distance Ratio (X) 1.0 Exclusion Radius Search Radius

Curve Parameters
Cruve No 1 2 3 Search Exclusion Radius Radius Cruve No Cruve No 1 2 3 1 2 3

Figure 3.5 This figure shows the blending relationship between distance decay functions relating to the exclusion zone and the search zone. A continuous function can be created by choosing complementary curves that will distance weight all point locations lying between the grid node and the search radius.

To create a grid using the Location Profiler model

1 2 3 4 Choose GIS Grid Analysis Create Grid Modeling. From the Select Modeling Method dialog box, choose the Location Profiler option. Click Next. In the Select Table and Column dialog box, click Open Table to add a table to the Table to Grid list if the table of points you want to use is not listed. From the Select Table to Grid list, choose the appropriate table of points that contains the geographic location of each point. From the Weight Column list, choose the column that contains the weighting value for each site record. If no weighting value has been assigned to the point records, choose the No Weight entry from the list. Weighting factors are values that specify the relative influence of each site compared with those surrounding it.

5 6

49

Chapter 3 Grid Analysis User Guide

To use an unmapped data file (an x, y, z file) that has not been converted to a vector point table (using the Create Points command), do the following:

From the X Column list, choose the column containing the x-coordinates for each point. From the Y Column list, choose the column containing the y-coordinates for each point. Click Projection and choose the coordinate system of the location data.

8 9

In the Data Description box, type an annotation (maximum 31 characters). This will be carried as a header in the grid file. Click Next, enter the required information in the dialog box, then click Finish. Once the grid is created, it appears in a Map window with a default color palette applied. You can change the color range assigned to the grid file. For more information, see To use the Grid Color tool on page 80.

Using the Trade Area Analysis models

The Trade Area Analysis models enable you to generate trade areas around stores or other services based on the probability of an individual patronizing a particular location. It is possible to identify market islands in a network of competing stores using criteria such as the attractiveness of a store and the distance one must travel to get to the store. Store attractiveness can be defined by parameters such as total floor space or shelf space, number of parking spaces, age of the store, or any combination of elements that defines its appeal. Using commercially available wealth and market profile information for potential customers located within each trade area, it is possible to estimate revenue for store locations and model the influence of competing stores or the effect of adding or removing stores.

Trade areas
Trade areas can be defined by simple circular boundaries around a store location. These trade areas are easy to visualize and construct and provide a standard for comparing stores. Trade areas determined in this way do not, however, account for the existence of competing stores and assume that the store has a monopoly over customers within the area. There is the common sense concept that, all other things being equal, a person is likely to shop at a

50

Creating Grids Using Spatial Models Grid Analysis User Guide

closer store rather than a more distant one. This concept of likelihood forms the basis for defining probabilistic trade areas. Defining the likelihood or probability that a given customer will patronize a certain store requires the use of a spatial interaction model. Examining movement over space for activities such as shopping trips, commuting to work, or migration patterns, became popular with the use of gravity models in the 1960s and 1970s. Gravity models are based on an analogy with Newtons theory of gravitational attraction. The degree of attraction between two objects is based on the size of the objects and the distance between them. Objects that are closer together will exert a greater pull on each other compared with objects that are farther away. Larger objects have a greater gravitational pull than smaller objects. Refinements to the gravity model were made by Professor David Huff. The Huff model is still one of the most popular models for predicting retail customer behavior. It enables the mapping of the neighborhoods from which each store derives its patronage. The answer is not a single circle or polygon but a probability surface or grid. This probability grid can be contoured to produce regions of patronage probability. The key feature implicit in the probability surface is that it accounts for overlapping trade areas. Using a Huff model, you can calculate the probable trade area of a single store (Figure 3.6) or compute the patronage probability values for every store and extract the maximum probability value at each grid location (Figure 3.7). You can then use this information to determine areas where people are least serviced or areas where there is great competition for a customers business, i.e., there is no preferred store location clearly winning a customers patronage.

51

Chapter 3 Grid Analysis User Guide

Figure 3.6 This map displays the probability that a consumer will patronize Store A. The influence of all stores is taken into account in this model so that the locations of stores surrounding Store A influence the shape of the contours. Patronage probability decreases as the distance from the store increases.

Figure 3.7 This map displays the maximum probability at every potential customer location that a consumer will patronize any one store when presented with all stores.

Retailers can modify the Huff equation parameters by defining the attractiveness of each store competing for a customers patronage and determining the rate at which the pulling power of a store drops off or decays with distance from the store. This produces a map estimating the probability that a potential customer (Customer) residing at a given grid location will patronize a particular store at a particular location (ReferenceStore) given a network of competing stores (Store1 to StoreN where N = the total number of

52

Creating Grids Using Spatial Models Grid Analysis User Guide

stores including the Reference Store). This relationship is presented in the following equation.

Equation 3.1 Customer probability Where

= attractiveness coefficient estimated from empirical observations = distance decay coefficient estimated from empirical observations

This formula ensures that all the probabilities are between zero and one and sum to one when considering all the shopping destinations. The and coefficients allow you to account for non-linear behavior when modeling the effects of store attractiveness or distance. In terms of a supply and demand model, the resulting probability surface spatially represents the supply side of the equation. It does not directly take into account any underlying demand data such as consumer households or census tract information. To process the demand data so that revenue calculations can be made, it is necessary to update the probability regions (produced by contouring the probability grid) with any underlying income information you have.
Instead of updating the probability regions with overlying demographic (population/wealth) point data, update the demographic data table with values from the probability grids. Open the grid files and use the Point Inspection command from the Analysis menu in the Grid Manager to add the grid values to the demographic data table.

To create a grid using a trade area analysis model

1 Choose GIS Grid Analysis Create Grid Modeling. The Select Modeling Method dialog box opens.

53

Chapter 3 Grid Analysis User Guide

Choose Trade Area Analysis and then choose either of the following options:

Single Site Multiple Sites

3 4

Click Next. In the Select Table and Column dialog box, click Open Table to add a table to the Table to Grid list if the table of points you want to use is not listed. From the Select Table to Grid list, choose the appropriate table of points that contains the geographic location of each store/site. From the Attractiveness Column list, choose the column that contains the attractiveness value for each site record. Store/site attractiveness is a relative number that can be defined by parameters such as total floor space or shelf space, number of parking spaces, age of the store, or any combination of elements which defines the appeal of an individual retail site. To use an unmapped data file (an x, y, z file) that has not been converted to a vector point table (using the Create Points command), do the following:

5 6

From the X Column list, choose the column containing the xcoordinates for each point. From the Y Column list, choose the column containing the ycoordinates for each point. Click the Projection button and choose the coordinate system of the location data.

8 9

In the Data Description box, type an annotation (maximum 31 characters). This will be carried as a header in the grid file. Enable the Ignore Records Containing Zero check box if you want to include only non-zero records.

10 Click Next and enter the required information in the Huff Model dialog box. For more information on the fields in the Huff Model dialog box, press the F1 key. 11 Click Finish.

54

Creating Grids Using Spatial Models Grid Analysis User Guide

12 If you chose the Single Site option in Step 2, in the Mentum Planet dialog box, click OK, and use the Grid Analysis Select Tool on the Grid Analysis toolbar to select a site in the Map window. The grid opens in a Map window with a default color palette applied. You can change the color range assigned to the grid file. For more information, see To use the Grid Color tool on page 80.

Suggested reading on trade area analysis

For a review covering the basics of probabilistic trade area analysis, see Dudley, G. Probability Primer: Spatial Insights into StoreBase Trading Areas. F.Y.I. Vol. 4, #2, August. A Publication of Blackburn/Polk Marketing Services Inc., Toronto, 1996. For further readings on the topic of spatial interaction, see Fotheringham, S. and O'Kelly, M. Spatial Interaction Models: Formulations and Applications. Kluwer Academic Publishers, 1989. Taylor, P. J. Distance Decay Models in Spatial Interactions. Concepts and Techniques in Modern geography. Norwich: Geo Books, 1975.

55

Chapter 3 Grid Analysis User Guide

56

4.
This chapter contains the following sections:

Chapter 4: Creating Grids Using Other Methods

Creating Grids Using Other Methods

Mentum Planet comes with a number of data preparation and grid creation methods that allow you to build grid files from data files and tables of regions, polylines, or points. You can also import a grid from several formats.

Introduction Creating point density grids Converting regions to a grid Buffering map objects to create a grid Preparing data using Poly-toPoint Importing grids

57

Chapter 4 Grid Analysis User Guide

Introduction
There are a number of ways you can create a grid other than by using the Interpolation or Modeling Wizards.

Create point density gridsUsing the Point Density command from the GIS Grid Analysis Create Grid menu, you can create a point density grid representing the density of events in an area. You can choose Point Density using Square Area or Point Density using Kernel Smoothing. For information on creating point density grids, see Creating point density grids on page 58. Convert regions to a gridUsing the Region to Grid command from the GIS Grid Analysis Create Grid menu, you can convert a table of regions to a grid. For information on converting regions, see Converting regions to a grid on page 60. Buffer map objects to create a gridUsing the Grid Buffer command from the GIS Grid Analysis Create Grid menu, you can create a buffer around a line, polyline, point, or polygon. For information on grid buffering, see Buffering map objects to create a grid on page 61. Use the Poly-to-Point toolThe Poly-to-Point tool does not create a grid, but it can be used to prepare line data for grid creation by extracting points from existing tables of polylines or regions. For information on the Poly-to-Point tool, see Preparing data using Poly-to-Point on page 64. Create grids by analyzing other gridsEach of the tools used to analyze an existing grid creates a new grid with the results of the analysis. For more information on analyzing grids, see Chapter 7: Using Grids for Spatial Analysis on page 113.

Creating point density grids

Point density techniques calculate the density of events in an area. Mentum Planet supports two techniques for calculating point density: Square Area and Kernel Smoothing. These options are available by choosing GIS Grid Analysis Create Grid Point Density.

58

Creating Grids Using Other Methods Grid Analysis User Guide

Calculating point density using square area

This method of calculating point density results in a grid where each bin value represents the number of events within a user-defined area around the bin. For example, if the specified area is one square kilometer, then each bin will have the number of events within one square kilometer around the bin. Two methods of calculating or summing the number of events are provided. The first method counts the number of symbol objects found within the user-defined area. This method assumes that each symbol represents a single event. The second method is more sophisticated in that it enables a single symbol to represent more than one event. It does this by accessing a column containing the number of events at each location in the database of the point file.

To create a square area point density grid

1 2 3 Choose GIS Grid Analysis Create Grid Point Density Square Area. From the Select Table to Grid list, choose the table from which you want to create a grid. In the Select Event Sum Column list, do one of the following:

If you want to count the points found inside the search area, choose Count Points Only. If you want to sum the values contained in a given field in the database, choose the name of the column in the database that contains the values.

4 5 6

Click Next. In the Point Density dialog box, enter the bin size and search area. Click Finish.

Calculating point density using kernel smoothing

The Kernel Smoothing method enables you to take a set of discrete points, such as specified locations of crime or disease, and transform them into a continuous surface. These discrete points are known as events. By studying the events occurring within a specified radius, you can determine the probability of the event occurring in adjacent regions. You can use two different variations of the Kernel Smoothing technique: weighted or classic. In the weighted variation, you can apply a weight to a point in order to reflect the occurrence of multiple events at a location. In the

59

Chapter 4 Grid Analysis User Guide

classic variation, all points are seen as single events and have the same weight.

To create a kernel smoothing point density grid

1 2 3 Choose GIS Grid Analysis Create Grid Point Density Kernel Smoothing. From the Select Table to Grid list, choose the table from which you want to create a grid. In the Select Event Sum Column list, do one of the following:

If you want to count the points found inside the search area, choose Count Points Only. If you want to sum the values contained in a given field in the database, choose the name of the column in the database that contains the values.

4 5 6

Click Next. In the Kernel Smoothing dialog box, enter the bin size and search area. Click Finish.

Converting regions to a grid

The region-to-grid conversion process involves extracting a text or numeric value from a column in the region table and assigning this value to all the bins that fall inside that region. If the assigned value taken from the table is a text string, the process automatically creates a classified grid file. If the assigned value is numeric, it generates a numeric grid file. In both cases, the resulting grid appears similar to the original region map. However, it often tends to look somewhat notchy or pixelated. For classified grids, the process assigns color to a bin when it extracts an attribute from the region table. Once you have converted a region table to a grid, you can analyze the data using processes that cannot be applied to standard vector regions. For example, using the Grid Query command accessible from the Analysis button in the Grid Manager, classified geographic information converted to grid format can be overlain with numeric grid data and queried using standard Boolean operators to create a new derivative grid map that meets the conditions of the query statement. Performing powerful spatial analysis routines using any type of geographic data is the single greatest advantage afforded by raster GIS.

60

Creating Grids Using Other Methods Grid Analysis User Guide

To convert a table of regions to a grid

1 2 3 Choose GIS Grid Analysis Create Grid Region to Grid. In the Region to Grid dialog box, from the Source Table list, choose the region file that contains the information to be converted. From the Column list, choose a column that contains the attribute information to be assigned to the bins. If you choose a column containing numeric data, you generate a numeric grid. If you choose a column containing text data, you generate a classified grid. 4 In the Bin Size box, type the bin size for the new grid. The conversion process builds a new grid file that, by default, covers the entire geographic extent of the region table. The grid file size and the grid dimensions will therefore increase as the bin size decreases. 5 6 7 In the Data Description box, assign an annotation (maximum 31 characters) that will be carried as a header in the grid file. In the File Name box, type a new file name or accept the default. Click OK. Once the grid file is created, it appears in a Map window with the same colors assigned to each bin, if any, that were assigned to the regions. You can change the colors assigned to the grid using the Color tool accessible from the Grid Manager. For more information, see To use the Grid Color tool on page 80 or Using color in classified grids on page 82.

Buffering map objects to create a grid

In a vector GIS such as MapInfo Professional, you can create regions around map objects at a predefined distance. You cannot, however, determine distances within a buffer region. For example, if a highway is buffered at a distance of one kilometer, you will not be able to determine the precise distances that fall within the buffered area (e.g., 750 m or 300 m). In Mentum Planet, the value of each bin in a buffer grid is calculated as the distance to the closest input object. As a result, you can determine precise distances within the buffered region. Input objects supported by this tool are line, polyline, point, and polygon objects. Polygon objects are treated as polylines, where the buffers are created

61

Chapter 4 Grid Analysis User Guide

using the objects boundary, and values are calculated on either side of this boundary (inside and outside of the polygon).

Figure 4.1 The Grid Buffer tool is used here to build a buffer grid of continuous distance from a road network.

You can choose one of two options in the Grid Buffer tool: Use Nodes Only and Use Maximum Distance. As shown in Figure 4.2, when you choose the Use Nodes Only option, buffers are created only at the junction of the lines segments but at no other places.

Figure 4.2 This graph shows the result when the Use Nodes Only option is chosen.

If you choose the Use Maximum Distance setting, you must specify the maximum distance between locations where a circular buffer will appear. If the maximum distance setting is too great, as shown in Figure 4.3, there will not be enough circular buffers to cover the length of the line.

62

Creating Grids Using Other Methods Grid Analysis User Guide

Figure 4.3 This graph shows the result when the Use Maximum Distance option is chosen and the distance setting is too big.

By reducing the distance setting, the buffers will cover the whole line. This is illustrated in Figure 4.4. Generally, the distance setting should be at least half of the buffer distance. For example, if you create buffers at 1000 m, the maximum distance setting should be at least 500 m.

Figure 4.4 This graph shows that the buffer covers the whole line when the maximum distance is reduced properly.

To create a grid using the Grid Buffer tool

1 2 Choose GIS Grid Analysis Create Grid Grid Buffer. From the Source Table list, choose a table for processing.

63

Chapter 4 Grid Analysis User Guide

In the Distance Between Points section, choose either the Use Nodes Only option or the Use Maximum Distance option and define the distance value between points in the Distance box. The maximum distance between points should always be equal to or less than the search radius.

4 5 6 7 8

Click Next. Define Bin Size and Buffer Distance settings. In the Enter Data Description box, assign an annotation (maximum 31 characters) that will be carried as a header in the grid file. In the Output Grid Name box, type a new file name or accept the default. Click Finish.

Preparing data using Poly-to-Point

Using Poly-to-Point, you can extract points from existing tables of lines, polylines, or regions to prepare data for creating grids.

To create a point table using Poly-to-Point

1 2 3 Choose GIS Grid Analysis Create Grid Poly-to-Point. From the Select Table list, choose a table for processing. In the Extract From section, enable any or all of the Include Existing Points, Lines, Polylines, and Regions check boxes to determine the type of objects in the table from which points are to be extracted. In the Distance Between Points section, choose either the Use Nodes Only option or the Set Maximum option and define the distance value between points in the Distance box. The maximum distance between points should always be equal to or smaller than the search radius. 5 6 In the File Name box, type a new file name or accept the default. Click OK. Once the point table is created, an information dialog box appears with a summary of the number of objects (points, polylines, or regions) processed and the total number of points created.

64

Creating Grids Using Other Methods Grid Analysis User Guide

Importing grids
Due to the ever-increasing usage of binary and ASCII-based grid files in desktop mapping environments, Mentum Planet imports a number of formats for the more commonly used grid types. Most of these grids represent the elevation of a land surface, commonly referred to as Digital Terrain Models (DTMs) or Digital Elevation Models (DEMs). Mentum Planet requires DEM data as numerical grid (.grd) files and clutter data as classified grid (.grc) files. If the data is in some other format, you must import it, using the import capabilities of Mentum Planet. A default file name will be created for each import file. If this file name conflicts with an existing one, you are prompted to supply a new one. It is impossible to overwrite existing files while importing grids.

Grid formats that you can import

The Grid Import utility can auto-detect and convert the following formats:

ASCII Grid ASCII Classified Grid CRC-500 Geological Survey of Canada Grid DTED GeoSoft

MONA UK Ordnance Survey Grid USGS DEM SDTS - Spatial Data Transfer Standard Grid MIG Grid

ASCII Grid

ASCII Grid is a simple text-based format consisting of five lines of header information followed by rows of space-delimited float point values. Due to its simple and relatively compact nature, you should consider writing your data in this format when you want to import your own grid data into Mentum Planet. When this file format is detected, a dialog box is displayed that enables you to set the projection, z-unit type, and a number used as a special flag indicating null value grid cells. The default null value is -9999; however, if the data includes -9999 as a valid numeric value, a new null can be assigned to bins with no value.

65

Chapter 4 Grid Analysis User Guide

Number of colums and rows

ASCII Grid Format

ncols 10 nrows 10 xllcenter 1520000.0 yllcenter 6490000.0 celsize 50.0 NODATA_value -9999 42.7 41.4 39.7 41.3 38.3 38.6 39.9 38.1 37.8 37.2 41.4 39.7 41.3 38.3 38.6 39.9 38.1 37.8 37.2 36.6 41.3 38.3 42.7 41.4 39.7 38.6 37.8 37.2 39.9 38.1 39.9 38.1 37.8 37.2 42.7 41.4 39.7 41.3 38.3 38.6 42.7 38.6 39.9 38.1 41.4 39.7 41.3 38.3 37.8 37.2 42.7 41.4 39.7 41.3 38.3 38.6 39.9 38.1 37.8 37.2 41.4 39.7 41.3 38.3 38.6 39.9 38.1 37.8 37.2 36.6 41.3 38.3 42.7 41.4 39.7 38.6 37.8 37.2 39.9 38.1 39.9 38.1 37.8 37.2 42.7 41.4 39.7 41.3 38.3 38.6 42.7 38.6 39.9 38.1 41.4 39.7 42.3 38.3 37.8 37.2

X end Y position of the center of the lower left node Cell spacing Null value Cell values

Figure 4.5 Example of a grid file illustrating the construction of an ASCII grid file.

ASCII Classified Grid

ASCII Classified Grid is a simple text-based format consisting of general grid information in the file header followed by rows of space-delimited integer values. The header of the file contains a lookup table of class values, which define the integer representing each class name. The value 0 is used to define the null bin (a bin that is not defined in the lookup table). The lookup table should always begin sequentially with the value 1. The maximum number of classes included in this format is 65 535.
Number of colums and rows X and Y position of the center of the lower left node Cell spacing Number of different classes Class name and RGB value Cell values ASCII Classified Grid Format
ncols 9 nrows 5 xllcenter 1520000.0 yllcenter 6490000.0 celsize 50.0 nClasses 4 1 "Forest" 255 255 0 2 "Open" 0 255 0 3 "Urban" 255 0 127 4 "Water" 0 0 255 111144111 121114113 122111431 111214331 111104111

Figure 4.6 Example of a simple classified import file.

66

Creating Grids Using Other Methods Grid Analysis User Guide

CRC-500

CRC-500 is a binary DEM for RF propagation modeling supported by Canadas Communications Research Centre. The bin resolution is 500 meters in a UTM projection using the NAD 27 data for Canada.
DTED

DTED is a binary DEM originally supported by the U.S. Defence Mapping Agency (now NIMA). The Grid Import utility recognizes both Level 1 (3 arcsecond) and Level 2 (1 arc-second) in the longitude/latitude coordinate system.
Geological Survey of Canada Grid

Geological Survey of Canada Grid is a text-based, delimited format grid file supported by the Survey for a number of geophysical data themes at a variety of bin resolutions and projections.
GeoSoft

GeoSoft is a proprietary binary grid format supported by the GeoSoft Inc. suite of industry-standard geochemical and geophysical data interpretation applications. The format is extremely flexible but does not support any projection information in the header. The Grid Import utility prompts you to enter a recognized coordinate system class. The Grid Import utility supports only uncompressed GeoSoft grids.
MONA

MONA is an x, y, z text-based, comma-delimited format that contains no header information. The data must be regularly spaced and sorted by row. The import projection is set to French Lambert II Carto. The format is supported by the French government for European DEM coverage.
UK Ordnance Survey Grid

UK Ordnance Survey Grid is an ASCII non-delimited format supported by the Ordnance Survey with DEM coverage throughout the United Kingdom. It comprises a diverse series of format variations in which Mentum Planet handles the grid-based DEM Panorama and Profile products. A variety of resolutions are supported.
USGS DEM

USGS DEM is an ASCII format DEM supported by USGS with coverage throughout the entire United States. The Grid Import utility automatically recognizes a variety of DEM coverages including the 7.5 minute, 30 arcsecond, and 3 arc-second longitude/latitude grids, as well as UTM-projected orthogonal 30-meter grids.

67

Chapter 4 Grid Analysis User Guide

SDTS

SDTS is designed to store several different forms of spatial data such as vector and raster information. The Grid Import utility extracts only the grid information from the file.
MIG

MIG is the native data format of MapInfo Corporation.

To import a grid
1 2 3 4 5 If the Grid Manager is not visible, choose View Show Grid Manager. In the Grid Manager, click the Tools button, and choose Import. In the Grid Import dialog box, click Add. Locate the grid you want to import, and click Open. Do any of the following. The dialog box that opens depends on the type of grid you are importing:

In the Choose Projection dialog box, choose the projection category and the category member. In the Grid Import Null Value dialog box, specify the null value you want to display for points with no associated data. In the Grid Import Z Unit dialog box, choose the unit of measure, or choose User Defined, and enter a unit of measure in the Enter User Defined Type box.

6 7 8 9

If you want to apply your setting to all grids of the same type, enable the Use For All Imports of Similar Type check box. Click OK. If you want to open each of the grid files in a Map window, in the Grid Import dialog box, enable the Open All Grids After Import check box. Click OK.

68

5.
This chapter contains the following sections:

Chapter 5: Working with the Grid Manager

Working with the Grid Manager

Gridding is the basic mapping technique used in Mentum Planet. The Grid Manager is the central dialog box from which you can open, sort, view, and manipulate grids. Mentum Planet also includes several tools that enable you to extract information from grids, create legends for grids, and create contour grids.

Working with the Grid Manager Using the Grid Manager Info function Using the Grid Info tool Using the Region Info tool Using the Line Info tool Using color in numeric grids Using color in classified grids Creating and displaying legends Modifying information about a grid Changing grid file projections Updating the projection file Reclassifying grids Converting grids Trimming grids Splicing grids Resizing grids Exporting grids

69

Chapter 5 Grid Analysis User Guide

Working with the Grid Manager

The Grid Manager enables you to obtain information about grid files, access the Grid Color tool or the Dictionary Editor, run the 3D Viewer, create a 3D drape file, create region and polyline contours, and analyze or modify your grid files.

To display the Grid Manager

1 Choose View Show Grid Manager. The Grid Manager opens. Open grids are listed in the window at the top of the Grid Manager. For information on the elements of the Grid Manager, see Table 5.1.

To view detailed information on the selected grid file, click the Info button.

70

Working with the Grid Manager Grid Analysis User Guide

The Grid Manager expands to display five tabs containing information on the grid file.

Table 5.1 Grid Manager functions Item Description Three check boxes are displayed next to the grid name. The check box at the left is the Active check box. By enabling this check, you can make a grid active and query it using the Grid Info tool. The check box in the middle is the Color Gradient check box. By enabling this check box, you can turn off the color profile for a grid and display it in gray-scale if relief shading has been applied. The check box at the right, is the Relief Shading check box. By enabling this check box, you can turn on the relief shading function if it has been previously generated for the numeric grid using the Grid Color tool. The Open Grid and Close Grid buttons enable you to open or close grid files. The Copy Grid As and Delete buttons enable you to copy or delete grid files. The Grid Info button expands the Grid Manager to provide information on a selected grid in terms of geographic characteristics, z-units, meta data, legend, and histogram.

71

Chapter 5 Grid Analysis User Guide Table 5.1 Grid Manager functions (continued) Item Description The Grid Color Tool button gives you access to the Grid Color Tool dialog box for numeric grids (e.g., elevation .grd files) and the Dictionary Editor dialog box for classified grids (e.g., clutter .grc files). Using the functionality offered in these dialog boxes, you can modify the color profiles associated with grids. The Grid View button enables you to make 3D drape files or to run the 3D Viewer. For more information, see Chapter 10: Creating 3D Views Using GridView on page 141. The Contour Grid button enables you to generate region and polyline contours for numeric grids. For classified grids, it creates region contours only. The Analyze Grid button enables you to analyze grid files using the Calculator or the Grid Query command, to create slope and aspect grids, or to create a cross section. You can also perform a point, line, or region inspection, calculate point-to-point intervisibility, and perform a viewshed analysis. For more information, see Chapter 7: Using Grids for Spatial Analysis on page 113. The Grid Tools button enables you to alter meta data, filter classified grid files, export grid files, create grid legends, and reclass or reproject files. You can also resize, splice, and trim grid files and launch the Classified Grid Filter tool.

Using the Grid Manager Info function

When you click the Grid Info button, the Grid Manager expands to provide detailed information on each selected grid file. Five tabs outline information such as:

the geographic characteristics of the grid file the z-unit characteristics of the grid file lineage and history (meta data) a data legend a grid histogram for both numeric and classified grids

The Info tab

The Info tab summarizes the spatial characteristics of the selected grid including bin size, geographic unit of measure, grid dimension (in bins), file size, x and y extent of the grid, light source settings that define the relief

72

Working with the Grid Manager Grid Analysis User Guide

shading of the grid (if applicable), and the name of the coordinate system to which the grid is referenced. Light source information does not apply to classified grids.

Figure 5.1 The Info tab

The Z-Units tab

The Z-units tab provides a summary of the grid data characteristics including the data value range, the data type, and data description. The data type and description are entered when the grids are built and can also be changed using the Alter Meta Data command. The Alter Meta Data command is accessible from the Tools menu in the Grid Manager If the Use Closest Node Value check box is enabled, the value of the closest node is returned whenever the grid is queried. Consequently, the value of any location within a bin will be the same. If the check box is cleared, the values of adjacent bins are used to estimate the value of locations inside a bin, and

73

Chapter 5 Grid Analysis User Guide

values of individual locations within the bin will be different. This setting affects results obtained when you are:

performing grid calculations querying grids analyzing point-to-point visibility using the Cross Section tool doing a point or line inspection using the Grid Info or Line Info tools

If you want to create a color profile with hard breaks between colors (as opposed to interpolated fading between colors), enable the Use Solid Band Colors check box.

Figure 5.2 The Z-Units tab

The Meta Data tab

The Meta Data tab provides access to historical information relating to the creation of the grid and is carried as a header string within the grid file. It includes important information such as the name of the original data file from which the grid was derived, the method of grid interpolation used to create it, including the interpolation settings, a date and time stamp of the original build, and the identity of the builder (if available). Additional information can be appended using the Alter Meta Data command accessible from the Tools menu in the Grid Manager.
If only one grid has been opened, you may have to click on that grid in the Grid Manager to view the meta data.

74

Working with the Grid Manager Grid Analysis User Guide

Figure 5.3 The Meta Data tab

The Legend tab

The Legend tab displays the color inflection values that define the gradient color scheme of the grid. These are the same values that are assigned in the Grid Color Tool and appear as defaults in the Legend Generator. If you query a classified grid file, the legend window displays the default color values applied to each class found in the grid.

Figure 5.4 Legend for a numeric grid.

Figure 5.5

Legend for a classified grid.

The Histogram tab

The Histogram tab displays a histogram of the grid data in a graph window. The histogram provides you with a visual display of the distribution of your

75

Chapter 5 Grid Analysis User Guide

data so that you can make better decisions for grid display and grid analysis operations. This window can also display classified grid data by class name, class count, percentage, and area. You can customize the display of the histogram by right-clicking on it.

Figure 5.6 Histogram for a numeric grid

Figure 5.7 Histogram for a classified grid

Using the Grid Info tool

The Grid Info tool enables you to query and return grid values from selected locations in the grid file and view the results in a Grid Info dialog box. If more than one grid is open and active, values will be returned from each one. Only grid information from those grids that are active in the Grid Manager and are in the same geographic location as the point being inspected are displayed. The grid files being inspected do not have to be layers in the current Map window. The Grid Info tool simply examines the x- and ycoordinates of the location being inspected and returns the information from those active grids in the Grid Manager that are geographically coincident. If you need to inspect a large number of predefined points, you can use the Point Inspection function. For more information, see Using the Point Inspection function on page 119.

76

Working with the Grid Manager Grid Analysis User Guide

To use the Grid Info tool

1 2 3 If the Analysis toolbar is not visible, choose View Toolbars, enable the Show check box for the Analysis toolbar, and click OK. Click the Grid Info button. Click in the Map window.

You can choose whether or not values are interpolated within bins by enabling or clearing the Use Closest Node Value check box on the z-units tab in the Grid Manager. For more information, see Working with the Grid Manager on page 70. You cannot use the Capture Data and the Cursor Tracking options at the same time.

Using the Region Info tool

The Region Info tool displays a statistical summary of the data within a selected region. The values displayed are for the grid highlighted in the Grid Manager. To display statistical information for a large number of regions contained in the same table, use the Region Inspection function. For more information about region inspection, see To perform region inspection on page 122.

To use the Region Info tool

1 2 3 If the Grid Analysis toolbar is not visible, choose View Toolbars, enable the Show check box for the Grid Analysis toolbar, and click OK. On the Grid Analysis toolbar, click the Region Info button. Select a region in the Map window.

The Volume calculation is the sum of all the values found in the region multiplied by the bin dimensions. This value is only meaningful when the z-unit is a linear measurement. The % null value indicates how much of the enclosed area contains null values.

77

Chapter 5 Grid Analysis User Guide

Using the Line Info tool

The Line Info tool displays a statistical summary of the data along a selected line. The values displayed are for the grid highlighted in the Grid Manager. To display information for a large number of lines contained in the same table, use the Line inspection tool. For more information about line inspection, see To perform line inspection on page 121.

To use the Line Info tool

1 2 3 If the Grid Analysis toolbar is not visible, choose View Toolbars, enable the Show check box for the Grid Analysis toolbar, and click OK. On the Grid Analysis toolbar, click the Line Info button. Click a line in the Map window.

The number of samples taken along the line is determined by the Cross Section: No of Samples parameter in the Preferences settings. The default is 100. To change this value, choose the GIS Grid Analysis Preferences command and enter a new number in the No. of Samples box. For more information about preferences, see Setting your preferences on page 18. You can choose whether or not values are interpolated within bins by enabling or clearing the Use Closest Node Values check box on the z-units tab in the Grid Manager. For more information, see Working with the Grid Manager on page 70.

Using color in numeric grids

The use of color is an effective way to give meaning to large stores of data. In Mentum Planet, color is assigned to numeric grid files according to ranges of grid values. For example, assume that a grid consists of values from zero to 100. Discrete color gradients are then assigned by range according to defined inflection points. The range from zero to 25 would be assigned colors varying from dark blue for the lowest value to medium blue for the highest value, while the range from 26 to 50 would be assigned colors varying from medium blue for the lowest value through green to yellow for the highest value in the range, and so forth. This technique is a standard method of presenting quantitative data in a continuous gridded surface using a gradient color model, where blue is commonly used to represent low values, and red is commonly used to represent high values.

78

Working with the Grid Manager Grid Analysis User Guide

When a grid file is opened, the color information in the header is read, and the color profile is assigned to the bins and displayed in the Map window. In this manner, the image displayed on-screen is a true representation of the grid file, where each bin of the image represents a bin centered on the grid node as illustrated in Figure 5.8.

Figure 5.8 Each bin is assigned a color value as defined in the header portion of the file.

Any numeric grid file generated in Mentum Planet is automatically assigned a default color palette that is applied when the file is first opened. The color inflection points are carried in the header portion of the grid file and can be easily modified using the Grid Color Tool. The settings available in the Grid Color Tool allow precise control of gradient color across the range of values found in the grid file. A histogram showing the distribution of values in the grid file assists you in the placement of color inflection points. You can define up to 64 inflection points. In addition to assigning color gradients to numeric grids, you can also apply variations in light and shadow. Relief shading creates 3D-like images from a continuous grid surface. The procedure calculates light and shadow effects based upon the angle of incidence of an imaginary point source of light to create gray-scale or color images. Mentum Planet also supports the generation of true 3D perspective views using GridView. This tool gives you the ability to compose a Map window of any vector, raster, or textual entity that coincides with the spatial extents of the grid file and then drape the window onto a 3D representation of the grid

79

Chapter 5 Grid Analysis User Guide

surface. Perspective 3D rendering is supported through implementation of the OpenGL graphics language.

To use the Grid Color tool

1 2 3 If the Grid Manager is not visible, choose View Show Grid Manager. In the Grid Manager, choose a numeric grid (.grd). Click the Color button. The Grid Color Tool opens.
.

If you want to load an existing color profile, in the Color Profile section, click Load to load a color profile (.vcp) file.

80

Working with the Grid Manager Grid Analysis User Guide

Do any of the following:

To add a color inflection point, click Add, define a value for the inflection point, and click OK. To delete a color inflection point, click a color inflection point to highlight it and click Remove. To define a new color for the inflection point, double-click on a color inflection point, choose a new color in the Color dialog box and click OK. To move an inflection point, click a color inflection point and drag it to the new location. This will update the value for this inflection point in the Color Scheme list. The calculated values in the Color Scheme List are automatically updated. To change color values and percentiles, click an entry in the Color Scheme List list to make the value editable and type a new value. This will move the inflection point to the appropriate location on the color ramp. To create a Quick Map Legend for the grid, click Quick. To create a customized legend for the grid, click Custom.

In the Legends section, do any of the following:

For more information, see Creating and displaying legends. 7 In the Color Profile section, do any of the following:

Enable the Solid Band check box if you want hard breaks between colors instead of interpolated fading. Click Flip if you want the colors associated with inflection points in reverse order. Click Revert if you want to return to the color pattern that was in place before you clicked Flip.

If you want to redefine the grid colors based on how they would be illuminated by a single light source, in the Relief Shading section, enable the Enabled check box, and click Properties. In the Relief Shading Properties dialog box, define the azimuth, the inclination, the contrast, and the brightness, and click OK.

10 If you want to save the color profile, in the Color Profile section, click Save to save color settings as a text file with a .vcp extension.

81

Chapter 5 Grid Analysis User Guide

11 In the Grid Color Tool dialog box, click OK.

In deciding whether to save color inflection points by value or by percentile, use the following guidelines: If it is more important to assign specific colors to specific values in a series of related grid files, then save by value. If it is more important to assign a particular color range to a series of related grid files where the value range may vary considerably, then save by percentile. You can add a color inflection point in the Grid Color Tool by doubleclicking on the color slider bar. Conversely, you can delete an inflection point by clicking on an inflection point to highlight it and pressing Delete.

Using color in classified grids

Each bin in a classified grid is assigned a color corresponding to its category. Unlike numeric grids, the data in a classified grid is not continuously variable, so there are no gradients or inflection points. You use the Dictionary Editor to assign a unique color to each category of bins.

To use the Dictionary Editor

1 2 If the Grid Manager is not visible, choose View Show Grid Manager. In the Grid Manager, choose a classified grid (.grc).

82

Working with the Grid Manager Grid Analysis User Guide

Click the Color button. The Dictionary Editor opens.

4 5

To define a new color for a category, double-click on it, choose a new color, and click OK. If you want to reduce the size of the grid file, enable the Pack on Exit check box to renumber the internal class numbering for all classes starting sequentially at 1. This option is most useful when the grid has fewer than 255 classes. In the Legends section, do any of the following:

To create a Quick Map Legend for the grid, click Quick. To create a customized legend for the grid, click Custom.

For more information, see Creating and displaying legends. 7 Click OK.

Creating and displaying legends

A legend adds meaning to grids by explaining visual elements in a map. In Mentum Planet, you can view a legend:

in a window using the Quick Map Legend. This is useful if you want to use the information from the legend in a report or create a print layout that includes the legend. You cannot customize the look or content of a Quick Map Legend in Mentum Planet. For
83

Chapter 5 Grid Analysis User Guide

information on creating print layouts, see Chapter 13: Generating Reports in the Mentum Planet User Guide.

in the Grid Legend dialog box. This is useful if you want to view legends for several maps interactively as you click in Map windows or if you want to customize the legend. on the Legend tab in the Grid Manager. This is useful if you want to view the legend along with a histogram, meta data, or information about the z-units of the grid. See Using the Grid Manager Info function on page 72.

When you create a legend, you create a MapInfo table (.tab file). You can modify the table and edit the labels (i.e., text objects). You can also create a new Browser window that incorporates the legend table or create a report that incorporates the table. For more information, see the MapInfo Professional User Guide or online Help.

To quickly create and display a legend

This method creates a MapInfo table (.tab file) and associated files (.id, .map, and .dat). 1 2 From either the Grid Color Tool for numeric grids or the Dictionary Editor for classified grids, in the Legends section, click Quick. In the Save Quick Map Legend As dialog box, type a name for the legend in the File Name box, and click Save. A legend window opens.

84

Working with the Grid Manager Grid Analysis User Guide

To open the Grid Legend dialog box

1 Choose View Show Grid Legend. The Grid Legend dialog box opens. By default, information on the active Map window is displayed.

85

Chapter 5 Grid Analysis User Guide

Do any of the following:

From the Grids list at the bottom of the Grid Legend dialog box, choose the grid for which you want to view a legend. Click in a Map window.

If you want the legend displayed in the Grid Legend dialog box to remain static even as you click in different Map windows, from the Options menu in the Grid Legend dialog box, choose Hold Legend. You can also access the Grid Legend dialog box by clicking the Tools button in the Grid Manager and choosing the Grid Legends command.

To customize a legend
1 2 From either the Grid Color Tool for numeric grids or the Dictionary Editor for classified grids, in the Legend section, click Custom. In the Legend Generator, define the range, text, and format options. For more information on the settings in the Legend Generator, press the F1 key. 3 Click OK to display the legend in a new Map window.

86

Working with the Grid Manager Grid Analysis User Guide

Modifying information about a grid

Metadata is data that describes characteristics of a grid. Information such as the name of the project of which the grid is a part or the name of the person who created the grid are examples of metadata. Using the Alter Meta Data tool, you can add a data description to the header of a grid. This is useful for identifying grids that originate from sources other than Mentum Planet. You can also change the data unit used in the grid. No mathematical conversions are applied, but you can swap one data unit for another (i.e., 20 feet can be changed to 20 meters). This is useful when a grid has been produced in meters but, in error, the data unit used is feet. Sometimes you have to define a specific data unit in order to use a specific tool such as the Point-to-Point Visibility and Viewshed functions. Both of these tools can be used only on grids in which the data unit is a distance measurement, i.e., meters or feet.

87

Chapter 5 Grid Analysis User Guide

To modify information about a grid

1 2 If the Grid Manager is not visible, choose View Show Grid Manager. In the Grid Manager, click the Tools button, and choose Alter Meta Data. The Alter Meta Data dialog box opens. 3 4 5 6 7 8 From the Grid list, choose the grid to which you want to append metadata. In the Meta Data box, choose the first blank line. In the Key box, type a label for the metadata information (e.g., Project Name). In the Value box, type the metadata you want to append (e.g., Downtown). In the Data Description box, type a description of the grid. This information will be added to the grid header. From the Unit Type list, choose a new unit for the grid. If you choose the User Defined option, you must type a value in the Custom Type Entry box. The User Defined data type is always used for grids with z-values measured in units other than distance units. This feature does not perform any mathematical calculation on the grid when you change the unit type. 9 Click OK.

Changing grid file projections

Some tools in Mentum Planet do not support unprojected data, so changing grid file projections is often necessary. Also, using unprojected data such as latitude and longitude can produce inaccurate distance and area calculations. Transforming a grid from its existing coordinate system to another supported system results in the interpolation of new values for each new bin. The process involves a complex coordinate transformation that makes use of quadratic surfaces to transform x, y locations in the original grid to the new coordinate system using four bin blocks. The procedure has been engineered to be accurate to within five percent of the new bin size at any location across the new grid extent. For example, if the new bin size is 100 m, the bin location

88

Working with the Grid Manager Grid Analysis User Guide

will be accurate to within 5 meters. The process uses a variation of the bilinear interpolator to recalculate new bin values at each transformed bin. Elevation data can be provided in any projection that Mentum Planet supports. Analysis layer files such as path loss or signal strength files are generated at the same resolution and in the same projection as the elevation file. There are some considerations worth noting when dealing with grids in the longitude/latitude coordinate system. Grid raster files are displayed differently from vector files when they are expressed in longitude/latitude format. Because raster grid files cannot easily be re-projected on the fly, they are displayed with an even x and y spacing between grid nodes. The farther one travels away from the equator, the greater the actual difference in distance between a degree of latitude and a degree of longitude. Depending on the latitude at which the map scene is viewed, vector objects can be compressed to maintain a valid aspect ratio of the grid. As a result, a circular area drawn over a raster grid map that is in a longitude/ latitude system appears flattened or elliptical. Since all signal strength grids generated in Mentum Planet are circular radius maps, they will look flattened if the elevation grid is not in a true projection. Although the signal strength grids can have a peculiar appearance, this has no effect on the accuracy of the predictions.
To avoid visual distortion and possible computational inaccuracies, it is recommended you work with projected elevation data, rather than in longitude/latitude format. Opening differently projected grids in the same Map window produces a visual distortion of one or more of the grids. It is recommended that you use a clutter grid that is in the same projection as the grid files open in the same Map window. The clutter grid file must be in the same projection and at the same resolution as the project elevation file when used for traffic modeling in some Mentum Planet modules.

To change the projection of a grid file

1 2 If the Grid Manager is not visible, choose View Show Grid Manager. In the Grid Manager, click the Tools button, and choose Reproject.

89

Chapter 5 Grid Analysis User Guide

3 4 5

From the Grid list in the Reproject Grid dialog box, choose a grid, and click Projection. In the Choose Projection dialog box, choose a projection category and category member, and click OK. Read the suggested range of bin size values displayed in the Mentum Planet dialog box, and click OK to close the dialog box. This range of values represents the minimum and maximum bin sizes to which any of the original bins will be transformed over the area of the new reprojected grid. If you specify a new bin size for the reprojected grid that is less than half the minimum recommended value, or if you specify a new bin size that is more than twice the maximum recommended value, an information message appears.

Click Browse, navigate to the folder where you want to save the reprojected grid, type a name in the File Name box or accept the default, and click Save. By default, the reprojected file has the original file name with _Reproject appended to it, and it is located in the same folder as the original file.

Click OK.

To change the projection of a vector-based .tab file

1 2 3 4 5 Choose File Save Copy of Table As. In the Save Copy As dialog box, choose the table from the Save Table list, and click Save As. In the Save Copy of Table As dialog box, type a new file name for the file, and click Projection. In the Choose Projection dialog box, choose the projection from the Category and Category Members lists, and click OK. Click Save.

Updating the projection file

The Info tab in the Grid Manager displays the projection type of a grid in the Coordinate System Name box. If the projection type is shown as not found in your projection file, the projection has not been given a specific name. You can add the name of a grids projection to the list of projection types in

90

Working with the Grid Manager Grid Analysis User Guide

Mentum Planet using the Projection File Updater tool, which you must add to the Tools menu.

To add the Projection File Updater to the Tools menu

1 2 3 4 5 6 7 Choose Tools Tool Manager In the Tool Manager dialog box, click Add. In the Add RF Tool dialog box, click Browse and navigate to the Mentum Planet program folder. From the Files of Type list, choose MapBasic Program (*.mbx). Choose showCheckProjection.mbx and click Open. In the Add RF Tool dialog box, click OK. In the Tool Manager dialog box, click OK.

To update the projection file

1 2 3 Choose Tools Projection File Updater. In the Projection File Updater dialog box, click Browse (...), navigate to your grid file, and click Open. Click Next. If the grids projection is already defined, an Information box is displayed. Click OK to finish. The projection file does not need updating. 4 5 6 In the Category box, type a projection category name or choose one. In the Member box, type the coordinate system name, and click Finish. In the Information dialog box, click OK.

Reclassifying grids
When you reclassify a grid, you generate a new numeric or classified grid. For example, you can view signal strength values in a numeric (.grd) grid either by gradual shadings or by a limited range of discreet bands where you do not need highly precise data divisions. In a classified grid, for example, you can view vegetation classes in broad categories or in greater detail by subcategories.

91

Chapter 5 Grid Analysis User Guide

To reclassify a numeric grid

You can reclassify the values contained in a numeric grid in order to make the data more suitable for analysis. This is useful, for example, when you want to create grids with consistent classes or label the class ranges contained in a grid. 1 2 3 4 5 6 7 8 Open the grid that you want to reclassify. If the Grid Manager is not visible, choose View Show Grid Manager. In the Grid Manager, choose the numeric grid (.grd) you want to reclassify. Click the Tools button, and choose Reclass. The Numeric Grid Reclassification Tool dialog box opens. Click Numeric at the top right of the dialog box. Double-click any Value field in the Ranges table to edit its value. Double-click the Color field to change the color of a class, and then click OK. To manually alter ranges, do any of the following:

Choose a row in the Ranges table, and click Insert to insert a row below the selected row. Choose a row in the Ranges table, and click Delete to delete the selected row.

To automatically assign ranges, click Intervals, in the Intervals dialog box, specify the minimum and maximum value for all ranges, and choose whether the interval is defined by an interval size or by a specific number of classes.

10 Click OK. 11 Click Save to save the reclass profile to a .pfr file which you can use with other grids. 12 Click Browse, navigate to the folder where you want to save the reclassified grid, type a name in the File Name box or accept the default, and click Save. 13 Click OK. The newly reclassified grid opens in a Map window.

92

Working with the Grid Manager Grid Analysis User Guide

To reclassify a classified grid

You can reclassify the values contained in a classified grid to make the data more suitable for your analysis. This is useful, for example, when you want to generate a clutter file with the same number of classes as another clutter file. 1 2 3 4 Open the grid that you want to reclassify. If the Grid Manager is not visible, choose View Show Grid Manager. In the Grid Manager, choose the classified grid (.grc) you want to reclassify. Click the Tools button, and choose Reclass To Classified to generate a new classified grid. The Classified Grid Reclassification Tool dialog box opens. 5 6 Click Add. In the Add dialog box, choose one of the following options:

Add Classenables you to add a new class to the output grid and assign a color to the class by clicking Color. Add Groupenables you to add a new group to the output grid. Once you have added a group, you will need to add classes to it.

7 8

Click OK. In the Original Classes list, choose the original class you want assign to the newly created class, and drag it onto the new class. The name of the original class appears in the Assigned Classes box. Repeat step 7 as required, and then click Verify to verify that all classes have been assigned.

10 Click Reduce to remove unassigned classes or groups from the New Classes list. 11 Click Save to save a reclass profile as a .pfr file which you can then use with other grids, and then click Save again. 12 Click Browse, navigate to the folder where you want to save the reclassified grid, type a name in the File Name box or accept the default, and click Save.

93

Chapter 5 Grid Analysis User Guide

13 Click OK.
To open a reclass profile for use with another grid, click Load, locate the .pfr file you want to use, and then click Load again. To edit the name of a class or its color, double-click the class in the New classes list. Rename the class in the Name box and double-click the color swatch and choose a color from the palette.

To reclassify isolated areas of a classified grid

Using the Classified Grid filter, you can reclassify small isolated areas of bins. This has the effect of smoothing the grid. The classes are processed individually, beginning with the class containing the smallest number of bins. 1 2 3 4 5 6 Open the grid that you want to reclassify. If the Grid Manager is not visible, choose View Show Grid Manager. In the Grid Manager, click the Tools button, and choose Classified Grid Filter. From the Grid list in the Classified Grid Filter dialog box, choose the grid to be filtered. In the Number of Stranded Bins box, type the number of bins that denotes an area to be reclassified. Clear the Use Largest Neighbor check box. If this check box is enabled, the Neighbor Uniformity Threshold box is not available and stranded bins are reclassified to the class of the largest neighbor. 7 In the Neighbor Uniformity Threshold box, type the percentage of bins in an adjacent class that must border the isolated bins in order for the isolated bins to be reclassified to this new class. For example, if you enter 75%, an isolated bin will be reclassified to match the class that has at least 75 percent of bins sharing a border with the isolated bin. If no adjacent class borders the isolated bins for the defined percentage, the class of the isolated bins does not change.

94

Working with the Grid Manager Grid Analysis User Guide

Click Browse, navigate to the folder where you want to save the file, type a name in the File Name box or accept the default, and click Save. By default, the file has the original file name with _filter appended to it, and it is located in the same folder as the original file.

Click OK. The newly reclassified grid opens in a Map window.

Converting grids
You can convert numeric grids to classified grids to view the data by category or class. It is often useful to view data by class rather than fine numeric gradations. You can also convert a classified grid to a numeric grid in order to perform grid math calculations. For each class, a numeric value is defined that is assigned to all bins of that class. The result is a new numeric grid. You can also choose the colors that represent each class in the new grid.

To convert a numeric grid to a classified grid

You can load and save your conversion settings so that they can be used in subsequent grid reclassifications. The save process creates an ASCII text file with a .pfr extension containing the range, the value, and the color of each class. You can change the default class names (e.g., Class 1) and specify the upper and lower values for each range directly in the Ranges table. 1 2 3 4 If the Grid Manager is not visible, choose View Show Grid Manager. In the Grid Manager, choose the numeric grid (.grd) you want to convert. Click the Tools button, and choose Reclass. In the Numeric Grid Reclassification Tool dialog box, click the Classified button.

95

Chapter 5 Grid Analysis User Guide

Do one of the following:

Click the Insert button to insert a row to the Ranges table. Click the Delete button to delete the selected row in the Ranges table. Click the Interval button to specify the minimum and maximum and/or number of intervals for the Ranges table.

6 7 8

Click the Save button to save the reclass profile to a .pfr file. In the File name box, enter a new file name or choose the default name. Click OK.

To convert a classified grid to a numeric grid

1 2 3 4 If the Grid Manager is not visible, choose View Show Grid Manager. In the Grid Manager, choose the classified grid (.grc) you want to convert. Click the Tools button, and choose Reclass to Numeric. In the Classified to Numeric Grid Converter dialog box, if you have more than one classified grid open, choose the grid you want to reclassify from the Grid list. If you want to use class names and values that are defined in a table, click Set from Table, choose the following settings in the Table Values for Classes dialog box, and click OK:

Tablethe table containing the class values Class Columnthe column containing the class names Valuesthe column containing the class values

6 7 8 9

To change a class value, double-click the Class Value field and edit the value. To change a class color, click a color field, choose a new color in the Color dialog box, and click OK. Click Browse, navigate to the folder where you want to save the file, type a name in the File Name box or accept the default, and click Save. Click OK.

96

Working with the Grid Manager Grid Analysis User Guide

Trimming grids
Using the Trimmer tool, you can create a new grid by trimming away portions of a larger grid. For example, if you have a grid that covers a geographically large city, but you need to generate a network analysis for only a small area of that city, you can trim the city grid, and create a new grid containing only the relevant area. This is useful when you want to focus an analysis on a particular area. Trimming a grid also reduces the size of a file and, therefore, the time it takes to open a project. You can only trim one grid file at a time.
You must create a region before trimming a grid.

To trim a grid
1 2 3 4 5 6 7 8 9 Open the grid you want to trim. In the Project Explorer, in the Windows category, expand the Map window containing the grid you want to trim. Right-click the Cosmetic layer, and choose Editable if the check box is not already enabled. If the Drawing toolbar is not visible, choose View Toolbars, enable the Drawing check box in the Show column, and click OK. Click the Polygon button on the Drawing toolbar, and draw a polygon over the area you want to appear in the new grid. Choose the region. If the Grid Manager is not visible, choose View Show Grid Manager. In the Grid Manager, click the Tools button, and choose Trimmer. In the Save Trimmed Grid As dialog box, type a new file name for the trimmed grid in the File Name box or accept the default, and click Save. Once the new numeric (.grd) file is created, it appears in a Map window with a default color palette applied. You can change the color range assigned to the grid file using the Grid Color tool.

97

Chapter 5 Grid Analysis User Guide

Splicing grids
You can splice multiple grids into a single grid file using one of two methods: merging or stamping. The difference lies in how each of these methods handles the values of grid cells that overlap other grid cells. Merging is the process where two or more numeric grids are combined and where in the overlapping areas a mathematical calculation is performed to determine the new grid value. For example, if two grids are merged using the average calculation, where the first grid has cells with a value of 20 and the second grid has cells with a value of 10, then overlapping cells will have a value of 15 in the resulting grid (see Figure 5.9). Merging is often used when several grid tiles cover a study area that needs to be combined into a single grid, or when you want to create a grid that has the highest/lowest value of all the input grids. Null values contained in any of the grids being processed are ignored.
Grid 1 (value 20) Grid 2 (value 20) Result 10 15 20

Figure 5.9 In the areas where Grid 1 (value of 20) and Grid 2 (value of 10) overlap, the resulting grid will have a value of 15 when the Average overlap calculation is chosen.

Stamping is the process where two or more grids are combined and where the overlapping cells of one of the input grids will overwrite the values of the other. Grids can be stamped only with a grid of similar type (i.e., classified or numeric). It is not possible to stamp a classified grid with a numeric or vice versa. Stamping is usually performed when you need to update areas of a grid. For example, you can stamp the land use values from an up-to-date clutter grid on an earlier one. Null values contained in any of the grids being processed are ignored.

98

Working with the Grid Manager Grid Analysis User Guide

Grid 2 Grid 1 (reference)

Grid 1 Result Grid 2 (reference)

Figure 5.10 Depending on which grid is the reference grid and the order in which the input grids are listed, different results will be obtained. On the left, Grid 2 is being stamped on Grid 1. Therefore the bin values of Grid 2 will override the values of Grid 1 in the resulting grid. On the right, Grid 1 is being stamped on Grid 2. Therefore the bin values of Grid 1 will override the values of Grid 2 in the resulting grid.

If Grid 1 and Grid 2 contained a clutter class with the same name but with different assigned colors, then when you spliced the two grids together the resulting grid would contain two clutter classes (e.g., cluttername_1 and cluttername_2) with the assigned colors taken from each grid.

To splice grids
The grids that you want to splice must have the same resolution and projection. Numeric grids must also have the same z-units.

1 2 3 4

Open the grids that you want to splice. If the Grid Manager is not visible, choose View Show Grid Manager. In the Grid Manager, click the Tools button, and choose Splicer. The Grid Splicer dialog box opens. From the Begin Splicing With list in the Grid Splicer dialog box, choose a grid. If the grids you are splicing have different units, the first grid you choose determines the units of the output grid.

From the Spliceable Grids list, choose one or more other grids, and click Add.

99

Chapter 5 Grid Analysis User Guide

Choose the type of splicing by doing one of the following:

Click Stamp to combine two or more grids. In the overlapping areas, the bin values of the grid listed first will overwrite the values of the other grids. Click Merge to combine two or more numerical (.grd) grids, and choose a calculation type from the associated list. In the overlapping areas, a mathematical calculation is performed to determine the bin values in the new grid.

Click Browse, navigate to the folder where you want to save the spliced grid, type a name in the File Name box or accept the default, and click Save. Click OK to close the Grid Splicer dialog box and create the spliced grid.

You can update specific areas of a grid by creating polygonal regions around the areas to be updated (i.e., the new subdivision or building footprint), attaching the appropriate information to these regions, (e.g., urban for the subdivision and the building heights) to the footprint, then, using the Region to Grid command, you can convert these regions to a grid. Be sure that the projection and the grid units are the same. Finally, you can use the Grid Splicer command to stamp these region grids on the grid to be updated. To attach information to a region, you must save an object drawn on the Cosmetic Layer to create a table. For more information on saving Cosmetic objects, see Chapter 3, The Basics of MapInfo Professional in the MapInfo Professional User Guide accessible through the Resource Roadmap in the Mentum Planet Help. For more information on changing tables, see Chapter 5, Enhancing Your Data in the MapInfo Professional User Guide.

Resizing grids
You can resize a grid to perform other tasks. For example, you might need to reduce the size and complexity of grid files in preparation for contour threading. While a very high resolution grid file is suitable for modeling and high quality output, a lower resolution version of the same file is adequate for generating contour lines or regions. When the bin size of an existing grid is increased, a new value must be interpolated for every bin in the new grid. You can visualize the original grid of smaller bins as a network of nodes overlain by a new network consisting of more widely spaced nodes. The value at the

100

Working with the Grid Manager Grid Analysis User Guide

new grid node is calculated using a bilinear interpolation (rectangular interpolation) of the four nearest nodes of the underlying, more closelyspaced grid. The file size increases when you resize a grid to a higher resolution. While a high-resolution grid is not more accurate, reducing the bin size of a grid will generate a more highly smoothed surface and shows a continuous gradient color fill over the grid area. There are two resizing tools available in Mentum Planet:

the Resizerresizes a grid and calculates new values for each bin using interpolation the Numeric Grid Filterresizes a grid and calculates new values for each bin using one of five smoothing methods as shown in Figure 5.11. This tool is particularly useful when you want to smooth out a grid consisting of many isolated bins, islands, or holes.

180 540

Original bin size Output bin size

3780

Filter Size area

Figure 5.11 The new bin value is calculated using all the points in the filter size area. The new value is then applied to the output bin.

When you use the Numeric Grid Filter, the calculations used to determine the bin value are different but the method used to calculate the value is the same. Using the averaging method, for example, given an original bin size of 180m,

101

Chapter 5 Grid Analysis User Guide

an output bin size of 540m, and filter size of 3780 m, the Numeric Grid Filter takes the bin values within the 3780m filter size, averages them and then applies this new value to the output bins.

To resize a grid using the Resizer

1 2 3 Open the grid that you want to resize. If the Grid Manager is not visible, choose View Show Grid Manager. In the Grid Manager, click the Tools button, and choose Resizer. The Resizer dialog box opens.

4 5 6

From the Grid list, choose the grid you want to resize. Type a new bin size in the New Bin Size box. Click Browse, navigate to the folder where you want to save the resized grid, type a name in the File Name box or accept the default, and click Save. By default, the resized file has the original file name with _Resized appended to it, and it is located in the same folder as the original file.

Click OK. The newly resized grid opens in a Map window.

To resize a grid using the Numeric Grid Filter

1 2 Open the grid that you want to resize. If the Grid Manager is not visible, choose View Show Grid Manager.

102

Working with the Grid Manager Grid Analysis User Guide

In the Grid Manager, click the Tools button, and choose Numeric Grid Filter. The Numeric Grid Filter dialog box opens.

4 5 6

From the Grid list, choose the grid you want to resize. From the Output Bin Size list, choose the resolution of the output grid. The output bin size is a multiple of the original bin size. From the Filter Size list, choose the area you want included in the smoothing calculations. The filter size should be larger than the output bin size; otherwise, no bin values remain constant and no smoothing effect is achieved.

From the Method list, choose one of the following:

Averagebin values within the filter area are averaged and applied to the output Minimumthe minimum bin value within the filter area is used in the output Maximumthe maximum bin value within the filter area is used in the output Medianthe median bin value within the filter area is used in the output Gaussiana Gaussian curve is applied to bin values within the filter area and a weighted value given to each point. The weighted values are then averaged to determine the bin value.

103

Chapter 5 Grid Analysis User Guide

Click Browse, navigate to the folder where you want to save the resized grid, type a name in the File Name box or accept the default, and click Save. By default, the resized file has the original file name with _Resized appended to it, and it is located in the same folder as the original file.

Click OK. The newly resized grid opens in a Map window.

Exporting grids
You can transfer numeric and classified grid data created in Mentum Planet to external applications. In most cases, the grid node x- and y- location coordinates and the z-value attribute attached to each node are extracted from the grid file and written to the new format. Currently, Mentum Planet supports ten export file types.

ASCII Point Export (.txt)x, y, z space delimited text file (numeric grid export only) 3D DXF Point Filex, y, z AutoCAD points file (numeric grid export only) 3D DXF MeshAutoCAD mesh file (numeric grid export only) 3D DXF Mesh (Microstation)Microstation-compliant DXF mesh format (numeric grid export only) VRML (.wrl)a simple Virtual Reality Markup Language file (numeric grid export only) MapInfo Point Table (.tab)point table with z-value attribute column Windows Bitmap (.bmp and .tab)image built according to the display settings of the grid. The file name is appended with _bmp. ASCII Grid Export (.txt)text file TIFF (.tif)Tagged Image File Format image USGS DEMan ASCII format DEM supported by USGS. The Grid Export utility supports only UTM, Transverse Mercator and Longitude/Latitude projections (numeric grid export only)

104

Working with the Grid Manager Grid Analysis User Guide

To export a grid
1 2 If the Grid Manager is not visible, choose View Show Grid Manager. In the Grid Manager, click the Tools button and choose the Export command. You must have a project or grids or both open. 3 4 5 From the Grid list in the Export Grid dialog box, choose the grid you want to export. In the Export Type section, choose the format of the export file. If you are exporting to either of the DXF mesh formats or the VRML format, specify a null value in the Null Value box. This differentiates a real zero value from a null value in the export file. 6 7 8 Click Browse to locate the folder where you want to save the exported file. In the File name box, type a new file name or accept the default. Click OK.

105

Chapter 5 Grid Analysis User Guide

106

6.
This chapter contains the following sections:

Chapter 6: Working with Graphs

Working with Graphs

This chapter describes customizing graphs, changing their viewing options, printing graphs, and exporting graphs in three formats.

Introduction Customizing graphs Changing graph viewing options Printing graphs Exporting graphs

107

Chapter 6 Grid Analysis User Guide

Introduction
In Mentum Planet, you can display grid information in graph format. The graphing tool appears as part of four features, namely the grid histogram in the Grid Manager, the Cross Section tool, the Point-to-Point Visibility tool, and the semivariogram in kriging interpolations. Each of these features employ some, if not all, of the available graphing capabilities. The grid graphing tools comprise the following:

The grid histogram for numeric and classified grids shows the distribution of your data. In a classified grid, for example, you can view the grid data across categories such as class name, class count, percentage, and area. For more information, see Using the Grid Manager Info function on page 72. The Cross Section tool displays the grids elevation profile and the signal strength of the best server along the length of a line. For more information, see Creating a cross section on page 119. The Point-to-Point Visibility tool displays a cross section of the elevation between two selected viewpoints. The graphs lines show the profile of the terrain and line of sight between the two viewpoints. For more information about point-to-point analyses, see Chapter 7, Generating a Signal Strength Prediction Between Two Points, in the Mentum Planet User Guide. The seriovariogram graphs are derived from Kriging interpolations: the variations of a grid surface over the distance between two points. The graphing tools assist in the modeling of north-south and east-west trends in your data. For more information, see Kriging interpolation on page 31.

Customizing graphs
To customize a graph

Position the cursor over a grid histogram, a cross-section graph, or a point-to-point solution graph, right-click and choose Customize Graph.

108

Working with Graphs Grid Analysis User Guide

Changing graph viewing options

You can enlarge and zoom in and out of graphs as well as display or hide graph legends and data.

To maximize a graph

Position the cursor over a grid histogram, a cross-section graph, or a point-to-point solution graph, right-click and choose Maximize. The Maximize command displays the graph window at full screen size. You can reduce the graph to its original size by pressing the ESC key or clicking in the upper-left corner of the maximized window.

To zoom in and out on a graph

To zoom in, position the cursor over a grid histogram, a cross-section graph, or a point-to-point solution graph, then click and drag. To zoom out, position the cursor over a grid histogram, a crosssection graph, or a point-to-point solution graph, right-click and choose Undo Zoom. The Undo Zoom command enables you to restore the size of the graph to the original zoom level at any time.

To show or hide the legend

Position the cursor over a grid histogram, a cross-section graph, or a point-to-point solution graph, right-click and choose Hide Legend or Show Legend. Because two or more graphs may be displayed in the same graph window, the legend indicates the color assigned to each grid file. You can modify the line or symbol color of the graph by double-clicking on the grid file name entry in the legend box. You can then make any changes in the Color dialog box.

To show or hide the graph data

Position the cursor over a grid histogram, a cross-section graph, or a point-to-point solution graph, right-click and choose Hide Graph Data or Show Graph Data.

109

Chapter 6 Grid Analysis User Guide

Printing graphs
The Print command prints the current graph directly to the default printer, as set by the computers operating system.

To print a graph

Position the cursor over a grid histogram, a cross-section graph, or a point-to-point solution graph, right-click and choose Print.

Exporting graphs
You have three options when exporting a graph:

The Text File optionsaves all the information used in constructing the graph to a comma-delimited ASCII text file with a .txt extension. The information saved includes the sample number, the distance from the starting point of the line, the xand-y coordinate of each sample, the overland distance between each sample, and the value at each sample location found on all grids plotted in the graph. The overland distance value is only available when the projection of the grids being graphed is in a Cartesian coordinate system and has not crossed a null data region. When this menu item is chosen, you will be prompted for the name of the text file. The Bitmap & Tab optionsaves the Graph window to a 24-bit bitmap image with an associated .tab file. This enables the graph to be opened and placed inside a Layout window. When this menu item is chosen, you will be prompted for the name of the text file. The Windows MetaFile optionsaves the Graph window as a Windows MetaFile (.wmf).

110

Working with Graphs Grid Analysis User Guide

To export a graph
1 2 Position the cursor over a grid histogram, a cross-section graph, or a point-to-point solution graph. Right-click, choose Export, and then choose one of the three Export options:

Text File Bitmap & Tab Windows MetaFile.

In the Save As dialog box, navigate to the folder where you want to save the file, type the file name, and click Save.

111

Chapter 6 Grid Analysis User Guide

112

7.
This chapter contains the following sections:

Chapter 7: Using Grids for Spatial Analysis

Using Grids for Spatial Analysis

Mentum Planet enables you to use grids in complex spatial analysis.

Introduction Using the Grid Calculator Understanding grid queries Creating a cross section Using the Point Inspection function Using the Line Inspection function Using the Region Inspection function

113

Chapter 7 Grid Analysis User Guide

Introduction
Mentum Planet gives you the ability to create grids based on geographic information and apply this information to complex spatial analysis problems. For example, annual variations in air temperature can be calculated by subtracting one years measurements from those of another year. In Mentum Planet, this arithmetic calculation is performed using the Grid Calculator. Using grids of each years temperatures, you can subtract the values in one from the values in another to create a new grid with the new values representing the variation. Using the Cross Section tool, you can draw or choose a line object that overlies a height grid of the area to construct a vertical profile of elevation along this line. Mentum Planet includes tools that you can use to solve problems that are virtually impossible to answer using only vector representations of geographic information. These can range from the simple Can I see that group of buildings from this tower? to the complex Given this range of land classifications, this distribution of soil types, these distances from major roads, and these county zoning boundaries, show me the most appropriate sites for a new communications tower.
When using functions that query grids such as the grid calculator, grid query, point-to-point visibility, cross section, point inspection, and line inspection, you can choose whether or not values are interpolated within bins in grids by enabling or clearing the Use Closest Node Value check box on the z-units tab in the Grid Manager. For more information, see Working with the Grid Manager on page 70.

Using the Grid Calculator

The Grid Calculator enables you to create mathematical expressions using an unlimited number of grids. You can modify or filter the gridded data in any one file, or you can generate a new derivative surface from two or more existing grids. The functions available in the Grid Calculator enable you to perform a number of mathematical operations either on a single grid file (for example, Grid1*sin(25) = Grid2) or on a combination of grids (for example, Grid1/ Grid2*Grid3 = Grid4). The new derived grid is the result of applying basic mathematical operations on all of the bin values in an existing numeric grid where each bin is assigned a value satisfying an expression you enter.

114

Using Grids for Spatial Analysis Grid Analysis User Guide

To use the Grid Calculator

The grids that you use in grid calculations must have the same resolution and projection.

1 2

In the Grid Manager, click the Analysis button and choose Calculator. In the Calculator dialog box, enter an expression by doing any of the following:

Double-click an item in the Grid/Value list to enter a grid in the Expression box, or choose a grid in the Grid/Value list and click the Grid key on the numeric keypad. Click the required numbers and functions to enter values or operators in the Expression box.

If you want to save the expressions to an .exp file, click Save As, type a file name, and click Save. Saving expressions to an existing file adds expressions to that file. In the Calculator dialog box, click OK. In the Grid Calculator Save dialog box, do the following:

4 5

In the Save the New Grid As box, type a file name for the new derived grid built from the math expression. In the Description box, type a description of the new grid. From the Z-Unit Type list, choose a z-unit type. If the unit type is user-defined, you can type a custom entry in the Enter User Defined Type box.

Click OK. The new grid file appears in a Map window with a default color palette applied.

115

Chapter 7 Grid Analysis User Guide

To use the Grid Calculator with saved expressions

1 2 In the Grid Manager, click the Analysis button and choose Calculator. To use a saved expression, click Load, choose an expression (.exp) file, click Open, and do one of the following in the Grid Math Expression Viewer:

To replace the current expression, choose an expression in the Expression list, then choose the Use as a Complete Expression option, and click Insert. To add an expression at the cursor position, choose an expression in the Expression list, then choose the Insert at Cursor option, and click Insert.

3 4 5

When you have finished inserting expressions, click Close to close the Grid Math Expression Viewer. In the Calculator dialog box, click OK. In the Grid Calculator Save dialog box, do the following:

In the Save the New Grid As box, type a file name for the new derived grid built from the math expression. In the Description box, type a description of the new grid. From the Z-Unit Type list, choose a z-unit type. If the unit type is user-defined, you can type a custom entry in the Enter User Defined Type box.

Click OK. The new grid file appears in a Map window with a default color palette applied.

When working with saved expressions, you can also create expressions using the Calculator dialog box, as described in To use the Grid Calculator on page 115.

Understanding grid queries

You can use grid queries to build new grids from existing grids where the new grid values are derived according to whether specific queries imposed on the existing grid files have been met.

116

Using Grids for Spatial Analysis Grid Analysis User Guide

For example, you can create an expression that will query a set of grid files and identify all coincident bins that meet these conditions: bins in Grid1 are >1049m AND bins in Grid2 are <=$250,754.50 AND bins in Grid3 are = Dense Urban OR bins in Grid4 are = Leeds County Therefore, by examining each grid and drilling down through the four grid layers on a bin-by-bin basis, a fifth grid can be built from the overlying and spatially coincident bins that identifies all the bins that meet (or do not meet) all of the query conditions.

Figure 7.1 The power of grid-based spatial analysis lies in the ability to create a query, using multiple themes of geographic information, that examines on a bin-by-bin basis each layer of data and derives a new grid based on the query conditions.

There is no limit to the number of grids that can be used in a grid query expression. However, there are restrictions on the specifications of grids that are used in a multiple grid query:

All grids used in a grid query must be in the same coordinate system. All grids must be spatially coincident or at least partially overlap. If grids have different bin sizes, the query will be controlled by the grid with the smallest bin size.

The Grid Query tool is comparable to the SQL Select command although, in the latter, the new Query table preserves all of the original data records that meet the SQL conditions. In the Grid Query tool, new bins are created in the

117

Chapter 7 Grid Analysis User Guide

derivative grid that are assigned a user-defined value according to whether the bin did or did not meet the conditions of the query expression. This is based on a simple true/false test: Does each bin in the new grid meet or not meet all of the query conditions?

Creating and editing conditional queries

The first part of creating a conditional query involves building a query expression from a selection of open grid files as a series of individual conditional clauses. The second part involves the assignment of values to the new grid, bin by bin, according to whether the query conditions are met (true) or not met (false) for each set of overlying bins.

To create a conditional query

1 2 3 4 5 In the Grid Manager, click the Analysis button and choose Grid Query. Click the New button. From the Grid list in the Select Condition dialog box, choose a grid. From the Operator list, choose an operator. Do one of the following:

Enable the Compare to Grid check box, choose a second grid from the Grid list, and click OK. This enables you to compare one grid with another. Clear the Compare to Grid check box, enter a value in the Value box, and click OK. This enables you to compare a grid to a constant value.

6 7 8 9

Click Verify to ensure that the query syntax is correct, and then click OK. Click Next. In the Output Results dialog box, choose the output format in the Output Format section. Specify the output when true in the When True section.

10 Specify the output when false in the When False section. 11 In the File Name box, type a name for the new derived grid, and click Finish.

118

Using Grids for Spatial Analysis Grid Analysis User Guide

Creating a cross section

You can create a query of grid values along a line or polyline constructed in a Map window using the Cross Section command. The values are displayed as an x, y line plot in a graph window.

To create a cross section graph

1 2 3 If the Grid Analysis toolbar is not visible, choose View Toolbars, enable the Show check box for the Grid Analysis toolbar, and click OK. On the Grid Analysis toolbar, click the Cross Section button. Draw the line of the cross section directly on the grid image, then doubleclick to end the line.

You can also create a line of cross section on a selected line or polyline by clicking the Analysis button in the Grid Manager, and then choosing the Cross Section command.

Using the Point Inspection function

The Point Inspection function updates a table of point data with a new column of values taken from one or more geographically coincident grid files. The process inspects the grid file at each point, returns the appropriate grid value, and writes the value to a new column in the point table. In many ways, the procedure is the reverse of creating a grid from a set of points. An example would be adding a column of income data to a point table of dwelling locations using a grid file of average family income. When you use the Point Inspection function, a new column is created in the point table for each open and active grid that contains the respective grid value corresponding to each point location.

To perform point inspection

1 2 3 Open a table of point data and the grid files from which the information is to be extracted. If the Grid Manager is not visible, choose View Show Grid Manager. In the Grid Manager, click the Analysis button and choose Point Inspection.

119

Chapter 7 Grid Analysis User Guide

From the Table to Update list in the Point Inspection dialog box, choose the appropriate table containing the data points to be updated. The Null Value box displays the value that will be used if the point lies off the grid or over a null bin.

Click OK.
Values from all open and geographically coincident grids will be added as new columns to the point table, assuming that all grids are active.

Using the Line Inspection function

The Line Inspection function displays several statistical parameters based on the grid values that a specified line overlays, such as the average elevation of a runway. The grid is sampled at several locations along the selected line. Each line is sampled a specified number of times regardless of the line length (the default is 100 samples). Because the number of samples taken influences the final results and each line is sampled the same number of times, statistics calculated for very long lines may be less accurate than those calculated for shorter lines. The way to get around this is to determine how long the longest line is and then base the sample number upon that. As shown in Figure 7.2, a simple SQL Query can be performed to do this, and the number of samples can be modified in the Preferences dialog box.

Figure 7.2 Sample SQL query

The Line Inspection function updates a table of polyline data with values taken from a grid file. The process inspects the samples between the

120

Using Grids for Spatial Analysis Grid Analysis User Guide

beginning and end of the line, calculates the selected number of statistical parameters, and writes the values to new columns in the data table.

To perform line inspection

1 2 3 Open a table of line data and the grid files from which the information is to be extracted. If the Grid Manager is not visible, choose View Show Grid Manager. In the Grid Manager, click the Analysis button and choose Line Inspection.

4 5

From the Table to Update list, choose the table to be updated with line data. In the Select Attributes to ADD for All Active Grids section, choose all of the appropriate statistical parameters (values) and the number of samples that will be calculated for each line and written into the data table. The default number of samples is 100. Enable any or all of the Start Value, Middle Value, and End Value check boxes to add columns to the table with the values at the start, middle, and end of the line. Click OK.

121

Chapter 7 Grid Analysis User Guide

Each statistical parameter calculated for each open and active grid is written into a separate column of the region table and given a default column name consisting of both the parameter name and the grid file name.
The number of samples taken along the line is determined by the Cross Section: No of Samples parameter in the Preferences settings. The default is 100. To change this value, choose GIS Grid Analysis Preferences, and enter a new number in the No of Samples box. For more information about Preferences, see Setting your preferences on page 18.

Using the Region Inspection function

It is important to understand the manner in which Mentum Planet calculates area using grids. Every grid file is constructed of bins of equal area, as determined from the unique coordinate system specifications to which the grid has been projected. However, a grid in any coordinate system is only truly accurate, in terms of real spherical (ground) distance, along the line of standard parallels. Therefore, as you move farther away from the standard parallel, the actual ground area of a bin becomes less accurate (usually it is larger than its real spherical area). While in most cases the error will not be significant, if area is important, you have two options to minimize this error:

Ensure that the grid is always projected to an appropriate coordinate system that is relevant to the geographic region of interest. Using the Contour function, convert the grid file covering the area of interest to a region and then query the region to obtain MapInfos spherical-based area calculation.

The Region Inspection function updates a table of regions with new columns of values taken from one or more geographically coincident grid files. The process inspects the grid file underlying each region, returns a selected number of statistical parameters calculated from the range of grid values lying within each region, and writes the value to a new column in the region table.

To perform region inspection

1 2 Open a table of point data and the grid files from which the information will be extracted. If the Grid Manager is not visible, choose View Show Grid Manager.

122

Using Grids for Spatial Analysis Grid Analysis User Guide

3 4 5

In the Grid Manager, click the Analysis button and choose Region Inspection. In the Region Inspection dialog box, choose the appropriate region table to update. In the Select Attributes to ADD for All Active Grids section, choose all of the statistical parameters that will be calculated for each region and written into the region table. Click OK.

Each statistical parameter calculated for each open and active grid is written into a separate column of the region table and given a default column name consisting of both the parameter name and the grid file name.

123

Chapter 7 Grid Analysis User Guide

124

8.
This chapter contains the following sections:

Chapter 8: Aggregating Data

Aggregating Data

Mentum Planet provides two methods of data aggregation: simple point aggregation and point aggregation with statistics.

Introduction When to aggregate data Techniques for data aggregation Simple point aggregation Point aggregation with statistics Building a table of standard deviation ellipses

123

Chapter 8 Grid Analysis User Guide

Introduction
Data aggregation is a mathematical process for reducing the number of points in a point file. Different methods are available to process a point file, but the underlying goal is to spatially group and statistically merge points that are in close proximity. This not only results in fewer points but also creates a more uniform distribution of points and point values. Mentum Planet provides two options for data aggregation.

Simple Point Aggregation quickly aggregates data when you do not need to keep track of the spatial or mathematical properties of the aggregation process at each point, i.e., knowing the average of the aggregated points is sufficient. Point Aggregation With Statistics tracks the statistical parameters associated with each aggregated point (e.g., standard deviation or coefficient of variation). This technique can also produce a region file displaying the actual circular or square bin region used to gather and aggregate points. Statistical information is attached as attributes to both the new point file and to the region file.

When to aggregate data

The following reasons for aggregating data are the most common ones.

There are multiple values at a given location. Data is erratic, and the average of the data area represents an appropriate estimate of a point. The sum of the data in a given area is required. A representative value at an average location is acceptable when the location of the data is not accurate. The original data set has too many points to permit efficient processing. The number of points can be substantially reduced without adversely affecting the accuracy of any operation. Values representing specific spacing or area criteria applied over the entire mapped area (most applicable to the square bin method) are required.

124

Aggregating Data Grid Analysis User Guide

Techniques for data aggregation

The following data aggregation techniques are available:

Simple Point Aggregation Point Aggregation with Statistics, which comprises the Forward Stepping, Cluster Density, and Square Bin aggregation techniques.

When you use these techniques, Mentum Planet creates a new file containing the aggregated points and leaves the original point file unaltered.

Simple point aggregation

Simple point aggregation is useful for quickly grouping points that are virtually coincident or that are separated by relatively small gaps. For example, when soil samples are collected, you may need to take several samples at each sample site. When creating a surface of this data, you may want to average all of the samples taken at each site before proceeding with the surface creation. It may also be desirable to aggregate data that has poor reproducibility over a large area and achieve a more smoothing or averaging effect. Simple point aggregation is also used as a preliminary data smoothing technique for TIN creation in the triangulation with smoothing interpolation method, for natural neighbor region creation in the natural neighbor interpolation method, and in kriging interpolation. Here, highly variable data points spaced closely together can be aggregated and new values calculated using a statistical expression. For example, if two points are very closely spaced, even a small difference in z-value will generate a steep slope between them. This slope affects the interpolated surface for a significant distance beyond the two points, creating unwanted rises and dips. Simple point aggregation uses a coincident point distance value for spatially aggregating points. This means you need to specify how close the points will be before they are considered coincident. The result of simple point aggregation is shown in Figure 8.1 and Figure 8.2.

125

Chapter 8 Grid Analysis User Guide

Figure 8.1 Original distribution of points.

Figure 8.2 Distribution of points after simple point aggregation.

To perform simple point aggregation

You can use simple point aggregation to quickly group points that are virtually coincident or that are separated by relatively small gaps. 1 2 3 4 5 6 Choose GIS Grid Analysis Data Aggregation Simple Point Aggregation. From the Select Table to Aggregate list in the Select Table and Column dialog box, choose a point table. From the Select Column list, choose a column of data to transfer to the aggregated file. Enable the Ignore Records Containing Zero check box to include only non-zero records. Click Next. In the Aggregation Technique and Distance dialog box, choose an averaging technique.

126

Aggregating Data Grid Analysis User Guide

7 8

In the File Name box, type a new file name or accept the default. Click Finish.

Point aggregation with statistics

The point aggregation with statistics techniques provide information about the mathematical and spatial characteristics of the aggregation process. To quantitatively measure the aggregation process, Mentum Planet offers three techniques: Forward Stepping, Cluster Density, and Square Bin. These methods are grouped under the category Point Aggregation With Statistics.

To perform point aggregation with statistics

1 2 3 4 5 6 7 8 9 Choose GIS Grid Analysis Data Aggregation Point Aggregation with Statistics. From the Select Table to Aggregate list in the Select Table and Column dialog box, choose a point table from the list of open tables. From the Select Column list, choose a column of data to transfer to the new aggregated file. Enable the Ignore Records Containing Zero check box to include only non-zero records, and click Next. To aggregate coincident points, clear the No Coincident Point Handling check box in the Select Coincident Point Technique dialog box. Choose an averaging technique option, and click Next. In the Aggregation Technique section of the Select Aggregation Technique and Statistics dialog box, choose an aggregation technique. In the Aggregate Point Attributes section, enable any of the check boxes. In the Aggregation Distance box, type the desired distance.

10 Enable the Create Regions Table check box if you want to build a table of regions, and the STD Ellipse Table check box if you want to build a table of standard deviation ellipses. 11 In the File Name box, type a new file name or accept the default. 12 Click Finish. Once the process is complete, an information dialog box opens stating the extent of the aggregation. The aggregated point file appears in a new Map window with a default symbol style applied to each point. If you choose the

127

Chapter 8 Grid Analysis User Guide

Create Regions Table option, the aggregation regions table appears in a separate Map window and is assigned the suffix AggRegion. The calculated statistical information is retained in the Browser window of both the aggregated point file and the aggregation regions file as shown below.

Forward Stepping Aggregation

Due to its speed and effectiveness, the Forward Stepping Aggregation technique is an appropriate method for any general aggregation application. This technique is useful when your data has a truly random distribution or when the other two techniques are not appropriate. This aggregation process aggregates points by moving through the data set from left to right and then top to bottom. The process begins by sorting the data points into rows that are generally three times the aggregation distance. This sorting is performed to determine where the aggregation will begin each left-to-right swath. There are no settings you can use to alter this process. Beginning with the data points in the upper left (northwest) corner of the data set, an inner circular search radius is created based on the Aggregation Distance setting. This search radius is then doubled and all data points that fall inside this outer search radius are selected and flagged. This prevents these data points from being aggregated to another location. The geocenter of the selected points is then determined. If the geocenter falls outside the inner search radius then values at the farthest edge of the outer search radius are dropped and the geocenter is re-calculated. The flag is removed from points that are dropped so that they can be used in other aggregations. This process is repeated until the geocenter falls inside the inner search radius. This becomes the location of the new aggregated point. The process ends by performing the aggregation calculations on the selected points as specified in the Point Aggregation dialog box, and the results are attributed to the new aggregated point.

128

Aggregating Data Grid Analysis User Guide

Once the first point is processed, the procedure sweeps from left to right and top to bottom across the study area, selecting and aggregating unflagged points. It is important to note that not every data point will be aggregated on the first pass through the data set. Normally a second pass is required to aggregate those data points missed on the first pass. The results are shown in Figure 8.3.

Figure 8.3 Circle aggregation using the Forward Stepping Aggregation technique. The shaded points are the original data points. The crosses represent the newly aggregated points, and the circles represent the aggregation region. The original points have been coded to show to which aggregation region they belong.

In Figure 8.3, you may notice that inappropriate aggregation decisions have been made in certain locations of the point file as well as the degree of overlap of the aggregation regions. In the upper left corner of the diagram there are two examples of an inappropriate aggregation, marked by the letters A and B. In both cases you would aggregate these points differently if you performed this process manually. The reason these points are aggregated this way has to do with the two aggregation passes this technique performs; the second pass aggregates the remaining unflagged points, resulting in a large degree of overlap of the aggregation regions. Some of the aggregation regions in the above diagram have been numbered to show the aggregation process order. The letters show which points were aggregated on the second pass.

129

Chapter 8 Grid Analysis User Guide

Cluster density aggregation

The Cluster Density Aggregation technique is typically used when a visual clustering effect occurs in the dispersion of the data points. For example, demographic data representing rural areas may often exhibit a shotgun pattern (Figure 8.4). Data from each small community is considerably more densely distributed than in the surrounding countryside. Cluster density does, however, process large data sets more slowly than the other aggregation techniques.

Figure 8.4 Original distribution of data points.

The Cluster Density Aggregation technique is usually most effective for small to medium size data sets, because it makes better decisions during the aggregation process, i.e., it aggregates the more densely clustered points first within a reasonable processing time. However, it is not appropriate if you need to know the number of points aggregated or if there are coincident points in the data. Coincident points are always aggregated first and are not included in the statistics appended to the new point file. The alternative is to use the Forward Stepping Aggregation technique. The Cluster Density Aggregation technique looks at the entire data set prior to aggregation and determines the single most densely populated area that would fall inside the user-specified search radius (aggregation distance). Points that fall inside the search radius are chosen and flagged. The geocenter of these points is calculated and that position becomes the location of the new aggregated point. Calculations are performed on the values of the selected points (as specified in the Point Aggregation dialog box) and the results are attributed to the new geocentered point. Then, the area with the second highest density of points is chosen and the process is repeated. At each stage, the entire remaining data set must be examined for its density patterns to

130

Aggregating Data Grid Analysis User Guide

avoid using previously aggregated points and to factor in the removed points in the density analysis. Cluster density aggregation always includes coincident point handling regardless of whether or not you request this type of handling, and the point density calculation used in the aggregation technique does not handle points that are coincident. If you do not choose any of the settings on the Coincident Point Handling dialog box, the default values will be used. A warning message appears reminding you when this technique is chosen.

Figure 8.5 Circle aggregation using the Cluster Density Aggregation technique.

There are a total of 330 points in the sample data set shown in the previous figure. After using the cluster density aggregator, you can see that the data is clustered in about ten separate areas. The more randomly distributed points lying outside these clustered zones (circles) are also aggregated, but with significantly fewer points included. Although many circles overlap, the degree of overlap is significantly less than when the forward stepping method is used.

Square bin aggregation

The Square Bin Aggregation technique is typically used when values are required that represent specific areas, e.g., a density map of the number of new housing units per square kilometer, or when you want to avoid overlapping aggregation regions. This method divides the area covered by the point file into adjacent squares determined by the aggregation distance. The points that fall inside any of these squares are aggregated to a new point created at the geocenter of the aggregated points (not at the center of the

131

Chapter 8 Grid Analysis User Guide

square). As with the previous two techniques, the specified statistical information is then attached as attributes to the new aggregated point.

Figure 8.6 Data points aggregated using the Square Bin Aggregation technique.

Although there is no overlap of the aggregation regions in Figure 8.6, points have been aggregated inappropriately in several areas. Therefore, the best results require a certain degree of overlap.

Building a table of standard deviation ellipses

Standard deviation is a measure of dispersion in point patterns. Typically, it measures dispersal in terms of a circle around the mean center. The circular model, however, takes no account of the fact that spread may be different in different directions. The standard deviation ellipse (Figure 8.7) summarizes dispersion in a point pattern in terms of an ellipse rather than a circle. The ellipse is centered on the mean center, with its long axis in the direction of the maximum dispersion and its short axis in the direction of the minimum dispersion. The axis of maximum dispersion in a point pattern is always at right angles to the axis of minimum dispersion. The length of the X axis is set to one standard deviation of the X values, and the length of the Y axis to one standard deviation of the Y values.

132

Aggregating Data Grid Analysis User Guide

Figure 8.7 An example of standard deviation ellipses. The ellipses are generated around the original data points, which are at the mean centers of the ellipses containing them. The X axis extends one standard deviation of the X values, and the Y axis extends one standard deviation of the Y values.

133

Chapter 8 Grid Analysis User Guide

134

9.
This chapter contains the following sections:

Chapter 9: Using Voronoi Diagrams

Using Voronoi Diagrams

Mentum Planet enables you to perform natural neighbor analysis. Using the natural neighbors technique, you can analyze point data that needs to be mapped to discrete regions with constant values assigned to each point.

Understanding natural neighbors Creating regions from points (Voronoi) Calculating the region area

135

Chapter 9 Grid Analysis User Guide

Understanding natural neighbors

While the strength of analysis lies in the ability it gives you to create a continuous grid from non-continuous data points, not all types of data are best represented as a continuously varying surface. Some types of point data should be mapped as discrete regions within which the values assigned to each point are constant. Point data such as this is referred to as having a natural neighbor. Examples include store locations that have natural areas of influence and point observations that record classified values (e.g., land use). In Mentum Planet, natural neighbors are built around data points using Delaunay triangulation. A network of Thiessen polygons is generated from the point locations, creating what is called a Voronoi diagram. As shown below, a Voronoi diagram is a network of regions. Any location within a region will always be closer to the enclosed point in that region than to the enclosed point of any other region. Each site exerts a natural area of influence in relation to adjoining sites. Assigning all the attribute fields from the original point data base to the new region file is a unique feature of Mentum Planet. Once the natural neighbors have been mapped, you can visually analyze and compare any column of information attached to the region table using thematic mapping tools.

Figure 9.1 A Voronoi diagram showing the natural neighbors of a series of point observations. One of the columns of data assigned to the region table has been thematically mapped.

The same natural neighbor relationships that are used to examine the distribution and proximity of point occurrences over space can also be applied to quantify point density. Of common concern through a wide range of disciplines is the examination of the spatial occurrence of certain phenomena.

136

Using Voronoi Diagrams Grid Analysis User Guide

Examples that reflect the extensive range of studies covering such phenomena include the location of towns in a state (geography) or seismic events over a continent (geology). Point distribution over space can be observed qualitatively simply by plotting the data on a map. However, producing a quantitative map of the distribution requires some analysis of the relative proximity of the points in relation to one another. Computing a natural neighbor (Voronoi) diagram for all points yields an excellent measurement of point density. A representative density surface grid can be produced by calculating the area of the natural neighbor region encompassing each point, attaching that area as an attribute to the point, and generating a grid of the new point file through interpolation. This function has been automated in Analysis and is a powerful tool in natural neighbor analysis. The following section describes the two main procedures in Analysis that make use of natural neighbor relationships: building Voronoi diagrams from point tables and calculating point density.

Creating regions from points (Voronoi)

Using the Voronoi technique, you can generate a region around each individual data point. The resulting network of regions is often referred to as a Voronoi diagram. Before running to completion, the point table is checked to ensure that no coincident points are present. If coincident points are found, an error message will appear warning you that the input table contains coincident points. You can edit the table or use data aggregation tools to fix this problem. The extended error code contains the row ID of the first coincident point found.

To create regions from points

1 2 Choose GIS Grid Analysis Natural Neighbor Analysis Create Regions From Points (Voronoi). In the Select Table dialog box, choose a table of points from the Create From Table list, and click Next. All data fields present in the point table are assigned to the new Voronoi region table. 3 In the Voronoi Options dialog box, choose the required settings to control the manner in which the Voronoi diagram is created.

137

Chapter 9 Grid Analysis User Guide

Click Finish. The Voronoi regions table opens in a new Map window.

Figure 9.2 In this Voronoi diagram, the Hull Boundary Margin width setting is very small (1). Therefore polygons are effectively cut off at the outermost limit of the point file.

Figure 9.3 For the same data, the Hull Boundary Margin width has been set to 1000, resulting in the construction of outer polygon edges 1000 meters away from the outermost points. The rounded corners of the outer boundary are controlled by the smoothness setting.

Calculating the region area

The calculate region area command provides a quick way of seeing how densely packed points are. It adds a column to the original data set specifying the area of the Voronoi region of each point. The closer together points are, the smaller the area of the Voronoi region will be.

To calculate the region area

1 2 Choose GIS Grid Analysis Natural Neighbor Analysis Calculate Region Area. In the Region Area dialog box, choose the appropriate table of points that contains the data to be processed in the Create From Table list and click Next.

138

Using Voronoi Diagrams Grid Analysis User Guide

Choose one of the following options from the Boundary Smoothness section:

Coarse Medium Fine

Choose one of the following options from the Hull Boundary Width section:

User Definedenables you to enter a value in the Width box. Select Region From Mapenables you to use the Grid Analysis Select Tool from the Grid Analysis toolbar to select a pre-defined MapInfo region as the Voronoi boundary.

Click Finish. The region area is added as a column to the grid table and displayed in a new Browser window.

139

140

10.
This chapter contains the following sections:

Chapter 10: Creating 3D Views Using GridView

Creating 3D Views Using GridView

GridView enables you to generate 3D renderings of one or more numeric grid files as well as adjust many settings such as the lighting, viewing angles, reflectance, and shading properties. Through the use of drape files, text, line work, colorfilled polygons, and raster imagery, you can render scenes in three dimensions. In addition, you can save any scene as a bitmap image with a corresponding .tab file. This allows you to incorporate 3D scenes within a Layout window.

Introduction Creating a 3D scene Adjusting scene properties Creating a drape file Adding a drape file to a scene Saving a GridView scene

141

Chapter 10 Grid Analysis User Guide

Introduction
Using GridView, you can

create a 3D perspective view of an area from height data, referred to as a scene manipulate the viewing perspective and appearance of the scene optionally, drape visually represented information, such as signal strength data, over the 3D scene save the scene as a bitmap and .tab file for later use

Creating a 3D scene
You can create a realistic 3D perspective view from any Digital Elevation Model (DEM) grid.

To create a three-dimensional scene

1 2 3 Open and display the grid from which you want to create a 3D scene. If the Grid Manager is not visible, choose View Show Grid Manager. In the Grid Manager, click the 3D View button and choose Run 3D Viewer. The 3D Viewer dialog box opens, showing the Scene Properties controls and a preview image of the scene. 4 5 Adjust the scene using the Scene Properties controls. For information on how to do this, see Adjusting scene properties. To render the image, click OK. The Rendering Scene dialog box opens, showing a progress bar. If you want to stop the rendering of the scene, click Cancel. When the scene has been rendered, the GridView window opens, showing a full-size view of the scene. If you want to make more adjustments to the scene, choose View Scene Properties.

142

Creating 3D Views Using GridView Grid Analysis User Guide

Adjusting scene properties

GridView provides a great deal of flexibility in setting the viewpoint, lighting, and surface appearance. You can adjust these characteristics interactively in the Scene Properties dialog box. This dialog opens when you start GridView and is accessible from the View menu after you have rendered a scene.

Setting the viewing mode

You adjust viewing properties on the Viewing tab of the Scene Properties dialog box. GridView provides three viewing modes:

From Your Locationyou specify the three-dimensional coordinates of the observers location and the direction in which the observer is looking. To Your Locationyou specify the three-dimensional coordinates of the location the observer is looking towards (focus point), the direction of the observer relative to the focus point, and the distance between the focus point and the observers location. From and Toyou specify three-dimensional coordinates of both the observers position and the focus point.

You can change the viewing mode at any time. The viewing mode affects only the parameters you can adjust to change the viewpoint. Use the mode that best facilitates the change you want to make. There are both text boxes and slider controls to adjust the parameter values. The preview window is updated immediately when you make changes. You can adjust the Camera Angle to vary the view of your scene from wide angle to close-up. Reducing the angle causes a smaller portion of the scene to fill the view, like using a zoom lens to get a close-up view.

Setting the scene lighting

The quality and direction of lighting can have dramatic effect on the appearance of the 3D scene. You adjust lighting on the Surface and Lighting tab of the Scene Properties dialog box. GridView provides variable amounts of three types of lighting:

Specularlight coming from a particular direction that tends to reflect off of the surface in a preferred direction. Increasing the

143

Chapter 10 Grid Analysis User Guide

amount of specular light results in strong shadow effects and more pronounced shiny or glare spots.

Diffuselight coming from a particular direction that tends to scatter uniformly in all directions when it hits the surface. Increasing diffuse light intensifies shadow effects. Ambientlight that has been scattered so evenly by the environment that its direction is impossible to determine. Increasing the ambient light brightens the scene without casting shadows.

You can set the azimuth and inclination of the light source for specular and diffuse light.

Setting the surface appearance

A rendered scene based on a low-resolution grid might have a grainy appearance. This is the result of the Flat shading option. Each bin is colored to represent its value, and there can be sharp changes in value between adjacent bins. If you choose the Smooth shading option, the color difference between adjacent bins is rendered as a gradient, resulting in a smoother appearance. You can also adjust

Shininessincreasing the shininess value of the surface will increase its overall gleam, while decreasing the shininess value will produce a duller, waxier looking surface. Specularityincrease to produce more surface highlights. Backdrop Colorchoose the color that appears behind the rendered scene. Backdrop Imageenables you to display an image behind the rendered scene. Null Value Colorenables you to set the color applied to areas of a grid where there are null values. A null value indicates that there is no data for that bin. Exaggerationmultiply the height values by a factor to overemphasize height variations.

These properties, except Exaggeration, are available on the Surface and Lighting tab of the Scene Properties dialog box. The Exaggeration property is available on the Loaded Grids tab.

144

Creating 3D Views Using GridView Grid Analysis User Guide

Creating a drape file

You can add visually represented information, such as signal strength data, to your 3D scene. The new information appears to lie on top of the scene and follow its contours. It is called a drape file. You create a drape file in Mentum Planet and then add it to your scene in GridView. These requirements must be met before you can create a drape file:

The grid file upon which the drape will be placed must be present as one of the layers in the current Map window, although it does not have to be visible. The layer or layers to be included in the drape file must be present as a layer in the current Map window. Only the data that falls within the extents of the grid file will be included in the drape. The grid on which the drape is to be placed must be highlighted in the Grid Manager.

To create a drape file

1 Open and display the grids you want in your scene. The elevation grid and the grid or grids that you want to make into a drape file are visible in the same Map window. 2 3 4 5 6 7 8 If the Grid Manager is not visible, choose View Show Grid Manager. In the Grid Manager, select the elevation grid on which the drape file will be placed. Click the 3D View button and choose Make 3D Drape File. In the Information dialog box, click OK. Click in the Map window. In the Drape File Resolution dialog box, from the Grid list, choose the grid on which the drape file will be placed, and click OK. In the Save Drape File As dialog box, navigate to where you want to save the file, type a name in the File Name box, and click Save.

145

Chapter 10 Grid Analysis User Guide

In the Information dialog box, click OK. This process creates two files: a bitmap (.bmp) of the Map window and a drape file (.drp) that contains the registration information needed by GridView.

Adding a drape file to a scene

When you have created a drape file, you can add it to a scene in GridView.

To load a drape file in GridView

1 2 3 4 5 6 7 Display a scene in GridView. For information on how to do this, see To create a three-dimensional scene on page 142. In the Scene Properties dialog box, click Load. In the Load Grid File or Drape File dialog box, choose Drape Files (*.drp) from the Files of Type list. Navigate to the folder containing the drape file, choose the drape file, and click Open. In the Grid list, enable the check box for the drape file so that it is displayed in the preview window. Adjust the scene using the Scene Properties controls. For information on how to do this, see Adjusting scene properties on page 143. Click OK to render the scene.

Saving a GridView scene

You can save your GridView scene, including any drape files you loaded, for later viewing. GridView saves the scene as a bitmap image (.bmp) file that you can view in many graphics programs. It also creates a .tab file in the same folder so that you can open and view the scene in Mentum Planet.

To save a GridView scene

1 2 3 4 In the GridView window, choose File Export to bmp & tab. Navigate to the folder where you want to save your scene. In the File Name box, type a name for the scene. Click Save.

146

Creating 3D Views Using GridView Grid Analysis User Guide

To open a saved scene in Mentum Planet

1 2 3 4 Choose File Open Table. Navigate to the folder where you saved your GridView scene. In the Files of Type list, choose MapInfo (*.tab). Choose the GridView scene file and click Open. The GridView scene opens in a new Map window.

147

148

Index

Index
Numerics C

3D view GridView 142

A

Aggregating data 124 cluster density aggregation 130 common reasons for 124 forward stepping aggregation 128 point aggregation with statistics 127 simple point aggregation 125, 126 square bin aggregation 131 techniques for 125 Alter Meta Data tool 87 Analyzing data calculating point density 58 using gravity models 50 using natural neighborhoods 137 using the Huff model 50, 51, 52 using the trade area analysis model 50 Analyzing grids 18 using spatial analysis 114 Anisotropic modeling 39 ASCII classified grids description 66 ASCII grid description 65
B

Bins isolated 93 Block kriging 34 Buffers creating 61, 63

Classified grids 15 adding color 82 ASCII 66 filtering 93 histogram 76 Cluster density aggregation 130 Color adding to grids 80, 82 Color gradient 71 Converting classified to numeric grids 95 regions to grids 18, 61 Copying grids 71 CRC-500 grids description 67 Creating grids by analysis of other grids 58 by conversion 18, 58, 60 by importing 18, 58 using grid buffering 61 using kernel smoothing 59 using point density 58 using square area 59 Cross sections creating 119 Custom Point Estimation 40 Customizing legends 86

149

Index Grid Analysis User Guide

Data Aggregation 124 cluster density aggregation 130 common reasons for 124 forward stepping aggregation 128 point aggregation with statistics 127 simple point aggregation 125, 126 square bin aggregation 131 techniques for 125 Delaunay criterion 26 Deleting grids 71 Distance Decay Functions 47 drape file GridView 145 DTED grids description 67
E

Estimation techniques 17 Exporting grids 103, 104

F

File types .grc 15 .grd 14 Formats grids 65, 67 Forward stepping aggregation 128 Functions line inspection 120 point inspection 119 region inspection 122
G

Graph data hiding or showing 109 Graphs exporting 111 maximizing 109 printing 110 setting the zoom for 109 Grid buffering 61 to create a grid 63 Grid Calculator 18 opening 114 Grid files description of 13 Grid filter 93 Grid legends creating and viewing 83 Grid Manager Histogram tab 75 Legend tab 75 Grid Query 18 derivative grid 117 Grid tools Alter Meta Data 87 Classified Grid filter 93 Grid Splicer 97

Geological Survey of Canada grids description 67 Geosoft grids description 67

150

Index Grid Analysis User Guide

Grids adding legends 83 analyzing 18 classified 15 converting classified to numeric 95 converting existing 18 copying 71 creating 17, 18, 53 creating from regions 61 definition 14 deleting 71 exporting 103, 104 filtering 93 formats 65, 67 getting information on 76, 77, 78 importing 18, 68 merging 97 numeric 14, 80 process of creating 14 querying 116, 118 reclassifying classified 92 reclassifying numeric 91 resizing 100 resolution 17 splicing 97 stamping 98 types of 14 viewing 82 GridView adding a drape file 146 adjusting scene properties 143 creating a 3D scene 142 creating a drape file 145 saving a scene 146
H

Hiding or showing graph data 109 legends 109 Histogram classified grid 76 numeric grid 76 Huff model 51 example 52
I

Importing grids 18 Information in unknown area 14 on grids 76 Interpolation 30 Interpolation techniques 17, 40 choosing 21 Custom Point Estimation 21 Inverse Distance Weighting 20, 27 Kernel Smoothing 59 Kriging 31, 32 Natural Neighbor 20, 28 Rectangular (bilinear) 21, 30 suggested readings 40 Triangulation with Smoothing 20, 25 understanding 20 using the wizard 24 Interpolation wizard 24 Inverse Distance Weighting 27 Isolated bins 93
K

Help accessing 6 getting technical support 4

Kernel smoothing 59 Kriging anistropic modeling 39 changing the semivariogram model 38 creating a semivariogram 34 interpolation process 34 modifying directional parameters 37 suggested readings 41

151

Index Grid Analysis User Guide

Kriging interpolation 31 how it works 32 ordinary 33 punctual or block 34 simple 33 universal 33 variations 33
L

Modeling Wizard 53 Modifying legends 84 MONA grids description 67

N

Legends classified grid 75 creating 83 customizing 86 hiding or showing 109 holding 86 modifying 84 numeric grid 75 Quick Map method 84 Line Info tool 78 Line inspection 120 Loading reclass profile 93 Location Profiler 44 creating buffers 61 decay curves 48 point weighting 46 setting the number of points 45 using 49
M

Natural Neighbor Analysis 137 calculate region area 138 Natural Neighbor interpolation 28 Numeric grids 14 adding color 82 display and color options for 80 histogram 76
O

Online Help 6 Ordinary kriging 33 Overshoots 24, 25, 28, 30

P

Maximizing graphs 109 Merging grids 97 Meta data altering 87 MIG grids description 68 Modeling techniques description of 17 Location Profiler 44

Planet products 1 Point Aggregation With Statistics 127 Point density calculating 58 using kernel smoothing 59 using square area 59, 60 Point Inspection 119 Points creating regions from 137 setting in the Location Profiler 45 weighting 46 Preferences setting 18 Printing graphs 110 Profiles for reclassified grids 93

152

Index Grid Analysis User Guide

Punctual kriging 34
Q

Querying grids 116 creating and editing 118 Quick Map Legends 84
R

Reclass profiles opening 93 Rectangular interpolation 30 Region area 138 Region Info tool 77 Region inspection 122 Region to Grid conversion 18, 60 Regions creating grids from 60 grid query 60 Relief shading 71 Resizing grids 100 Resolution 17
S

Simple kriging 33 Simple Point Aggregation 125, 126, 127 Slope and Aspect 18 Spatial analysis 114 Splicing grids 97 Square Area creating point density grid 59, 60 Square bin aggregation 131 Stamping grids 98 Standard Deviation Ellipses 132
T

Technical support 4 TIN interpolation 20, 25 Trade Area analysis creating a model 53 description 50 suggested readings 55 Triangulation with Smoothing interpolation 25 Delaunay criterion 26 Types of grids 14
U

scene properties GridView 143 SDTS grids description 68 Semivariograms changing 38 creating 34 tuning the model 38 Showing or hiding graph data 109 legends 109

UK Ordnance grids description 67 Undershoots 24, 25, 28, 30 Universal kriging 33 USGS DEM grids description 67
V

Viewing grids Dictionary Editor 82 Voronoi technique 137

Z

Zoom 109

153

Index Grid Analysis User Guide

154