# Introduction

Online WhatIf? (OWI) is a GIS-based online, open source planning support system (PSS) that has been developed as part of the Australian Urban Research Infrastructure Network (AURIN) workbench. The model implements the desktop WhatIf PSS developed by Richard Klosterman (1999, 2008).

OWI uses existing GIS data to support some of the most important and difficult aspects of the land planning process: analyzing the suitability of land for different uses, projecting future land use demands, and allocating the projected demands to the most suitable locations. It can also be used to prepare long-term projections for future land uses, population and housing, and employment for a user-defined area.

OWI provides a single integrated package that allows planning tasks that previously required weeks or even months of effort to be completed quickly and easily. It is easy to use, customized to a community’s data, concerns, and desires, and provides outputs in easy to understand maps and reports which can be used to support community-based collaborative planning.

#### System Overview

OWI does not attempt to predict future conditions exactly. Instead, it is an explicitly policy-oriented planning tool that can be used to determine what would happen if the user’s policy choices are made and their assumptions concerning the future are correct. Policy choices that can be considered in the model include the staged expansion of public infrastructure and the implementation of alternative land use plans or zoning ordinances. Assumptions for the future that can be considered in the model include future population and employment trends, assumed household characteristics, and anticipated development densities.

OWI implements a “bottom-up” model which begins with homogeneous land units or uniform analysis zones (UAZs), applies alternative public policy choices to these units, and then derives information for the study area by aggregating the smaller land units. UAZs are GIS-generated polygons which are homogeneous in all respects considered in the model. Thus, for example, all locations within a UAZ have the same slope, are located in the same municipality, have the same environmental value, are within the same distance of an existing or proposed highway, and so on.

UAZs are created by using a GIS to combine all of the relevant layers of information to be used in an OWI project. The map layers can contain information on natural conditions (e.g., slopes, soils, and scenic vistas), existing and proposed infrastructure (e.g., the proximity to major roads and the availability of sewer and water service), and land use controls such as zoning districts and planned land uses. OWI combines these layers to produce UAZs which contain information from each of the constituent layers, i.e., each UAZ contains information on the slope, the availability of sewer and water service, planned land use, and so on for the locations inside it.

OWI projects future land use patterns by balancing the supply of, and demand for, land suitable for different uses at each location. Alternative visions for an area’s future can be explored by defining alternative suitability, demand, and allocation scenarios. For example, a “Trend” scenario could determine the effects of continuing current development policies. An “Environmental Protection” scenario might consider the impact of policies that severely limit growth in scenic areas and on land highly suitable for agricultural uses. The assumptions underlying a scenario can be easily modified to incorporate the full range of alternative visions for an area’s future.

The results generated by considering these alternative scenarios provide concrete and understandable expressions of the likely results of a scenario’s underlying policy choices and assumptions. For instance, the model might demonstrate that there is insufficient land simultaneously to accommodate high growth, low residential densities, and strict agricultural protection policies, thereby forcing the community to choose between highly desirable, but inconsistent, policy goals.

#### Differences between Desktop What If? and Online WhatIf

Users of the desktop What If? should be aware that Online WhatIf incorporates many differences from the desktop version. For instance, the What If? Desktop version requires the user to customize the program to reflect the kind of analysis the user aims to perform, i.e. the Suitability Analysis, Land Use Analysis, Land Use/Population Analysis, or Land Use/Population/Employment analysis modes. The Analysis mode will have an impact on the input data requirements and on the project setup. Online WhatIf does not require the user to specify upfront the kind of analysis that will be performed. For simplicity, the steps followed during Project Setup in Online WhatIf are identical regardless of the kind of analysis the user wishes to perform. If a certain type of data is not available in the input UAZ file, then the user may skip or specify “none” in that particular step of the Project Setup.

Online WhatIf does not include the U.S.-specific concept of population housed in “group quarters” nor does it support multiple different levels of data aggregation (for example, using sub-area boundaries such as political jurisdictions and Enumeration District) for land use, population or employment projections. This requirement can be accommodated to some extent in OWI by exploring allocation mechanisms.

Finally, we have amended some of the computational procedures found in the land demand and land allocation algorithms. For a more detail description of such amendments, see Appendix A.

#### Online WhatIf System Components

The Online WhatIf system includes two components. The first component, the Setup component, is used to define the program settings and values for a particular WhatIf project. The second component, the Scenarios, is used to create the scenarios for determining the suitability of land, projecting the demand for land, and projecting future land use, population and employment patterns.

This guide will first describe how to access the OWI system. It then illustrates the procedure for using the Setup component to create an OWI project. Finally, it describes the process for using the Scenarios component to create a collection of suitability, demand, and allocation scenarios.

# Accessing the Online WhatIf System

OWI is available at the following Web address: https://whatif.aurin.org.au/whatif/login . The system requires a user account which can be arranged upon request with the AURIN team. Please note that the OWI system is supported by Firefox, Chrome or Safari natively only.

The OWI start up screen is displayed in Figure 1. If the Projects window is not visible, click on the Project List icon at the top left corner of the OWI start up screen. The project list will include the Wanneroo Demonstration project that will be used in this Guide. Any new projects you create will be stored on your Project List and can be accessed anytime by using your login credentials.

[Click to Enlarge]

# OWI SETUP

The OWI Setup component includes the following options:

• Project Setup
• Demand Setup
• Allocation Setup
• Demographic Trends

These four options are described in the following sections.

## Using Project Setup

The following steps are required to create a new OWI project:

• Defining project parameters;
• Specifying UAZ data fields;
• Defining existing land uses;
• Defining suitability land uses;
• Defining suitability factors; and
• Finalizing Setup.

The procedure for conducting each of these setup steps is described below.

#### Defining Project Parameters

This Quick Start Guide will use sample data from the Wanneroo Local Government Authority (LGA), a political division of Perth Metropolitan Area in Western Australia. The GIS data for the demonstration project is contained in the Wanneroo.zip file which you should download from here and store on your computer. Wanneroo_UAZ.zip is a compressed (zip) file containing a polygon shapefile containing the parcel-level information for the Wanneroo LGA.

You should begin by clicking on New Projects option at the top of the Projects screen. This will display the OWI Setup screen shown in Figure 2 below. To define the project parameters for your new project you should

• Enter Wanneroo for the Project Name;
• Enter Wanneroo WA in the Study Area text box.

[Click to Enlarge]

• Click on the Browse button to locate the Wanneroo_UAZ.zip file on your computer and click on Open to add the file path and name to the New UAZ file text box;
• Click on OK when the Success box pops up;

#### Specifying UAZ Data Fields

You will then see the UAZ Data Fields screen shown in Figure 3. You can use this screen to identify the fields in your UAZ that contain the existing land use and area measurement data to be used for your project. To do this you should:

• Select the ExistingLU field from the Select existing land use field dropdown list;
• Select Area_ha from the Select area measurement field; and
• Click on Next.

[Click to Enlarge]

#### Defining Existing Land Uses

You will then see the Existing Land Uses Setup screen shown in Figure 4. The first column, Existing Land Use Label, identifies the label that is used in the existing land use field you selected. You can use the second column, New Land Use Label, to define new labels for any of the existing land uses. You can use the Not Dev. (Not Developable) check boxes to identify land uses such as highways and waterways that cannot be developed. You can also use the Not Dev. check boxes to identify land use types that will not be developed.

[Click to Enlarge]

To enter these assumptions you should:

• Click on Initialise Values and Yes on the Overwrite? message prompt to clear the current project settings;
• Click on the Not Dev. check box for (1) Commercial, (2) Education, (3) Hole, (4) Industrial, (5) Infrastructure / Public Purposes, (6) Nature / Passive Recreation, (7) Private Recreation, (8) Railway existing, (9) Railway Proposed, (10) Recreation, (11) Residential, (12) State Forest, (13) Transport, and (14) Water land uses; and
• Click on Next at the bottom of the screen.

#### Defining Suitability Land Uses

You can now use the Suitability Land Uses Screen shown in Figure 5 to define the land uses that will be used in the OWI suitability scenarios. You can define as many suitability land uses as you wish; this quick study guide will define four suitability land uses. Each suitability land use is associated with a single existing land use.

[Click to Enlarge]

To define the suitability land uses you should:

• Click on the Add New Item button;
• Key Residential into the Suitability Land Use text box and select Residential from the Associated Existing Land Use drop down list;
• Key Commercial into the Suitability Land Use text box and select Commercial from the Associated Existing Land Use drop down list;
• Key Industrial into the Suitability Land Use text box and select Industrial from the Associated Existing Land Use drop down list;
• Key Medical and Educational into the Suitability Land Use text box and select Education from the Associated Existing Land Use drop down list; and
• Click on Next.

#### Defining Suitability Factors

You can then use the Suitability Factors screen shown in Figure 6 to define the suitability factors that will be used in your analysis.

Suitability factors are the characteristics of the land that will be considered in determining the relative suitability of different locations for each suitability land use. The suitability factors to be considered are specified by the user and can include the full range of natural features including slopes, soils, flood plains, and landslide prone areas. They can also include other suitability factors such as the distance to amenities such as roads and parks or to disamenities such as hazardous waste sites.

Factor types are the possible values for a particular suitability factor. For example, a study that used a suitability factor Distance to Transport might include three distances thresholds: (1) < 400m; (2) < 800m; and (3) > 800m.

This example will consider four suitability factors: (1) ActivityCentre; (2) UrbanValue, (3) Environmental Value; and (4) UrbanExpansion.

[Click to Enlarge]

To define these suitability factors you should:

• Click on Add New Item to add a new suitability factor;
• Select the ActCentre value from the UAZ Field drop down list and enter ActCenter in the Factor Label field;
• Select the UrbanValue value from the UAZ Field drop down list and enter UrbanValue in the Factor Label field;
• Select the Env_Value value from the UAZ Field drop down list and enter EnviroValue in the Factor Label field; and
• Select the UrbanExpan value from the UAZ Field drop down list and enter UrbanExpansion in the Factor Label

You can use the red delete buttons to delete any factor types that you may have mistakenly added to the factor list.

If you select one of the suitability factors added, its factor types will be displayed on the right side of the screen upon clicking Initialize Values. The UAZ Records field lists the unique values stored in the selected UAZ field. You can enter a new label for the selected factor value in the Values Label textbox. You can click on Initialize Values to reset any changes you made to the factor value labels. Click on Next when you are finished specifying the suitability factors and factor values.

#### Finalizing Setup

You will then view the Finalize Setup screen shown in Figure 7. Click on Finalise Setup to complete the project setup process.

[Click to Enlarge]

## Using Demand Setup

The OWI has a model to estimate future residential land demand and employment-related land demand. However, to do this, you must have past and current information related to population, housing and employment – by sector – for the study area. Such information can be derived for the Australian Bureau of Statistics (ABS). In addition to the current demographic and employment data, the demand projection model uses two past years point data entries.

The Demand Setup is a seven-step process designed to help the user to input current and past information related to the population, housing and employment. These steps are:

• Specify Current Demographics;
• Specify Residential Current Data
• Residential Past Trend Data: Data for two previous years
• Projection Years
• Employment Sectors
• Current Employment
• Employment Past Trend Data: Data for two previous years
• Finalize Setup

#### Specifying Current Demographics

You should begin by entering the current demographic data for your study area, as shown in Figure 8. The following data input should be entered through the user interface:

• Enter 2011 for the Current Year.
• Enter 152083 for Total Population
• Enter 56362 for Housing Units
• Enter 51987 for Households
• Enter 0.1 for Vacancy Rate
• Click on Next

[Click to Enlarge]

#### Residential Current Data

You will then see the Residential Current Data screen shown in Figure 9. Please note that this interface allows you to input data regarding current residential land uses and their respective breakdown densities (e.g. Low, Medium and High residential densities). First, you need to identify the Land Use types in your UAZ land use field that is associated with residential use only. Then, you should provide information related to the Number of Housing Units, Density, Breakdown Density and Vacancy Rate for each existing residential Land Use type. Breakdown Density is the proportion of housing that falls within each density band category. For simplicity, in this example we will consider one single residential land use, with an average density. In this screen you should:

• Click on the Add New Item button and Select Residential from Existing Land Uses drop box.
• Enter the values:
• 56362 for Number of Housing Units;
• 15.12 for Density;
• 1 for Breakdown Density; and
• 0.1 for Vacancy Rate
• Click on Next.

[Click to Enlarge]

#### Residential Past Trend Data (enter data for two previous years)

As shown in Figure 10, you can use this screen to input historic population and household data for your study area. These data are essential for estimating future land demand for residential use. Trend projections assume that growth rates that are observed in the past will continue into the future. Hence, you will need to enter two different years of observed/recorded data values for population and households. To do this, you should click on Add New Item twice and add the values as displayed in the table below. Click on Next.

#### Household

2001 81215 27366
2006 110941 38290

[Click to Enlarge]

#### Projection Years

You can then use the Projection Years screen to specify the years to which future land uses will be projected. To enter this information you must click on Add New Item and key in the future projection years, as shown in Figure 11. After entering this information, click on Next.

NOTE: the intervals between current year (baseline year) and projection years need to be constant.

[Click to Enlarge]

#### Employment – Sectors

This screen is used to specify the assumptions related to the study area’s existing employment sectors and their associated land use types, as shown in Figure 12. These assumptions will be used in computing the future demand for employment-related land uses, i.e., land use demands that are dependent on the number of people employed in the study area, not on the number of people who reside in that area.

To specify employment sectors and associated them to land use type(s) you should:

• Enter the value 01 in the Code field;
• Key Manufacturing in the Label field;
• Select Manufactur from the UAZ Field drop down list;
• Associate the sector above with land use type Industrial by selecting the values from the Associated LU drop down list;
• Repeat steps 1 to 5 for all the sectors as shown in Figure 12;
• Click on Next.

[Click to Enlarge]

#### Current – Employment

This screen is used to enter the current number of employees per sector. For each of the employment sectors entered in the previous step you should enter the number of employees in the current year, as shown in Figure 13. Click on Next.

[Click to Enlarge]

#### Employment: Past Trend Data

You can then use the Employment Past Trend Data screen to input historic employment numbers for each sector in the study area. Similarly to residential trend projections, employment trend projections assume that the growth rates that are observed in the past will continue into the future. Hence, you will need to enter two different years of observed/recorded data values for the employment per sector, as shown in Figure 14. To do this, you should click on Add New Item twice and enter the values as displayed in the table below. Click on Next.

#### Year 2006

Manufacturing 10438 14494
TranspUtility 1825 2442
Recreation 437 585

[Click to Enlarge]

#### Finalise Demand Setup

You will then view the Finalize Demand Setup screen shown in Figure 15. Click on Finalise Setup to complete the demand setup process.

[Click to Enlarge]

## Allocation Setup

WhatIf projects future land use, population and employment patterns by allocating the projected land use demands derived from a user defined demand scenario to different locations on the basis of their relative suitability as defined by the assumption in a user-selected suitability scenario and any allocation controls specified in the allocation scenario assumptions.

For instance, the projected residential land use demand can be assigned first to the most suitable sites for residential development, then to the second-most suitable residential sites, and so on until all of the projected residential demand in each projection year have been satisfied. The projected demand for employment is allocated in an identical way. If desired, the demand allocation process can be guided/restricted/prioritized by one (or a combination of) user-selected public policies such as the implementation of land use plan, planning schemes, infrastructure expansion, growth patterns, etc.

The Allocation Setup is used to specify possible assumptions which may underlie an allocation scenario. These are the steps to be followed in the Allocation Setup:

• Specifying Planned Land Use field;
• Specifying Land Use colours
• Associating Planned Land Use with Existing Land Use;
• Specifying Infrastructure Control fields;
• Specifying Growth Pattern Controls fields; and
• Finalizing Setup.

The procedure for conducting each of these setup steps is described below.

#### Planned Land Use – Select Field

On opening the Allocation Setup you will see the Allocation Configuration Planned Land Use screen as shown in Figure 16. If there is previously defined Land Use Plan or “vision plan” that will be assumed to control the location of future demand then you should select the field in the UAZ file that represents Planned Land Use. When this option is selected, the projected land use demands will only be allocated to locations for which they are planned in the Land Use Plan. In this case study you should select the Plan_LU field from the Planned Land Use Field Name dropdown list. Click on Next.

[Click to Enlarge]

#### Land Uses Colour scheme

The screen shown in Figure 17 can be used to specify the colour scheme used to produce maps. WhatIf automatically assigns the colours for each Land Use Type using values previously stored in the UAZ database (the field can be named colour or color, and it is not case sensitive). If no colour code scheme (HEX Colour Code) is stored in the UAZ file then the user is required to manually enter a unique HEX Colour Code for each Land Use Type via the user interface, as displayed on Figure 17. Click on Next.

[Click to Enlarge]

#### Planned Land Uses – Details

You will then see the Planned Land Uses – Details screen shown in Figure 18. You can use this screen to associate Planned Land Use with Existing Land Use. The first column, label, identifies the label that is used in the planned land use field you selected in the previous step. The second column, associated ALUs, is a drop down box with a list of all existing land uses. You can use the drop down box to select all existing land uses that correspond to each of the planned land uses (see Table 3 for a specification of the Existing Land Uses that correspond to the Planned Land Uses). Click on Next.

[Click to Enlarge]

#### Existing

Activity Centre
• Commercial
• Urban Land
Commercial
• Commercial
• Rural
• Urban Land
Developed
• Commercial
• Open Space – Passive Recreation
• Rural
Education / Training
• Education
Established Residential
• Residential
Hole
• Hole
Industrial
• Industrial
• Urban Land
Infrastructure / Public Purposes
• Infrastructure / Public Purposes
• Education – Convertible
• Industrial – Convertible
• Open Space – Passive Recreation
• Rural
Nature / Passive Recreation
• Nature / Passive Recreation
Private Recreation
• Private Recreation
Railway Existing
• Railway Existing
Railway Proposed
• Railway Proposed
Recreation
• Recreation
Rural
• Rural
State Forest
• State Forest
Urban Land
• Urban Land
• Commercial
• Education – Convertible
• Industrial – Convertible
• Open Space – Passive Recreation
• Rural
Transport
• Transport
Water
• Water

#### Infrastructure Control

You can now use the Infrastructure Control Fields Screen shown in Figure 19, to define infrastructure plans affecting regions within the study area that will be provided with a particular kind of infrastructure in a given year. Examples of Infrastructure Plans are Sewer Plan, Water Plan, Utility Plan, Major Roads Plan, etc. In this case study we will consider the impact of the Major Roads Plan on the land allocation process. This means that each land parcel in the UAZ file contains information regarding the year from which Major Roads will be accessible from that particular parcel. If this attribute is empty, that means the parcel of land already has access to major road. WhatIf can use this information to allocate future land uses to parcels beginning with the year in which infrastructure becomes available. To add the road infrastructure control you should:

• Click on Add New Item button;
• Enter InfrastructureControl into the label text box and select Inf_Trans from the field name drop down list;
• Click on Next.

[Click to Enlarge]

#### Growth Patterns Control Fields

The Growth Patterns Control Fields screen shown in Figure 20 allows you to specify a pattern in which different locations will be developed. In this case study we have set up the data for two possible growth patterns: (1) the “ActTranspGrowth” growth pattern which allocates future growth to buffers lying alongside public transportation and Activity/Commercial centres, concomitantly; and (2) “TransportCorridor” growth patterns which allocates future growth to buffers around public transport corridors. To define Growth Patterns you should:

• Click on the Add New Item button;
• Key TransportCorridor into the label text box and select TransGrowt from the field name drop down list;
• Click on the Add New Item button;
• Key ActTranspGrowth into the label text box and select GrowthUtil from the field name drop down list;
• Click on Next.

[Click to Enlarge]

#### Finalizing Allocation Setup

You will then view the Finalise Allocation Config screen shown in Figure 21. Click on Finalise Setup to complete the allocation setup process.

[Click to Enlarge]

## Demographic Trends Setup

WhatIf provides two approaches for projecting the residential population and employment that will determine the future demand for various land uses. The Growth Trend, accessible through the Demographic Trends Setup component (see Figures 22 to 24), is automatically generated by the system by extrapolating user-specified past population and employment values (refer to item Demand Setup) into the future.

[Click to Enlarge]

[Click to Enlarge]

[Click to Enlarge]

There is also the option for the user to manually enter a different population and employment projection than has been previously estimated. This option may be useful when forecasts for the area’s growth are available and you want to consider the implications these values would have on the area’s future development patterns. The following steps are required to manually create a new Demographic Trend:

• Add a New Demographic Trend;
• Input residential-related trend values;
• Input employment-related trend values;
• Finalise Setup.

Each of these steps is described below.

#### Add a New Demographic Trend

You can use the Demographic Trend Names Screen shown in Figure 25 to create a new demographic trend. You can create as many demographic trends as you wish. To do this you should:

• Click on the Add New Item button;
• Key External Trend into the demographicTrend text box.
• Click on Next

[Click to Enlarge]

#### Residential Trends

You can now use the Demographic Trends screen shown in Figure 26 to input the residential related demographic data trend. To do this you should:

• Select the External Trend value from the Choose trend drop down list;
• Key the values shown in Figure 26 into the totalPopulation, housingUnits and averageHouseholdSize fields for each projection year;
• Click on Next.

[Click to Enlarge]

#### Employment Trends

You can then use the Employment Trends screen shown in Figure 27 to input employment projections. To do this you should:

• Select the External trend value from the Choose Trend drop down list;
• For each employment sector listed in the Item Column, enter the projected number of employees for each projected year into the appropriate text boxes. Example values are shown in Figure 27;
• Click on Next.

[Click to Enlarge]

#### Finalise Demographic Trends Setup

You will then see the Finalise Demographic Trends Setup screen as shown in Figure 28. Click on Finalise Setup to complete the Demographic Trend Setup process.

[Click to Enlarge]

# Using OWI Scenarios

The Scenarios component provides procedures for the users to create new scenarios or modify/delete existing ones. As shown in Figure 29, there are five types of scenarios that can be created in WhatIf. They are:

• Suitability Scenario;
• Demand Outcome;
• Demand Scenario;
• Allocation Control Scenario;
• Allocation Scenario;

The procedure to create each of these scenarios is described hereafter.

[Click to Enlarge]

## Suitability Scenarios

The Suitability Scenarios component provides procedures for determining the relative suitability of different locations for different land uses. The scenario suitability analysis implemented in WhatIf uses Multi-criteria Evaluation techniques to calculate the Suitability Score, a numeric value, indicating a location’s overall suitability for a given land use. Creating and using the OWI suitability scenarios involves the following steps:

• Creating a suitability scenario;
• Specifying land use conversions;
• Specifying importance weights and suitability ratings;
• Computing suitability scores;
• Viewing suitability maps; and
• Viewing suitability reports.

The procedures for performing each of these steps are described below.

#### Creating a Suitability Scenario

The following steps are required to create a new suitability scenario:

• Expand the tree for the Wanneroo project in the Projects window to show the Setup and Scenarios branches;
• Right click on the Scenarios branch and click on the New Suitability Scenario option as shown in Figure 29; and
• Enter Suitability_Scenario_1 in the Scenario Name form and click on OK.
• Right click on the Suitability_Scenario_1 to open the suitability scenario.

[Click to Enlarge]

After opening a Suitability Scenario you are presented with the Suitability Analysis screen shown in Figure 30. This screen has three tabs:

• Convertible LUs;
• Suitability factors; and
• Map

#### Specifying Land Use Conversions

The first step of the Suitability Analysis is to specify the permitted land use conversions, i.e., the land uses that may be converted from their current use (e.g. Urban Land) to another use (e.g. Residential) during the land use allocation process. You should use the Convertible LUs tab to specify the land uses that can be converted from their current use to a different use during the land development process. Figure 30 shows land use conversions allowed in the Wanneroo Project.

N.B. Land use types that have been defined as “Not Developable” during project set up (refer to section 3.1.4 and Figure 4) will be automatically set to not convertible. Thus, their checkboxes will be inactive.

#### Specifying Suitability Factors

The Suitability Factors screen shown in Figure 31 can be used to specify:

• The importance weights for each suitability factor;
• The suitability ratings for each factor type.

Suitability Factors are the characteristics of the land which will be considered in determining the relative suitability of different locations for a particular land use. The suitability factors to be considered are specified by the user and can include natural features (e.g. slopes, soils, flood plains, landslide prone areas) and other suitability factors such as the distance to amenities such as roads and parks or to disamenities such as hazardous waste sites.

Factor Types are the set of possible values for a particular suitability factor. This, for example, the Wanneroo Project considers six different Urban Values types: (1) Very High; (2) High; (3) Medium; (4) Low; (5) Very Low; and (6) Unknown.

Importance Weights are numerical scores indicating the relative importance of different suitability factors for determining the suitability of different locations for a particular use. Thus, for example, the Urban Value factor could be given an importance weight of 100 and Environmental Value 50 to indicate that UrbanValue is twice as important as the Environmental Value for determining the suitability of locations for that land use. An importance weight of zero (0) can be used to indicate that a suitability factor should not be considered.

Suitability Ratings are numerical values indicating the relative suitability of locations with a particular factor type for locating a specified land use. Thus, for example, suitability ratings can be specified for each of the six Urban Value (Very Low; Low; Medium; High; Very High; Unknown). In this case Urban Value – Very Low will be given a low suitability rating for commercial, residential and Medical/Educational land use types but high rating for the Industrial land use type (see Figure 31). Please notice that a rating of 0 is given to Urban Value – Unknown, indicating that land with unknown Urban Value should be excluded from any development, regardless of other Factors, their Types and Weights.

Figure 31 shows the importance weights and ratings assigned to all suitability and suitability factors in the case study.

Different Suitability scenarios can be created by assigning different weights and/or ratings.

[Click to Enlarge]

#### Computing Suitability Scores

Suitability scores are numerical values indicating different locations’ suitability for a given land use when all of the suitability factors are considered. The suitability score for a particular UAZ is determined by: (1) multiplying the UAZ’s suitability rating for each factor by the corresponding importance weight; and (2) summing these products. Suitability scores are computed for all UAZs and all land uses. OWI automatically computes the suitability scores when you click on the Suitability Map tab.

#### Viewing Suitability Maps

Clicking on the Suitability Map tab displays the suitability map shown at different zoom levels in Figure 32 a. and b. You can use the radio buttons at the top of the form to select the land use to be displayed and modify the basemap on which the suitability map is overlaid. For example, the Commercial radio button has been selected in Figure 33 and so the map shows the relative suitability of different locations for accommodating commercial development. As shown in Figure 32 areas that are not suitable for the selected land use are shown in yellow and areas that are not developable / convertible are not displayed on the map at all.

[Click to Enlarge]

[Click to Enlarge]

#### Viewing Suitability Reports

You can now view the following reports on the suitability scenario’s assumptions and results:

• Land Use Conversions report;
• Factor Weights and Ratings report; and
• Factor Scores report.

The procedures for viewing these reports and their contents are described below.

##### Viewing Land Use Conversions Report

You can click on the Convertible Report button to download the Land Use Conversions Report as shown in Figure 33. This option creates an Excel-compatible file showing the land uses conversions you specified on the Land Use Conversions screen shown in Figure 4. The report also shows the project name and suitability scenario name.

[Click to Enlarge]

##### Viewing Factor Weights Report

You can click on the Factor Weights button to view the Factor Weights report shown in Figure 34. This option creates an Excel-compatible file showing the factor importance weights and factor ratings you specified on the Factor Weights and Ratings screen shown in Figure 9. The report also shows the project name and suitability scenario name.

[Click to Enlarge]

##### Viewing Suitability Score Report

You can download and view the Suitability Score report in Excel or PDF formats by clicking on the Excel Report button or PDF Report button, respectively. Figure 35 shows the Excel-compatible file showing computed suitability scores for each suitability land use. It also shows the amount of the land in a qualitative scale that ranges from Low to High Suitability.

[Click to Enlarge]

## Demand Scenarios

The Demand Scenarios function is an important part of WhatIf. It provides procedures for projecting the future demand for land, based on different assumptions regarding population and employment growth.

There are two different ways to create Demand Scenarios in the Online WhatIf:

• Demand Outcome; and
• Demand Scenario.

The procedures for performing each of these steps are described hereafter.

#### Demand Outcome

The demand Outcome is the simplest and most straightforward way to create a Land Demand Scenario. It consist of manually inputting the amount of land required for each of the land use types in each of the projection years considered. To create a Demand Outcome Scenario you should:

• Expand the tree for the Wanneroo project in the Projects window to show the Setup and Scenarios branches;
• Left click on the Scenarios branch and click on New Demand Outcome option;
• Enter DemandOutcomeScenario1 in the Scenario Name pop-up text box and click on OK. The DemandOutcomeScenario1 form will automatically open up, as shown in Figure 36.
• Key in the amount of land (Area) required for each land use type and respective projection years as shown in Figure 36;
• Click on Save and close the window.

[Click to Enlarge]

#### Demand Scenario

The Demand Scenario function takes the population and employment growth projections defined in the Demand Setup (see section 4.2) and computes the amount of land necessary to accommodate that projected growth in both types of land demand. The first land demand, residential-related land use demand, future demand for residential land is estimated for each projection year. The second land demand, employment-related land use demand, land demand to accommodate workplaces is estimated for each projection year. Examples of the employment-related land uses include industrial, commercial, educational, medical and agricultural uses.

To create a Demand Scenario you should:

• Expand the tree for the Wanneroo project in the Projects window to show the Setup and Scenarios branches;
• Right click on the Scenarios branch and click on New Demand Scenario option;
• Enter DemandScenario1 in the Scenario Name pop-up text box and click on OK. The DemandScenario1 form will automatically open up, as shown in Figure 37.

[Click to Enlarge]

There are 5 steps to create a Demand Scenario. These steps are:

• Information;
• Land Use;
• Employment;
• Density;
• Report

##### Information

After creating a new (or opening an existing) Demand Scenario you are presented with the Demand Scenario Form as shown in Figure 37. The first tabbed sheet, Information, is used to specify the population and employment projection as the underlying assumption for the calculation of future land demand.

As shown in Figure 37, the Information sheet contains a drop-down list labeled Choose Demographic Trend that can be used to select one of a number of previously computed projections for the future population. The number of projections listed is dependent on the information that was specified in the Demand Setup, described in Section 3.2. In this Project, we have specified two alternative population and employment projections: the Growth Trend and the External Trend.

To specify the population projection for the DemandScenario1 you should:

• Click on the drop-down list of the Information tab sheet
• Select “ExternalTrend
• Click on the Save button at the top of the Screen.

[Click to Enlarge]

##### Land Use

You can then use the second sheet, Land Use, to specify the residential land use information and assumptions that will help determine the future demand for residential land. To do that, you should:

• Click on Add New Item button;
• Select from the Land Use drop-down list a land use type associated with residential land use, as shown in Figure 38;
• Key in a number between 0 and 1 in the column Future Breakdown by Household Type. This number specifies the future percentage breakdown by housing type. In our example, we are only considering one single (average) residential land use. Thus, you should enter the number 1. However, if more than one Residential Land Use type is considered (for instance, Low Density, Medium Density and High Density Residential Land Use types) then the sum of all Future Breakdown by Household Type must equal 1.
• Key in the value for current residential density (Number of housing units/Residential Land UseArea) in the Current Density column, as shown in Figure 38.
• Key in the value for future residential density (Number of housing units/Residential Land UseArea) in the Future Density column, as shown in Figure 38.
• Key in a value between 0 and 1 that represents the Future Vacancy Rate. Vacancy Rate is the proportion of housing that will be vacant and available for sale (i.e. surplus housing). In this case we will consider a vacancy rate of 10%, thus, you should enter the value 0.1.
• Key in a value between 0 and 1 that represents the Infill Rate. Infill Rate is the proportion of new housing units that are assumed to be allocated as “infill” to areas that are already devoted to residential uses. The land use demand for these “infill units” are deducted from the total residential demand in any given year. In this case we will consider an infill rate of 10%, thus, you should enter the value 0.1.
##### Employment

You can then use the Employment sheet to specify the assumptions to be used to compute the future demand for employment related land uses, i.e., demand whose size is dependent on the population that works in the study area. The first assumption is related to Infill Rate, i.e. the percentage of the employment related demand that will be used to infill areas that are currently devoted to employment related land uses. The infill demand will be deducted from the projected land demand in each projection year.

Figure 39 shows the Employment-related Infill Rates considered in the DemandScenario1.

[Click to Enlarge]

##### Density

The Employment Density sheet shown in Figure 40 can be used to specify the future density (employees per hectare – or any other area measure) for each employment sector. Current densities are automatically calculated by the system.

[Click to Enlarge]

##### Report

After specifying all the values for the residential and employment-related land use demands, you can click on the Save button to compute the land use demand for all land use investigated and all projection years. The computations will be completed and a Demand Report will be generated, as shown in Figure 40. The Demand Report identifies the projected demand for all land uses in each projection year, given the values you specified on the Demand Scenario Form.

[Click to Enlarge]

## Allocation Control Scenario

Allocation Control Scenarios can be used to guide the process by which the various land use demands are allocated to different locations.

To create a New Allocation Control Scenario you should:

• Expand the tree for the Wanneroo project in the Projects window to show the Setup and Scenarios branches;
• Left click on the Scenarios branch and click on New Allocation Control Scenario option;
• Enter AllocationControl_Scenario1 in the Scenario Name pop-up text box and click on OK. The AllocationControl_Scenario1 form will automatically open up, as shown in Figure 42.
• Select the allocation control (or a combination of them) that you wish to use in the Allocation Scenario;
• Click on Save and close the form.

[Click to Enlarge]

Please notice that the Infrastructure Control is enforced at the land use type level. Assume, for instance, that: (1) a public transport service infrastructure control has been selected which identifies areas where public transport services will be provided in different years; and (2) it is assumed that new residential and commercial development can only be located in areas that have public transport services. In this situation, the model will allocate the projected residential and commercial demand to areas that have public transport services in a given projection year. However, this control will not be applied for industrial or educational demand. Thus, the model will allocate projected industrial and educational demand regardless of infrastructure provision; as no other control is enforced, suitability will be the only factor considered in the allocation process.

## Allocation Scenario

The Allocation Scenario projects future land use in each projection year by: (1) allocating the projected land use demands derived from a user-selected Demand Scenario to different locations on the basis of (2) their relative suitability, as defined by the assumptions in a user-selected Suitability Scenario, and (3) the allocation controls defined in a user-selected Allocation Control Scenario.

OWI begins by allocating the projected land use demand for each projection year in turn, i.e., for the first projection year, then the second, and so on. If there is not enough land to satisfy the projected demand for a particular land use type, the system will not terminate the allocation routine. A message will be displayed to the user at the end of the allocation procedure warning about land demand not satisfied. If the amount of land allocated to a particular use (shown in the Allocation Report) is lesser than the land use demand at a given projection year, it is very likely that there is insufficient suitable land to satisfy its demand.

The following steps are required to create a new allocation scenario:

• Expand the tree for the Wanneroo project in the Projects window to show the Setup and Scenarios branches;
• Right click on the Scenarios branch and click on the New Allocation Scenario option;
• Enter Allocation_Scenario1 in the Scenario Name form and click on OK. The Allocation_Scenario1 form will open automatically, as shown in Figure 40

[Click to Enlarge]

The Allocation Scenario form has two tabbed-sheets:

• Information; and
• Map.

#### Information

The Information sheet requires you to select the suitability, demand and control scenarios that will be used in the allocation scenario indicating that the allocation scenario results reflect the combined effect of these three scenarios. As their labels suggest, the three drop-down list on the form can be used to select previously defined suitability, demand and controls scenarios. For the Wanneroo Project we will select from the drop-down lists:

• Suitability_Scenario_1 as the Suitability Scenario;
• DemandScenario1 as the Demand Scenario; and
• AllocationControl_Scenario1 as the Control Scenario;

Next, the user must specify the order in which different land use demands are to be satisfied, i.e., the land use demand to be satisfied first, the demand to be satisfied second, and so on. You can do this by dragging and dropping the Land Use types listed under Select LandUse Order (use Drag and Drop). In this project we will order as follow:

1. Commercial;
2. Residential;
3. Educational;
4. Industrial

After providing the required information in the Information sheet you can click on the Save button and then on the Compute Analysis button to project future land use, population and employment patterns in your study area.

If there is not sufficient land to satisfy the demand for a particular land use, a message box is displayed at the end of the allocation process indicating the land use demand that could not be satisfied and the Projection Year that it first happened.

You can then download the Allocation Report in either Excel or PDF formats, by clicking on the Excel Report button or PDF Report button, respectively. Figure 44 shows an example of the Allocation Report in PDF format.

[Click to Enlarge]

#### Map

The current and projected land use maps can be viewed by switching to the Map sheet, as shown in Figure 45.

[Click to Enlarge]

The Layers checkboxes allow you to turn the layers for each projection year on and off. Figure 45 has the year 2011 turned off (baseline year) and all the other projection years on. This allows the user to identify the parcels of land that are likely to change use over time.

The Basemap radio buttons allow you to choose different basemap styles. In Figure 45 we have chosen Google Satellite as the basemap, which can be quite useful in identifying the current land cover, as well as land cover of neighbouring parcels.

When the user clicks on a particular parcel of land, an information box will pop-up allowing the user to quickly and easily access the attribute data of the parcel. The attribute data fetched in the information box are: (1) Land Use Type for the parcel in each projection year (ALU_Year); and, (2) the calculated Suitability Scores.

.

# Appendix A: Algorithm Changes

#### A.1 Online WhatIf – Revised Approach to calculate Land Demands

##### A1.1 Residential Land Demand

$$residentialDensity = \frac{N_{Houses}}{Area_{Residential}}$$

$$residentialLandDemand = \frac{Growth – Infill}{residentialDensity}$$

##### A1.2 Employment-Related Land Demand (one general equation for all relationships)

Let L be the Land Use Class and L = (1, 2,..,m)

Let i be the employment sector and i = (1, 2,..,n)

Let Li be the Land Use Class associated with sector i, and Li = (1,..,k)

Let iL be the employment sector i on Land Use Class L, and iL = (1,..,r)

Let AL be the Total Area of the Land Use Class L

$$LandDemand_{L} = \frac{EmploymentDensity_{L}}{Density_{L}}$$

.

$$EmploymentGrowth_{L} = \sum_{i_{L}=1}^{r}\left [ \left ( futureEmployment_{i_{L}} – currentEmployment_{i_{L}} \right )*\left ( \frac{A_{L}}{\sum A_{L_{i}}} \right ) \right ]$$

.

$$Density_{L} = \frac{\sum_{i_{L}=1}^{r}\left [ currentEmployment_{i_{L}}*\left ( \frac{A_{L}}{\sum A_{L_{i}}} \right ) \right ]}{A_{L}}$$

.

#### A.2. Online WhatIf: Allocation Algorithm Modifications

##### A.2.1 Objective Function:

$$Total\ Area\ Land\ Use_{i, t+1} = \sum Area_{i,t} + Land\ Use\ Demand_{i,t+1}$$

##### A.2.2 Iterative Process:

The procedure will go through 3 iterations, which we found to be effective to achieve a desirable outcome.

##### A.2.3 Remember the output of the previous time steps:

Any surplus or deficit of land allocated from previous time steps will be balanced out in the next time step.

##### A.2.4 Check the Area of the parcel allocated against the Land Demand

Find the land from the pool with closest size (area) to the land demand. The procedure has 2 objectives:

• Meet the land demand at a given time step, as accurately as possible;
• Minimize fragmentation of land allocated (i.e. will maximize the compactness of the land assigned to any particular use).

Using pseudo-SQL, this is implemented as:

[Click to Enlarge]