Appendices

Appendix 1: Bibliography

Listed below is a partial bibliography of published and unpublished work on defining built-up areas. The analyses and methods in the literature and case studies below apply to varying scales - some at the large scale of a country and others at the very local scale of zoning.

All weblinks cited below were confirmed as active on April 2, 2008.

Batty, Michael Yichun Xie & Zhanli Sun (1999); The Dynamics of Urban Sprawl; CASA Working Paper Number 15. Centre for Advanced Spatial Analysis (CASA), University College London. 
http://www.casa.ucl.ac.uk/publications/workingPaperDetail.asp?ID=15

Batty, Michael, Elena Besussi, Nancy Chin (2003); Traffic, Urban Growth and Suburban Sprawl; CASA Working Paper Number 70. Centre for Advanced Spatial Analysis (CASA), University College London. 
http://www.casa.ucl.ac.uk/publications/workingPaperDetail.asp?ID=70

Burchfield, Marcy, Henry G. Overman, Diego Puga, Matthew A. Turner (2003); Causes of Sprawl: A Portrait from Space; Working Paper, University of Toronto Department of Economics.

Federal Register (2002); Urban Area Criteria for Census 2000; Washington DC, Bureau of Census.

Chin, Nancy (2002); Unearthing the Roots of Urban Sprawl: A Critical Analysis of Form, Function and Methodology; CASA Working Paper Number 47. Centre for Advanced Spatial Analysis (CASA), University College London. 
http://www.casa.ucl.ac.uk/publications/workingPaperDetail.asp?ID=47

Churchman, Arza (1999); Disentangling the Concept of Destiny; Journal of Planning Literature, Vol.13, No. 4, (May) pp.389-411

City of Toronto, (2000); Clean, Green and Healthy: A Plan for an Environmentally Sustainable Toronto; Final report of Environment Task Force. 
http://www.toronto.ca/council/etfepfin.pdf

DRCOG/APA Planning Commissioners Workshop (2002); Civitas.

Australian Department of Primary Industries and Energy / Department of Human Services and Health, (1994); Rural, Remote and Metropolitan Areas Classification 1991 Census Edition; Australian Government Publishing Service, Canberra, Commonwealth of Australia.

United States Department of Agriculture Economic Research Service, (2003); Measuring Rurality: New Definitions in 2003. 
http://www.ers.usda.gov/Briefing/Rurality/NewDefinitions/

Ewing, R., Pendall, R., & Chen, D. (2002); Measuring sprawl and its impact: the character & consequences of metropolitan expansion; Washington DC: Smart Growth America.

Hess, George R. et.al. (2001); Just What is Sprawl, Anyway?; North Carolina State University, Raleigh, NC.

Miron, John R (2003); Urban Sprawl in Canada and America: just how dissimilar?; University of Toronto at Scarborough. 
http://citieslab.utsc.utoronto.ca/Papers/UrbanSprawl.pdf

Mitchell, Bill (2002); Bibliography of Sources; taken from Compact Cities and Urban Intensification: Desirable, Acceptable, Achievable?

Moglen, G.E., S.A Gabriel and J.A. Faria (2003); A Framework for Quantitative Smart Growth; Land Development's Journal of the American Water Resources Association; Vol. 39 no. 4, pp.947-959. 
http://www.eng.umd.edu/~sgabriel/Research/publications/JAWRA-SmartGrowth.pdf

Northwest Environment Watch (2004); The Portland Exception. A Comparison of Sprawl, Smart Growth, and Rural Land Loss in 15 US Cities; Northwest Environment Watch, Seattle, Washington. 
http://www.sightline.org/

Oregon Health and Science University, (2004); Office of Rural Health; Definitions of Rural & Assessing Healthcare Needs.

Rural Doctors Workforce Agency, (2004); Rural, Remote, Metropolitan Areas Classification.

San Francisco League of Conservation Voters (2004); Definitions & Calculations: This View of Density. 
http://www.sflcv.org/density/

Statistics Canada (2001); Census Dictionary- Geography Section. 
www12.statcan.ca/english/census01/Products/Reference/dict/geotoc.htm

Torrens, Paul M. and Marina Alberti (2000); Measuring Sprawl- CASA Working Paper Number 27; Centre for Advanced Spatial Analysis (CASA), University College London. 
http://www.casa.ucl.ac.uk/publications/workingPaperDetail.asp?ID=27

United States Census Bureau (2003); Differences Between the 1990 Census and Census 2000 Urbanized Area Criteria.

US Census Bureau, (2004); Census 2000 Urban and Rural Classification.

Waddell, Paul (2000); Monitoring and Simulating Land Capacity at the Parcel Level, in Monitoring Land Supply with Geographic Information Systems: Theory, Practice and Parcel-Based Approaches; Vernez-Moudon, A. and M. Hubner, eds., John Wiley & Sons, Inc.: New York., (pp. 201-214).

Wassmer, Robert W. (2000); Urban Sprawl in a U.S. Metropolitan Area: Ways to Measure and a Comparison of the Sacramento Area to Similar Metropolitan Areas in California and the U.S.; Social Science Research Network.

Zhang, Wanquing, Angella Bowman & Keith Mueller; Rural / Urban Definitions: Alternatives and Numbers by State; Nebraska Centre for Rural Health Research.

Appendix 2: Overview and Description of Municipal Property Assessment Corporation and Ontario Parcel Alliance Datasets

For the purposes of this paper, the two datasets which make up the parcel land-use database are referred to as the OPA dataset for the geographic location and outline of each parcel, and the MPAC dataset for the land use and residential unit count information within each parcel.

Included in these two information sources are geospatial locations, land-use designations and the number of residential units on the property. In theory, the location of every parcel is known, as are its land uses and the number of residential units. Shortcomings in data quality mean that not all these attributes are always known or accurate for all of the approximately 2.4 million parcels within the Greater Golden Horseshoe. The datasets do not include a parcel fabric for roads and water bodies.

The OPA and MPAC datasets obtained by the Ministry of Public Infrastructure Renewal contained only land use, unit count, and parcel number information. The datasets used to develop this methodology and those applied in this analysis did not contain any confidential, personal and financial information. For a full list of attributes contained in the MPAC dataset obtained by the Ministry of Public Infrastructure Renewal and used for this analysis, please refer to Appendix 3.

Brief overview of MPAC dataset

Every municipality in Ontario is a member of the Municipal Property Assessment Corporation (MPAC), a non-share capital, not-for-profit corporation whose main responsibility is to provide its customers – property owners, tenants, municipalities, government and business stakeholders – with property assessments. MPAC administers a uniform, province-wide property assessment system based on current value assessment in accordance with the provisions of the Assessment Act.

MPAC receives all the registered Land Transfer Tax Affidavits (LTTA) within the Province of Ontario. Upon receipt of this information, MPAC investigates and codes this information into its database.

MPAC data is classified into nine key property types:

• Commercial
• Industrial
• Residential
• Farm
• Multi-residential
• Managed Forest
• Pipeline
• Special and Exempt
• Vacant Land

Within each of these types, the following fields were used for compiling the database used to create the built boundary:

• Assessment Roll number
• Property code
• Property code description
• Location address including postal code
• Site Area
• Site Area Unit of measure (Acres or Square Feet)
• Realty Tax Class
• Realty Tax Qualifier
• Number of residential units for:
  • Multi-residential properties
  • Non-residential properties

Brief overview of OPA dataset

The Ontario Parcel Alliance (OPA) has been created jointly by Teranet Enterprises Inc., MPAC and the Ministry of Natural Resources.

The OPA data is a standardized, Ontario-wide, geospatial dataset of assessment, ownership and Crown parcels of land. The OPA database includes parcel boundaries, assessment roll numbers, and property identification numbers where applicable.

The OPA data consists of three parcel layers - assessment, ownership and Crown. Of these, the assessment and Crown layers were used in the creation of a built boundary.

Assessment parcels are areas defined by a boundary and an assessment roll number (ARN), for property assessment purposes as determined by MPAC. The assessment parcel is mapped for the entire province. Crown Land Parcels are those owned by the Crown, and are also mapped for the entire province, except in areas where more detailed sub-ownership mapping already exists.

Benefits and challenges in using MPAC and OPA data for determining a built boundary for the Greater Golden Horseshoe

While the MPAC and OPA datasets have been selected as the most appropriate and useful to identify a built boundary for the Greater Golden Horseshoe, the data is not without its challenges.

The key reasons why the MPAC and OPA data are most appropriate are:

• Data is available in a consistent format for the entire geography of the Greater Golden Horseshoe.
• Data is updated regularly and is available annually at a minimum.
• Greater Golden Horseshoe upper-, single- and lower-tier municipalities have full access to these datasets.
• Data tracks residential units each year, which allows for measurement of the Growth Plan intensification target.

Challenges in using this data have included:

• Currency of MPAC data can vary across the Greater Golden Horseshoe. Municipalities report that the data can be as many as two years out of date and that there are errors. The primary reason for this delay is that the reporting processes can take time since there is often a need for individual validation of records. These issues are addressed in Step 4 of the methodology.
• There are some challenges in simply building a combined dataset that has the land use, the number of residential units and the parcel boundaries on it. Part of this is the historical way the files have been built since many uses are made of the same property and thus multiple files have been developed. In addition, linking files between the two datasets can be challenging. These issues are addressed in Step 1 of the methodology.
• There is no parcel or land-use information for roads and highways.
• Land use data on the parcel records may not fully reflect the actual use of the land. There are also often multiple uses on single parcels and though MPAC records list primary and secondary uses, these may not reflect all the actual uses on the parcel. An example of this can be commercial and industrial land where some owners may own more land than they have yet to develop.

The consultation and verification process and application of Step 4 is intended to identify and address such shortcomings in the datasets.

Implications with regard to sub-division developments

The use of parcel data according to their recorded land use by MPAC means that land that has been approved for development, such as registered sub-divisions, but is not yet built, will be considered unbuilt. Similar situations involve newly-built condominium structures. Parcels that are built-up prior to June 16, 2006, the effective date of the Growth Plan, that are not captured in the MPAC dataset and thus not included as part of this analysis, will be identified and accounted for in Step 4 of this methodology.

Appendix 3: Assignment of Summary Land-use Codes to Municipal Property Assessment Corporation Property Codes

Source: MPAC, May 2006

Appendix 4: Other Data Sources Evaluated for the Built Boundary Methodology

Various sources were examined for comprehensiveness, applicability to the policy application of a built boundary for the purposes of the Growth Plan, level of geographic detail, currency of the data, accuracy of the data, availability of the data and its expected availability over time.

Statistics Canada's quinquennial census data

This is the most exhaustive source of demographic information available for very local areas of geography. Statistics Canada does define the 'urban area' for every urbanized place in Canada. The geographic specificity of the data as stored and as published is too broad in areas that are on the fringe of the urban area. Fringe area boundaries encompass extensive areas of undeveloped land and thus this data is inappropriate for the level of detail required to measure the objectives of the Growth Plan.

Private sector annual updates of the census data

A number of companies in Canada prepare annual 'updates' of the Census on the same geographic basis as the Census is published. These data are not based on actual fieldwork but are based on models developed using data trends at larger levels of geography and thus do not necessarily reflect actual changes on the ground. They also have the same geographic challenges as Statistics Canada data.

Canada Post

Canada Post maintains a count of residential and business delivery points for every postal code in Canada and updates these counts on a monthly basis. A residential delivery point is very similar to the household concept used by the Census and is also very similar to a residential unit as used to measure intensification in the Growth Plan. However, the geography of postal codes on the fringes of urban areas and in rural areas is inexact and very large and thus cannot be used for defining a built boundary.

Utilities

Electric, gas and telephone utilities keep track of where their customers are located. Theoretically, this information could be assembled into a database that would permit an address-based dataset to be compiled that could be used to define built-up areas. Unfortunately, there is no such compilation and the prospects of being able to work with all the organizations involved to derive such a composite database, even within one utility sector, are unlikely. There would also be no guarantee that the information would be available annually. Geographic boundaries may also vary from utility to utility.

Aerial and satellite imagery

The automated and/or manual analysis of aerial and satellite imagery for defining urban areas has become a well-recognized and commonly accepted method for defining the urbanized extent of a city or a large region. The methods and data sources involved are complex. They are publicly available but are rarely undertaken over large areas. There are also some accuracy issues and a significant amount of ground validation is needed. The key shortcoming is that this approach has no direct ability to count one of the Growth Plan's key measures, that of residential units (since under any one rooftop visible on aerial photos could be very few or very many residential units). Also, while the approaches can differentiate between urban and non-urban uses, they have difficulty in differentiating different types of unbuilt land within existing urban uses (conservation areas from golf courses from parks, etc.) Such differentiation is important in setting where boundaries should actually be drawn.

Importantly, these images are the best indication of what is actually happening on the ground since they represent a real-life 'picture' at a point in time. They can also be repeated at regular intervals and can be expected to be available for the life of the Growth Plan. They will always be useful sources of information for refining a built boundary, but would not be used as the primary source for defining it.

Planning designations and related records

Most planning departments keep track of the use of land within their jurisdiction. They use a variety of data sources and organize and display the information according to their own particular needs. Land-use classifications can vary as can the determination of what the land is currently being used for as compared to what the land is zoned for. In many cases, actual geographic limits of a use are not pinpointed – only an approximation of the location and size. The currency and update schedules for these maps and data can vary. Also, these efforts focus on land and usually do not include a count of the number of residential units involved.

Some attempts have been made to assemble a composite picture of land-use planning designations across all the jurisdictions in the Greater Golden Horseshoe but none has yet been assembled that covers all municipalities in the Greater Golden Horseshoe. The Ministry of Municipal Affairs and Housing has a process in place to assemble a composite map that shows where urban land use is permitted according to official plans. However, this project does not identify where development currently exists.

Building permits and housing starts and completions

Since new residential units and most sizeable conversions require building permits, a system that tracks where these permits are located and then monitors them through to completion (starts and completions) could be used to identify built-up areas. Following building permits through to construction is important since it should be noted that many permits are issued that are subsequently not used to build or to convert a residential unit. At this time, no such consistent, composite monitoring is done across the entire Greater Golden Horseshoe. Very few municipalities geocode their building permits – i.e. assign a latitude and longitude position for each permit. It also appears that even fewer municipalities then track by location the start and completion of the new building and/or conversion. No provincial or federal organization does this. The Canada Housing and Mortgage Corportaion and Statistics Canada collect summary statistical information from local authorities but it is at the scale of each municipality. Thus this approach cannot be used.

Appendix 5: Sources and Currency Dates of Data

Please note that the refinement methodology described in Step 4 allows corrections for errors, omissions, and out-of-date elements in these datasets.

Related

Growth Plan for the Greater Golden Horseshoe, 2006

A Place to Grow: Growth Plan for the Greater Golden Horseshoe, 2019