Methodology
Dataset
This project has required our group to find and create/use a dataset that is the focus of our results. The DMTI Spatial dataset is our choice, as it is readily available to us, and it is standardized because of its incorporation of Statistics Canada Census data. The age of the data is a concern, however. Data from 2001 would be preferable, but 1996 data is currently the most complete and accurate data set. We have used a land use coverage from the GVRD, as it is a more comprehensive residential classification than the DMTI layer, as housing type is categorized (single family, townhouse & low-rise apartment, high-rise apartment, rural residential). In addition, agricultural zoning has been incorporated into our methods. To account for mixed use areas in downtown Vancouver, a commercial/residential layer from 2001 has been included in this report.
There are only a few files that we required to create the data that we used. One of them is CANuep.xxx. This is a Unique Enhanced Postal Code File for all of Canada. It is unique because there is only one point for each postal code file. There is also a Multiple Enhanced Postal Code File as well, but that theme has multiple points per postal code in the same Shapefile. DMTI created the postal code files with “superior positioning based on CanMap’s Streetfile enabling users to map distribution of their data for spatial analysis” (Unique Enhanced Postal Code File User Manual, UEP.pdf, page 8). This has given us added confidence in our results. The attribute table for this theme contains population counts for each postal code. This field is the primary attribute for our analysis.
Tied into our study is the desire to map population accurately, meaning that population be placed only where people reside. In order to do that, we have incorporated a land use coverage of the region. DMTI produces one. It contains the residential areas of the GVRD, but only at such a classification level. This file contains some misrepresentations. For example, Barnet Highway is incorrectly classified as a residential area. This, we suspect, is because it is used to link two residential areas (Burnaby and Port Moody) together. The greater accuracy of GVRD land use layer’s classification scheme made our decision to select it straightforward. That is not to say that it is completely accurate. We believe it was still necessary to modify the zoning with additional information such as aerial imagery (Triathlon, 1995) and local knowledge. Land use themes, no matter the source, do not take into consideration all factors. Much of the world in which we live is covered with roads for the purpose of transporting people. People do not live on roads, therefore the space they occupy should not be included as areas within residential zones. DMTI’s road network was used to decrease the size of inflated polygons.
The DMTI dataset is a collection from various data sources and not the primary source therefore, our results can only be as accurate as the data we are working with. More up to date data could be acquired from varying sources (various municipalities), however, there is a risk of inconsistencies between municipalities. For example, a duplex may be classified as a single family dwelling in Vancouver, but may be classified as a townhouse/low-rise apartment in Burnaby.
Methodology:
Area of Study .
Dataset .
Modus Operandi .
Cartographic Model .
Map Discussion
Errors & Limitations