Automated Land Area Tests for the 2001 Census: Preliminary Results Using the 1996 Digital Cartographic Files - ARCHIVED

Prior to the 2001 Census, the Geography Division manually calculated land area using a planimeter. The manual approach was necessary since the division did not have a national digital database to support an automated method. The manual land area calculation was a labour intensive process as well as very time consuming, expensive and very prone to error. With the anticipation of having a national digital base for the 2001 Census (to be called the National Geographic Base or NGB), the automated land area calculation becomes a simple GIS process.

The objectives of this paper are to show the effects of data aggregation and different map projections on automated land area calculation. The testing is a two- staged process; the first stage uses the 1996 Digital Cartographic Files (DCFs) and the second stage will use the NGB. Although the DCFs are not ideal for testing land area, this two-staged approach is necessary since the NGB will not be completed until Autumn 1998. This report presents results on the first stage using the 1996 DCFs.

The results reveal that the effects of data aggregation are minimal; thus land area can be stored at the census block level and then aggregated to higher level geographic entities without any serious ramifications for the 2001 Census. The results also indicate that an equal-area projection, specifically the Albers Equal-Area Conic projection, is more appropriate for calculating land area.