Why Doesn't The Property Boundary Line Up With The Aerial Image?

Opinions, facts, and data quality.

If you're a real estate agent, a lawyer, or perform any kind of on-location services for property management, you'll understand my frustration with maps that just look wrong. A common combination is orthophoto and parcel data. If they're both correct, why does it look like this in some areas when you put them together?

Related: Top Three Questions About Parcel Mapping [Cheat Sheet]

St. Thomas, 2010, 20cm orthophoto
The entire neighbourhood is shifted out of alignment in this Pre-BIM section of parcel mapping.

Vaughan, 2011, 15cm orthophoto
The boundaries are in precise alignment in this POLARIS section of parcel mapping.

It's because of data quality. Where your orthophoto and property lines don’t align perfectly, it's not that one is right is one is wrong. It’s due to different accuracy standards used to create the data sets. A few factors come into play when creating the data that have a direct effect on the success and usability when combining data from different sources.

The orthophoto you find on MapWarehouse come in various sub-metre resolutions. The smaller the number, the more detail you can see in the photo and the more accurate the geographic position of the features you see in the image will be compared to the real location. The orthophoto from FBS are spatially accurate to double the photo resolution value. Simply put, if a photo has resolution of 20cm, any point on the photo will be accurate within 40 cm of its true location.

Related: Where Are My Zoom2It Images?

If you compare several years of imagery in the same location - no matter what company produced it - you'll see features that should be in a fixed position, like the corner of a building, jump around within a limited range of a few pixels. That's locational accuracy at play. Each year of photo compared in a time series could have different resolutions, different processing applied, different acquisition methods etc, which means the acceptable error range will be a bit different year over year.

How about the parcel? Well, that depends on where in Ontario the property is. Different counties in Ontario were originally drawn to different standards and then later integrated into one data set that’s now maintained by Teranet. The most accurate being POLARIS (Province of Ontario Land Registration Information System) which is based on surveys, followed by BIM (Basic Index Mapping) which is based on 1:10,000 scale mapping or smaller, and Pre-BIM, which is compiled in areas where good control data is less available, making it the least positionally accurate.


Related: What Are You Planning To Do With It? [Infographic]

In an area where pre-BIM standards are used, it’s not unusual for the property lines to be off by 10 m or more from where you would expect them to lie on the photo.  Because of the legacy issues surrounding the compilation of the parcel data, it's not possible to know the positional accuracy or error of any part of the parcel mapping. It comes to us in "as is" condition, but keep in mind that the positional error has no relationship with the accuracy of the attributes like the PIN or legal description for the property.

For more information and to purchase parcel mapping or orthophoto, please visit firstbasesolutions.com