Click here for the full list of articles
The Don River Valley Historical Mapping Project is a unique effort by Map and Data Library (MDL) staff at the University of Toronto to collect, document and map the historical and industrial change in the Don River Valley Watershed between the mid 1800s and the mid 1900s using Geographic Information Systems (GIS). The project is a collaboration between the MDL and Jennifer Bonnell, a PhD student in the History of Education at OISE/UT.
GIS integrates computer hardware, software, and data for capturing, managing, analyzing, and displaying all forms of geographically referenced information. GIS are used for visualizing, understanding, questioning and interpreting geographic data to reveal geospatial relationships and patterns. Output can range from maps and globes, to reports and charts. For further details on GIS and GIS Services at the University of Toronto’s Map and Data Library, please visit http://mdl.library.utoronto.ca
While GIS applications were the obvious solution for this research, one key feature missing from the equation was suitable data. The Map and Data Library holdings contain hundreds of datasets and thousands of data layers, but historical data are still quite rare. Even with all the recent open data initiatives by governments, including the City of Toronto, historical data are still scarce. Luckily, because of the Map and Data Library’s rich paper map collection and thanks to a wonderful wide-format, high-resolution map scanner, users are now able more than ever to create their own data by digitizing or extracting data from the paper maps in a digital and geospatial environment. Of course, the only barrier to creating the data was time and the labour to do the work. As a result, we applied for funding from the Network in Canadian History and Environment (NiCHE) to pay for the costs of hiring a student to undertake the data creation. The Map and Data Library would oversee and manage the project through the joint direction of the GIS librarian and the PhD student.
NiCHE accepted our proposal and funded our project through two rounds of grants totaling close to $10,000 and a computer system on which to conduct some of the work. The Map and Data Library matched the funding through in-kind support.
Historical GIS Methodology
Historical GIS, or HGIS, is a very new concept. The main premise of HGIS is to create digital geospatial data that has historical significance or that has a temporal component in order to undertake historical geospatial analyses.
The main process for creating historical geospatial data using this methodology is by georeferencing, or geopositioning historic maps; that is, to scan a paper map, load it into GIS software with established (already built) geospatial data, and to then match the digital map’s pixels to the geographic location on the surface of the earth where they belong, using the established data as a reference point. Once georeferenced, map data can then be extracted from the paper map into geo-enabled databases. The results are geospatial data created from features from the paper map.
Most of the paper sources of cartographic information for this project came from our collection, with other sources originating from the Ontario Archives, the National Archives, the City of Toronto Archives, the Toronto Public Library, and the Thomas Fisher Rare Book Library.
The main types of cartographic sources used for the project are described below.
Fire Insurance Plans and Atlases
Fire Insurance plans and atlases are highly detailed urban maps which grew out of the need for fire insurance underwriters to understand the physical characteristics of a structure to be insured. These maps show with detailed colour drawings and symbols the character of the outside and inside construction of buildings. The years of publication for the variety of Fire Insurance plans and atlases used for the project are from the 1880s until the 1950s.
City Directories contain key information about industries at a given address and were invaluable for determining not only an address for industrial sites, but also to confirm the existence of properties whenever a map for a certain year could not confirm the existence or the ownership for specific industrial sites. Toronto City Directories are available as early as the 1870s.
Other Map Sources
We extracted historical information from a variety of other map sets. Sources include private commercial publisher maps as well as public organizations such as the Toronto Harbour Commission, the Toronto Transit Commission, the City Engineer’s Office, etc. We also scanned many of Toronto’s most famous maps and extracted some information from them. These include James Chewett’s famous 1834 map, Tremaine’s Map of the County of York from 1860, James Cane’s map of 1842, and A.P. Coleman’s maps of the 1890s. While many of these maps or copies of them were available at the Map and Data Library, some were obtained from the City of Toronto, the Ontario Archives and the National Archives.
Geospatial Data Created
Upon completion of the project following just over a year’s work, the project team had created the following GIS datasets for the Don River Watershed:
• Historical and physical change between 1858 and 1918 of
°The Don River’s course
• Industry and industrial footprint in 1825 and between 1858 and 1950
All these data are available for download in both GIS format for use in GIS software and in Keyhole Markup Language (KML) for use in Google Earth and Google Maps applications.
The Don Valley Historical Mapping Project was a success on many levels. We completed many of the datasets we had planned to create but there were, nonetheless, many challenges to overcome.
It was often difficult for us to obtain clearance to use the Fire Insurance plans we intended to use for the project.
Also, another challenge was obtaining digital copies of originals of some of the sources we felt were crucial to the project but that were not available at the University of Toronto. Consequently, we had to purchase many images from other institutions.
Yet another challenge was HGIS itself. Being a relatively new field, there are few examples to draw from and even fewer guidelines to follow. As a result, we dealt with issues as they came along using our own needs and requirements as guides for solutions. For example, if a map from 1857 had the Don River shown in one location and a map for 1858, at the same location on the river, showed the river in a different way, we either tried to build the data for both years, or instead described the data as being drawn from a map in 1857.
The physical features of the Toronto area also make it difficult to document properly the land historically. Ashbridge’s Bay, for instance, before becoming a landfilled area, was mostly marshland susceptible to change from one season to another. By the time the marsh was mapped one year and published, massive changes in its makeup would have taken place.
With this project we endeavoured to create the best metadata possible. This meant that much student time was spent documenting and authoring metadata.
Future of the Project and Data
We were invited to present the results of our project to both the City of Toronto’s Task Force to Bring Back the Don, and Lost Rivers, a community-based urban ecology organization focused on building public awareness of the city’s river systems. Both groups were very receptive and interested in possibly funding further development of the datasets for use by the public.
A second phase of our project would expand the current datasets to a larger geographical base than the Don Valley Watershed. The industrial history of Toronto spans a much larger area than this and to document it further would be very useful for historical research.
Data from the project are currently available in both GIS and Google Earth format. It is hoped that we will, in the future, build online mapping applications for querying and displaying the data online rather than simply for visualization.
Thanks and Acknowledgements
Thanks are due to NiCHE for their funding and support. Without their vision for non-traditional academic research, the project would not exist. As well, valuable research support for the Points of Interest web pages that support our project came from Lost Rivers. This project would not have been possible without Jordan Hale, a University of Toronto Geography student, who conducted much of the digitization and database work. Visit the Don Valley Historical Mapping Project at http://maps.library.utoronto.ca/dvhmp/