When working as a GIS Analyst at Agriculture Canada, the main project I worked on was the update of a major national dataset called the Soil Landscapes of Canada (SLC). The SLC is a polygon layer compiled at a 1:1 million scale that covers all of Canada and is representative of the underlying soil. With this dataset, researchers can get an idea of variables such as soil moisture, pH, clay content, stoniness, etc, which can be important in planning and modeling. There are close to 50,000 polygons for the entire country.
However, when I first started working on this project in 2010, I learned that this dataset had many issues and needed to be updated. Boundaries were off for many areas, misrepresenting the data and adding to the inaccuracy and unreliability of the entire dataset as a whole. This is a problem that could normally be resolved using GIS. Someone with GIS knowledge, like me, could adjust the boundaries accordingly and produce a new, more accurate version of the dataset. But the problem is that those that knew how to fix these errors, soil scientists known as “pedologists”, were older and were very limited in their computer technical skills. Many of them were near retirement or already retired, and training them on how to learn an advanced software such as ArcGIS was simply not feasible. I was therefore called upon to find a creative solution to this problem.
I had to figure out how to tap into the expert knowledge of pedologists in terms of soil composition and geography in order to update an important GIS layer, but without needing for them to learn GIS.
The solution that my supervisor and I devised was to utilize Google Earth. Google Earth is not a true GIS, but rather more of a visualization and location application. Although it cannot do what a GIS software could do (e.g. make edits to the SLC boundaries), it could provide a familiar “look and feel” for pedologists to base themselves on when drawing out new boundaries. After all, everyone knows how to use Google Maps and Google Earth! Luckily, Google Earth can save files in KML format, which can then be imported into a GIS. Once we made this decision, the time came to create a Web based application that would allow pedologists to assess each polygon, one by one, and be able to make their recommendations of where boundaries should fall, should these be inaccurate.
This involved learning new Web languages, such as HTML, JavaScript, PHP, and CSS. Using these languages, I developed an application that was embedded in a KML point file that could be opened into Google Earth. This point file looked like blue question marks that, when clicked on, would have a popup window that would show the current SLC values for the polygon. There was also a link in this popup table that would lead pedologists to a webpage where they could determine if the polygon in question needed to be altered or not, and attach any KML files that they drew to “correct” the boundaries.
In the end, this project was very successful and the pedologists were able to make updates to the dataset that GIS Analysts were then able to implement to create the new SLC layer.
To learn more about this dataset, see this link from Agriculture Canada: http://sis.agr.gc.ca/cansis/nsdb/slc/index.html.