A Problem Well-Spread: Investigating arsenic in Nova Scotia’s well water


By Mikkel Frederiksen

In Nova Scotia, a province on the east coast of Canada, many homeowners rely on a private well for their drinking water. Much of that water has a high likelihood of having arsenic in it, and many therefore risk exposure to a chemical element that’s linked to a variety of cancers. I’m not a Nova Scotia native, and I didn’t know this, but as I would later find out, a great number of Nova Scotian well users didn’t know either. Looking at a problem spread across an entire province, I put together an investigative piece that revealed a pervasive ignorance of arsenic among well owners, and worse yet, a complete lack of effort by the provincial government to inform and protect its citizens.

The story was the conclusion of my journalism Masters degree at the University of King’s College in Halifax, Nova Scotia. We had free reign to pursue whatever story we wanted, as long as the story was based on data, and had legs to sustain a lengthy investigative piece.

I knew early on I wanted to deal with an issue particular to Nova Scotia. Not only for logistical purposes, but also because I wanted to work on something relevant to my local community. After looking through federal, provincial, and municipal open data sites, I found the Nova Scotia Well Logs Database, a dataset compiling logistical data on every single well, private or public, drilled or dug in the province since 1940 up until 2012. Substantial, to say the least.

The well log was just one of many datasets I pulled, so it wasn’t until I came across a series of maps produced by Nova Scotia Environment detailing different chemical elements presence in groundwater, that I saw a possibility. They had also produced an interactive called Groundwater Atlas, which I saw was produced with GIS files. Many of the layer-files were free to download, but the files for a map of arsenic in groundwater was missing.

So, I called the department, and they agreed to send me the files used to produce the map. The complete set of files included a shape-file that detailed the likelihood of arsenic in the groundwater around the province, breaking risk levels down into “Likely” and “Very Likely”. In ArcMap, I joined that with the shape-files from the well database, and the result was a map detailing the higher risk areas of the province, and how many wells were drawing water from those areas.

This was of course a quick way to get an eyeball-glance of how many wells were drilled on land considered “very likely” to have arsenic in its groundwater. With the risk levels a dichotomous “Likely/Very Likely”, it was already at this point clear to me how common arsenic was in Nova Scotians’ well water, and the sheer number of wells convinced me of the project’s scale and scope. The map itself was just a visual confirmation, and didn’t give me more information beyond that. Essentially, it was just telling me there were a lot wells, and many of them drew from groundwater that was likely to have arsenic in it. I wanted to get numbers, and get a feel for which parts of the province, and which communities, were most likely impacted.

Image: The Nova Scotia map produced in fusion tables; the green-shaded area is "Very Likely"-risk level. 

By joining the two map layers in ArcMap, I also generated a new dataset of the wells that now included a value of either “Very Likely” or “Likely” in terms of arsenic risk, and I loaded it into a SQL-programme to sift through the data. The information was quite detailed, including geographic data on each well, going right down to the civic address a well was registered to. Querying the data allowed me to find out what areas of the province had the highest number of wells, highest number of wells at a “very likely” risk-level, and whatever else I could think to ask. All in all, a comprehensive breakdown of the province’s well users and their likely exposure.

I had originally planned to use the street addresses of each well as a tip sheet by identifying the owners of a given well by matching the address to property records from Property Nova Scotia. I had a small list compiled, but never got around to making any calls, as my professor and editor on the project, Fred Vallance-Jones, suggested I instead get in touch with local councillors who may have an immediate knowledge of their constituents and whether arsenic has caused any problems in their districts. Calling around, I came into contact with Michael J. Gunn, a councillor from Annapolis County, whose political platform centered on water safety. Him and his wife Michelle became the central characters of my story. 

This project wasn’t short on data. The vast catalogue of wells meant the majority of the work went into finding something to combine it with that could suggest an issue, and then organizing the data into something meaningful. Ultimately, it went from a sweeping overview to a detailed breakdown, providing a roadmap of a province-wide public health concern that led into every citizens’ backyard. 

About the author

Mikkel Frederiksen is a freelance journalist. He earned his Bachelor of Journalism and Masters degree at the University of King’s College in Halifax, Canada, and currently resides in Denmark. 

Explore the project here.