Wednesday, June 21, 2017

A citizen's map of Industrial Wind Turbines in Ontario

I have produced and put online an interactive map of Industrial Wind Turbines in Ontario. Having done so feel I should explain why, and how, and invite others to participate in improving data.

The map is embedded later in this post, as is a YouTube clip of me talking to its design and functionality.  I talked for longer than I expected, and I suspect I shall write on this longer than I expect too, because I think the issues of public data availability is important, and I have a particular fascination with data mining, manipulation, presentation and technology.

There was a paper published late in 2013 titled Mapping Ontario’s Wind Turbines: Challenges and Limitations. The abstract began:
Despite rapid and vast development of wind turbines across the Canadian province of Ontario, there is no map available indicating the location of each wind turbine. A map of this nature is crucial for health and environmental risk research and has many applications in other fields.
These seem like noble goals, so it might be surprising there's not a single official source of spatial data 3-4 years later.
The scope of the desired map was described in the paper:
A map showing individual wind turbines would be a valuable resource for researchers to examine health effects, as the distance that a resident lives from a wind turbine could act as a proxy indicator of a "dose" and could be used...
This is one example, but it could be measured against a lot of things - bat counts, bird carcasses, house and farm values, household income. The possibilities need not be known, only the data requirements, and most of these were described well years ago.

The 2013 paper noted its sources of data, and the severe restriction on the utility of the produced map that came from using some of those sources:
the map was initially imagined to be a product that could be shared publicly, but due to data sharing agreements with a variety of public and private groups, this is not possible. For other maps like this, the map should be made public and as a crowd-sourced public resource, community members can keep the map updated.
I didn't ask any official sources for data.

I asked Wayne Gulden, whose Ontario Wind Turbines site was cited in that 2013 paper, and he provided the files containing most of the turbine location that can be found in my map. The Ontario Wind Turbines site currently has a significant advantage in that it uses Google maps, with the available satellite imagery.

The 2013 paper noted the problem of determining whether turbines actually were constructed as planned. This issue is much more manageable today as a list of contracts is now publicly available: the active generation contract list produced by the system operator (IESO). The list doesn't include all generators, but it should omit very few. Non-utility generators (NUGs), contracted in the previous century, are not in the list but I don't recall any wind that old. Also omitted from the IESO contract list are "micro-FIT" contracts, but in it's latest report on contracted supply the IESO showed only 4 microFIT contracts with a total capacity of 0 (rounded to the nearest tenth of a megawatt).

The advantage of my map is it's produced after pulling the data into Microsoft's Access database, and loading it into their Power BI platform - which produced the map using bing. I understand the platform has just been updated to allow satellite views on the map - but I do not yet understand how to do so.

I did think of one turbine not listed by the IESO, or on the Ontario Wind Turbines site. I have now added OPG 7, which is a 1.8 MW capacity turbine on the Pickering nuclear generation site, and as far as I know the first IWT to enter commercial operation, in 2001. Within a year on OPG 7 producing power the first commercial wind facility, Huron Wind, entered operation outside of the Bruce Nuclear site. A side-benefit of this map building has been discovering public power started industrial wind in Ontario - before being blocked from pursuing it.

I've been meaning to try creating very short videos explaining some graphics I produce for some time - I have produced one about 3 times longer than I intended for this one. If you have 4 minutes you may wish to open up the one window with the chart as I talk through some of the functionality in another window.

I'll conclude this post with some comments on data. A tool I think people use to appear smart is to criticize the quality of data - smarter people should recognize data is never perfect. The question is always whether data has utility - not perfection. This data is good enough to map, and now that it is people can choose to improve it where they see fault. I've shared data in a spreadsheet *, which you can take and improve - or help me improve my sending me an e-mail with updates to a turbine location (I am using 6 decimal places on the longitude and latitude).

I have some known data gaps and discrepancies, but far fewer than I had expected. There are some sites where the counts of turbines are slightly off what other sources indicated. 6 of the 8 sites not listed by the IESO as having achieved commercial operations by the end of 2016 have no turbine detail, but all sites in commercial operation do - although there are some where the turbine count is not as expected. These issues are listed in the final 2 tabs of the spreadsheet.

One additional data detail that would be useful is turbine model. I think most facilities use a single model but know of at least one planning on using two varieties. I'd welcome input on whether to track that detail at facility or turbine level... and if you happen to have a list of turbine models for sites lying around, please send it my way.


*Update: if you desire the latest co-ordinates of turbines, please e-mail me at
Since the aerial view was added a large number of (mostly small) changes have been made, so the originally posted spreadsheet is now outdated.

No comments:

Post a Comment