So after a reasonably long hiatus, I'm back to blogging; this time about population access and playgrounds in Toronto.
Why I am keen on this topic is: (1) As a new parent, I am starting to get interested in play and play spaces. Seeing as how I will be visiting Toronto in June, I (non-altruistically) wanted to find out where is the best place to visit with regards to playground access for my energetic toddler; (2) I wanted to showcase some of the new (and awesome) Census data; and (3) I wanted to draw attention to what can be accomplished with Toronto Open Data.
So given the above interest, I was able to generate some very interesting questions around playgrounds in the City. For instance:
Where are all the playgrounds in the City of Toronto and how do I get that data?
Where are the best and worst parts of the City for play access for young people (0-14yrs old)?
Which parts of the City are the particularly challenging from a play access standpoint (ie where is there much potential walking travel along busy roads)?
Which areas of the City may benefit from new infrastructure?
Thankfully, getting this information mapped and analysed was not too great of a challenge, thanks, once again to the good folks at open data Toronto. In any event, I needed a few things to get started with this analysis:
First, I searched for a playground dataset - which I found here. However, this was not all the playgrounds in the City, only ones located on City land. To complete my set, I added in all of the non-overlapping locations of junior or elementary schools in the City. (operating on the assumption that most, if not all, elementary schools have playgrounds). I combined the two datasets, and spent about 10 minutes reviewing the data. Check out the map below, (and please let me know if I missed any or added in an unnecessary locations).
WHERE ARE THE CHILDREN
Once I had the playgrounds mapped, I then set about figuring out how kids would possibly walk to these play spaces. Safe walking to school is a big deal these days, however, safe walking to playgrounds - because they are often co-located on school grounds - is less well analyzed. Nevertheless, the general consensus is that a safe walk to a playground should be short, protected and accessible. Given the rising number of pedestrian fatalities in the City of Toronto, I thought that by taking an overview level approach to access, I may be able to isolate less safe areas for walking and see how these intersected with potential walks to playgrounds.
To generate these walks, I used the newest release of census data (Age by dissemination areas!), and mapped out the concentration of young people (age 0-14) in the City. In response to some other commentators out there, there are still A LOT of young people in the City of Toronto, with the highest densities being located in the Yorkville, Thorncliffe, and Woodbine Gardens Areas...
The next step is to get "geo-nerdy" and figure out access. As mentioned in previous blog posts, this is best accomplished through the creation of pedestrian walking network and running a nearest location algorithm. Thankfully open data Toronto to the rescue once again, a great singleline road and trail network set exists here. Building the network was a snap. and from that I was able calculate how far each population point was in time (as the kid walks - 77m/minute) from each playground location. The findings are heartening. 44% of Children live within a 5 minute walk of a playground and 88% of all children in Toronto Live within a 10 minute walk of a playground. These are excellent numbers... but there is more to analyze here!
DIVERSITY OF ACCESS
As we all know children do not walk to their closest playground but typically pick from what is local. I therefore also mapped which parts of the City have the access to highest number of playgrounds in the City (see figure below). The findings are pretty interesting:
88% of children aged 0-14 live within a 10 minute walk of at least 1 play facility;
66% of children live within a 10' walk of at least 2 play facilities;
42% of children live within a 10' walk of at least 3 play facilities;
10% of children live within a 10' walk of at least 5 play facilities; and
0.7% of children live within a 10' walk of 8 or more play facilities
So, now we have a reasonable idea of access. But as I mentioned, what I am really after is safe access to playgrounds. What I wanted to highlight then was the shortest paths to playgrounds that run alongside, or cross a major arterial. Thankfully there are not many of these but they bear investigation. First, based on the theoretical 3,702 trips to their closest playgrounds, I noticed that 10% of all these trips in terms of distance run alongside major arterial roads, 17% along collectors, 52% along local roads and only 7% along trails. In terms of trips, 18% of all shortest path trips travel some appreciable distance (50m+) along major arterial roads.
That being said, walking alongside a major arterial, is not nearly as dangerous as completing a crossing. I recently read an interesting article that basically states that Children up to their early teenage years had difficulty consistently crossing the street safely. I'll let you read the article, but suffice to say, crossings are important... Therefore, based on my preliminary analysis, it seems that 15% of all trips involve at least one major arterial crossing to get to a playground. Now, full caveat here, I am looking at theoretical shortest path routes and not actual routes. In reality, many children will not take the shortest path to their local park but the safest path... However, with that in mind the figure below is still indicative of potentially dangerous crossings.
BONUS: WHICH AREAS ARE SPOILED FOR CHOICE AND WHERE COULD THE CITY BENEFIT FROM A NEW PLAY FACILITY OR TWO
If you managed to read this far, then you are certainly into geospatial analysis, so here is one more for you. Basically, with all of these data we can also find out which areas of Toronto have more playgrounds to access within 10 minutes given the population of children in the City. If that didn't make any sense, let's try this way: I am trying to locate neighborhoods where the expected density of playgrounds is less than the expected density of children using a 10 minute walkshed.
To make this analysis happen, I found out for each dissemination area what is the total population of young people within a 10 minute walk as well as the total number of playgrounds. I then divided these two numbers (and divided population by 100) to get a measure of playgrounds per 100 children within a 10 minute walking distance. I then layered on locations of under-serviced densities to gain a full picture of access plus playground and population density. The output is pretty cool, if a little busy. To read the map, you can either look for green areas that have many playgrounds per 100 children (these could be areas with too many facilities), or look for the black areas that have no playgrounds and therefore have under-serviced populations (these are ares that may require facilities). Areas that are orange are basically at the mean of 0.7 playgrounds per 100 children within a 10 minute walk!
As with all of my work, please let me underscore that this all preliminary analysis, subject to change. yes, it looks professional and cool, but this is work done on my own time, between projects, and therefore may be subject to validation and analytical errors that I have not been able to suss out. Please recognize this when you review or share this work. As always, comments are more than welcome!