M II A II R II K
Senior Member
The less denser parts including the 905 is what you see from Lawrence and on up with a lack of townhomes going up in already existing areas.
|
|
|
Density and Subways, revisited
- Thread starter M II A II R II K
- Start date
k10ery
Senior Member
memph: Interesting but, um, what the hell does a population-weighted average population density mean? (Do the math. If all the people in the city lived on one street, it might be meaningful I guess.)
Palma
Senior Member
Because the townhomes at Bayview are of a different quality than those built in other areas of the suburbs. Thats the problem. People have seen cheap looking townhomes with poor quality.
k10ery
Senior Member
Which is of no relevance to transit planning. You want a measure of dispersion in density of population by census tract. Population-weighted population density is (approximately) the second raw moment of the distribution. That's not a measure of dispersion because it is not a central moment. It overweights high-population places by construction. A better measure of dispersion in this case is the percentage of tracts with population density over a given threshold ... which is what the transit planners use.
Memph
Active Member
Fair enough, but a population distribution graph doesn't paint a particularly different picture.
The bulk of the population of North York, Scarborough, Etobicoke, Mississauga and Brampton live at pretty similar densities with a little bit of high density in each (except Brampton which has the same moderate density throughout). Old Toronto has a lot more high density... which is not surprising since the weighted density is higher as well. There are a few differences between those 5 suburbs, but all in all, they're pretty similar, as their weighted densities suggest.
k10ery
Senior Member
Fair enough, but a population distribution graph doesn't paint a particularly different picture.
The bulk of the population of North York, Scarborough, Etobicoke, Mississauga and Brampton live at pretty similar densities with a little bit of high density in each (except Brampton which has the same moderate density throughout). Old Toronto has a lot more high density... which is not surprising since the weighted density is higher as well. There are a few differences between those 5 suburbs, but all in all, they're pretty similar, as their weighted densities suggest.
Cool picture, thanks. But the logarithmic scale makes the cities look more similar than they really are, doesn't it? You could also plot Lorenz curves of population density.
M II A II R II K
Senior Member
There could be a move for some short subway lines and elevated RTs to relieve downtown transit as well as make it a lot easier to zip around downtown.
Like a line that includes stops at Spadina, Harbord, Kensington, Chinatown, St. Patrick, Atrium, Dundas Square, Jarvis, Sherbourne, Regent Park, Cabbage Town, St. Jamestown, and end at Castle Frank.
Like a line that includes stops at Spadina, Harbord, Kensington, Chinatown, St. Patrick, Atrium, Dundas Square, Jarvis, Sherbourne, Regent Park, Cabbage Town, St. Jamestown, and end at Castle Frank.
Memph
Active Member
I don't know... it sounds like the stop spacing would be quite small. I think using streetcars and buses with signal priority would mostly be able to serve the need for transit within downtown. Many of the densest parts of Old Toronto aren't actually inside downtown but in Parkdale, High Park North, Midtown, around Cosburn. Admittedly we're talking about residential density, not residents+jobs though...
And k10ery, the logarithmic graph does make Old Toronto look closer to the others, but you can tell that the bulk of Old Toronto is basically about 2 times as dense and the low and high density portions of Old Toronto are about 3-4 times as dense as the low and high density portions of the suburbs. However, using a logarithmic graph also allows to distinguish between the suburbs better, since they would otherwise basically be squashed together at the left side of the graphs. I can make a Lorenz curve sometime later.
Glen
Senior Member
Planners, professional and armchair alike, should be careful of giving equal accord to both population and employment density. Employment density is a much larger factor in determining ridership potential than residential density. NY and LA have similar population densities yet LA would never be able to reach the ridership levels of NY with an equal amount of track miles.
In Toronto's case, other than possibly the DRL, most transit expansion plans are foolish.
http://www.ppic.org/content/pubs/report/r_211jkr.pdf
In Toronto's case, other than possibly the DRL, most transit expansion plans are foolish.
http://www.ppic.org/content/pubs/report/r_211jkr.pdf
M II A II R II K
Senior Member
Paleo
Active Member
I had to look very closely before I realized why the old city looked so similar to the old suburbs. If the point was to show how similar they were, it's very misleading visually.
Memph
Active Member
The old city doesn't look very similar to the old suburbs... if you look at the graph you'll see that the bulk of the old city lives at 5000-15000people per km2 while the bulk of the suburbs (new and old) live at densities of 2000-8000 people per km2. Meanwhile the old city has hardly anyone living at densities below 4000/km2, which is still a common density in the suburbs (even 2000/km2 is relatively common in the suburbs). The point was to show that the old city is a fair bit denser than the old suburbs, and that the old and new suburbs are pretty similar. Also, if the graph wasn't logarithmic, you'd either have to hide everything above 20,000 people per km2, or squash the bulk of the population into the left side of the graph so that the suburbs are even less distinguishable from each other.
@Glen, I most definitely do agree that employment density is important, and I agree that there is not enough focus on that. However, I'm not sure what makes you say New York and Los Angeles have similar population densities... New York has way more dense housing than Los Angeles, assuming the threshold for high density is set quite high (Los Angeles might come out ahead in the 10,000-15,000ppsm range, but I don't consider that particularly dense). Also, downtown Toronto is quite well served by transit, I'm not sure how many people commute to the CBD, but once the portion of people commuting into downtown by car reaches a low enough level, it's not really worth making improvements.
Meanwhile, there are something like 2million jobs outside downtown where most people drive, so there's a lot of room for improvement there. The improvement doesn't have to be made only through transit investments of course, a lot could probably be done by adding employment around subway stations and maybe GO stations, bus terminals, etc. Not all the transit investments would have to be subways either, you could have BRTs or GO improvements too. Finally, it's still worth considering the potential for future development, as the land around existing transit lines gets intensified, new rapid transit lines can be considered in areas with more moderate densities that have land that can easily be intensified.
Memph
Active Member
So I think this is what k10ery wanted. Again, the old city clearly has more density. If you measure rapid transit suitability as "population above x density", North York is second most suitable... although several of its high density areas already have subways. The third most suitable depends on the density threshold. I think it's still fair to say though that the suburbs aren't all that different from each other. I have a version on the graph that's linear on the x axis in addition to logarithmic since that's what some people seem to prefer (I still prefer logarithmic).
nfitz
Superstar
Excellent work Memph!