Transit Fantasies

[waves]

Newton-Guildford Skytrain?

[Migrant_Coconut]

Yup.

[fredinno]

Originally Posted by waves

Fredinno, I have thought about that route too. The issue with it technically is that the inlet is deeper there. For my alignment in green, I have to start the bored tunnel at 1st avenue in order to get it deep enough (supposedly*) under the ocean floor. If we were to go with your route, you would have to do about the same (meaning the majority of that section has to be bored tunnel anyways and can't be at grade). Migrant Coconut also makes good arguments from a planning perspective.

Edit: Also, about the curve at the Quay on to Chesterfield, you might be able to have the line turn eastward sooner and follow that ridge before turning north-east straight up Chesterfield.

*I say supposedly because I am not a geologist and I haven't seen the stratigraphy along the route. This publication has some good information: https://www.gac-cs.ca/publications/J...verGeology.pdf.

https://www.google.com/maps/@49.2673.../data=!5m1!1e4
I couldn't find any topography off of the stuff you showed me. I don't know how you managed to calculate the grades; But you're at least right with the issue that you'd have to start digging (probably near Venables St. or 1st St. at worst) just go get deep enough into the inlet.

My goal with the purple Line was just to provide a high-speed feeder on the Expo, Millennium, and a future Hastings Line/extension, with stations only transferring to them, so that the line could be made more cheaply, not to penetrate deep into East Vancouver.

It's already going to cost a fortune to get under Burrard; we need all the savings we can get to make it more palatable (allowing to more cheaply reach its primary goal of North Vancouver), and not put it in development hell.


It's not a bad line (it's actually great), I'm just trying to think less 'fantasy-like'.

Originally Posted by waves

Oh the list wasn't complete or in any order. I have no idea what should come next. Right now as far as I can see things politically, in order:

1. Langely Skytrain Extension + SNG-White Rock BRT.
2. UBC Skytrain Extension
3. SFU Gondola
4. Something for the North Shore, idk what
5. Canada Line Reno's and Extension (3 car trains, 50m long platforms, extension to Ladner)

If the line goes to 49th, that would be 4km and 2 stations, so $500 million extra.

Nothing for Hastings?

[Migrant_Coconut]

Originally Posted by fredinno

It's already going to cost a fortune to get under Burrard; we need all the savings we can get to make it more palatable (allowing to more cheaply reach its primary goal of North Vancouver), and not put it in development hell.


It's not a bad line (it's actually great), I'm just trying to think less 'fantasy-like'.

Waves added a cost estimate with the map: $2.4 billion. In comparison, we're spending about $2.8B on the Broadway extension and $2.93 on the Langley extension & BRT, both at the same time. Realistic enough for all practical purposes.

Originally Posted by fredinno

Nothing for Hastings?

Not before all the others on the list. Surrey, UBC and the North Shore are going to be hogging most of the funds for the next 10-20 years.

[Jasoncw]

I saw in an article that McCallum claims that the Skytrain extension to Langley could be much cheaper by running it on the ground for a portion of the route. Does anyone know of any cost breakdowns, either for running it right on the ground like in an old rail ROW like much of the Skytrain, or in a slightly elevated structure like this (https://goo.gl/maps/ygrU3XW1tVp EverLine Line in South Korea)?

The old Surrey RRT3 alternative was useful because it only contained elevated guideway, and the alternatives analysis was detailed enough to be able to roughly separate the cost of the guideway compared to the cost of vehicles/stations/O&M center, while the Evergreen extension was a mix of at, below, and above grade, without separate cost information for the separate chunks.

Basically I'm looking for any kind of reliable generalized Skytrain cost numbers. I'm very interested in cost numbers, especially cost numbers for running it on the ground, since I've never been able to find information about that. Presumably the expensive part is the elevated guideway structure, and not the rails/power collection system/etc, but there's not information about that in my little collection of .pdfs.

Originally Posted by waves

You could yes. To complete your alignment below from Victoria to Bridgehouse. So at $114 million per km above grade (11.4km), $270 million per km bridge (1.0km), $30 million per station (8), you result in $1.81 billion (2015 dollars) to complete that southern portion. If you only went to Marine Drive, it would be only $800 million (6km, 3-4 stations).

Do you have a source for this breakdown? Your Lonsdale Line is wonderful.

[fredinno]

Originally Posted by Migrant_Coconut

Waves added a cost estimate with the map: $2.4 billion. In comparison, we're spending about $2.8B on the Broadway extension and $2.93 on the Langley extension & BRT, both at the same time. Realistic enough for all practical purposes.



Not before all the others on the list. Surrey, UBC and the North Shore are going to be hogging most of the funds for the next 10-20 years.

Yeah, I have no idea how he came to that number for the Lonsdale Line when Broadway is $2.8B.

The bored section of Broadway is close to the same length as just the Lonsdale Line Segment all the way up. (+ - 0.1km) [unless he plans on Cut-and Cover- which would be impractical due to high densities in the region] , includes a bored tunnel though East Vancouver, and a tunnel into Burrard Inlet, and is LESS than the Broadway Extension?

I don't buy it.

[waves]

There are many ways to get topography for land. Caltopo.com provides one way that is quick and easy: just plot a line, then right click and select profile. For underwater it's harder, bathymetry maps are harder to come by. You can look at the ground profile yourself on my caltopo map for the lines I have below.

For a more complicated way, you could probably find lidar data or use the 1m contours. http://geoweb.dnv.org/data/metadata....=ElvContours1m

As for getting the grades, once you have the ground profile, and the station locations you can design their depth and the alignment grades. Grade % is calculated by rise/run.

Nothing for Hastings?

My list was based on the what in the news/on people's minds politically. Hastings isn't really in that category yet.

[Migrant_Coconut]

Originally Posted by fredinno

Yeah, I have no idea how he came to that number for the Lonsdale Line when Broadway is $2.8B.

The bored section of Broadway is close to the same length as just the Lonsdale Line Segment all the way up. (+ - 0.1km) [unless he plans on Cut-and Cover- which would be impractical due to high densities in the region] , includes a bored tunnel though East Vancouver, and a tunnel into Burrard Inlet, and is LESS than the Broadway Extension?

I don't buy it.

Two words: property values. Right now, Broadway is practically part of downtown, and even a Denny's or IHOP sells for over $20M; the estimate for the subway was $1.9B just last year.

The Canada Line cost about $2B ($2.3B in 2018 CAD) in part because it went through pre-density suburban and demiurban areas; even if Commercial is higher because of TBM costs, we're not going to see one of the LRT shills' moronic $3B+ figures.

[waves]

My secondary source for the cost numbers stem from Daryl's Blog post here: https://blog.daryldelacruz.com/2015/...ansit-systems/. If he's reading this and has a newer post out I would be interested to see it

If you want to take a closer look at my calcs you can follow along in my crude spreadsheet here: https://1drv.ms/x/s!AjWWltw_HSwzh8lAGO2et-WqzET3EA

I then took the spreadsheet and expanded it out to include km of each grade type then used Excel Solver to minimize an RSS. This sounds complicated but it isn't. Skip the bullets below if you are not interested in the process.
- Each line is made up of 4 components (with the data I have): (a) Above Grade (b) Bored Tunnel (c) Cut and Cover (d) At Grade.
- Each line has a number of km that it travels at each type, so we have variables a, b, c, d.
- The total cost = aw+bx+cy+dz, where w, x, y, z is the fixed cost per km for each grade type.
- If we set w, x, y, z arbitrarily at $500 mill per km. Then we can calculate the cost for each line using these fixed variables.
- We then compare the cost of each line to the stated full cost. To do this, we take the difference.
- Now, with the wxyz variables arbitrarily set, the costs will be wildly different, which means the difference in cost will be high. The RSS method, takes the sum of the squares of all the differences of the lines you are using, then takes the square root. You can't just sum the differences because some might be positive and some might be negative. Also the RSS method produces a number more sensitive that just summing the absolute values.
- The last step is to use the Excel Solver. What the solver does is iterate through potential wxyz values to try and get the RSS number as close to 0 as possible. If it does get to 0, that means there is a perfect correlation between all the lines. The larger the RSS, the more variability and error there is in those wxyz values. That can be random, or it could be systemic in that maybe there is a factor in the cost that we are ignoring that would account for that variability.

So, now the questions is, which lines in the table can we use? It's a good question. I had been using the "Only BC" numbers since my RSS results were the lowest, and most Vancouver related anyways. The table below however shows the range of $mill/km values you can get and the cost estimate associated with each assumption:



The Option 2 for the Langley Extension is based off of the map from Global BC here at about 55 sec. Technically there is about 9km of At-Grade described in the map, but I reduced that to 8km for an overpass at Hwy 15.

Also fredinno, Cut and Cover for 13th to 23rd for the Lonsdale Line is suggested on the profile.

[fredinno]

Originally Posted by waves

My secondary source for the cost numbers stem from Daryl's Blog post here: https://blog.daryldelacruz.com/2015/...ansit-systems/. If he's reading this and has a newer post out I would be interested to see it

If you want to take a closer look at my calcs you can follow along in my crude spreadsheet here: https://1drv.ms/x/s!AjWWltw_HSwzh8lAGO2et-WqzET3EA

Also fredinno, Cut and Cover for 13th to 23rd for the Lonsdale Line is suggested on the profile.

Thanks.

Though, looking through the numbers, I can't help but feel that this is a low-ball estimate.

I'd be iffy on Cut-and-cover on 13-17th; it's just too dense yet to start cutting up the road. Starting the cut on 17th avoids the remaining skyscrapers

You also list Broadway to Artubus as $1.8 Billion, clearly too low, even for 2010 numbers. The actual cost is increased heavily due to real estate costs- and Lonsdale is not cheap.

I'll try to reflect this with a 25% contingency cost for both yours and my lines.

Also, the number of Stations for the Lonsdale is 9, not 8. Unless it's a spur of the Millennium?

Lastly, there's no cost estimate for At-grade- BC. I'll assume $50M/km, to make the comparison more fair.

Where is the section above-grade? I want to make this as fair as possible, so I didn't touch it.



With these numbers:

With a 25% contingency for RE costs, the Lonsdale Line is $3080M, and the Purple Line is $2498M.

I doubt either would be built all at once, considering both would have comparable costs to the Broadway Extension, which is already an unprecedented amount of money on one extension.

The Purple Line is about 20% cheaper for the entire route. Is that worth bypassing Commercial St?

I'll leave that for everyone else to decide.



In conclusion: Admittedly, the Commerical-Broadway Line, though underpriced, is not as bad as I thought. Definitely some advantages, especially in a high-growth scenario where the land around current Skytrain Lines gets built up.


PS: Made a few errors, the Lonsdale line should be $3.15 Billion; see my post below for more details.

[Sheba]

Here are a couple other ideas for getting Skytrain to the North Shore (I'll leave the NS route / stops to people who know the area better than I do).

1) A Willingdon Line from Metrotown, past BCIT and Brentwood to Kootenay Loop and then a bridge next to the Iron Workers Bridge to Phibbs Exchange and beyond

2) A Hastings Line starting at Kootenay Loop to downtown Van, through Stanley Park and then a bridge next to Lions Gate Bridge and east to Lonsdale Quay and beyond

[Migrant_Coconut]

Originally Posted by Sheba

Here are a couple other ideas for getting Skytrain to the North Shore (I'll leave the NS route / stops to people who know the area better than I do).

1) A Willingdon Line from Metrotown, past BCIT and Brentwood to Kootenay Loop and then a bridge next to the Iron Workers Bridge to Phibbs Exchange and beyond

2) A Hastings Line starting at Kootenay Loop to downtown Van, through Stanley Park and then a bridge next to Lions Gate Bridge and east to Lonsdale Quay and beyond

Can't find the study itself, but the findings are that a Downtown-Lonsdale crossing gets the most ridership.

[fredinno]

I made a mistake above: it drastically changed my numbers. Forgot to add the extra bored tunnel cost for up to 17th St. I also accidentally left an extra station on the purple line, and a few other minor things.



The Commerical-Lonsdale line is now 22% more costly than the Purple Line.


The difference of 2% isn't a deal-breaker though, so I'll still just leave my previous analysis intact.

[Sheba]

Originally Posted by Migrant_Coconut

Can't find the study itself, but the findings are that a Downtown-Lonsdale crossing gets the most ridership.

...the longest route with the deepest section - no wonder it has the Seabus.

[waves]

Originally Posted by fredinno

Though, looking through the numbers, I can't help but feel that this is a low-ball estimate. I'll try to reflect this with a 25% contingency cost for both yours and my lines.

The numbers I presented are not biased to my feelings. They are a derivative of past projects. You need justification for increasing cost/km on a certain section. For example, the bored tunnel will be going under water so maybe you add a 25% price increase for bored tunnel section under water. You can't just add 25% contingency to increase the final cost to a number you like: that's biased.

Originally Posted by fredinno

I'd be iffy on Cut-and-cover on 13-17th; it's just too dense yet to start cutting up the road. Starting the cut on 17th avoids the remaining skyscrapers

Just because there are taller buildings beside the road doesn't mean that you can't have cut & cover in the road. Buildings don't go under roads typically and the ones on Lonsdale are not atypical. I stand by the ability to use cut and cover from 13th upwards.

Originally Posted by fredinno

You also list Broadway to Artubus as $1.8 Billion, clearly too low, even for 2010 numbers. The actual cost is increased heavily due to real estate costs- and Lonsdale is not cheap.

This does not include property acquisition costs, as MC correctly noted.

Originally Posted by fredinno

Also, the number of Stations for the Lonsdale is 9, not 8. Unless it's a spur of the Millennium?

The number of stations is 8. Commercial Broadway exists already. Re-look at the graphics. With the purple route, there'd be no point in having a station at 1st (or Venables really) so you'd have 6-7 stations.

Originally Posted by fredinno

Lastly, there's no cost estimate for At-grade- BC. I'll assume $50M/km, to make the comparison more fair.

This was a mistake. The At-Grade cost for BC only is $39 mill/km. Spreadsheet should be updated now. Doesn't matter though. Lonsdale Line doesn't use any at-grade sections.

As for your numbers I can't verify them because you didn't use the spreadsheet properly. You should have a column for each option like this:



As you can see, cost for the Green line is still cheaper than the purple line, even with more stations.

https://caltopo.com/m/15Q5

[waves]

Also saw in the Arbutus Thread that Commercial/Victoria is the third busiest bus route in Vancouver (1st is 99, 2nd is 41). With connection to the North Shore, that Commericial Corridor has a very high ridership potential.

http://dailyhive.com/vancouver/trans...est-bus-routes

Top 10 busiest bus routes in Metro Vancouver in 2016
Based on annual boardings (operating cost per boarded passenger):

99 UBC/Broadway Station B-Line: 17,288,000 boardings ($0.60 per passenger)
41 Joyce Station/Crown/UBC: 9,019,000 boardings ($0.94 per passenger)
20 Victoria/Downtown: 8,371,000 boardings ($0.93 per passenger)
49 Metrotown Station/Dunbar/UBC: 7,506,000 boardings ($0.87 per passenger)
25 Brentwood Station/UBC: 7,409,000 boardings ($1.07 per passenger)
16 29th Avenue Station/Arbutus: 7,200,000 boardings ($1.19 per passenger)
9 Boundary/Broadway Station/Granville/Alma: 7,137,000 boardings ($1.12 per passenger)
135 Burrard Station/SFU: 6,513,000 boardings ($1.08 per passenger)
3 Marine Drive Station/Main/Downtown: 6,249,000 boardings ($1.01 per passenger)
106 Metrotown Station/New Westminster Station: 5,948,000 boardings ($0.94 per passenger)

[Migrant_Coconut]

Fun fact: all of the top ten outperform San Fran's entire cable car network.

[GMD]

Originally Posted by Sheba

Here are a couple other ideas for getting Skytrain to the North Shore (I'll leave the NS route / stops to people who know the area better than I do).

1) A Willingdon Line from Metrotown, past BCIT and Brentwood to Kootenay Loop and then a bridge next to the Iron Workers Bridge to Phibbs Exchange and beyond

2) A Hastings Line starting at Kootenay Loop to downtown Van, through Stanley Park and then a bridge next to Lions Gate Bridge and east to Lonsdale Quay and beyond

Ideally, these could connect as part of a single line (which would cross itself around Kootenay Loop - Hastings/Willingdon), which would of course, then extend from Metrotown, west along 41st and up to UBC... in my transit fantasy I refer to this as the 'Sea to Sky Line' (starting at the sea in UBC, and ending, via Gondola from Hastings/Duthie in the sky at SFU)

[Bdawe]

Originally Posted by Migrant_Coconut

Fun fact: all of the top ten outperform San Fran's entire cable car network.

I mean, it's a cable car network? It's two lines with cars that go no more than 10 km/h that have been technically obsolete for all purposes since the development of the trolleybus, and cost an extra fare?

That the cable cars still out-preform a lot of intended-as-transit new build streetcars whats really notable.

[Migrant_Coconut]

Originally Posted by Bdawe

I mean, it's a cable car network? It's two lines with cars that go no more than 10 km/h that have been technically obsolete for all purposes since the development of the trolleybus, and cost an extra fare?

That the cable cars still out-preform a lot of intended-as-transit new build streetcars whats really notable.

Hah, fair enough. Was trying to avoid ragging on Portland again (their Streetcar has similar numbers).