31 ms·

North/South can be pretty painful, but across Lake Washington isn't nearly as bad. I have a view of the entirety of the I-90 bridge which I see every day all da

North/South can be pretty painful, but across Lake Washington isn't nearly as bad. I have a view of the entirety of the I-90 bridge which I see every day all day and it doesn't get nearly as bad as coming into Seattle via I-5 or Aurora. Whether you go with roads or rail, you still have the problem of building across those geographic chokepoints. Either way, you've got the cost of building bridges/tunnels regardless of the mode of transportation.

bobthepanda 6y ago
The issue at hand is capacity. Free-flowing highway lanes move people at ~2000 people per hour per direction per lane, and the Seattle light rail has a capacity closer to ~30000 pphd. So light rail is 15x more space efficient than car travel; you might need to build a bridge, but you will generally need to build less bridge. Autonomous cars may improve road capacity but not 15x. More importantly, roads have a very noticeable capacity curve as they become congested and become less efficient; trains don't really get slower with crowding, they mostly just get more crowded inside. And I don't know, I-90 is kind of a crapshoot (especially on weekdays) when it comes to congestion. Even worse, it's bad both ways, because you have both heavy inflows to jobs in Seattle and jobs at the various Eastside centers like Redmond and Bellevue. 520 is much better, but it's also tolled, which is also why I-90 is so much worse.
- malandrew 6y ago
  pphd was a new metric for me so I did some googling because I wanted to know more. From that searching I came across the following article. Now this article has a clear bias (but so do the people advocating for rail), but bias doesn't mean that there aren't valid points being made even if exaggerated. Curious what your response is to their counter claims: https://www.washingtonpolicy.org/publications/detail/sound-transits-ridership-capacity-sounds-exciting-in-theory-but-it-means-nothing-in-practice https://www.washingtonpolicy.org/publications/detail/sound-t... One thing that stands out is that ST3 is saying it is a theoretical limit of 16,000 pphd not 30,000. Where did you get the 30,000 figure from?
  - bobthepanda 6y ago
    I can respond to some of the statistics used in this article. > The problem is that Sound Transit officials are comparing 16,000 humans to 2,000 vehicles. This is actually not that bad of a comparison; most Americans drive alone. The FHA, hardly some anti-car agency, estimates car occupancy at 1.54 persons a vehicle, and overally light-vehicle occupancy at 1.67 persons a vehicle. Unless American social mores are going to change significantly, I don't see why self driving cars would really change the fact that people like being in a car alone. > Light rail is, by definition, light capacity transit. The theoretical 16,000 people per hour requires a “target max load” of 200 people per car on four-car light rail trains at three-minute headways for a full hour, a concept unheard of in the United States. This is not unheard of at all. We have bigger light rail systems running smaller headways, to say nothing of very large traditional rail systems. New York City is running 600 foot trains every 2 minutes with an ultimate capacity of >60000 pphd. Also, this article is a bit dated; we now know that Sound Transit is purchasing the Siemens S70. The identical Portland model carries 240 people per car. And while the current service plan calls for trains every four minutes in the busiest segments, the signaling system actually supports trains every 2 minutes so long as you have the vehicles. (This is not the lower limit for rail signaling systems; trains every 90s are run on many systems around the world.) Buying new trains is cheap compared to building more tracks or lanes. > Sound Transit officials also claim that they will carry 561,000 to 695,000 unlinked daily trips by 2040 if ST3 passes. Let’s assume Sound Transit carries its highest projection of 695,000 daily trips with their estimate that most (91%) will be trips taken by existing transit riders. This means that only 62,550 will be trips taken by new transit riders. This assumes the only utility of a transit project is new trips or new riders. This isn't true; there is also value in speeding up existing commutes. The bus gets caught up in the same highway traffic and has very high labor costs per trip; moving bus passengers to a light rail train free from traffic will increase reliability and reduce commute times. And every person on a light rail train is a person not taking up space on a road. Which isn't to say that it relieves congestion, but rather it frees up space and provides an alternative to congested roads. It's also worth noting that Sound Transit's recent record shows that it is most likely underestimating ridership: in 2016, the brand new University Link extension had already doubled initial estimates of its first year ridership, and the line well exceeded its 2020 ridership goals in 2018. https://www.historylink.org/File/20720 https://www.historylink.org/File/20720 > This means that under ST3, each new transit rider will cost over $1 million dollars. It's worth noting that this figure is not qualified by how much road projects cost per road user, because this is not meant to be a comparative talking point. Roading projects are just as expensive, if not more due to higher land requirements; yet they are not scrutinized to the same degree. If you were to even try and match half of the capacity in a road expansion, it would be prohibitively expensive. The SR 99 tunnel we just built cost $3.2B for two lanes in each direction with no exits. This is double the cost of the aforementioned University Link, which is nearly a mile longer than the road tunnel to boot. The reality of the situation is any additional capacity in the transport system is going to cost a lot of money now that the city is rapidly growing and densifying, and roads cost more because they require more land.

bobthepanda 6y ago

The issue at hand is capacity. Free-flowing highway lanes move people at ~2000 people per hour per direction per lane, and the Seattle light rail has a capacity closer to ~30000 pphd. So light rail is 15x more space efficient than car travel; you might need to build a bridge, but you will generally need to build less bridge. Autonomous cars may improve road capacity but not 15x. More importantly, roads have a very noticeable capacity curve as they become congested and become less efficient; trains don't really get slower with crowding, they mostly just get more crowded inside. And I don't know, I-90 is kind of a crapshoot (especially on weekdays) when it comes to congestion. Even worse, it's bad both ways, because you have both heavy inflows to jobs in Seattle and jobs at the various Eastside centers like Redmond and Bellevue. 520 is much better, but it's also tolled, which is also why I-90 is so much worse.

malandrew 6y ago
pphd was a new metric for me so I did some googling because I wanted to know more. From that searching I came across the following article. Now this article has a clear bias (but so do the people advocating for rail), but bias doesn't mean that there aren't valid points being made even if exaggerated. Curious what your response is to their counter claims: https://www.washingtonpolicy.org/publications/detail/sound-transits-ridership-capacity-sounds-exciting-in-theory-but-it-means-nothing-in-practice https://www.washingtonpolicy.org/publications/detail/sound-t... One thing that stands out is that ST3 is saying it is a theoretical limit of 16,000 pphd not 30,000. Where did you get the 30,000 figure from?
- bobthepanda 6y ago
  I can respond to some of the statistics used in this article. > The problem is that Sound Transit officials are comparing 16,000 humans to 2,000 vehicles. This is actually not that bad of a comparison; most Americans drive alone. The FHA, hardly some anti-car agency, estimates car occupancy at 1.54 persons a vehicle, and overally light-vehicle occupancy at 1.67 persons a vehicle. Unless American social mores are going to change significantly, I don't see why self driving cars would really change the fact that people like being in a car alone. > Light rail is, by definition, light capacity transit. The theoretical 16,000 people per hour requires a “target max load” of 200 people per car on four-car light rail trains at three-minute headways for a full hour, a concept unheard of in the United States. This is not unheard of at all. We have bigger light rail systems running smaller headways, to say nothing of very large traditional rail systems. New York City is running 600 foot trains every 2 minutes with an ultimate capacity of >60000 pphd. Also, this article is a bit dated; we now know that Sound Transit is purchasing the Siemens S70. The identical Portland model carries 240 people per car. And while the current service plan calls for trains every four minutes in the busiest segments, the signaling system actually supports trains every 2 minutes so long as you have the vehicles. (This is not the lower limit for rail signaling systems; trains every 90s are run on many systems around the world.) Buying new trains is cheap compared to building more tracks or lanes. > Sound Transit officials also claim that they will carry 561,000 to 695,000 unlinked daily trips by 2040 if ST3 passes. Let’s assume Sound Transit carries its highest projection of 695,000 daily trips with their estimate that most (91%) will be trips taken by existing transit riders. This means that only 62,550 will be trips taken by new transit riders. This assumes the only utility of a transit project is new trips or new riders. This isn't true; there is also value in speeding up existing commutes. The bus gets caught up in the same highway traffic and has very high labor costs per trip; moving bus passengers to a light rail train free from traffic will increase reliability and reduce commute times. And every person on a light rail train is a person not taking up space on a road. Which isn't to say that it relieves congestion, but rather it frees up space and provides an alternative to congested roads. It's also worth noting that Sound Transit's recent record shows that it is most likely underestimating ridership: in 2016, the brand new University Link extension had already doubled initial estimates of its first year ridership, and the line well exceeded its 2020 ridership goals in 2018. https://www.historylink.org/File/20720 https://www.historylink.org/File/20720 > This means that under ST3, each new transit rider will cost over $1 million dollars. It's worth noting that this figure is not qualified by how much road projects cost per road user, because this is not meant to be a comparative talking point. Roading projects are just as expensive, if not more due to higher land requirements; yet they are not scrutinized to the same degree. If you were to even try and match half of the capacity in a road expansion, it would be prohibitively expensive. The SR 99 tunnel we just built cost $3.2B for two lanes in each direction with no exits. This is double the cost of the aforementioned University Link, which is nearly a mile longer than the road tunnel to boot. The reality of the situation is any additional capacity in the transport system is going to cost a lot of money now that the city is rapidly growing and densifying, and roads cost more because they require more land.