Often a center turn lane does more to increase safety and capacity than a through lane does, and you can go from a four lane configuration to a five lane by changing one of the through lanes each way to a bike lane, shift the left lanes outwards, and narrow the lanes to 11-12' max.
The phrase “5-lane” is misleading IMO, because two of those 5 lanes are bike lanes.
Just because the lane is mode- or turn-specific doesn't mean it's not a lane or doing it's part to improve capacity, and capacity isn't the only factor that needs to be accounted for. Safety needs to be the first and foremost concern for city streets. If it can't be safely used by nearly everyone 8-98 regardless of means, the design is fundamentally flawed. The irony that your average, modern interstate-standard freeway with it's hard, clean shoulders allowing for adequate separation between modes and good sightlines
like this tends to be safer than
your average american boulevard today is not lost on me (and I've driven and bicycled on both roads pictured in the links).
Also, I’m curious how removing a lane can increase capacity.
We'll use Lewis Avenue in Tulsa, for example. Currently, the left lane in both directions from the Joe Creek overpass north is four lanes, two north, two south. The left lanes are typically blocked by queues turning left, and people caught in the queue trying to continue on Lewis. A road diet layout would add a lane by consolidating the two left lanes as a single bidirectional left turn lane, which would also free space up to add bike lanes along the curb. Through traffic would not be subject to getting caught behind bicycles or left turning traffic, ergo, improved throughput. Having a dedicated lane for cycling reduces traffic conflicts, and having a dedicated turn lane increases visibility for drivers turning across oncoming traffic, as does having a buffer between the bike lane and general access lane.
"Road diets" have not yet made their way to Oklahoma, so any bike lanes that are installed are as part of a widening or elimination of on-street parking.
This is not accurate, there have been at least three street diets in Tulsa. 4th Street did not eliminate on-street parking, nor did 31st Street, nor Detroit Avenue. Nor were any of these streets widened. 4th Street and 31st Street did not have on-street parking in the first place. Detroit Avenue reduced lane width slightly and moved parking away from the curb to create a rudimentary Dutch cycletrack. Out of these, only 31st Street is currently an incomplete street, but they're also still working on it, so we'll see if it gets sidewalks as well, since they're not going to be done until a year from now come June.
City of Tulsa and TyPROS experimented with a new arrangement on 11th Street, temporarily changing the curb lane from Peoria to Harvard to a cycletrack for a weekend to demonstrate what could be done there (last I heard the most seriously considered course of action would permanently change the curb lane to a 24-hour combined bus/bike lane in anticipation of the 11th Street AERO, not that different from, say,
Hastings Street/BC 7A, save that no parking would be lost since you can't currently park on 11th Street anyway).