Bike Lanes on Hudson Street from Indianola across I-71: An Inexpensive Paint and Barrier Proposal
Last week, I was almost run off the road by a car on Hudson Street. This happened in the portion of westbound Hudson Street, just after Fourth Street, where the painted bike lane stops at the intersection, and bicyclists must merge with traffic.
It's a dangerous place to ride a bike, but it's also the quickest way for a cyclist to get to the protected bike lane on Summit Avenue. For most of Hudson, the only acknowledgement of non-car traffic is some painted sharrows — yet Hudson Street is a major east-west route, part of a planned connection between the Olentangy and Alum Creek trails. Hudson connects the Summit and 4th street bike lanes to the planned bike lanes on Indianola Avenue.
Despite this key position in Columbus' bikeway network, this stretch of Hudson is built only for cars and trucks.
This essay presents a proposal for building protected bike lanes on Hudson Street from the Indianola bike lane all the way across I-71 to the Hudson Street Shared Use Path. This proposal uses a combination of Jersey barriers and precast curbs for fast, low-cost installation of protected bike lanes, in combination with a road diet and a lot of paint.
TL;DR: Remove one travel lane from Hudson, and replace it with two 5' bike lanes protected from traffic by concrete barriers. The remaining lanes are generally configured as one lane in each direction with a center turn lane.
I think Columbus could do it for $550,000.
Skip to cost estimate Skip to FAQ section Comment on this post on Reddit
From Indianola towards I-71
We're going to scroll from the western end of this project to the eastern end. I've broken it up into separate images for easier annotation, but you can also view this as a single image.
In this map, west is up, north is right, east is down, and south is left.
From I-71 to Indianola
Now that you've traveled the whole route from west to east, scroll up to travel from east to west. Or view this whole map as a single image.
Note how westbound scooters and bikes don't have to merge into vehicle lanes at all, with this new route. Nor are they forced to turn onto 4th Street in order to achieve safe passage to Summit or Indianola: these road users are now given the same privileges as motor vehicles.
Cost estimate
Here's my back-of-the-envelope cost estimation math:
Item | Cost per each | Cost |
---|---|---|
0.5 mile of Jersey barriers | $1000 per 8' | $333,000 |
0.5 mile of precast concrete curbs | $103 per 6' (source) | $45,320 |
Paint | unknown | $100k? |
2 new curb ramps | $3,600 (2013) | $8,000 |
4 loads of asphalt to create elevated bus stops in bike lanes | $1,200/load | $4,800 |
Revised signage | unknown | $10,000? |
Installation costs | included | |
Subtotal | rounded up | $500,000 |
Consultant fees, to make the hand-drawn maps into architectural drawings | 10% | $50,000 |
Total | $550,000 |
Half a million dollars for two half-mile segments of protected bike lanes, not including traffic signal work. Does that seem reasonable? I don't know.
For comparison:
- Chicago spent $17 million for 100 miles of bike lane.
- Columbus spent $20 million to completely repave Hudson from I-71 to Cleveland Avenue (1.4 miles) including adding a shared-use path, rebuilding buying slivers of a bunch of houses' yards, rebuilding 4 traffic lights, and reconstructing several dozen driveways and yard stairs. And installing a new water main.
My cost estimate seems kind of high.
Frequently Asked Questions
You almost got run off the road?! Are you okay?
I'm uninjured but I'm pissed off.
What is Columbus doing for cyclist safety?
Columbus built one (1) protected bike lane on Summit Street in 2016, covering 1.4 miles. Every other bike lane in the city is unprotected, with at most a painted buffer zone separating multiton vehicles from bike and scooter riders' bodies. Columbus has spent so little effort on expanding bike lanes that the city webpage for biking describes "protected bike lanes" with a link to a webpage that describes the Summit lanes (built in 2016) in the future tense. Columbus hasn't updated their bike webpages since before 2016. Columbus hasn't built any protected bike lanes since 2016. The current year is 2022.
What about Columbus' Vision Zero plans?
To be fair to Columbus, the 2021 Vision Zero Annual Report shows that Columbus has been upgrading traffic signals and crosswalks, and they might finally complete the unprotected Indianola Avenue bike path and the Hudson Shared-Use Path by 2024.
The delays in implementing the Indianola bike lane have resulted in poor reviews of the planning process:
So three years of data collection, public meetings, consultant fees and staff time for less than three linear miles of roadway improvements. Columbus is a big, spread-out city, with 5,678 lane miles of roadway, according to the Department of Public Service. As they like to say in the venture capital world, this process is just not scalable.
Columbus' public messaging around Vision Zero principles is similarly lacking. A shining example is this tweet from Columbus' Vision Zero account: A reminder that riding bicycles on sidewalks is illegal, accompanied by a photo of a weedy sidewalk, a trash-filled gutter, and scrap metal in a painted bike gutter approaching an intersection where signage doesn't remind drivers that cyclists may be present. Columbus' Vision Zero messaging appears to be focused on changing cyclist and pedestrian behavior, rather than installing physical protections for non-drivers.
"Driver flees scene after hitting bike" is a headline that happens about once a month in Columbus, in 2022. Chicago takes visible, rapid action to address danger spots. I'm not sure what Columbus does.
It seems like you're asking a lot of Columbus.
Meanwhile, the city of Chicago announced plans on June 29, 2022, to upgrade most bike lanes in the city to have physical barriers between bikes and vehicles. Chicago will add 25 miles of protected bike lanes by the end of 2022, and finish the upgrade process in 2023. Since 2020, Chicago has added more than 125 miles of bike lanes to their network. Columbus paints a few miles a year, at most.
Columbus can do better.
Who is your design user for this project?
Referencing section 3.2.1 "User Profiles" of ODOT's Multimodal Design Guide, my design user profile is the "interested but concerned" person:
Often not comfortable with bike lanes, may bike on sidewalks even if bike lanes are provided; prefer off-street or separated bicycle facilities or quiet or traffic-calmed residential roads. May not bike at all if bicycle facilities do not meet needs for perceived comfort.
This profile represents half of the total population in Ohio. By selecting this user as the design target, this design promotes a bikeway that is accessible to everyone.
Is this just for bicyclists?
No. These bike lanes are a safety improvement for bicycles, scooters, rollerblades, wheelchairs, wagons, micromobility devices, unicycles, skateboards, strollers, dogs, children, parents, pedestrians, and anyone who isn't ensconced in a multiton steel box traveling at a high rate of speed.
Don't you worry that cars might collide with the Jersey barriers?
If a car collides with a human being, that human might die. If a car collides with a Jersey barrier, the car can be repaired, and the Jersey barrier has done its job to prevent the car from going where the car should not go. My only worry is why the car hit the barrier. I'm not worried about the car; cars are replaceable and humans are not.
Don't you worry that giving up road lanes to entitled cyclists and crazy scooter riders will make drivers mad? And mad drivers might try to collide with bikeway users?
If a driver is driven to homicidal rage at the thought of sharing the road with other people, than that person should not be a driver. Homicidal rage should not control a vehicle moving at 35mph, or indeed at any speed.
To help prevent collisions between motorized vehicles and unprotected humans, this bike lane proposal uses physical barriers between traffic modes: Jersey barriers and precast concrete curbs of the sort you see in parking lots.
Why do the bike lanes on Hudson need physical separation?
35mph, more than 6000 vehicles per day based on Indianola traffic surveys: The traffic characteristics of Hudson exceed the National Association of City Transportation Officials' guidelines for protected bicycle lanes. Hudson's street conditions also meet the criteria in the Federal Highway Administration's Bikeway Selection Guide and in ODOT's guide to install a separated bike lane or shared-use path. Hudson meets the criteria to protect its bike lane.
There's an actual legal justification for physical separation?
Section 2.5.2.2 "Bikeway Feasibility Assessment" of the ODOT Multimodal Design Guide says, "There are a variety of conditions that may indicate the need for greater separation between motorists and bicyclists, which could increase the width of the bikeway or materials used in the buffer," listing 9 criteria. Hudson meets at least 6:
- High percent of heavy vehicles (trucks and buses)
- Vehicle speeds exceed posted speed (frequent speeders in a 35)
- Presence of vulnerable populations (old folk)
- Network connectivity gaps (gap between Summit/4th and Indianola)
- Proximity to transit (many COTA and non-COTA bus routes on Hudson)
- Frequent driveways.
Why do you propose using Jersey barriers or precast concrete curbs for lane separation?
Precast concrete curbs are used by Chicago for most bike routes.
Jersey barriers provide more protection against larger vehicles, and are why I recommend them at intersections and on the I-71 bridge. The Jersey barriers' large vertical surfaces are an excellent art program opportunity; NYC DOT has operated a "Barrier Beautification" program on protected bikeways in the past.
Anything smaller than a 6" curb is going to get smashed to pieces by drivers who are oblivious, inconsiderate, or malicious. Look at the flexible plastic bollards installed on the Summit bike lane: After 6 years, many bollars are missing, and they do not stop cars from parking in the bike lane. Columbus needs better protection for road users who aren't in a multiton armored steel cage. Knee-high concrete barriers successfully stopped a car in Toronto; Columbus would benefit from that approach on parking-less Hudson Street.
Why are all your crosswalks using ladder-style crosswalk lines instead of Columbus' most-common method: two parallel lines marking the crosswalk edges?
In my experience as a pedestrian and cyclist and driver in Columbus, Columbus drivers will treat straight lines running across the road as stop bars, instead of as a crosswalk. Drivers will advance until the furthest-forward line crossing the road is somewhere under their vehicle. High-visibility crosswalks like the ladder style don't look like stop bars. Perhaps drivers won't treat them as stop bars.
Why aren't the lines in your illustrations straight? Why do you use two different colors of green for the bike lane paint?
Because I don't have access to professional transit planners' software. I did the linework by hand in paint programs. This is an amateur's impression of what a bike lane would look like.
What's with the weird raised bus stop designs?
Putting the bus stop in the bike lane, with bikes yielding to pedestrians, is a variant of the common "floating bus stop" pattern already used on Summit. For a US implementation of the specific design referred to in this plan, see the 4th Street Corridor in Charlotte, North Carolina.
Are you worried about turning traffic at alleys and driveways?
Yes. But that's also why the bike lane is protected, and heavily marked: So that turning vehicle traffic is aware of the bikeway traffic. The protected bike lane's physical barriers will require vehicle traffic to slow down to cross the bikeway, which reduces the risk of collisions for sidewalk and bikeway users.
What are potential obstacles to building this bike path?
- Car drivers may object on the grounds that this will slow down traffic in the area. They should be politely listened to and then ignored, because Hudson is unsafe. The posted speed limit is 35 and driver speeds are often as high as 45mph, depending on congestion and whether someone is trying to run a red light.
- The police department and fire department and ambulances will want to be able to bypass traffic in this corridor. Part of the reason I included a center turn lane for most of the route was to provide a safe place for emergency vehicles to pass.
- The snow plow department will say that they won't be able to plow the bike lanes. They already don't plow the bike lanes, so I don't see why this is a valid objection. The city should do do what New York did and buy a fleet of Holder C70 and Holder S100 articulated compact tractors equipped with plows and scrapers. Other options exist; check out StreetsBlog's guide to clearing snow from protected bike lanes. This would cost a few million dollars in vehicle and staffing costs, and as a bonus could be used to clear out sidewalks from the snow barriers created by Columbus' larger plows.
- Business owners in the area may object, but unlike the Indianola bike lane, this plan doesn't remove any parking. This proposal merely makes it easier for people to get around on bikes and scooters and skateboards. That should help bring customers to these businesses.
- Clintonville and SoHud residents may object to "ugly" Jersey barriers. This can be accounted for with a neighborhood art program, such as an "Adopt a Jersey barrier!" scheme. Raffle some off, sell others, and donate some to charity. Columbus might even turn a profit on this venture.
- Changes to the Hudson bridge over I-71, and to the stretch of US-23 that runs along Hudson between Indianola and 4th, may require coordination with federal agencies.
- The two-stage bicycle turn boxes depicted in this plan are allowed under FHWA interim approval, according to section 1.2.2 of the ODOT Multimodal Design Guide. However, additional regulatory approvals may be needed.
What can we do to get this built?
- If you're a normal person like me, you can share these suggestions with your contacts in city government and with city council. Talk to your local area commission. Find your neighborhood liaison and drop them a line. Fill out the 2022 Capital Investment Input Form and let the city know your spending priorities.
- If you work for city government, you can try to get this plan implemented, or a plan like it.
- If you're a city council member or if you're Mayor Ginther, you could take a page from Chicago. Read the Chicago Streets for Cycling Plan 2020 and steal all you can from it, to give Columbus a bicycle accommodation within 1/2 mile of every Columbus resident. Put it in the 2022 Capital Investments Budget. Get those revisions to the standards manuals complete and start building new bike paths.
Anything else?
Go read Brent Warren's editorial at Columbus Underground: Indianola Bike Lane Saga Shows the Need for a New Approach.
Watch this entire video:
Credits
This blog post would not have happened without:
- Google Maps and Bing Maps, whose aerial imagery contributed to these drawings. But neither is up-to-date enough to capture the new crosswalk on the north leg of the intersection of Hudson and Silver Drive.
- The Indianola Complete Streets Survey authors.
- Photos of a car colliding with a concrete barrier along a Toronto Bike Lane, which inspired me to recommend actual Jersey barriers instead of flimsy plastic bollards.
- Not Just Bikes' video Why Cars Rarely Crash Into Buildings in the Netherlands.
- The Pedestrian and Bicycle Information Center cost estimates.
- Krita and The GIMP, and a shell tutorial on
for i in {1..11}; do cp combined.jpg "$i.jpg"; done
. - Brent Warren's editorial.
- Editing and review passes from friends.
- That driver that almost ran me off the road.