I remember my dad telling me that when I was in my late teens. That it’s the fourth power startled me.

To give an idea of what this means:

For a 50 tonne semi-trailer with five axles, (⁵⁰⁄₅)⁴ = 10,000 units of damage.

2 tonne vehicle with two axles, 1 unit of damage.

That one truck, only 25 times as heavy as a pretty heavy car, is doing 10,000 times as much damage to the road.

Thus indeed if you have basically any heavy freight at all on a road, the rest is negligible.

The last time this was posted I dug in to this and it appeared as though 2 vs 4 tires per axle was not tested, in the original or follow up paper usually referenced. They were testing trucks vs trucks with the same axle type, but different numbers of axles (hence axle weight instead of vehicle weight).

It could be that four tires on a truck axle are much more effective at spreading the weight than two car tires. Or it could be that it's not much better, as the forces are still placed on a relatively small area of asphalt.

How this makes cars and trucks compare is, as far as I could understand from the literature when I dug in to it, an untested open question.

There could be a 10,000 factor difference. Or if two tires at each end are spreading the weight over 2x the surface area and you assume that matters, then you have to divide that 10,000 by 2^4 and it's now only 625x as damaging.

I suspect the reason nobody cares enough to test this is that it would still be a gap large enough to make cars insignificant.

FWIW, if we assume it's based on tire weight, a 50-tonne 18-wheeler actually does (50/9)^4 ≈ 952.6 times as much damage, not 625.

> I suspect the reason nobody cares enough to test this is that it would still be a gap large enough to make cars insignificant.

That seems likely, yes.

Right, because you don't double up the two front wheels which are used for steering. I suspect that more weight is put on the rear and trailer axles than the front as well, and that two tires in close proximity aren't as good as two tires separated by greater distance. So even 1000x is probably optimistic.

My main point though is just that the study happened to use truck axles as a measurement and that can't blindly be substituted for car axles. The back of the napkin math is just there to show the difference it can potentially make.

Also, truck tires are also much larger (than an average passenger car) and have a larger contact area.

Passenger off-road vehicles with knobby tires might be harder on roads too in a different way than weight.

I actually thought this was the case but a quick search indicated about twelve inches wide for both. Was I missing something?

It would make a significant difference to the calculations, eg 50% wider tires would mean dividing truck severity by 5.

The 4th power claim is alarming, yes. But are we focused too much on road wear and not enough on the money spent to increase the capacity of roads? My gut says that widening a road is at least an order of magnitude more expensive than maintaining one, so whatever we can do to reduce the number of cars is more worthy of focus than just getting all the trucks off the road.

100 kg bicycle (including rider), 0.00000625 units of damage.

So, getting people out of their cars onto bicycles cuts road maintenance cost.

Because of that, building bike lanes/paths can be cheaper than not building them.

(And 50 tonne trucks are very rare on most city roads, I hope)

A fuel truck that refuels most gas stations in the US is something like 40 tonnes so it’s not rare at all.

As long as the roads are built to sustain those trucks the cars are negligible. So until cars are the big wear item, moving people to bikes is pointless from a road wear perspective.

i love bikes and there are many reasons to get people on them, but assuming the 4th power claim is correct you are right. road damage isn’t a reason. math checks out.

How many car miles per large truck mile is driven?

Looks like about 10x car miles per truck mile. So it's a rounding error in terms of road damage, it's trucks by a landslide.

https://www.bts.gov/content/us-vehicle-miles

It’s a nice real world example of big O :)

There is one issue though. Bike lanes tend to require better road conditions for people to want to use them. Roads do degrade over time due to weather. This means that the difference between a vehicle at x weight vs no vehicle at all will be negligible. I don't know what the x is though.

the vast majority of road degredation isn't due to weather, its due to weight wear. As heavy vehicles travel down the road, they create a "ripple" on the road surface in front of the axle. You can create a similar effect if you run your hand across a loose sheet on top of your bed (you will see the wrinkle in the sheet in front if your hand). This bending is the method of road wear that causes like 99% of the maintenance need.

If that were the case, then roads would be almost equally bad everywhere regardless of the weather.

As a Michigander, the dictionary does not contain words to express how strongly I disagree with that implication.

Weather alone does very little. Weather + road ware makes things dramatically worse much faster. You can see this effect with private driveways in the north vs public roads in the south assuming similar construction.

As someone who migrated to upstate new york from Florida, I disagree. People here are constantly repaving, repairing, and resurfacing their driveways and sidewalks. While you need an initialization crack, that can come from a host of issues other than heavy vehicles including thermal cycling and initial defects. I will note that the asphalt and concrete construction may be different in the south vs. the north, but there is much much more driveway repair up here than there was in Florida.

As a pedantic side note, this is why your batteries die, the copper leads begins to fatigue and fracture due to stresses from thermal cycling.

No, the batteries "die" because cold weather requires more energy to start the car and batteries are far less effective when cold. (Physically they die either from grid corrosion increasing resistance so they can't provide enough current to start or the plates shedding active material to the point where they can't hold enough charge). There is normally not any copper in a standard car starter battery.

Copper? I thought batteries were primarily lead.

Northern roads are often (but not always) designed with more substantial shoulders to mitigate frost heaving on the edges of the road bed and plow damage.

While I may disagree with GP about the exact method of wear, regional differences are mostly due to construction not weather or wear. Roads ~50km west of us look almost new, while ours usually develop potholes or at minimum cracks, within a year. It's a different country, so different standards.

Frost heaving is also a significant contributor to road wear in winter areas.

Confirmation bias to the fourth power. A person just told you that an average truck of which you see plenty on the roads, does a couple thousand times the damage of a car and your response is: "Let's get all the cars of the road because a bicycle does no damage at all compared to cars"

Who says all? I’m just pointing out that there can be economic incentives for improving bicycle infrastructure.

That isn’t an original idea. See for example https://www.vabike.org/vehicle-weight-and-road-damage/ (2009)

Also, if in Google “50 ton 5-axle truck” I get trucks that I rarely see on roads in Europe. Reading https://www.jpisla.es/resources/Download+JPIsla+20130106+Pes..., that’s because t most countries don’t allow them on roads.

A quick search suggests that a semi truck without trailer weighs between 10 and 25 tons[0]. Those are usually 3-axle, so that's between about 125 and 4800 units of damage. That's still a massive amount when compared to the 1 unit of damage for a sedan.

Add an empty trailer and you get 35 tons (5 axles, 2400 units of damage). US max allowed is 80 tons, for a whopping 65,536 units of damage.

I don't know how full they are, but I see 5-axle tractor-trailers on highways all the time in the US, and also locally doing last-mile deliveries to larger businesses like supermarkets and home improvement stores.

I guess the differences in allowed weights might account for why US roads are often in worse shape than many in Europe, though I assume vastly different maintenance schedules play a large part as well.

[0] https://www.tcsfuel.com/blog/truck-weight-classification/

No, a ton is 2000 lbs or 1000kg, you’re using ton as 1000 lbs.

Max US ordinary large truck loading is 80k lbs, or 40 tons.

(Source, I trained as a civil engineer. Units in us practice were all over the place, but kip was generally the one that got used the most.)

The usual limit is Europe is 44 tonnes.

https://en.wikipedia.org/wiki/Large_goods_vehicle

https://assets.publishing.service.gov.uk/government/uploads/...

Nowhere did they suggest that they want to get rid of all cars ever.

The principle that shifting traffic from cars to bikes saves money for the government is a sound one. There'll always be a need for cars for some things -- probably not gonna have ambulance bikes, and of course deliveries for anything big needs a car/truck -- but you can certainly reduce the need. Especially with ebikes making biking more convenient and accessible.

I'll keep my car thanks

Americans moving to Tokyo largely give up cars and switch to public transport and walking. This isn't because there's a welcome committee of Japanese people shaming them out of driving; it's because driving works less well there, and public transit much better.

The point isn't to individually shame unwilling people out of their cars. The point is to make biking a real, viable transportation mode for short/medium distance trips for most people in urban or suburban areas. It's entirely possible to do this with the right infrastructure.

Once you do that, people will choose biking of their own accord, because it makes sense.

Most cyclists are also drivers, so yeah, I'd expect you to. Noone was telling you to get rid of your car.

But not every journey needs to be done by car.

I noticed while driving to Tahoe the last time that the right lane on the mountain roads was absolutely trashed in the tire path. I assume this is mostly due to chain usage, but it was incredibly more present in the right lane where semis drive.

And, by implication, they're a massively higher contributor to local pollution. I was surprised to learn that most road pollution is no longer generated by tailpipes, but by the grinding of brake, tire, and road into dust through friction. Vehicle weight has an outsized effect there too.

Wow, I always knew the semis are ruining roads but this 10k factor just blew me away, crazy! Thanks for a great post.

> Thus indeed if you have basically any heavy freight at all on a road, the rest is negligible.

There's many more passenger cars than freight trucks on the roads.

It doesn't change the math all that much, but the 1-to-1 comparisons in the thread didn't account for this factor.

> Thus indeed if you have basically any heavy freight at all on a road, the rest is negligible.

Racetracks don't carry freight, but the surface does goes bad in a couple decades.

Except we need trucks. We don’t need private cars.

A very substantial portion of truck freight in countries like Australia and the USA is inter-city—almost all, for larger semi-trailers and B-doubles. But for inter-city freight, rail is much cheaper than road, once you account for all costs, rather than allowing society and passenger cars to subsidise the roads for trucks.

We need transportation networks for goods. Rail is at least one alternative.

Maybe you don't, don't assume to know my needs

That is insane. I will remember this for life now.