Sound walls and parking garages, 3/17/2019

For decades, Vienna has required a 6′ masonry wall to separate commercial property from residential neighborhoods.  This page explains what that wall does — reduce the noise from the parking lot.  It also explains what it doesn’t do (or, really, what it isn’t needed to do) — reduce the noise from Maple Avenue.

This analysis ends with what I hope is an obvious conclusion.  Where MAC development replaces parking lots with (two-story) garages adjacent to residential areas, the old 6′ masonry wall doesn’t cut it any more.  Vienna needs to change the building code to require that those MAC garages be enclosed, not open-sided, when they are adjacent to residential areas. Otherwise, MAC strips away a decades-old noise protection that the Town has offered to its residential neighborhoods.

The argument in brief

The parking lot of a commercial establishment can generate a significant amount of noise.  Cars starting, car doors slamming, car trunks and tailgates slamming, every idiot who sets off a car alarm, every car that chirps its horn when you lock it, motorcycles, people who talk loudly when they are drunk, people who talk loudly when they aren’t drunk, people who crank up the music  and crank down the windows, guys who like loud mufflers, noise from use of the dumpster.

You get the idea.   Hubbub.  We’re talking the potential for significant hubbub.

And if we are talking about (e.g.) a restaurant or bar, it will generate a lot of that hubbub late at night.  That is, against an otherwise quiet backdrop.

That’s why the Town (usually) requires a six-foot masonry (brick, block, or stone) wall to separate commercial from adjacent residential property.  A wooden fence would block the sight of the parking lot, but not the sound.  The point of a masonry wall is to reduce the amount of noise that makes it from the parking lot to the adjacent residential neighborhood.  (That, and to keep the rats on the same side of the wall as the restaurant dumpster, but that’s a separate issue.)

OK, now do a little thought experiment: Take that noisy commercial parking lot.  Put a sound-reflecting concrete roof on it.  Put sound-reflecting concrete walls around most of it.  And then stack another parking lot on top of it, also mostly surrounded by concrete.

Question:  Is that change going to make the parking lot noisier, or quieter, on the sides where there aren’t solid walls?  Answer:  Noisier, of course. 

If you can grasp that, then you know what I’m going to say next.  If the noise from the parking lot is enough that the Town requires a 6′ masonry wall, then you can’t just replace that with something even  noisier — a double-decker parking garage — and do nothing about the noise.  I mean, sure, you can if you don’t give a darn about the adjacent neighborhoods.  But if you even want to pretend that you care about the qualify of life in the adjacent neighborhoods, you really need to do something.

And the obvious thing to do is to require that any such parking garages, adjacent to residential areas, be fully enclosed. That is, enclosed by walls that provide at least as much sound reduction as the standard 6′ masonry wall does, for surface parking.

This isn’t some new benefit for the protection of the neighborhoods.  This is merely updating the existing protection — the 6′ masonry wall for surface-level parking — to accommodate the changes being introduced by MAC zoning.

What a 6′ masonry wall does and doesn’t do.

I realize most people are not going to read the details, so let me get to the point here, and put the underlying science in a separate section.

A 6′ masonry wall cuts the parking lot noise in half.  At best.  More-or-less.  Being behind such a wall will typically reduce sounds that are generated close to the other side of the wall by between 6 and 10 decibels (db).

At that upper limit of 10 decibels, people would perceive that as “half as loud”.  And while that doesn’t sound like much, if you look at a chart of typical sound levels, you can see that a 10 decibel difference is substantial.  It’s the difference between standing 50′ off the interstate during rush hour (70 db) and having a conversation in a restaurant (60 db).

As importantly — and a surprise to me — those walls are NOT there to block the sound of Maple Avenue traffic.  As I will explain at length below, they don’t need to.  The sheer distance from Maple is enough to reduce that sound to a tolerable level.  Maple Avenue traffic generates high noise levels near the road, as I documented here.  (Or, more informally, here.)  But sound intensity drops off rapidly with distance, following an “inverse square law”.  Sounds that are annoyingly loud 20′ from Maple avenue will barely register (say) 280′ off the road (at the rear lot line for the proposed 380 Maple West condos).

You know this intuitively if you’ve ever eaten at (say) the outdoor seating area gathering space at Tequila Grande.  On the same day when the noise levels 20′ from the road were intense (see second link above), the noise just 100′ off the road at the Tequila Grande patio gathering space was not an issue.  I could hear the traffic, but I rarely felt as if I had to talk over the traffic.

Let me be clear that I am not the only one who didn’t understand what those walls do and don’t do. At recent meetings of Town Council/Planning Commission and Board of Architectural review, it was clear that most-if-not-all of the people making decisions about those walls don’t really understand the acoustics of them, either.

So now that I have my head straight about what those masonry walls do and don’t do, I’m putting together this brief summary of the acoustics of sound walls.

Decibels, inverse square law, and all that.

It would probably be adequate to take my recording decibel meter out and just demonstrate this next part empirically.  But first, I’ll lay out the theory.  (And at some point, I’ll actually go out with a sound meter and test that out empirically.)

I searched for a good, short explanation of this topic, and didn’t find one.  So let me keep this short by presenting the rules of thumb and skipping the nuances.

  • An increase of 10 decibels = 10x as much sound power = “twice as loud”.
  • Twice the distance from a sound source = 6 decibel reduction in sound level.

You can calculate the “6 decibel” reduction if you know that a) the decibel scale is a (base 10) log scale, and that sound power follows an inverse-square law.   (Sound power falls off as one over the square of the distance to the sound source).

So, log(1/(2×2)) = -0.6 = 6 decibel reduction.

I only present that much math because that’s how to understand the effect of distance on sound intensity, and so understand that you don’t need these masonry walls to reduce the sound of Maple Avenue traffic.  The distance from Maple is quite enough to do that.

Let me quickly calculate the difference in noise level 20′ from Maple (MAC streetside dining area) versus 280′ from Maple (back lot line of 380 Maple West proposed condos).  Instead of twice the distance (2 in the formula above), it’s now 14 times the distance.

So, log(1/(14×14)) = -2.3 = 23 decibel reduction.

The typical human ear would perceive that as more than four times quieter. (Because every 10 db = “twice as loud”).

You can look at my chart of actual noise levels near Maple, and see that a 23 db reduction would take that noise level “off the charts”.  It would reduce the loudest noises as measured 20′ from Maple down to 57 decibels —  a typical figure for conversation in a restaurant.

The upshot is that at the back lot line of a property like (the proposed) 380 Maple West, you would still hear the traffic noise, but it would truly be background noise.  Normal conversation would not be drowned out by the noise.

Walls and sound diffraction.

Some amount of sound will pass right through a brick wall — literally be transmitted from one side to the other.  But most of the sound you hear (from one side of a 6′ brick wall to the other) is sound that is “diffracted” — that is, that bends as it passes over the top of the wall.

Skipping the nuances for a moment, the rest of the story is mostly just geometry, as applied to the top of a 6′ masonry wall.  The closer the sound source is the wall, the less sound energy will be bent down to ear height on the other side of the wall, and the more effective the wall is at blocking the sound.  And conversely, the farther away the sound source, the greater the portion of the sound energy that will be bent down to where you can hear it, and the less effective the wall is at blocking the sound.

Diffraction in this case depends strongly on the frequency of sound, but that’s a nuance I’m going to ignore. If you care, low frequencies (bass notes, truck rumble) diffract more strongly around a wall than high frequencies (soprano notes, brake squeal) do.  So what you end up hearing on the other side of a 6′ wall is both lower in volume and lower in pitch.

For this one, a picture is worth a thousand words.  Arguably, the most complete explanation I found of sound barriers is this document (.pdf).  But the bottom line is that as sound passes over the top of a sound barrier wall, some part of it gets bent down toward the ground.  That’s “diffraction”. And the further the sound source and sound receiver are from the wall, the less effective the wall is at stopping noise, due to that diffraction.   Here I have taken a picture from the .pdf able (Technical Report 2017-02, State of the art in managing road traffic noise:
noise barriers, Conference of European Road Directors, January 2017).

This source provides the means to do detailed calculations.  From that, for different frequencies of sound, you can see how effective a sound wall would be, in theory.  The results look like this (sound source assumed to be located at ground level, on the other side of the 6′ tall sound barrier wall):

For reference, somewhere between 3 and 5 decibels is typically described as a “just noticeable difference”.   The upshot is that for sounds 200′ away (bottom line), for the frequency most used to simulate traffic noise (500 hertz), the presence of the wall makes a little better than a “just noticeable difference” (6.5 decibels).  But for sounds located directly on the other side of the wall (top line), the wall reduces the perceived loudness of a mid-range sound (500 hertz) more than in half (12.75 db).

If you are standing behind the wall along the current 380 Maple West lot line, the reduction in the noise level of Maple Avenue traffic is due almost entirely to the distance (-23 decibels), and only to a minor degree to the wall (somewhere around an additional -6.5 decibels). 

For reducing parking lot noise directly adjacent to the wall, the wall is crucial.  For reducing Maple Avenue noise, the wall is nice, but not necessary.  The distance from Maple accounts for most of the sound reduction.


We need these walls to protect us from parking lot noise, not from the noise of traffic on Maple Avenue.  Accordingly, as MAC converts what used to be parking lots into double-decker parking garages, we need a wall to stop the sounds emanating from those parking garages.  The garages need to be enclosed.

As an aside, this impact of distance-from-the-wall is why Wawa needs to build a sound wall, not the affected neighbors.  The sound wall will be more effective the closer it is placed to the source of the sound.  (In this case, the Wawa parking lot.)

Finally, if I can find the time, I’ll record some sounds to demonstrate this empirically.  At that point, I’ll add those recordings to this page.