What do decibels actually mean? A plain English guide to road noise levels

Decibels are not a normal scale

The decibel (dB) scale is logarithmic, which trips people up. A 3 dB increase represents a doubling of sound energy, yet it is only just perceptible. A 10 dB increase is roughly what most people judge as 'twice as loud'.

This is why small-sounding differences matter. Going from 60 dB to 63 dB doubles the acoustic energy reaching your home, even though the number barely moves.

A-weighting: measuring what we hear

Road-noise figures are usually given in dB(A) — A-weighted decibels. The A-weighting adjusts the measurement to match the sensitivity of human hearing, which is less responsive to very low and very high frequencies. It is the standard for environmental noise assessment.

Typical levels to anchor the numbers

As a rough guide: a quiet rural night sits around 30 dB; a quiet residential street around 45–50 dB; ordinary conversation around 60 dB; a busy road frontage 65–70 dB; and standing beside a motorway can exceed 80 dB.

The World Health Organization recommends road-traffic noise stays below 53 dB Lden and 45 dB Lnight to protect health and sleep. Many homes near major roads exceed both.

Why a barrier that cuts 10 dB is a big deal

Because of the logarithmic scale, a well-designed acoustic barrier delivering a 10 dB insertion loss roughly halves the perceived loudness — a clearly noticeable improvement, not a marginal one. This is why the exact decibel reduction from a fence is worth calculating properly rather than guessing.

Adding two noise sources together

Because decibels are logarithmic, you cannot simply add them. Two sources each producing 60 dB do not make 120 dB — they make about 63 dB, because you are adding energy, not numbers. Two equal sources always add roughly 3 dB. This is why a second nearby road, or a railway on top of a road, raises your overall level far less than intuition suggests, but still matters.

It also explains why removing one of two similar sources only cuts the total by about 3 dB. Meaningful reductions usually require tackling the dominant source directly.

How distance changes the number

For a single vehicle (a point source) noise falls by about 6 dB each time you double the distance. For a steady stream of traffic (a line source) it falls more slowly, roughly 3 dB per doubling. So moving from 25 m to 50 m from a busy road might only cut a few decibels — which is why simply being 'a bit further back' often helps less than people hope, and why a barrier that breaks the line of sight can outperform extra distance.

Why the same dB can feel different

Loudness is not the whole story. A constant motorway drone and an intermittent railway pass-by can share the same average level yet feel very different to live with, and low-frequency noise from heavy vehicles travels further and penetrates walls more than higher-frequency car noise. Metrics like Lden and Lnight capture the average energy over time, which is the right basis for health and planning, but your personal experience also depends on the character and timing of the sound.

The reference point behind the number

The decibel is not an absolute unit like a metre; it is a ratio, expressed relative to a reference sound pressure close to the quietest sound a healthy young ear can detect. That is why 0 dB does not mean 'no sound' but rather the threshold of hearing, and why the scale can climb so high — a jet engine at close range is around 140 dB, a level that causes immediate harm. For environmental noise the interesting range sits between roughly 30 dB (a quiet bedroom at night) and 80 dB (beside a motorway), and it is within that band that the decibel scale's quirks matter most.

Because the scale is logarithmic and referenced this way, small numerical changes represent large changes in sound energy, which is the single most important thing to keep in mind when interpreting any noise figure.

Peak, average and the metrics that matter

A single decibel figure can mean very different things depending on how it was measured. A momentary peak as a lorry passes is not the same as an average over 16 hours, and environmental assessment almost always uses time-averaged metrics — because it is sustained exposure, not one loud event, that drives annoyance and health effects. This is why road noise is reported as Lden and Lnight, long-term averages, rather than as the loudest moment of the day.

When you see a noise figure quoted, it is always worth asking what period it covers. A number without a timeframe is close to meaningless, and comparing a peak to an average is a common source of confusion.

What a given level actually feels like

Translating numbers into lived experience helps. Around 40 dB, noise is a soft background you would barely notice; at 50 dB you are aware of it but can converse and sleep normally for the most part; at 55–60 dB conversation outdoors starts to require effort and sleep can be disturbed; at 65–70 dB the environment feels clearly noisy and outdoor life is affected; and above 70 dB the noise is intrusive and dominates. These are broad guides — individual sensitivity varies — but they anchor the abstract figures to something tangible.

The WHO's recommended limits for road traffic — 53 dB Lden and 45 dB Lnight — sit at the point where the evidence shows adverse effects beginning to accumulate, which is why many homes near busy roads, sitting above these figures, warrant a closer look.

Frequency: why HGVs sound different

Loudness is not only about level; the frequency content matters too. Heavy goods vehicles produce more low-frequency energy — the deep rumble you feel as much as hear — and low frequencies travel further, bend around obstacles more readily, and pass through walls and windows more easily than the higher-frequency hiss of car tyres. Two roads with the same overall decibel level can therefore feel quite different if one carries a lot of HGVs.

This is also why acoustic barriers and glazing are specified with frequency in mind, and why an HGV-heavy route can be harder to screen than a car-dominated one at the same level. AcoustaPlot's calculation accounts for the HGV proportion precisely because it changes the character and reach of the noise.

Using decibels to judge a mitigation

Because of the logarithmic scale, the decibel reduction a measure delivers is the key figure to ask about. A 3 dB reduction halves the sound energy but is only just perceptible; a 5 dB reduction is clearly noticeable; and a 10 dB reduction is judged by most people as roughly halving the loudness — a transformative change. So when a barrier or glazing upgrade is quoted as delivering, say, a 10 dB insertion loss, that is a genuinely significant improvement, not a marginal one.

This framing helps you spend wisely. A measure that moves you a couple of decibels may not be worth the cost, while one that delivers a solid, well-evidenced reduction can change how a property feels to live in. An Acoustic Barrier Design report expresses barrier performance in exactly these terms — the predicted decibel reduction at your seating position, using the line-source method rather than the infinite-barrier assumption that overstates a garden fence — so you can judge the value before you build.

Common misunderstandings to avoid

A few decibel misconceptions recur and are worth clearing up. Doubling the traffic does not double the decibels — it adds only about 3 dB, because the scale measures energy logarithmically. Halving your distance from a road does not halve the noise — for a line of traffic it adds only around 3 dB. And a barrier that cuts 3 dB has not made a trivial difference to the physics even though the number looks small; it has removed half the sound energy, though you would only just notice it. Keeping these relationships in mind stops the numbers from being either over- or under-interpreted.

The practical upshot is to focus on the size of any change in decibels and what it means perceptually, rather than on the raw figure alone.

Why a range beats a single number

Finally, it is worth remembering that any modelled decibel figure carries uncertainty, and a responsible assessment says so. Real levels vary with weather, traffic on the day, and factors the model cannot see, so a single precise-looking number can give false confidence. Reporting a central estimate with an honest range — and, better still, translating it into a clear risk band — is more truthful and more useful than a spurious exact figure. That is the approach AcoustaPlot takes, and understanding how decibels work is what makes that range meaningful rather than mysterious.

The bottom line on decibels

If you take one thing away, let it be that the decibel scale rewards careful reading. Because it is logarithmic and A-weighted and almost always time-averaged, the same number can carry very different meaning depending on the reference, the weighting and the period, and small numerical differences represent large changes in sound energy. Treat any figure as a guide to a risk band rather than a precise verdict, pay attention to the size of any change a measure delivers, and remember that a well-evidenced 10 dB reduction is a genuinely transformative improvement.

With that framing, the numbers on a noise report stop being intimidating and become a practical tool — which is exactly how AcoustaPlot is designed to present them, as clear risk bands with honest ranges rather than false precision.