This is the single most important and most misunderstood point in the whole subject of noise disputes. A company commissions a measurement. An accredited laboratory carries it out. The report states that limit values are not exceeded. And from that a conclusion is drawn: there is no problem, the complaints are unfounded.
That conclusion is wrong — not because the measurement is wrong, but because the measurement answers a different question from the one residents are asking.
Two different questions
The measurement answers: at a specific point, at a specific height, at a specific time, does the equivalent sound level exceed the figure set in the regulation?
The resident asks: can I sleep in my flat at night with the window open?
The answer to the first can be "no, it does not exceed". The answer to the second, at the very same moment, can be "no, I cannot". There is no contradiction here.
Why this happens — six reasons
1. A limit value is not the threshold at which nuisance begins
Limit values are an administrative compromise between environmental protection and economic activity. They are not a medical boundary beyond which noise stops affecting people. World Health Organization recommendations for night noise are lower than the statutory limits in many countries. Compliance means legality, not comfort.
2. The equivalent level averages away exactly what disturbs
Measurements normally use LAeq — the energy-average level over a period. A continuous monotonous hum and the same average level produced by a few passing cars look identical in the measurement. To a human being they are nothing alike: with intermittent noise people habituate and fall asleep in the gaps; with a monotonous one they do not.
3. A-weighting suppresses precisely the low frequencies
Standard A-weighting imitates the sensitivity of the human ear at quiet levels and therefore cuts low frequencies substantially — roughly 30 decibels at 50 Hz. If the source noise is a low-frequency hum, the A-weighted result systematically understates it relative to how a person perceives it indoors.
4. The measurement point is not where people live
Environmental noise measurements are usually taken at ground level, 1.5 metres up, at the property boundary. If a fence and trees stand between source and microphone, they screen that low point. On upper floors that screening does not work — sound arrives directly. A measurement at the fence says nothing about those flats.
5. A measurement is a snapshot
Three days in particular weather conditions is not a year. Low-frequency sound propagation depends heavily on wind direction and air temperature layering. This is exactly why residents so often say "some days you cannot hear it at all, on others it is unbearable" — and exactly why a measurement on particular days can land in favourable conditions.
6. Background means one thing by day and something else at night
By day, industrial noise merges into the traffic background and its contribution may not be detectable at all. At night, when the background subsides, that same unchanged noise becomes the dominant sound in the area. A measurement reporting a single figure for both periods does not capture that shift.
How this looks in a typical case
To make it concrete, take an invented but typical example. Suppose an industrial installation runs continuously next to a residential building, and measurements are taken in all three periods of the day — twice in each, with the installation running and switched off.
| Period | Installation running | Installation off | Difference | Limit |
|---|---|---|---|---|
| Day | 55 dB(A) | 55 dB(A) | 0 | 65 dB(A) |
| Evening | 52 dB(A) | 50 dB(A) | +2 | 60 dB(A) |
| Night | 47 dB(A) | 45 dB(A) | +2 | 55 dB(A) |
Formally the conclusion is unambiguous: no limit is exceeded in any period, and by a comfortable margin. On paper the case is closed.
But look at the table again:
- By day the installation's contribution is not detectable at all — it vanishes into the traffic background. That does not mean it is not running; it means it cannot be distinguished.
- At night the same unchanged installation adds a measurable difference. The background has fallen by 10 decibels; the installation has not. It becomes the dominant sound in the area.
- The overall night level is the lowest of the three — 47 dB(A) — and that is the period that disturbs people most. The smallest number in the table corresponds to the worst situation in real life.
What such a measurement actually proves
Not that there is no noise. But that the installation does produce noise, and that its contribution becomes detectable specifically at night — which is exactly when it disturbs people most.
And if no measurement with the installation switched off was taken in the evening or at night, then nothing at all can be said about those periods.
What follows from this
The measurement result carries real legal weight: if limit values are not exceeded, it is harder for an authority to apply enforcement measures. We understand and accept that.
But it does not follow that there is no problem or that nothing needs to be done. What follows is only that the solution is unlikely to arrive as an official order — it has to be a decision the company takes itself. Which is precisely why the request to install sound insulation is addressed to the company, not only to the authorities.
What this site documents
This article is published by the Jūrkalnes and Irlavas neighbourhood community. We document the industrial noise produced by the outdoor Nestro filter at SIA Livonia Print, Jūrkalnes iela 15/25, Riga. The installation has run 24 hours a day since 2021, and its hum reaches around 12 streets. The residents’ request to the company is a single one: fit an acoustic enclosure around the filter and sound insulation on the air ducts. The board member of SIA Livonia Print is Trond Erik Isaksen, and it is to him that the neighbourhood’s letters about the noise have been addressed.
