What each figure means

Overflight

The continuous passage of an aircraft above the outline of a zone, below the altitude ceiling. It is not a single dot on a map: it is the whole passage, with its time, its lowest altitude, its duration and its speed.

The same aircraft crossing again later leaves a new overflight. Those are two passages above your home, and counting them as one would hide the second.

Zone

The outline being watched, which is not the whole municipality. What is counted is what passes above that outline: an aircraft going by a kilometre from the edge, low, is perfectly audible from your kitchen and does not appear in the overflight count.

What is recorded here is what flies over, not what is heard. Those are two different questions and this number only answers the first. The second is answered by the noise monitors, which is why there are days with far more noise readings than passages.

One flight above two zones

It leaves one overflight in each, with their own altitudes and their own duration. It is not a duplicate: they are two observations of the same flight taken over different places, and the altitude it shows in one can differ a great deal from the one next door, because the aircraft keeps descending while it crosses them.

The practical consequence: the number of overflights of one zone is not added to its neighbour’s to reach a total. Each zone counts its own.

Minimum altitude

The lowest point of the whole passage. It is the figure that decides whether a flight is noticed from the ground, and that is why it is the one published and the one the filter uses.

When you ask for «those that came below 300 metres», 300 is a ceiling and not a floor: what comes out are the ones that passed below that height, not above it. Raising the number widens the list; lowering it narrows it.

Barometric altitude

The one the aircraft computes from atmospheric pressure, referred to the standard level. It is the one air traffic control uses and the one that appears in official records, so it is the one published.

It is not the height above your roof. With pressure other than standard it can sit tens of metres away from the real height, and that is one of the reasons an isolated number does not sustain a complaint: what sustains it is the series.

Manoeuvre

Whether the aircraft was landing, taking off, or crossing through without doing either. It is not inferred from whether it was climbing or descending: it comes declared with the flight’s own record.

While the definitive record of a passage is not yet published, its manoeuvre shows as «under review». That is not an error nor a permanent gap: it is the time it takes to arrive, and it fills in on its own.

Night period

From 23:00 to 07:00, local time. It is the period the airport’s rules treat separately, with a preferential runway configuration different from the daytime one.

On the charts it is marked in amber. It is a time reference, not a verdict: a flight falling inside it is not thereby irregular.

A-weighted decibel — dB(A)

The scale noise is measured on, corrected to resemble how a person hears: it gives less weight to very low and very high tones, which the ear perceives less intensely than a microphone does.

It is logarithmic, and this matters more than it seems. Three decibels more is twice the sound energy. Ten decibels more is perceived, very roughly, as twice the noise.

LAmax

The maximum level reached during an aircraft passage: the peak. It is the figure published next to each flight and the one that answers «how loud was that aircraft».

SEL

Sound exposure level. It concentrates all the energy of the event into one equivalent second, so that passages of different duration can be compared —a slow, low-pitched aircraft against a fast, sharp one—.

It always gives a larger number than the LAmax of the same passage. They are not two measurements of the same peak: they are two different questions, how loud it got and how much noise it brought in total.

Noise monitor

Each fixed station of the official network that measures noise around the airport. A noise level without saying which station measured it means nothing: the same aircraft gives very different readings at each one depending on distance and orientation.

That is why every noise figure on this dashboard comes with its station beside it.

Which aircraft a peak belonged to

A monitor measures the noise reaching a point, wherever it comes from. That a particular peak was an aircraft and not a motorbike is an inference, and it is made by comparing the instant of the peak with the moment that flight passed its closest point to the station, also counting how long the sound takes to come down.

The reading is always stored with how confidently it was attributed. When there were several possible candidates, it says so. A doubtful match presented as certain contaminates everything built on top of it.

Noise with no aircraft nearby

It happens, and it is not a fault. Noise does not stop at the zone boundary: the stations record what reaches them, and it reaches considerably further than where the aircraft goes by. There are days with dozens of readings and barely a handful of passages over the outline.

Those readings are not deleted or hidden. If only the noise that matches a passage were published, half the phenomenon would disappear from the record.

Why noise appears after the flight

The noise measurement is published hours later than the aircraft position, and its classification starts out provisional. An hour is taken as closed when it is confirmed to have stopped changing, not when it ends.

Hence a flight from this afternoon can be on the list with no noise figure and have one tomorrow. The figure is not being withheld: it has not arrived.

Runway configuration

Which end aircraft come in and go out by at a given moment. It is the decision that determines whether your zone gets thirty passages that day or a thousand, and it changes with the wind through the day.

It is not a small variation around an average: they are two distinct regimes. That is why this observatory publishes no «daily average» of overflights, which would be a figure describing no real day.

Preferential runway

The configuration the rules set by default for each period, precisely in order to spread the noise. The daytime one is not the night-time one.

That the preferential one is not in use during an hour is not by itself an irregularity: the rule itself writes down under which conditions it may or must be changed. What this observatory does is check whether those conditions were on record.

Tailwind

The wind blowing from behind the aircraft along the runway direction. It is the one that forces the change of configuration to be considered, because it lengthens the take-off run and the braking.

Crosswind

The wind blowing across. It has its own threshold, far higher than the tailwind one, and it is the other written reason to depart from the preferential configuration.

Knot

One nautical mile per hour: 1.85 km/h. It is the unit the rules write their thresholds in and the one the airport bulletin publishes the wind in, so it is the one used here in order to compare without converting anything.

The written thresholds

They are the only numbers the rule sets down in writing. At 7 knots of tailwind the change of configuration may be considered. At 10 of tailwind, or 20 across, the preferential runway ceases to be required. Gusts count the same as sustained wind.

On the charts they are the two vertical lines. Everything to the left of the first is an hour in which none of the thresholds was reached.

Hour with no published cause

An hour in which there were aircraft above the zone under a configuration that was not the preferential one, and the weather bulletin for that hour recorded none of the foreseen causes: no sufficient wind, no thunderstorm, no wind shear, no low ceiling, no poor visibility.

It is marked in amber. It does not say anything was done wrong: it says the foreseen justification is not on record in the bulletin, which is a legitimate question and is exactly how the draft puts it.

Other operational causes

Besides the conditions with a threshold, the rule allows departing from the preferential runway for operational needs, without setting any number for them.

It is the clause a weather bulletin can neither confirm nor refute, and that is why this observatory never asserts that an hour was irregular: at most, that none of the measurable causes was on record. That distinction is the reason what is demonstrated here is the design of the rule and never anybody’s conduct.

The airport bulletin

The aerodrome’s official weather report, issued every half hour, plus the extraordinary ones published when the weather changes abruptly. From it come the wind, the visibility and the phenomena of each hour.

Each overflight is matched to the bulletin nearest in time, and how many minutes separate them is stored. That number decides whether the bulletin really describes that moment: at five minutes, yes; at fifty, it describes the mood of that stretch of the day. There is no interpolation between two bulletins, because averaging two observations would invent a third that nobody made.

The forecast

The aerodrome’s forecast bulletin reaches twenty-four hours beyond its issue, so it sees a different slice of tomorrow depending on when you look. That is why tomorrow’s figure counts hours and always says out of how many looked-at hours it comes.

With the forecast wind and the written rule, the configuration can be anticipated at day granularity. It is a forecast, not a promise.

Reportable episode

An overflight that meets the conditions for filing: the passage is on record, the altitude is on record, and the bulletin for that hour publishes none of the causes foreseen for the configuration in use.

It is an axis apart from the count. Everything recorded counts in the statistics, always and unfiltered; an episode also being reportable changes not one of the figures published.

Draft

The letter the site prepares with the day, the time, the zone, the aircraft model and the measured minimum altitude, plus the wind paragraph when the conditions of that hour support it. Nobody has to write that at eleven at night.

The complaint is always personal: you file it and it goes in your name. The draft can be copied and pasted, or sent from the site itself with a copy to your own inbox. This observatory does not file on anyone’s behalf.

The time shown

Everything is shown in local time, the one on your kitchen clock, with the March and October changes accounted for. The day a flight belongs to is the local day: without that distinction, a flight at 00:30 would appear on the previous day.

Hour without coverage, empty hour, and hour not yet reached

They are three different things and they are drawn differently on the charts. An hour without coverage —with a diagonal hatch— is an hour that could not be looked at. An empty hour is an hour that was looked at and nobody passed. An hour that has not yet elapsed is left blank.

All three would read «zero aircraft» if counted together, and they mean opposite things. Being able to say «nobody passed that afternoon» instead of «we did not look that afternoon» is the reason a record is kept for every hour.

Median

The value leaving half the cases below and half above. It is used instead of the mean whenever an extreme case could drag the number: a single very low flight moves a mean and does not move a median.

Why there is no average of overflights per day

Because there is no typical day. Depending on the runway configuration, the same zone goes from the order of tens of overflights to the order of a thousand the next day, and the average of those two numbers describes neither of them.

Measuring a single day says little in itself either: it says which regime that day was in. What means something is the distribution across the period.

The period

The window of days the dashboard figures are computed over. It is long enough for both configuration regimes to fall inside, which is the condition for a comparison to mean anything.

Nothing recorded is deleted

Every overflight and every noise reading is stored permanently. There is no purge, no expiry, no window emptying from behind.

What does have a limit is how far back it was possible to look at the start: only the previous sixty days could be recovered, because beyond that the data was no longer available anywhere. From that date onwards nothing is lost.

The reference station

Air values are hourly and come from a fixed station of the monitoring network, which may sit in another municipality: not every zone has one inside it. The dashboard always says which one and how far away.

Each overflight is matched to the measurement of the hour it happened in. When that hour is not published yet, the last measurement prior to the flight is shown and when it was taken is stated. Never a later one: tomorrow’s air does not describe yesterday’s flight.

Unvalidated

Values marked this way are provisional: the network publishes them live and revises them afterwards. They update on their own when the revised version arrives.

Coincidence, not cause

An overflight coinciding with a high pollution value does not say it caused it, nor does it allow any share of it to be attributed. The station breathes the air of everything going by: road traffic, industry, the port, the weather and aviation.

Separating aviation’s contribution from the rest requires an attribution study this observatory does not carry out, and which is not replaced by putting two numbers side by side on a screen.

A flight’s absence does not prove it did not exist

A passage can be missing and a reading can arrive late. Hours without coverage are marked precisely so that a gap is not read as a quiet sky.

What does hold is the converse: what appears, appeared. No figure is adjusted, rounded to convenience, or delayed so that it looks better.

This is not an official register

It is an inventory of our own, for observation purposes. It does not replace any official register of flight paths or noise, and it does not rule on any infringement.

Publishing how many passages there were, at what height and with how much noise describes an exposure. Qualifying it is for whoever has the competence to do so.