Observing · August 12, 2026

How to Spot the Space Station From Your Own Garden in August 2026

Let us start with the answer. According to the orbital elements of 18 August 2026 at 12:02 UT, the best opportunity of the period from Paris is the early morning of Sunday 23 August: the Station appears at 05:45 in the south-west, culminates at 82° altitude at 05:48, almost overhead, and disappears at 05:53 in the east-north-east. Seven and a half minutes, 422 km above your head, for a computed magnitude of −2.6: brighter than Sirius. The nights that follow stay good, and 25 August is better still.

Why is the ISS only visible a few nights a month?

A visible pass has nothing to do with the Station being « overhead ». It very often is. Three things have to happen together.

One: the Station has to be above your horizon. It orbits at a mean altitude of 418 km and travels at 7.658 km/s; in one orbit of 92 min 56 s the Earth turns 23.3° beneath it, which shifts each pass westwards. Most of the time it is simply on the other side of the globe.

Two: it has to be lit by the Sun. The Station produces no light of its own, it reflects the Sun’s. If it enters the Earth’s shadow cone it goes out, sometimes in the middle of a pass.

Three: your sky has to be dark. We use the threshold common to tracking software: Sun below −6°, that is, after the end of civil twilight. Above that, the sky background swamps a magnitude 0 point even though it is geometrically there.

The last two conditions contradict each other: the Sun has to be down for you and up for the Station. That window drifts through the Station’s orbit over the weeks, which is what produces visibility seasons: ten or so good nights, then a fortnight in which every pass falls in broad daylight. In late July, Paris was in the trough. In late August, Paris is at the peak.

Two rows deserve an alarm. The 23rd for the altitude, 82°, which puts the Station practically vertical and gives you eight minutes end to end. The 25th for the brightness, −2.8, the best of the series: lower down, but better lit. All these times fall before dawn, between 05:00 and 06:00: this is the morning season. Readers in the UK and Ireland should subtract one hour throughout.

How is an ISS pass time computed?

It all starts from a measurement: the two-line orbital elements, published several times a day by CelesTrak from observations by the American space surveillance network. They are two lines of 69 characters describing the orbit at a precise instant, called the epoch. For the Station, the epoch used here is 18 August 2026 at 12:02:50 UT, with an inclination of 51.633°, an eccentricity of 0.00076 (a near-circular orbit) and 15.4949 revolutions per day.

These elements are propagated by SGP4, the standard in the field since the 1980s, which accounts for the Earth’s oblateness (the orbital plane drifts) and atmospheric drag (the Station slowly descends). The resulting position is converted to azimuth and altitude for your location, then tested against two criteria: is the satellite in the Earth’s shadow, and where is the Sun.

Apparent magnitude is no constant. We start from a standard magnitude of −1.8, the Station’s brightness at 1,000 km and half illumination, corrected for distance by 5·log₁₀(d/1000) and for phase angle by a diffuse-sphere function. On 23 August at culmination, the distance of 422 km contributes −1.87 magnitude and the phase angle of 98.8° takes back 1.03: what remains is −2.6. It is the phase angle that explains the paradox in the table, where the pass of the 25th is brighter than the one on the 23rd while standing twenty degrees lower.

Are pass times the same everywhere?

No, and for a reason that is not the obvious one. Over three weeks from 18 August we count 28 visible passes from Paris, 28 from Marseille, 34 from Montreal and 4 from Papeete. Yet the same computation run in late July gave the opposite: 3 from Paris and 19 from Papeete. In one month the ranking inverted, while the length of the nights barely moved.

Night length is therefore not the dominant factor, contrary to what we wrote in an earlier version of this article. What governs it is where the orbital plane sits relative to the day-night line. That plane turns slowly under the Earth’s oblateness while the Earth moves along its own orbit around the Sun, and the agreement between the two is made and unmade over a few weeks. When it is favourable for your latitude you get a pass almost every night; when it is not, you get nothing for a fortnight, however long your nights are. There is no national timetable, and not even a stable ranking between places: only a computation for each point on the globe, week by week.

What does a Space Station pass look like?

At culmination on 23 August the Station covers roughly the apparent diameter of two full Moons every second. It crosses the visible sky in seven and a half minutes.

• Go out five minutes before the time of appearance and look in the direction given. The Station appears low, often fainter than expected, and strengthens as it climbs.
• Look for a point that does not blink. Aircraft blink, the Station does not; it moves with a perfectly steady motion, and in silence.
• No instrument helps: binoculars narrow the field and will make you lose it. A phone on a stable support, with a 10 to 20 second exposure, does record a fine trail.
• You will see a point. It is 109 m from one end of the solar arrays to the other and it is 422 km away: that works out at 53″ of arc, at the very limit of what the eye can resolve.

Can you trust a forecast made three weeks ahead?

This article supplies its own demonstration. In its first version, published on 12 August from elements dated 29 July, we announced for 23 August a culmination at 80° and a magnitude of −3.9. Recomputed on elements one day old, the same pass gives 82° and −2.6. The altitude moved by only two degrees, but the brightness lost 1.3 magnitude, because the phase angle had drifted from 81° to 99°: the Station now presents itself less well lit. A pass forecast is not a planetary ephemeris; it rests on a measurement of an orbit that ages.

• At three days, the forecast is good to a few seconds. At three weeks, it drifts by a minute on the time and by more than a magnitude on the brightness. Always confirm the day before.
• This figure does not cover manoeuvres. The Station raises its orbit several times a year and performs debris avoidance decided within hours. After a manoeuvre, any earlier forecast is void.
• The standard magnitude of −1.8 comes from visual observation tables, not from a photometric measurement: allow ±0.5 magnitude, plus the real attitude of the solar arrays, which can swing the brightness by several magnitudes for a few seconds.
• The −6° threshold for the Sun is a convention. From a brightly lit city centre a low, faint pass will stay invisible even below it; from a dark site, a bright pass can be seen slightly before.

Updated on 19 August 2026: orbital elements refreshed, pass table recomputed, and the explanation given for the differences between places corrected. This article was translated from the French edition by AI, then reviewed before publication.

• Beginners
• August 2026 sky