The 2 August 2027 total solar eclipse in Casarabonela (Málaga)

How long totality lasts in Casarabonela

Totality in Casarabonela lasts 10s, the shortest of the entire Spanish path: the other 104 municipalities in totality all spend longer inside the shadow. The longest is Ceuta, with 4m 48s.

The path of totality is about 238 km wide as it crosses southern Spain. Casarabonela sits roughly 119 km north of the centre line and about 1.2 km from the north limit of totality: driving that far north would drop you out of totality and into a partial eclipse. Going south, on the other hand, lengthens it.

At 10s, Casarabonela sits in the third of the path that spends least time in darkness. This far from the centre line, moving a few kilometres south lengthens totality noticeably.

Below half a minute you are on the very edge of the path: the shadow only clips the town and totality shrinks to little more than a blink. A few kilometres of error in choosing your spot costs you the whole thing.

The Sun, the shadow and the terrain

Maximum eclipse reaches Casarabonela at 10:48:27 mainland Spanish time. The Sun will be 39° above the horizon at azimuth 97°, that is, towards the east. The magnitude of the eclipse will be 1.07.

That altitude is the deep difference from the 2026 eclipse, which happened with the Sun almost touching the horizon and the terrain to the west deciding who saw totality and who did not. Mid-morning on 2 August the problem disappears: in Casarabonela the margin between the Sun and the terrain profile in that direction is 39.1°, and across the whole path the tightest margin is Benalauría, at 27.8°. All 105 municipalities clear the visibility test.

The shadow, however, does not arrive from where the Sun is: the umbra comes in from the west (259°), crossing in from the Atlantic. From Casarabonela the line of sight in that direction stops at 0.4 km, which at this scale means the terrain blocks it right there: you will not see the shadow front coming, darkness will simply arrive.

The sky at eclipse time, from 30 years of data

Across the 1996-2025 series, the sky over Casarabonela was clear during the 10:00-11:00 window —the hour that brackets maximum— in 86.7 % of years. Under a stricter test, less than 20 % cloud cover, the figure is 83.3 %; under 10 % cloud cover, 76.7 %.

The whole-day average gives 93.3 %, clearly better than the eclipse window, and that gap is not noise: mid-morning the low cloud off the Strait and the coast has not burned off yet, and it does so as the morning wears on. In that hour the mean low cloud over Casarabonela is 0.3 %. That is why this page publishes the window figure rather than the daily one: the daily average paints a sky that does not exist yet at eleven in the morning.

The mean maximum temperature for 2 August in this grid cell is 32.8 °C and the mean minimum 20.2 °C, and it rained in 0.0 % of the 30 years. None of this is a forecast: it is climatology, an answer to what is normal here at that hour, not to what the weather will do on the day.

Observing index for Casarabonela

Casarabonela scores 28.9 out of 100 on the observing index for this eclipse and ranks 105 of 105. The index weighs three things measured on the spot: the duration of totality (52.94 %), the sky in the window around maximum (29.41 %) and the line of sight in the direction the shadow arrives from (17.65 %).

Its three components, normalised from 0 to 1, come out at 0.042 for duration, 0.906 for sky and 0.000 for view. The heaviest drag is the view towards the shadow, and that is what you would have to make up for by choosing the exact observing spot well.

The index ranks the path; it does not replace a forecast. It also knows nothing about access, parking or permits: that fourth component was in the design at 15 % and is not published because the dataset does not exist, so the other three are renormalised rather than padded out with an invented estimate. The horizon margin, which carried 30 % of the 2026 index, does not weigh here: with the Sun at nearly 40° it came out identical in all 105 municipalities.