In the total eclipses where it has been measured, the air temperature has fallen by a few degrees: AEMET, Spain’s state weather agency, notes that drops of 2 to 5 °C have been documented, and during the 2024 eclipse a study using New York State’s Mesonet network of weather stations measured cooling of 2.8 °C on average and up to 6.8 °C. There is no figure for Spain yet. During last summer’s eclipse AEMET found no significant drops at its stations, because the Sun was very low and totality was short. The eclipse of 2 August 2027 comes in mid-morning, and AEMET itself sees it as a good opportunity to measure the effect.
What AEMET measured last summer
AEMET looked at radiation at nine observatories, from Santander to Palma, and at temperature at three stations of the EMEP/VAG/CAMP network inside the path of totality: O Saviñao (Lugo), Campisábalos (Guadalajara) and Els Torms (Lleida). It also checked the Burgos Villafría station.
Radiation fell during the eclipse, although not very sharply, because at that time of day it was already low. Temperature did not: AEMET writes that no significant drops can be seen. After totality there was a small plateau, and at O Saviñao a slight rise. In Burgos the drop was somewhat larger than on the previous day, but AEMET urges caution because the weather was not the same on both days.
AEMET gives two reasons. The eclipse came shortly before the Sun went down, when it has little effect on the temperature at the surface, and totality was short: a little under 2 minutes where it lasted longest. For comparison it cites a total eclipse in Zimbabwe in 2001, with 5 minutes of totality at around 3 in the afternoon. There the temperature fell by about 4 °C.
What was measured in the United States
During the 2017 total eclipse, the Office of the Washington State Climatologist compared three stations in the north-west. The biggest drop was at John Day (Oregon), 4.8 °F, about 2.7 °C; Seattle recorded 4.0 °F and Goldendale 3.2 °F, although there the readings came every 15 minutes and probably fall short. An analysis of National Weather Service stations, with readings every 5 minutes, found drops of up to 10 °F (about 5.6 °C) in the east of the country, probably because the eclipse reached the east at a later and warmer time of day.
In 2024, the study by Wang and colleagues published in Geophysical Research Letters, using the New York Mesonet stations, measured surface cooling of 2.8 °C on average and 6.8 °C at most. The coldest moment came about 17 minutes after totality, and the air went still, with a surface inversion similar to the one that forms at night.
Why 2027 is different
Last summer’s eclipse came with the Sun already on its way down. The one on 2 August 2027 arrives in mid-morning: in the path of totality, the Sun will be between 37 and 41° up, towards the east, and still climbing. At that time of day the Sun is heating the ground, and the eclipse takes away a good part of that energy for more than two hours. The length of the eclipse is very similar from one city to another, between 129 and 141 minutes. What changes is how long almost the whole Sun stays covered.
Minutes of eclipse in each city, and how many of them pass with more than half, or more than 90%, of the disc covered:
| City | Minutes of eclipse | Over 50% covered | Over 90% covered | Maximum |
|---|---|---|---|---|
| Málaga | 140 | 61 | 17 | total |
| Seville | 138 | 59 | 16 | 98.4% |
| Valencia | 141 | 60 | 3 | 90.4% |
| Madrid | 137 | 56 | 0 | 86.1% |
| Barcelona | 141 | 55 | 0 | 81.6% |
| Bilbao | 133 | 50 | 0 | 74.4% |
| A Coruña | 129 | 47 | 0 | 73.4% |
| Las Palmas de Gran Canaria | 129 | 49 | 0 | 77.2% |
Málaga will also get 1 min 49 s of totality, and the centre of the path more than four minutes (our learn section explains what happens in those minutes). Outside the path the Sun is never fully covered: in Madrid the maximum is 86.1%.
None of these sources says how many degrees it will drop in your town, and we are not going to make it up. If you plan to measure it, keep the thermometer in the shade and carry on taking readings for a good while after totality: in New York the coldest moment came 17 minutes later. And check the forecast for the day (in Spanish) beforehand: with cloud about, comparing with the previous days gets complicated. The drop in light also shows up in solar panels, which we cover in the eclipse and the power grid.
Sources: ¿Hubo cambios en las variables meteorológicas durante el eclipse? — AEMET Blog (in Spanish: stations, results from last summer’s eclipse, the drops of 2 to 5 °C in other total eclipses and the Zimbabwe example); On the Great American Solar Eclipse — Office of the Washington State Climatologist (the 2017 drops at John Day, Seattle and Goldendale, and those of up to 10 °F in the eastern US); Characterizing the Impacts of 2024 Total Solar Eclipse Using New York State Mesonet Data — Geophysical Research Letters, on the UCAR portal (average and maximum cooling, the 17-minute delay and the still air); eclipseando data (minute-by-minute obscuration curves and the Sun’s altitude from the skyfield + DE440s engine; duration of totality from the IGN).
If you would like something to keep from the eclipse, the shop prints T-shirts, mugs and posters with the name of your town, the time of its maximum and its seconds of totality. See the merchandise →
Remember: never look at the Sun without proper eye protection (filters made to EN ISO 12312-2). The official reference for safe viewing in Spain is the IGN (in Spanish): eclipses.ign.es.