27

Summer in the Northern Hemisphere starts on the day of the summer solstice. This is the day that the Northern Hemisphere receives more light from the Sun, due to the Earth's tilt. To my knowledge, the amount of light we receive is related to the temperatures that we have. That's why summer is the warmest period in most regions of the Northern Hemisphere.

Then, shouldn't the solstice be the central day of the summer instead of the initial day? That way, summer would be constituted by the days of the year when more light is received in the Northern Hemisphere, which is related to the higher temperatures in practice.

Peter Mortensen
  • 301
  • 2
  • 8
DaniPM93
  • 395
  • 3
  • 4
  • 1
    I'm not sure how it works elsewhere, but throughout most of the US, the coldest months are Jan and Feb, and the warmest July, Aug, which are the months between the solstices and equinoxes. So, I think it does make sense from that point of view. – Greg Miller Feb 04 '23 at 20:36
  • @GregMiller yes, it works in the same way here in Spain. July and August are usually the warmest months. But I don't see the actual reasons for that, I think the warmest month should be the one when the solstice occurs (i.e. June). – DaniPM93 Feb 04 '23 at 20:51
  • 28
    As an amateur observer, I'd think that the mechanism is that it takes some time for things to warm up... or cool off ... – paul garrett Feb 04 '23 at 21:05
  • It is only meteorologists who say the solstice starts summer. That is due to the hysteresis in the warming of the earth. – Chenmunka Feb 04 '23 at 21:58
  • 2
    https://en.wikipedia.org/wiki/Midsummer – PM 2Ring Feb 04 '23 at 22:49
  • 3
    Seems there are different definitions. The Wikipedia entry for Summer says it's centered on the solstice. Merriam-Webster says it's June, Jul, Aug or from solstice to equinox. – Greg Miller Feb 04 '23 at 23:32
  • Summer is the warmest period in both hemispheres, not just the northern half (except for tropical regions near the equator of course). – Martin Kealey Feb 05 '23 at 05:21
  • 16
    Summer in the northern hemisphere starts on the day of the summer solstice. - According to whom? – user985366 Feb 05 '23 at 13:59
  • 1
    Very little to do with Astronomy. A meteorological definition (that varies). Uninteresting. – ProfRob Feb 05 '23 at 15:54
  • 1
    "To my knowledge, the amount of light we receive is related to the temperatures that we have." Here's a related answer : https://earthscience.stackexchange.com/a/21242/12617 It explains why in many places in the northern hemisphere, August is hotter than June even though it receives less irradiance. – Eric Duminil Feb 05 '23 at 15:56
  • https://en.wikipedia.org/wiki/Seasonal_lag – llama Feb 06 '23 at 18:35
  • As already said, there are different conventions on whether summer begins at solstice, or has its midpoint at solstice (midsummer). The former makes some sense because of that "seasonal lag" just mentioned. You can infer from the names of the solar terms ( jiéqì), that the tradition in Chinese calendars is to regard (northern) solstice as the midpoint of the summer. Also, the Chinese New Year (on a new moon close to February 4 (average)) marks the beginning of spring (northward equinox being its midpoint). – Jeppe Stig Nielsen Feb 06 '23 at 23:56

3 Answers3

51

The English word "summer" means the season of the year that is associated with higher temperatures and shorter nights. There is no official "first day of summer" and different groups of people take different conventions.

One possible convention is to take "June, July and August" as summer, so the first day of summer is June 1st. This is the convention taken by the Met Office in the UK, and roughly corresponds to the warmest temperatures in the UK.

(The reason that the warmest temperatures are not around the solstice is nothing to do with astronomy, it is because the surface takes some time to warm up, so there is a lag between the longest day and the highest temperature)

Another possible convention is to take June 21st to Sept 20th as "summer". This fits the solstice and equinox and still roughly corresponds to the warmer days of the year in the Northern Hemisphere. This is the convention in many modern calendars.

Another possibility is to take the "cross-quarter days" (named in Gaelic Samhain, Imbolc, Beltane, and Lughnasadh) So summer would be from Beltane/May day to Lughnasadh/Lammas day: May 1st - August 1st. This matches the shortest nights, but generally, May is cooler than August in the UK, so is not consistent with summer meaning "warmest season of the year".

The big point here is that, there is no official definition of summer.

James K
  • 120,702
  • 5
  • 298
  • 423
  • 11
    For context, here in Australia I’ve always used and understood “summer” to mean December, January, and February, at least when it refers to a bounded time period. To my recollection, it wasn’t until I was an adult that I encountered alternative definitions, or realised in hindsight that some past encounters with the term may not have meant what I’d assumed! (And then on some website or other I got downvoted for disagreeing with the solstice-to-equinox definition…) – Tim Pederick Feb 05 '23 at 07:08
  • 5
    Don't forget the US "unofficial summer", which runs from Memorial Day (last Monday in May) to Labor Day (first Monday in September). – Barmar Feb 05 '23 at 20:36
  • @TimPederick that's very strange indeed, when I first moved to Australia I found it (using the 1st day of the month) very confusing... and the BOM still uses it: http://www.bom.gov.au/climate/glossary/seasons.shtml – Gerardo Furtado Feb 06 '23 at 03:52
  • 4
    In Spanish class in the far north of Sweden (Kiruna), we had a great laugh when a student listed the months of winter in Spanish: octubre, noviembre, diciembre, enero, febrero, marzo. Another definition (in use by the Swedish meteorological institude based on WMO definitions) is any month with the 24-hour average temperature below 0°C (by this definition, the UK has no winter except for parts of the Scottish Highlands). – gerrit Feb 06 '23 at 07:48
  • 1
    Historically, farming activity has been important in use of the names of the seasons. With the Irish calendar definitions, the various activities tend to fit with the definitions and places the solstices and equinoxes pretty centrally within their seasons. – Dannie Feb 06 '23 at 12:53
  • @Gerardo, what can possibly be "confusing" about the 1st day of the month? This is the most plain and simple definition, and it is used in many (probably most) places. Generally, the trend (worldwide) is to use stricter/fixed definitions instead of "floating" ones like solstice (it just so happens our calendar is not exactly tied to solstice). Contrary to what the answer states, there are official definitions of summer, the BOM one is as official as it gets for Australia. There is just not one worldwide definition. – Zeus Feb 07 '23 at 00:45
  • @Zeus Where I come from (Brazil) everybody says the "summer" goes from the 21-ish of December to the 21-ish of March. The same goes for all other seasons: you would see on the newspapers on 21st of September: "The first day of spring...". This is the government website, even if you don't speak Portuguese you can see the 21st, the official date for starting all seasons: https://www.gov.br/observatorio/pt-br/assuntos/noticias/verao-2021-2022-comeca-em-21-de-dezembro – Gerardo Furtado Feb 07 '23 at 01:04
  • @Zeus I'm also aware that the solstice/equinox dates vary, as early as the 20th or as late as the 23rd. But in Brazil the official date for the beginning of the seasons is fixed on the 21st. – Gerardo Furtado Feb 07 '23 at 01:11
  • @Gerardo, understandably, it can be unfamiliar, but hardly confusing. On the contrary, 21st Dec/Jun can be confusing as solstice doesn't always fall on that day. Either definition is arbitrary (like most of them), but the one tied to the calendar is at least simpler (and arguably, makes more sense: June is colder than September (even in Brazil); similarly in Europe June is warmer). – Zeus Feb 07 '23 at 01:16
  • @GerardoFurtado, I wonder if folks in (for example) Boa Vista in the North use spring etc much. It seems it has a two season climate of "wet" (May to August) and dry, Kind of the opposite to Manaus (wet from December to May) In fact in Amazonia and Roraima you might think the notions of spring, summer, autumn, winter (as derived from Europe/Portugues) are redundant or inappropriate - they don't correspond to tropical weather patterns – James K Feb 07 '23 at 05:58
  • @JamesK you are absolutely right, there's just "wet" and "dry" season. Not only that, in the whole area east to the Amazonia (a semi-desert area around the size of Mexico) the temperature is pretty much the same the whole year, 30°C, so the whole idea of seasons, despite being learnt at school, has little meaning in everyday life. To make things worse, the wet season over there is called "winter" ("inverno"), because it cools down a bit, but it happens during the southern hemisphere summer months! – Gerardo Furtado Feb 07 '23 at 07:17
  • @Zeus: In what sense is the "solstice" floating? Do you mean it is floating with respect to the background of the Gregorian calendar which subdivides a year into sometimes 365 and sometimes 366 days, grouped into months having anywhere from $28$ to $31$ days that are named after ancient Roman gods and politicians? I guess we should be glad that the French Revolution fizzled out or we would all be celebrating the summer solstice on around Messidor III. – Lee Mosher Feb 07 '23 at 18:46
10

As others have suggested, the definition of when seasons change is arbitrary. The advantage that using solstices and equinoxes as the dividing points of the seasons is that it's easy to determine these precise dates. They correspond to specific astronomical phenomena that can be measured and predicted.

If you call them the middles of the seasons, then how do you determine which day each season begins? You could use the halfway point between each pair of solstice and equinox, but this feels less meaningful to astronomers. Earth's motion around the Sun isn't uniform, so this halfway point doesn't correspond to anything actually happening physically.

Glorfindel
  • 4,790
  • 3
  • 25
  • 40
Barmar
  • 975
  • 7
  • 12
  • On the flip side, I think a "new moon" is often defined as the midpoint between the time the moon ceases to be visible and the time it becomes visible again, since judgments of when those events occur may depend upon the sensitivity of the observer, but an observer that perceives the moon as becoming invisible sooner would perceive it as becoming visible later. – supercat Feb 06 '23 at 15:57
  • I was thinking about that analogy. While it does depend on the observer's sensitivity, it's still easier than trying to determine the precise day of a full moon. – Barmar Feb 06 '23 at 16:00
5

A day of just the right length will warm the Northern hemisphere just as much as it cools over the following night, as measured by thermal Joules. The summer solstice is longer than that, or else a year would see us cool; by the same logic, this sweet spot is also longer than the winter solstice. After the summer solstice, the days shorten, but remain long for a while, so warming continues. The NH's hottest time of year is therefore quite a while afterward; there is a similar delay in comparing the coldest time to the darkest.

J.G.
  • 325
  • 1
  • 7