The Met Office has forecast a warm start to the week across much of the United Kingdom, followed within roughly thirty-six hours by the possibility of air frost.
Monday brought highs of about 24C in southern areas and 22C further north, under dry and sunny skies in the south with cloud and patchy rain in the north.
Tuesday is expected to be the peak, with the south-east reaching 25 to 26C. A cold front moves south-eastwards across Scotland and Northern Ireland during the day, with rain arriving in the north-west later.
After that the picture changes. "The forecast becomes more changeable from Wednesday as a series of weather fronts move across the UK," said Tony Wisson, Met Office deputy chief forecaster.
Heavy rain is possible on Wednesday, with eastern and south-eastern Scotland and northern and north-eastern England the areas being watched. Thursday brings long sunny spells across England and Wales, around 22C in London, and the potential for strong winds in the far north-west.
Between those two days sits the detail that makes the week unusual. The Met Office expects a notably cold night from Wednesday into Thursday, with localised air frost in rural areas: air temperatures falling to freezing or below.
So a single week in late September is forecast to deliver 26C in one part of the country and a frost in another, with a spread of roughly twenty-six degrees between them.
None of this will be recorded as anything. The Met Office defines a UK heatwave as a period of at least three consecutive days on which the daily maximum temperature meets or exceeds a threshold that varies by county, calculated from the 1991-2020 climatology of daily maximum temperature at the mid-point of meteorological summer, 15 July.
The warm spell has two days. It is therefore not a heatwave, and an overnight air frost confined to rural areas in September will not be announced as anything either. The week will pass through the headline categories as an absence.
This newspaper regards the arrangement as sound and wishes to defend it. A threshold exists precisely so that weather can be compared with itself, and a threshold that is met by anything warm is not a threshold but a thermometer.
The mid-point of meteorological summer is the correct benchmark for the same reason. Late September is being measured against 15 July, a day it is not, and is failing to be it by a margin the Met Office has calculated to the county.
The consequence is a country that can produce a twenty-six degree spread inside three days and emerge from the exercise with nothing on its permanent record. The Department of Unnecessary Research considers this the finest piece of climate management in Europe: not a reduction in extremes, but a definition under which extremes decline to count.