Before we had central heating and well insulated homes, there would be wonderful and intricate ice patterns on windowpanes when we pulled back the curtains in the morning following a hard air frost.
I remember them well and the images here are examples that I took on 4 January 2009 when we experienced nine consecutive nights with an air frost including a night when the thermometer plummeted to -13.3C on January 9.
With the subject of climate change being widely discussed and a major conference being held this month in Paris, it is interesting to look at one aspect that affects us as autumn advances into winter: the onset of air frosts.
Details of the first air frost to occur each autumn since 2000 are: 2013 – November 11, 2000/04/05 – November 13, 2011 – November23. This year the first air frost was on the November 22 (-2.6C), which almost equaled the record for a late start to air frosts set in 2011.
Air frosts can mean attractive landscapes and frozen car windscreens when they occur. Their severity tends to deepen as the winter moves into January and February.
Extracting and analysing data since my records began in 1984 reveals a changing pattern. I will take the period from 1985 for this article and use the data for when the air temperature drops to or below -0.1C.
There is a distinct trend over the period for the first occurrence of air frosts to get later and later into autumn. Using September 1 as a datum and averaging the days since that date when an air frost occurs, I find the following trend. The first and last periods are for five years whilst the central periods cover ten years.
We all realise that the mild weather experienced in autumn tends to last longer and delay the more severe weather of winter. Looking at the overall trend, these figures show the quite dramatic shift in the advent of an air frost each year, whilst recognizing that there are considerable variations from year to year.
While air frosts occurred three times in September 2003 and five times in September 1986, Septembers since 2004 have been frost free.
An air frost always occurred during October up to 1999 when there were five with the coldest recorded at -1.7C. The last three consecutive years have been without an air frost in October.
November is quite a different month in that the overall trend is flat for the number of air frosts. However, there are wide variations. We experienced an air frost fifteen times in 2005 contrasting with four, nine, eight and four over the last four years.
December provides quite a contrast in that the trend for the frequency of air frosts has increased over the years from around eight a month to twelve a month. At the extremes there were 25 air frosts in 2005 and only two in 1988.
The data above provides evidence that the climate in Marlborough has changed over the past thirty years or so but is this short-term change or evidence of more permanent climate change?