I recently received an advance copy of a fascinating weather book, Oxford Weather and Climate Since 1767, authored by Stephen Burt and Tim Burt (no relation to me or, for that matter, to each other!) and published by Oxford University Press. The Radcliffe Observatory in Oxford has one of the longest continuous weather records in the world, and thus its data contributes in a significant way to the climate change record.

Oxford Weather and Climate is probably the most detailed analysis of any one city’s weather and climate ever published. Over the course of its 544 pages every imaginable weather statistic is sliced and diced into an array of permutations, all beautifully illustrated in charts, tables, maps, and photographs, many in full color (the production quality of the book is outstanding!)

One might fairly ask: why all this for the small city of Oxford? As the Burts explained in a press release, “The Oxford temperature, barometric pressure and rainfall records have all been made with the best instruments of the time, on the same site, within a few meters of each other, for over 200 years. Almost every change of instrument or exposure was documented, so we have excellent metadata to go with the records themselves: such complete documentation is exceptional. Oxford also has what we believe to be the world’s longest unbroken record of sunshine duration, with daily records back to February 1880.”

Longest periods of record (POR) for weather observations

The longest series of temperature records in the world is that for Central England, a collection of data from several different sites that date back continuously to 1659. The Oxford records form a significant part of the Central England series between 1815 and about 1920.

In their book Stephen and Tim publish a list of sites that have the oldest “mostly continuous” temperature records in the world. The “mostly” means perhaps a few months or years of data are missing, often because of military conflicts.

Precipitation records began in 1767 at the Radcliffe Meteorological Station, with temperature observations starting in 1815. The only sites with longer temperature POR’s than Oxford’s are the following:

1722 to date: Uppsala, Sweden (located about 40 miles north of Stockholm), although with several changes of site within a mile or so. Anders Celsius invented his thermometer scale while serving as an observer at Uppsala.

1725 to date: Padova (Padua), Italy.

1756 to date: Stockholm, Sweden.

1763 to date: Milan, Italy.

1773–1980: Kew Observatory, London, England, although observations are continuous only from 1848.

1775 to date: Prague, Czech Republic.

1781 to date: Hohenpeissenberg, Germany (located 50 miles southwest of Munich). This is the oldest mountain observatory in the world, sited at the elevation of 977 meters (3220 feet).

1794 to date: Armagh Observatory, Northern Ireland; although records are older than Oxford, there are breaks until the 1830s.

In the U.S. the oldest near-continuous weather records are those for Charleston, South Carolina, with precipitation data dating back to 1738. Chapel Hill, North Carolina, and Minneapolis, Minnesota (originally Ft. Snelling), have continuous temperature records since 1820 but from different locations within each city. New York City’s Central Park has the longest series of weather observations (since 1869) for a site that has not been relocated across a town or city at one time or another. (The Central Park site was moved about a mile^[1] across the park, from The Arsenal to Belvedere Castle, in 1920.)

A few standout statistics from Oxford

For those (like myself) who are interested in weather extremes, here are some of the site’s weather superlatives:

—The hottest day yet recorded in Oxford was 19 August 1932, when the temperature reached 35.1°C (95.2°F); this was equalled on 3 August 1990. The coldest night was Christmas Eve 1860, when the temperature fell to -17.8°C (0°F). Update: Oxford reported a high temperature of 36.5°C (97.7°F) on July 25, 2019.

—The longest spell of continuously below freezing temperatures was the 17 days from January 11 to 26, 1823; almost as long was a spell of 14 days in February 1947. In contrast, the summer of 1976 saw temperatures surpass 30°C (86°F) every day for 14 consecutive days from June 25 to July 8, more than twice as long as any other similar spell before or since.

—In February 1888, a 20-hour snowstorm dumped over 600 mm (24”) of snow, the heaviest snowfall in Oxford’s records. (Interestingly, this came just weeks before the heaviest and deadliest snowstorm in U.S. history, the Great Blizzard of 1888^[2].) Remarkably, in that same year, snow reportedly fell for 30 minutes in Oxford just after dawn on July 11 (!), the only record in over 200 years of a summer snowfall in the city. In the entire Oxford record this is the only instance of snow falling between mid-May and mid-October. Other locations in southern England also reported snow showers that day, confirming the account, although with a minimum temperature that morning of 5.1°C (41.2°F) it is likely that the frozen precipitation was ice pellets or graupel rather than real snow. [“Sleet” actually means “mixed rain and snow” in the UK!]