“…since records began” – Christopher Wren’s first automatic weather station

Giles Harrison Department of Meteorology, University of Reading, UK

Weather events have been written about for millennia, so the human urge to describe, catalogue and assess the natural world is not new. But historical weather measurements now have a modern importance well beyond the original investigators’ reasons for acquiring them, in providing benchmarks against which climate change can be assessed. Good quality atmospheric data exist from the seventeenth century onwards, with the longest consistent set of thermometer measurements, the Central England Temperature series, beginning in 1659 (Manley, 1974). Those measurements were made by hand – but how and when did the recording of weather parameters begin to be carried out automatically, rather than by an individual? It turns out to be over 350 years ago.

This first automatic weather recording system was almost certainly devised by Sir Christopher Wren (1632-1723, knighted in 1673). Wren is perhaps best known for rebuilding St Paul’s Cathedral following the Great Fire of London, but, amongst very many other things, he was also a founder of the Royal Society and the Savilian Professor of Astronomy at Oxford. Meteorology was only one of the interests he actively pursued. Wren sketched his plans for a recording rain gauge in 1662 (Biswas, 1967). Although many rain collectors had been made before then (Symons, 1891; Strangeways, 2010), Wren’s designs brought the innovation of recording. One method proposed was based on a flat box, segmented into compartments and shunted by clockwork to present a different receiving section under the rain funnel each hour. The possibility of evaporative losses led to a second design, for a self-emptying collector. In a later summary provided by Nathaniel Grew, the measuring approach was described as:

“…a triangular tin vessel in a frame…whereby the water, as it fills, spreads only on one side from the centre till it fills and empties itself. Which being done, a leaden poise, on the other side, immediately pulls it back to fill again”  (Grew, 1681)

This would now be known as the “tipping bucket” principle. Such a self-emptying approach overcomes the fundamental problem of collector rain gauges – what to do when their collecting reservoir becomes full – and allows both cumulative rainfall and rain rate to be determined.

A recording rain gauge formed part of Wren’s “Weather Clock” – a design for the first automatic weather station. At its core was a pendulum clock, both to keep time, and to turn drums carrying charts on which the variations of meteorological parameters were recorded (see below).

Christopher Wren’s weather clock. (Image RS14257 of the Royal Society Picture Library, reproduced with permission).

Whether this device was ever implemented practically by Wren has been debated, and a recent investigation of the diagrams has concluded that it is unlikely (Mihailescu, 2021). However, after many years of refinement, a version devised by Wren’s friend Robert Hooke (1635-1703) was operated successfully, in 1679 (Biswas, 1967). Five meteorological properties were sensed – from a barometer, thermometer, hygrometer, rain recorder and anemometer – and, every quarter of an hour, the pendulum clock operated a hammer to punch the position of mechanical linkages from the different sensors on paper charts. Nathaniel Grew’s description said:

“…First a pendulum clock, which goes with three-quarters of 100pound weight, and moves the greatest part of the work. With this, a Barometre, a Thermometre, a Rain-Measure, a Weather-Cock…and a Hygroscope. Each of which have their Register, and the Weather-Cock hath Two, one for the Points and the other the Strength of the wind” (Grew, 1681).

These aspects can of course be recognised in a modern Automatic Weather Station, although the sensing and recording methods are rather different. Digital technology has brought many advantages and ready access to high quality measurements (eg Harrison, 2021), but the earlier analogue methods of recording on paper charts, photographic paper and even scratched metal plates have long provided measurements yielding fundamental meteorological findings. Furthermore, analogue traces are often directly linked to the atmosphere in ways which can offer more information than the originally measurement intended. For example, the width of a sunshine recorder’s burn on its record card, rather than just the scorch mark duration, provides additional data on the daytime sky (Wood & Harrison, 2011). Archive historical charts and autographic records therefore deserve attention to their preservation beyond solely keying the data, as new retrieval methods requiring the original material may yet come along.

Wren’s Weather Clock is an outstanding and inspiring piece of meteorological instrumentation using the technology of the 1660s to provide continuous measurements, just as we expect today.

The tercentenary of Wren’s death this year is being marked with a series of events, including an exhibition on his ideas about Sign Language, Beehives, Anaesthesia, Astronomy, Microscopy, Urban Design, Sheltered Living and Weather Recording, at the Old Royal Naval College at Greenwich https://ornc.org/whats-on/christopher-wren-what-legacy-now/

References

Biswas, A. K. (1967). The automatic rain-gauge of Sir Christopher Wren, F. R. S. Notes and Records of the Royal Society of London, 22(1). https://doi.org/10.1098/rsnr.1967.0009

Grew, N. (1681). Musaeum Regalis Societatis (or a catalogue and description of the natural and artificial rarities belonging to the Royal Society and preserved at Gresham College). Rawlins.

Harrison, R. G. (2021). Make your own met measurements: build a digital barometer for about £10. Weather, 76(2). https://doi.org/10.1002/wea.3857

Manley, G. (1974). Central England temperatures: Monthly means 1659 to 1973. Quarterly Journal of the Royal Meteorological Society, 100(425). https://doi.org/10.1002/qj.49710042511

Mihailescu, I. (2021). Graphical details: the secret life of Christopher Wren’s drawing of the weather clock. Notes and Records: The Royal Society Journal of the History of Science. https://doi.org/10.1098/rsnr.2021.0048

Strangeways, I. (2010). A history of rain gauges. Weather, 65(5). https://doi.org/10.1002/wea.548

Wood, C. R., & Harrison, R. G. (2011). Scorch marks from the sky. Weather, 66(2). https://doi.org/10.1002/wea.657