History of the Rothamsted Weather Station

History of the measurement of meteorological data at Rothamsted

Meteorological data has been measured at Rothamsted's Harpenden site since 1853. The earliest records are of rainfall and wind direction. Rainfall was originally recorded in a 5 inch (12.7cm) rain gauge in the garden of Rivers Lodge, a house next door to the Rothamsted Russell Building. The water collected was carried across to the laboratory and measured in a graduated cylinder. The gauge was moved to the meteorological enclosure in Barnfield in around 1880.

Thomas Wilson of Rivers Lodge was recording weather from at least 1873. In August 1876, he offered to 'keep a Meteorological register' for the Met. Office. There is, as far as we know, a continuous record from then until 1899. He undoubtedly submitted the vast majority of these returns but as he often failed to sign them it is possible that someone else (most likely Edwin Grey) was responsible for some of the later ones; the last one that Wilson signed was, we think, that for January 1897. The Rothamsted Archive has Wilson's original records (MET3.1) because he asked the Met. Office to return them after copying the figures. He inconsistently gave the location as Harpenden, Harpenden Common or Rivers Lodge but never (or, perhaps, very seldom) wrote Rothamsted.

Wilson made various observations (at Rivers Lodge) including temperature but his rainfall figures are from Rothamsted (i.e. the Barnfield site). This is not always specified but there are more than enough references to the fact to leave no doubt that this was invariably true.

Data from the Rothamsted rain and drain gauges are in MET2.2. From January 1878, the tables also record minimum and maximum temperatures, which are from Rivers Lodge. Again this is not always specified but it is indicated sufficiently frequently to be sure that it was invariably so. We checked figures for November 1884 in the two documents ( i.e. Wilson's return to the Met. Office and the Rothamsted rain and drain gauge tables) and confirmed that, for both rainfall and temperature, they are identical. It is likely that the temperature data in e-RA were copied from MET2.2 because both start on 1st January 1878; if the source was Wilson's original returns, the transcriber(s) would have been aware that the records started in August 1876.

There is potential to extend back by seventeen months the daily maximum and minimum temperatures in e-RA but it would be wise to first examine in more detail the correspondence in MET4.2, which is thought to make reference to the acquisition of a screen; it's possible that Wilson did not, at first, have a screen. As Wilson's figures also include wet and dry bulb temperatures there is even greater scope to add to the humidity data in e-RA, which currently start in 1915.

What happened after 1899 has not been established but nothing has been found to suggest that Wilson continued making returns to the Met. Office after that date. It is, however, almost certain that it was still Rivers Lodge data that were being submitted to them because Edwin Grey stated (in his 'Reminiscences'): 'The present collection of meteorological instruments now in use were obtained and set up in their present position by Sir Daniel Hall. Before these were acquired we used the records of Mr. Wilson's private instruments, the thermometer screen being in the grounds of River's (sic) Lodge and the barometer in the hall'. Hall came in 1902 and left in 1912 so, assuming that 'the present collection of meteorological instruments' to which Grey refers was on the Barnfield Met. Site, they can't have been in use before 1902. MET5.5 contains Edwin Grey's notebooks of 'Daily Meteorological Records' 1891-4 and, perhaps more significantly, 1897-1910; figures for the years up to 1902 can be assumed to be for Rivers Lodge even if that is not explicitly stated. We have found nothing to indicate when the Barnfield instruments were installed but, interestingly, there are in MET5.6, 'River's (sic) Lodge meteorological records, 1910-27'. Thus the Rivers Lodge instruments were being read until at least 1927, thus implying that observations were for some years being made on both sites (i.e. Rivers Lodge and Barnfield). It is, therefore, possible that, for some time after the new instruments had been installed on the Barnfield site, it was still Rivers Lodge data that were being sent to the Met. Office.

With thanks to John Jenkyn and Liz Allsopp who provided this background information, Jan 2018