Skip to main content

Department of Meteorology – University of Reading

Show access keys
  • Image Copyright © Stephen Burt


    A teaching and research facility - with over 100 years of weather records

Atmospheric Observatory

Meteorological observations have been made almost continuously at the University of Reading (known until 1926 as University College) since 1901. Although records for some of the earlier years have been lost, an almost complete daily record of many elements can be assembled from January 1908 to date.

Current weather conditions

The weather station has occupied two major sites – from 1901 to 1967 on the London Road campus and then at the Whiteknights campus since January 1968 (and at the current site since 1 January 1970). The Atmospheric Observatory has been a centre for atmospheric measurements, micrometeorological research and meteorological observations since 1968. More details of the London Road sites and records are given in One Hundred Years of Reading Weather by Roger Brugge and Stephen Burt, published in 2015; these pages give more details of the current Atmospheric Observatory site together with details of instruments currently and recently in use.

Manual observations continue to be made 365 days per year. Between 1918 and 2017, the observations were the responsibility of just five principal observers – most recently Mike Stroud from 2009 until his retirement in December 2017. Today most of the observations are made by the current principal observer, Cahyo Leksmono, one of the technician team within the Department of Meteorology. Cahyo is assisted by a small team of mostly volunteer ‘deputy observers’ including PhD students and postdocs who stand in for occasional weekday and weekend duties.

A short brochure about the Atmospheric Observatory is available as PDF download.

During each semester, many of our local observations are discussed at our weekly WCD meeting:

Site details

Latitude 51.44136°N, Longitude 0.93807°W

National Grid Reference (41) 739 719

Altitude 66 m above MSL

Per standard practice, this is the location of the standard raingauge, located towards the western edge of the observatory enclosure.


Over the standard 30 year average period 1981-2010, the mean temperature at the Observatory site was 10.6 °C (mean daily maximum 14.5 °C, mean daily minimum 6.7 °C). The coldest month of the year is January (mean temperature 4.8 °C) and the warmest July (mean temperature 17.6 °C). Since climatological records commenced in 1908, the lowest observed air temperature has been -14.5 °C (on 14 January 1982) and the highest 36.4 °C (on 10 August 2003).

The annual average precipitation is 634 mm, falling on 154 days per year. The driest period of the year is late winter to early spring (February average 41 mm) and the wettest autumn (October average 72 mm), but with wide variations from year to year. The wettest day, calendar month and calendar year on the record since 1901 are, respectively, 76 mm (on 22 September 1992), 180 mm (October 1903) and 961 mm (in 1903). Snow can be expected to fall on 11 days in a typical year, with the ground snow-covered on six mornings. The greatest observed snow depth since records of this element began in 1950 has been 31 cm, on 3 January 1963. Thunderstorms occur on around nine days per annum, most frequently during the summer half-year.

The mean annual cloud cover at 0900 GMT is 72%, and the average annual sunshine duration 1522 hours, about 34% of the possible duration of daylight at this latitude. Contrary perhaps to perceived wisdom, sunshine is recorded on twice as many days per year (just over 300) as measurable rainfall (154).

Atmospheric Observatory tours

The Department of Meteorology offers free public tours of the Atmospheric Observatory during the summer. For more information, dates and to register for a tour, please see the Observatory tours page.

Observatory metadata

Brief details of measurements made are given in the links below, organised by the parameter recorded. More detailed technical information and specifications for the sensors in use can usually be found in the manufacturer’s website links, which are given where possible. (Manufacturer-specific product information and links are given solely to provide additional details on the particular sensors in use at the observatory, and should not be inferred as commercial endorsement.) Photographs of the instrument and its siting are given where available. The period of record available is also given by element where known in the table following. Calibration details of individual sensors can be found in the ‘Calibration data’ tab at the end of this section, or on application to the department’s technicians.

More information on meteorological instruments and standard observing practices can be found in the following publications, copies of which are available in the department library:

Meteorological Measurements and Instrumentation by Giles Harrison (Wiley, 2014)
The Weather Observer’s Handbook by Stephen Burt (Cambridge University Press, 2012)

There are also further details of certain instruments given in published papers using the data generated – see the Cited references tab for latest information.

There are two types of measurements made:

  • ‘Climatological’ – a manual observation made once daily at 0900 UTC, with the values entered in registers by hand. This observation includes ‘eye’ observations of cloud amount and types, present weather, visibility and other elements, as well as manual readings of the thermometers, manual raingauge etc. Almost all climatological observations therefore relate either to the time of the observation itself (i.e. 0900 UTC) or to the 24 hours ending at 0900 UTC. By convention, the maximum temperature and the rainfall read at 0900 UTC are ‘thrown back’ to the previous date. Exceptions are sunshine duration, ‘days with’ thunder, snowfall, hail etc, mean and maximum gust wind speeds and all solar radiation values, all of which refer to the ‘civil day’ (i.e. 0000-2359 UTC). Some climatological records extend back more than 100 years – for details see the individual element description.
  • ‘Automatic’ – regular recording of electrical sensors, at sub-daily frequency. A display of the derived values to the department has been a central aspect, also providing information or data for teaching. Automatic records (sub-daily digital data) commenced for some elements in 1996, other elements more recently: not all of these records are continuous. The three different logging systems which were individually constructed by staff in the department during the period of continuous automatic recording are summarised in Table 2 below. These systems were named after the location of the display – “corrdisp” was the original corridor display system used in the old department buildings at Earley Gate, and “Fidas” – an acronym generated from “Foyer Information and Display System” – was implemented in the Meteorology Building to which the department moved in March 1997. Later, METFiDAS became the acronym, generated from “Meteorology Field Data Acquisition System”. A custom data reception room (GU04) was part of the specification for the new building, cabled to a pillar in the entrance foyer to allow a display to be mounted. (An earlier corridor automatic weather system “CAWS” was operational in the early 1990s and used for teaching, but almost no data survives.)

Currently the automatic sensors are polled and logged once per second using a Campbell Scientific CR9000X logger (Table 2). Data are available at 1 s resolution for research purposes, whilst standard ‘meteorological’ output consists of period averages of the polled values in compliance with World Meteorological Organization specifications – for example, wind speeds are normally quoted as 10 minute averages, and wind gusts as the highest 3 second running mean wind speed. Further details are given in Quantities and Units.

A separate self-contained automatic weather station operated by the Met Office was installed within the enclosure in August 2012; this description does not include this equipment. In addition, there are occasional experimental or test instruments located within the observatory which are not described here.

Because of differences in recording methods and instrumental response times, slight differences can occur between ‘climatological’ and ‘automatic’ records. The ‘climatological’ records are subject to quality-control measures and are normally regarded as the definitive value.


See the links below for the details of each of the instruments in use, together with key metadata and a selection of cited references.

Quantities and units
Air temperature and humidity
Surface temperatures
Earth temperatures
Barometric pressure
Sunshine and solar radiation
Wind speed and direction
Trace gases
Atmospheric electricity
Lightning detection
Soil heat flux
Calibration data
Cited references

Period of record by element

Table 1 summarises the period of record by main element; in all cases, records were made at London Road until 1967 and at Whiteknights from 1968 to date. In a few instances, missing data or incorrect values have been estimated using data from other local observations. The records are not completely homogeneous, in that the current Whiteknights site has less urban influence, and is 20 m higher, than the London Road site and therefore slightly cooler, although careful comparisons have shown that there is no significant difference in the precipitation record.

The full daily observation database and scanned registers are available for research on application to the department; see also One Hundred Years of Reading Weather by Roger Brugge and Stephen Burt, published in 2015. In most cases, a few minor gaps have been estimated from other local observations.

TABLE 1 – Period of record by element

Record commenced
Record ceased
Precipitation Monthly totals – 1901
Daily totals – January 1908
Continues Records missing 1904-1907 (estimated from Forbury Gardens record), together with a few minor gaps in early years
Air temperature – daily maximum and minimum temperatures January 1908 Continues
Daily climatological observations at 9 a.m. GMT – cloud cover and type, wind direction and speed, present weather, barometric pressure January 1908 Continues
‘Days with’ thunder, snowfall, snow lying at 0900 UTC, etc January 1908 Continues Snow depth records commenced late 1950
Earth temperatures at 1 ft / 30 cm January 1908 Continues
Earth temperatures at 4 ft / 120 cm January 1908 December 1971  Measurement depth ‘metricated’ to 100 cm – q.v.
Earth temperatures at 5, 10 cm and 20 cm June 1961 Continues
Earth temperatures at 50 cm and 100 cm January 1972 Continues  100 cm depth measurements replaced 122 cm (4 ft)
Grass minimum temperature January 1920 Continues
Daily sunshine duration (Campbell-Stokes recorder) April 1956 Continues
Autographic precipitation records (Dines tilting siphon recorder) November 1960 Continues
Autographic wind records January 1968 Continues
Solar radiation January 1968 Continues

TABLE 2 – Automatic logging record details

Logging system Equipment Features Supporting staff
1 Jan 1997 to
11 Jan 2009
Keithley metrabyte 12 bit logging cards installed in a DOS PC, with radio clock synchronisation. Initially a UHF radio link transmitted serial data to the corridor display (a further DOS PC) in the old building and, initially, to the Meteorology Building. Shortly after the new building was commissioned, a cable link from the Observatory site was installed. Data transmission necessitated a 20 s interruption in the data logging every 5 min.
  • 5 min average Excel compatible data files, generated daily from 1 s sampling
  • 1s raw data stored for three days
  • 5 min update of graphical display PC
  • Radio clock time synchronisation, also used to ensure display PC “listened” for data at the correct time it was transmitted.
  • Basic auto-refreshed html text web page and graphs.

Hardware: Andrew Lomas

Software: Giles Harrison
8 Aug 2005 to 27 July 2014
METFiDAS I Amplicon 12 bit logging cards and Windows NT PC. Optical fibre link to Meteorology Building allowed mounting of hard disk by operating system, and bi-directional data flow.
  • 5min average Excel compatible data files, generated daily from 1 s sampling
  • 1s raw data stored indefinitely
  • Radio clock time synchronisation
  • 5 min update of graphical display PC
  • Basic auto-refreshed html text web page
Hardware: Andrew Lomas Software: Iain Russell
METFiDAS II METFiDAS II followed as a back-up running alongside the original and eventually taking over. Software for this used National Instruments LabVIEW, and was written by Iain Russell
30 Aug 2014 to date
METFiDAS3 Campbell Scientific CR9000X data logger Level 0 and Level 1 data – Andrew Lomas
Level 2 data – Roger Brugge

This page updated February 2018

We use Javascript to improve your experience on, but it looks like yours is turned off. Everything will still work, but it is even more beautiful with Javascript in action. Find out more about why and how to turn it back on here.
We also use cookies to improve your time on the site, for more information please see our cookie policy.

Back to top