Inside the Collection

Meteorology at Sydney Observatory

Aerial photo with a weathervane, mounted on top of a large yellow time ball, with various sandstone buildings below. The Sydney Harbour Bridge, The Rocks area, Sydney Harbour and Luna Park can be seen in the background.
The weathervane atop Sydney Observatory’s time ball tower, in 2018, with a view over Sydney Harbour.

Meteorology, the study of the Earth’s atmosphere and weather conditions, plays a significant role in determining climate trends and patterns and their impact on the environment. It is a branch of science that has become even more relevant as we observe the devastating consequences of climate change all over the world.

In 2019–20 Sydney has seen extreme smoke engulf the city as a result of the Gospers Mountain megafire, during which Penrith briefly became the hottest place on the planet, and the storms and flooding that followed in February. The 200-plus years of scientific weather data recorded across Sydney and NSW, including from the weather station on Observatory Hill, will become even more important in coming years as we continue to understand our changing climate and weather, and the effects on the city’s environment and inhabitants.

Climatologist Linden Ashcroft explains the profound history between the Bureau of Meteorology and Sydney Observatory in this extract from The Story of Sydney Observatory.

A brief history of weather observations in Sydney

Sydney Observatory is home to one of the longest continuous weather records in Australia. Weather data has been recorded at Observatory Hill almost every day since July 1858, and the Observatory has long been a hub for the development of Australia’s meteorological network.

The site of Sydney Observatory is historically very fitting, as it is less than 700 m away from Australia’s first weather and astronomical observatory. Lieutenant William Dawes of the First Fleet set up his observatory, at what is now Dawes Point, to monitor the passing of Halley’s Comet. Here, in a little wooden structure, Dawes dutifully recorded the temperature, pressure, wind and clouds experienced from September 1788 to December 1791, the first three years of British settlement in Australia.

Weather observations in Sydney were taken at a range of places across the city between Dawes’ original observatory and the opening of the current building in 1859. Governor Thomas Brisbane kept a weather journal at his residence and observatory in Parramatta in the 1820s, and records were taken at the city hospital. Sydney newspapers also reported on the weather for much of the 1830s and 1840s, but not much is known about how they were recorded.

Government weather stations were set up at several coastal sites along eastern Australia in the 1840s, including at South Head at the entrance to Sydney Cove, William Stanley Jevons, an assayer at the Mint and one of Australia’s last ‘amateur meteorologists’, monitored the weather from his residences in Petersham and Double Bay in the late 1850s.

Black and white photograph depicting a lightning strike at night. The top two-thirds of the image is filled by sky with multiple, branching bolts of lightning. In the foreground can faintly be seen various buildings and a body of water beyond.
Photograph of lightning taken at Sydney Observatory by Henry Chamberlain Russell, 1892. MAAS Collection: 95/239/25

A suitable home

The placement of the Observatory was chosen primarily for astronomical reasons, but its exposure to its wider surroundings was also suitable for recording accurate weather data. William Scott, the Observatory’s first government astronomer remarked that the location ‘occasions some inconvenience in windy weather but on the other hand, there is considerable exemption from smoke and dust, and the building, standing in the midst of a large Government reserve, is not liable to be encroached by neighbouring buildings.’

A barometer was housed inside the Observatory for measuring pressure while an anemometer, a device for recording wind speed, was placed approximately 20 m above the ground, on the Observatory roof. One rain gauge, or pluviometer, was located in the grounds, and another with the anemometer, to measure rainfall. Thermometers were kept on the south side of the Observatory in a thermometer house known as a Glaisher stand, which was common practice at the time. Other instruments within the site measured different variables, such as ground temperatures, sunshine and evaporation.

When George Smalley took over from Scott as government astronomer in the 1860s, he devised a new thermometer shed that he and his contemporaries felt was more suited to the warm Australian climate. Observations were taken from there between 1865 to 1917. The shed fell into disrepair in the early 1900s, and observations began in a Stevenson screen, the type of thermometer screens used today. In 1917 the Stevenson screen and most other instruments were moved to the Bureau of Meteorology’s current site, about 135 m further south near the Fort Street School.

Sepia-toned photograph mounted on an aged piece of paper. The photograph depicts complex measuring machine, approximately as tall as a person, with various weights, pulleys, a water-collecting container and a rolled measurement chart.
Recording parts of the anemometer and pluviometer at Sydney Observatory, C Bayliss, Sydney, 1880–90. MAAS Collection: P3549-61

A weather nerve centre

For many years, the Observatory was the operational centre for the wider NSW weather network. Scott, for example, set up 12 regional weather stations in his role as government astronomer. When Smalley took over he apparently closed many of them, but after his retirement in 1870 the reins were given to Henry Chamberlain Russell, a tireless advocate for expanding Australia’s weather observing network. Under Russell’s care, NSW’s network of weather stations grew to around 1800 and included many volunteer rainfall stations. Instruments would travel to and from the Observatory by rail or horse and cart, along with detailed calibration instructions to ensure they remained standardised with the official Observatory equipment.

Silent observer

Over the years, Sydney Observatory has recorded the varied and often wild conditions of Sydney’s weather. The wettest day at the Observatory was during the infamous east-coast low event in 6 August 1986, when over 300 mm fell in just 24 hours. The mercury reached a maximum of 45.8°C on 18 January 2013 during the ‘Angry Summer’ of 2012–2013, when many other extreme temperatures records were broken. The Observatory’s coldest night was back in 22 June 1932, when temperatures fell below 3°C.

The end of era

An expansion of the Fort Street School means that the weather station will soon have to move. In 2017 the Bureau of Meteorology, in conjunction with the Observatory, started a new chapter in the history of weather on Observatory Hill. Equipment has been set up less than 200 m north of the current station, near the Observatory offices. Parallel observations will now be taken to monitor any differences between the two sites, to ensure that the old and new records can be used together, continuing the weather records at this important location.

Lawn area in which is mounted a white box, approximately 40-50cm in size, on top of a metal post. The four sides of the box have louvres, fixed on an approximately 45 degree angle.
Stevenson screen used by Bureau of Meteorology on Observatory grounds. Photo: MAAS


‘Meteorology at the Observatory’ written by Linden Ashcroft, extract from The Story of Sydney Observatory, published by MAAS Media, 2018. The Story of Sydney Observatory can be purchased online or from the MAAS Store.

4 responses to “Meteorology at Sydney Observatory

  • Is there a reason the BOM closed the old station?

    It has one of the longest streams of consistent data of weather in Australia.

    It could be because it has some inconvenient contradictory data. This is worth checking for oneself.

    The new station will faithfully report records every other year because there is no history.

    It is sad to see history erased in this way…

    • Hi Rick,

      Thanks for the question. As mentioned in the post, the old station was closed to make way for expansion of the Fort Street School. The new station (built 200m away, on the Sydney Observatory site) was established in 2017, and the two stations took parallel data until 31 August 2020, when the Observatory station took over as the primary station, and the old one was decommissioned (if you’re looking at the BOM website, the actual changeover took place at 4pm on 31 August). These parallel observations allowed any small differences between the data streams to be monitored, and a small calibration was required prior to the changeover to ensure the continuing of climate data from Observatory Hill.

      From a historical context, of course, there have been many changes in procedure and instruments over the years. For example, the Fort Street station went fully automatic in 1990; prior to that, it used self-recording (paper-chart) devices. So nothing new there – and no doubt we will continue to upgrade the instruments and procedures into the future.

      Hope that helps!
      Sarah Reeves, MAAS

      • I cannot see how an expansion of a school can be more important than ” the greatest moral challenge of our time”.
        Seriously. If anyone cared at all about climate change they wouldn’t consider closing the longest accurately maintained data stream for a building extension.
        Unfortunately the “parallel streams to be monitored” will not be examined by the media – instead new “records” will be reported every other year.
        The historic changes you mention are nowhere near the impact of the location change. The location change deliberately erases history just like Orwell’s 1984…

  • Rick, your concern about moving thermometer locations is justified to some extent, but is routine, and it should not cause any significant distortion of temperature histories, when properly reconstructed. In fact it is undesirable to have long records at sites undergoing urbanization, as they are guaranteed to have distortion from increasing urban heating.

    Actually it is not at all necessary to have an overlap between data at two different sites, and the BoM ACORN-SAT corrected-data sometimes goes wrong when it relies on overlap to join data segments, rather than using data from neighbouring stations.

Leave a Reply