Melbourne's highest temperature on record, 117°F (47.2°C), on Black Thursday, 6 February 1851—but is it valid?

Melbourne's Argus newspaper of Saturday, 8^th February 1851, reports: “Thursday (6^th February 1851) was one of the most oppressive hot-days we have experienced … In the early morning the atmosphere was perfectly scorching, and at eleven o'clock the thermometer stood as high as 117 degrees (47.2°C) in the shade; at one o'clock it had fallen to 109 degrees (42.8°C) and at four in the afternoon was up to 113 degrees (45.0°C). The blasts of air were … impregnated with smoke … in the evening, after an hour's battle for the supremacy, the cool breeze came …”

Andrew May at http://www.emelbourne.net.au/biogs/EM00199b.htm writes:

“On 6 February 1851, three months after news of Separation from New South Wales had reached Melbourne, severe bushfires gave the soon-to-be-proclaimed state of Victoria a fiery baptism and are remembered as one of the State's worst natural disasters. Fires around Macedon, the Dandenong Ranges, the Pentland Hills, Portland, Port Fairy, Geelong and in the Plenty Ranges showered the city with charred matter, while a blasting wind carried singed leaves as far as Tasmania, scattering ash upon the decks of ships in Bass Strait …

'From the appearance of the sky', noted the Argus, 'and the occasional brilliant bursts of light that were perceptible, one might have been almost led to imagine that it had been some one of the numerous residences in the vicinity of the Botanic gardens that had taken fire'.

William Strutt's painting Black Thursday (1864) graphically depicts a melee of terrified people and animals fleeing ahead of the conflagration. Strutt described in his journal the overwhelming heat of 'that scorching Thursday' where at breakfast 'the butter in the butter dish was melted to oil, and the bread when cut turned to rusk'.

Stock, fencing and property losses were severe, and fatalities were recorded in the Barrabool Hills near Geelong and elsewhere. In a harrowing episode, the bodies of Bridget McLelland and her five young children were taken by bullock dray from their Diamond Creek station near the Plenty Ranges to the Travellers Rest Inn at Collingwood for an inquest before Coroner W.B. Wilmot.”

Melbourne's Age newspaper of 18 January 2007 describes these fires as being the most severe in terms of size to hit Victoria in recorded history, burning 5 million hectares (50,000 square kilometres) or nearly one quarter of the state's land area.

William Strutt, artist; Black Thursday, February 6th (1851); La Trobe Picture Collection.

However, the Abstract of the Meteorological Journal kept at Melbourne Port Phillip, during the month of February 1851 as recorded by the New South Wales Government Gazette did not suggest that temperatures reached such extreme levels. It reported temperatures of 96°F (35.6°C) at 8.30am, 108°F (42.2°C) at 2.30pm, 106°F (41.1°C) at sunset (~7pm) and 88°F (31.1°C) at 9pm.

A regression analysis was performed on 1979-2008 9am, noon, 3pm, and 6pm temperature data to yield Melbourne's maximum temperature as a function of temperatures at 9am, noon, 3pm, and 6pm, and the highest of these four temperatures. This proved to be quite an effective method of deriving the likely maximum temperature, the standard error of estimate being only 0.65°C.

Maximum=0.491+0.020*9am+0.087*Noon-0.023*3pm+0.026*6pm+0.929*Highest

A regression analysis was then performed on the temperature data reported by the New South Wales Government Gazette to yield Melbourne's temperature on 6^th February 1851, as a function of time of day. It is important to note that the clock times reported needed to be adjusted 1 hour 20 minutes later, in order to compare with Eastern Daylight Saving Time.

Temperature=-0.0205*time³+0.758*time²-8.035*time+60.863

The second equation was then solved to obtain estimates of 9am, noon, 3pm, and 6pm temperatures for 6^th February 1851 (these being, respectively, 35.0°C, 38.2°C, 41.7°C, and 42.2°C), the solutions so derived being inserted in the first equation. This yielded an estimate of Melbourne's maximum temperature on that day of 43.9°C. This new estimate is 3.3°C (or about five standard deviations) below that suggested by the newspaper report.

In conclusion, if one assumes that the temperature data reported by the New South Wales Government Gazette at 8.30am, at 2.30pm, at sunset, and at 9pm are valid, it seems highly unlikely that Melbourne's temperature reached 47.2°C (117°F) on 6^th February 1851 and that a better estimate of the day's maximum temperature is 43.9°C.