The Saguenay Fire . . . May 1870

The 'strange' appearance of the sun and sky . . .
 John Oram was not the only person in western Europe to observe the 'strange' appearance of the sun and sky on the 22nd and 23rd May, 1870. His careful record (22nd) is remarkable though as it is made by a man with many calls on his time - and thus he must have been describing something of great interest: for the sake of completeness, his statement is repeated here . . . " Strange appearance of the sun rose quite red - got dark red at 10 o.c. pink at 1 fading into yellow after - so dim that it looked like the moon. "

 Over the period 22nd / 23rd May (1870), there were several reports of unusual coloration to the sun; the exact colours are described differently, but initially white (not long after sunrise - this is at odds with Oram), then " purple side of red " or " dark red ", then as the sun climbed higher in the sky . . . " pink, inclining to purple ", seems to sum up the observations from a wide area of Ireland, Britain and western mainland Europe; the phenomenon lasted several hours at any one location and the overall sky was described as 'hazy' (implied high dust loading); observers described the sun as if " shining through smoke " and " so dim it looked like the moon ".

 Some reported seeing sunspots on the solar disk - looking directly at the sun, even using opera glasses and as no ill-effects were noted, we must assume that the sun's radiation intensity was significantly reduced.

 Some specific reports: from Dunmurry, County Antrim (Ireland) . . . " The Sun shone pink inclining to purple through a haze during the afternoon ". The observer wrote, " I never before saw (the Sun's) colour on the purple side of red ". Elsewhere, a pink Sun was widely seen in England, and also observed at Rohrbach on the Mosselle on the 23rd, where the Sun appeared 'rose-coloured' through an afternoon mist. [Meteorological Phenomenon, Joseph John Murphy]

 A letter to 'Nature' has this (composite extract) . . . " on the afternoon of Sunday, the 22nd, a very curious appearance was noticed by many. The sky was hazy, and the sun was seen through the haze of a pink colour, inclining to purple. I see by a newspaper that the same was noticed at Dublin. A red or orange sun is common, but I never before saw its colour on the purple side of red ".

  Looking at Symons' "British Rainfall" publication for 1870 (available now on the Met Office web site), on the 22nd the observer at Norwood Cottage [unclear as to the exact location, but probably in Ireland] records . . . " sun's disc visible through a thin covering of cloud ". And from the same publication, at Ballinasloe (Co. Galway) . . . " curious appearance of sun through red clouds ".

 Here are more reports of the 'strange' optical phenomenon, obviously widely observed over Ireland and one or two other parts of the British Isles. J.W. Moore writes: . . . " Through the canopy of cirrostratus the sun now came into view, being at first of a pale white colour, but soon assuming a pinkish or carmine tint. A strange lurid light spread over the landscape and it seemed as though a total eclipse was in progress. For some hours the sun was seen under these circumstances and several spots were noticed scattered over the disc [...] the cloud which produced the appearance just described, no doubt was a vapour-fog suspended in mid-air, the motion of which was extremely slow. The extent of this cloud formation was remarkable. It was noticed in Wicklow, Dublin, Meath, Cavan, Connemara, Louth and Antrim ". Unfortunately, the location of Mr. Moore's report is not stated in the entry I have!

 Another report, again without a location though thought to be in Ireland . . . " On Sunday last it was remarked that the sun had a strange bright copper colour and when looked through an ordinary opera glass, eight spots were distinctly visible on lower half of disc. The atmosphere was thick and indeed the sun seemed as if shining through smoke ". As noted elsewhere, the fact that the sun could be viewed through a magnifying instrument suggests a significant reduction in solar radiation.

 From Bailieborough (Co. Cavan) . . . " From a very early period of the day until late in the evening, groups of people might be seen gazing heavenward, some of them betraying by their looks evident apprehension of danger and all of them more or less emotions of interest or astonishment, while they continually inquired from one and the other: 'What is the matter with the sun?' Young, old, rich and poor were attracted by the strange and varied attire with which 'mighty Sol' decorated himself ever and anon exhibiting almost lightning rapidity, dresses of every tint and colour, simple and compound: light pink, blood, red, purple, green, blue, yellow and then bright like silver ".

 Then at Inniskeen Rectory, Dundalk (Co. Louth) . . . " The obscuration of the pink veil lasted until 5 o'clock. During its prevalence all terrestrial objects presented a strange lurid appearance, like exhibited during a solar eclipse. There were many spots on the face of the sun, some of them large enough to be seen with naked eye, which required no defence of darkening glass even in the focus of a three inch glass telescope. The phenomenon was very beautiful and not alarming ". Again, note the implied marked reduction in solar intensity!

 Another (Irish but location unspecified) report . . . " 9am to 3pm curious appearance of the sun changing from red to pale violet, there was a dense and peculiar cloud passing high up from the earth; horizon clear ".

 And finally, times only - Dublin: 1pm to 5pm, Tynemouth: 5pm and Cambridge: 10am on 23rd.

 These reports were all made at the time - and published largely independently of each other - and it is obvious that for some decades afterwards, the phenomenon was not attributed to a significant fire; for example, the following is taken from 'Astronomical Curiosities: Facts and Fallacies', by J. Ellard Gore (London / Chatto & Windus) which was published in 1909 . . . " A remarkable instance of sun-darkening recorded in Europe occurred on May 22, 1870, when the sun's light was observed to be considerably reduced in a cloudless sky in the west of Ireland, by the late John Birmingham; at Greenwich on the 23rd; and on the same date, but at a later hour, in North-Eastern France: " a progressive manifestation . . . " Mr. Birmingham says, " that seems to accord well with the hypothesis of moving nebulous matter. "
The physics of smoke & light in the atmosphere . . .
 But why should the phenomenon observed by Oram and many others be due to smoke? The theories attributing coloration of sky, sun, moon & other atmospheric phenomena were not fully refined until many decades after the reports laid out above. I have adapted the following explanation from Wallace & Hobbs:

 Smoke, dust, haze particles etc. have radii circa 0.1 micro-metres. The scattering of sunlight by such particles falls within the "Mie" regime. Usually such particles exhibit a spectrum (or range) of sizes wide enough such that the differential scattering of incident light renders what the eye sees as a neutral or whitish hue - the common case.

 However, if the particles are quasi-uniform in size, as would be the case from a single, intense source such as a brush / forest fire, the scattered sunlight due to such particles will be either bluish or reddish in hue as in the case outlined here - with variations that are dependent upon the solar angle, the particular part of the smoke plume that is passing overhead and the concentration of same - with time of course, the concentration decreases and the effects lessen.

 There can be no other explanation for the particular observations recorded other than a major fire - and given the events below, we can confidently tie the 'cause' to the 'effect'.
The Fire and its consequences . . .
 Spring 1870 was notably dry across Quebec. It was so dry for example that farmers across the Province had mostly ploughed their fields by early May of that year. The Saguenay area [Figure 5] in the nineteenth century was populated by people performing the tasks of farming and forestry - with relatively small, well-scattered settlements.

 How the fire started, on the 18th May, is not clear - but the source was thought to be in an area where forest clearing operations were in progress; it's possible that a small fire lit either for cooking purposes or for scrub / debris clearance was allowed to get out of control. The fire, once started, developed rapidly amongst the dry timber - fanned by an already strong breeze - and generated its own heat-driven wind. The fire then spread quickly through the 18th and much of the 19th May and was so intense & spread so rapidly that many families only had time to reach safety or the nearest area of water - abandoning their possessions. Some people had to stay in water for hours whilst the fire burnt down, or they were able to be rescued - the intensity of the heat was such that the fire jumped alarmingly from one area of scrub, brush or timber to another. [Figure 6]

 The area affected lay between the Mistassini River, near Lac St-Jean and Baie des Haha covering some 150 kilometres (just under 100 miles): the area is estimated at some 4000 square kilometres (or 400 000 ha, nearly 1 million acres); over 500 families, representing nearly one third of population in the Saguenay region lost all their possessions - as everything in the path of the fire was destroyed - personal property as well as public bridges and buildings.

 By evening on the 19th, the fire had largely burnt itself out and an 'eerie calm' settled over the devastation; but not before it had produced huge quantities of fine ash - the plume of smoke / soot particles was carried towards the east (and eventually to Europe) on the upper winds (see below).

 Surprisingly few lives were lost (5 deaths is the figure generally quoted), but many people were seriously burned (many due to the intense heat radiation as well as the more normal contact burning by flame). Cut off from the rest of Canada & the near USA, and with poor communications, the victims were reduced to building crude huts from scorched tree trunks and sleeping on the ash-covered ground. A disaster relief committee was eventually set up and began distributing food, seed, clothing and other supplies sent in from parishes along the St. Lawrence and from the large dioceses in Montréal and Québec City. As a result of newspaper coverage describing the terrible plight of the victims, aid from across the Province began to come in. However, it was many years before the region recovered fully from this disaster.

 As is common in these types of fire, the lower atmosphere assumes a 'super-adiabatic' lapse rate [the temperature falls rapidly with height from the hot surface for a 1000 m or so] due to intense surface heating - allowing vigorous upward currents to carry the smoke / soot particles (along with other, larger debris) high into the troposphere; the larger particles will fall out not far from the source but the tiniest products of the combustion process will be carried well clear of the seat of the fire - and then moved in the direction of the prevailing mid / upper tropospheric wind - circa 600 to 400 mbar [roughly 4 to 7 km altitude].

 As regards the path of the minute smoke (or soot) particles, the distance Saguenay - Mayo is approximately=2350 nmi / 4300 km. The fire was declining rapidly by the evening of the 19th May (see above), yet had 'occurred' on the 18th, and given the obvious destruction, it must have been an intense / hot source - pushing a large quantity of smoke/ash into the atmosphere within a period of 24 to 30 hours. John Oram (along with others in Ireland) observed the effects on the morning of the 22nd - from sunrise (deep red) throughout the morning, and well into the afternoon. Assuming a nominal 'start' time of 0900 GMT 18th, and a nominal 'arrival' time of 0900 GMT 22nd, this gives an elapsed period of 4 x 24 hours=96 hours. The plume, travelling 2350 nmiles in 96 hours has an average speed of 24 knots. If the plume actually arrived overnight (which, given Oram's report of the effect at sunrise is quite likely: say 0300 GMT), then this increases the speed to 2350/90=26 knots: a working figure of around 25 knots would then seem to be the best approach.

 This sort of speed is quite low for the middle part of the troposphere (the 'weather' zone) - and implies that we're not dealing with an active jetstream or anything approaching that strength. But at the same time, the flow must have been close to being 'zonal' in character (i.e., a wind from due west to east); any large-scale meanders in the flow would imply a good deal of turbulence which would disperse the sooty particles to such an extent that any optical effect would be small.

 Nowadays of course, we would be able to analyse the trajectory, altitudes, composition etc., very quickly - but in the 1870s no such methods were available. It was many decades before the link was made between the optical phenomenon and the Saguenay fire.
 [ Much of the basic information for this section taken from:, accessed 07 MAY 2014 ]