Snow in Western Australia – part two

A special thank you to Graham Barker for generously granting permission to reproduce this article from his webpage: Graham's Paddock.

The images used in this article remain the property of the copyright holder and may not be used without prior written permission.

List Of All Known Snowfalls

This page contains a list and some details of all known snowfalls in Western Australia since European settlement. Information was obtained from mostly the Bureau of Meteorology, but has been supplemented with Graham's own findings from other sources, particularly reports from old newspapers in the State Library of WA.

Snow has almost certainly fallen on many other occasions and in other places that are not recorded here. In the table below, only town names explicitly mentioned in the sources are included. In some cases snow could have been seen in neighbouring areas, but if not specifically stated, they are not included. "Stirlings" refers of course to the Stirling Ranges, and "Porongorups" to the Porongorup Range. 

Year Date Affected Locations Comments
2015 Sep 1 Stirlings Several cm of snow on Bluff Knoll and Toolbrunup Peak
2015 Aug 9 Stirlings Light dusting on Bluff Knoll
2015 May 4 Stirlings Light and short-lived early-morning dusting on Bluff Knoll summit
2013 Aug 16 Stirlings Very light snow on Bluff Knoll
2012 Sep 4 Stirlings Several cm of snow on Bluff Knoll
2012 Jun 13 Stirlings Dusting of snow on Bluff Knoll
2011 Sep 27 Stirlings Snow showers reported on Bluff Knoll
2009 Sep 29 Stirlings Snow fell but did not settle during the morning on Bluff Knoll
2009 Jul 25 Stirlings Bluff Knoll experienced rime plus snow showers which did not settle
2008 Jul 19 Stirlings Light dusting of snow on Bluff Knoll
2007 Oct 27 Stirlings Snow on Bluff Knoll from 8am, continuing throughout day. Snowflakes reported in Katanning for about 30 seconds at 11am.
2005 Sep 27 Stirlings Light dusting of snow on Bluff Knoll
2005 Aug 30 Stirlings Patchy snow up to 1cm on Bluff Knoll
2005 Aug 18 Stirlings Snow 5cm deep remained on Bluff Knoll after overnight falls
2005 Jul 06 Lake Johnston Falling snow and sleet sighted by truck drivers at Lake Johnston, west of Norseman
2005 Jul 05 Stirlings Snow fell on the Stirling Ranges but did not settle on ground, though some remained in bushes
2004 Aug 28 Stirlings Snow up to 1cm deep covered Bluff Knoll above 900m level
2003 Aug 22 Stirlings, Ongerup, Gnowangerup, Jerramungup, Nyabing and near Lake King Snow fell in various parts of the Great Southern around 7am including Gnowangerup, Ongerup, Jerramungup, Nyabing and near Lake King. Up to 7cm of snow at a farm 35km SE of Nyabing. The Stirlings had 2.5cm of snow at 1pm. Most widespread fall since the 1992 snow events.
Details including charts (at Australian Weather News website - will open in new window)
2003 Jul 22 Stirlings There was 2cm of snow reported on Bluff Knoll
2002 Jul 12 Stirlings Hail was widespread across the southwest and snow fell on the Stirling Ranges for the first time in several years
1999 Sep 15 Stirlings A blanket of snow covered the upper reaches of the Stirling Range; some of the deeper portions of snow persisted for most of the day
1999 Sep 03 Stirlings A light covering of snow was observed on Bluff Knoll
1998 Jul 26 Stirlings. See below for a detailed report on snow falls for 1998. 20cm of snow on Bluff Knoll lasting up to three days.
1998 Jun 11 Stirlings Light fall of snow on Bluff Knoll
1993 Sep 18 Porongorups Porongorups Caravan Park from about 12.45pm
1992 Nov 19 Stirlings, Porongorups, Greenbushes, Wagin, Darkan, Narrogin, Manjimup, Kojonup, Rocky Gully, Boyup Brook, Mt Barker. See below for a detailed report on snow falls for 1992. There were light falls recorded at Greenbushes, Wagin, Darkan, Narrogin, Manjimup, Kojonup, Rocky Gully, Boyup Brook and Mt Barker. There were heavier falls on the Stirling and Porongorup Ranges. This is the latest during a year that snow has ever been reported in WA.
1992 Oct 06 Stirlings, Katanning, Broomehill, Kojonup, Cranbrook, Borden, Gnowangerup, Arthur River Snow fell in the early morning at many towns including Katanning, Kojonup, Cranbrook, Borden, Gnowangerup and Arthur River. At Broomehill there were reports of 40 per cent of crops damaged with 5 cm of snow. Katanning registered a maximum temperature of just 7°C. Snow was 20 cm deep at the top of Bluff Knoll, and remained 2-3 days. Nothing like it had fallen in the Stirlings since June 1981; elsewhere it was the heaviest snow since 1952. >
1992 Aug 28 Stirlings, Mt Barker, Katanning A crisp layer of snow at Bluff Knoll remained on the ground until the next day. Light falls were recorded at Mt Barker (two falls of ten minutes each) and Katanning. The temperature at Mt Barker at 3 pm was just 5°C.
1992 Aug 16 Stirlings About 5 cm snow fell at Bluff Knoll.
1991 Sep 15 Stirlings Light snow fell on the higher peaks
1991 Jun 27 Stirlings Snow was reported on the higher peaks. Albany reported a daytime maximum of 7.8°C
1990 Jul 23 Stirlings, Kojonup, other areas About 25 mm of snow fell on Bluff Knoll while light falls were reported further north towards Kojonup. Reports of snow from throughout the south of the state
1989 Jul 07 Stirlings A light fall of snow was reported
1988 Oct 19 Stirlings A light fall of snow was recorded on Bluff Knoll
1986 Aug 12 Stirlings, near Kambalda A moderate fall was reported at Bluff Knoll, down to 850m. South of Kambalda, snow was reported falling and melting on reaching the ground. The temperature at Kalgoorlie at midday was just 6.4°C
1986 Jul 15 Stirlings, Porongorups A moderate fall of snow was reported, down to 700m level on Bluff Knoll
1981 Jun 22 Stirlings A heavy snow fall was reported
1981 Jun 08 Stirlings, Porongorups Light snow was reported
1979 Sep 26 Stirlings Snow was recorded
1979 Jul 20 Stirlings Moderate to heavy snowfall was reported on Bluff Knoll
1978 Jul 25 Stirlings, Tambellup Snow was reported
1978 Jul 21 Stirlings, Porongorups, Greenbushes Greenbushes recorded snow at 8 am. Other reports were from the Stirling and Porongorup Ranges
1976 Nov 12 Stirlings, Porongorups Snow was recorded between 6 and 9 am in the district
1976 Jul 27 Esperance Snow was reported
1975 Jul 31 Stirlings Snow was reported
1973 Aug 06 Cranbrook, Tambellup, Narrogin Heavy snow was reported at Cranbrook between 2 and 4 am, and also at Narrogin and Tambellup
1971 Sep 24 Gnowangerup, Ongerup, Newdegate, near Pemberton Heavy falls of snow were reported from Gnowangerup and Ongerup in the late morning. Other reports extended from near Pemberton to Newdegate
1970 Sep 02 Stirlings, Porongorups, Ongerup Snow was reported extending to Ongerup
1970 Apr 20 Kojaneerup Light snow was reported
1969 Jul 03 Kojaneerup Light snow was reported
1968 Oct 05 Stirlings, Porongorups, Perth Hills Apart from reports in the Stirlings and Porongorup areas, snow was reported in the Perth hills suburbs of Karragullen, Bickley, Carmel and Carilla
1968 Sep 23 Stirlings Snow was reported
1968 Aug 27 Tenterden, other Great Southern Many towns in the southern Great Southern district received about 1 cm of snow although Tenterden near Cranbrook reported one inch (2.5 cm)
1968 Aug 17 Stirlings Snow was reported
1968 May 01 Kojaneerup Snow was reported
1967 Aug 27 Kojaneerup Snow was reported
1967 Aug 09 Kojaneerup Snow was reported
1966 Sep 19 Porongorups Snow was reported
1966 Jul 26 Norseman A light fall of snow was reported, melting on reaching ground
1966 Jul 23 Stirlings A light fall of snow was reported
1966 Jun 27 Coolgardie A light fall of snow was reported
1965 Jun 29 Porongorups, Manjimup, Greenbushes, Bridgetown, near Dwellingup Apart from the Porongorup district, snow was reported in the southwest towns of Manjimup where the depth of snow was one inch (2.5 cm), Greenbushes, Bridgetown and even as far north as near Dwellingup
1956 Jun 26 Perth Hills, Goomalling, Bencubbin, Toodyay, New Norcia and throughout wheatbelt. See below for a detailed report on this event. Snow was reported over a large area, ranking as one of the most widespread events since European settlement. There were widespread reports throughout the wheatbelt between 7 and 10 am including Goomalling, Bencubbin, Toodyay and New Norcia in addition to southern areas. There were reports of snow from the Perth hills suburbs of Roleystone, Mt Helena and Bedfordale. At Geraldton there were reports of snowflakes falling with rain. The maximum temperature was just 6.3°C at Wongan Hills, and 8.8°C at Perth, the lowest ever maximum temperature since records began in 1897.
1946 Aug 10 Stirlings Snow was reported
1946 Jun 26 Stirlings Snow was reported
1944 Sep 20 Borden, Amelup Light snow was reported from Amelup and Borden
1943 Jul 27 Borden, Amelup A heavy fall of snow was reported from Amelup and Borden
1943 Jul 13 Borden A light fall of snow was reported from Borden
1942 Oct 03 Dumbleyung, Tambellup, Kukerin Dumbleyung, Tambellup, and Kukerin reported snow during the day
1941 Oct 17 Porongorups, Mt Barker A light snowfall was reported
1941 Sep 26 Porongorups Snow was reported
1939 Oct 06 Porongorups
1939 Oct 04 Porongorups Heavy snowfalls were recorded on the 4th and 6th. In the latter case snow covered the ground to a depth of 2 to 3 inches (5 - 7.5 cm)
1937 Sep 17 Stirlings Snow covered the ground
1937 Jul 03 Stirlings Snow was reported
1936 Sep 13 Kendenup, Kojonup, Kukerin, Williams, Quindanning Snow was reported from a number of sites including Kendenup, Kojonup, Kukerin and Williams. At Quindanning, near Williams, the ground was reputedly covered for "several miles with an inch or two of snow"
1932 Aug 02 Greenbushes, Mt Barker, Pingelly, Brookton Light snow was reported from a number of areas including Pingelly, Greenbushes, Mt Barker, and as far north as Brookton
1928 May 31 Stirlings, Porongorups Generally light falls were reported in the district. There was enough snow at the Porongorups to build a snowman
1926 Sep 02 Porongorups Light snowfalls occurred at a few places around the Porongorups
1924 Oct 02 Arthur River Snow was recorded at 3.15pm
1923 Nov 18 Dumbleyung Enough snow fell from 6 to 8 am and enough remained on the ground to make snowballs
1923 Sep 19 Narrogin, Lake Grace, widespread throughout Great Southern Snowfalls were widespread throughout the region between 5 and 9 am. Snow fell as far north as Narrogin and covered the ground over a wide area. There was sufficient snow at Lake Grace for snowballing to be prevalent
1923 Jun 29 Stirlings, Broomehill, Nyabing, adjacent Great Southern area Heavy falls occurred between 5 and 9 am covered the Stirling Ranges and surrounding areas with a blanket snow. Snow remained on the Stirling Ranges for four days. There were many other reports from the adjacent Great Southern area. At Broomehill snow fell for half an hour at 9 am. Further north at Nyabing, a light fall melted very quickly
1923 Jun 08 Katanning, Wongan Hills, other areas Light falls were reported throughout many areas of the southwest. Snow fell in small quantities throughout the day at Katanning but most reports indicate brief falls of snow that quickly melted. There was even a report of light snow east of Wongan Hills, one of the most northern reports of snow ever recorded
1920 Aug 07 Kojonup, Nyabing, Dumbleyung, Tammin, east of York Falls were reported at Kojonup lasting over half an hour in total. Other light falls were reported around Nyabing, Dumbleyung and as far north as Tammin and east of York
1920 Jun 29 Mt Barker, Tambellup, Kendenup Snow fell in the Mt Barker-Tambellup districts during the morning. About one inch (2.5 cm) of snow covered the ground at Kendenup. The effects of these falls were noticeable in the year by the retardation of the growth of crops
1910 Aug 08 Albany, Collie Snow was recorded and hailstorms were experienced in Albany. A very light fall of snow was reported in Collie
1910 Jun 30 Mt Barker Snow and sleet fell for 2 hours in the afternoon and then later at 10pm
1909 May 29 Mt Barker Snow fell
1908 Sep 08 Tambellup The ground and trees were covered in snow
1908 Broomehill, Kojonup A heavy fall of snow was recorded in August
1907 Broomehill, Katanning A light fall of snow was reported in August
1903 Oct 10 Snow fell at a few places in the southwest
1901 Oct 23 Narrogin Snow and frost affected crops
1901 east of Esperance Snow was reported inland from Israelite Bay early in early June
1900 Sep 17 Kojonup, west of Katanning, Narrogin, Nyabing, Bridgetown, Broomehill, Williams, Greenbushes, Wandering, Arthur River, Kurnalpi Snow was reported in many areas of the southwest. A heavy fall was reported from Kojonup and other localities west of Katanning between 5 and 7am. Other reports include: Narrogin snow fell for 8 or 9 mins and remained on the ground for 20 mins; Nyabing between 6.35am and 7.15am, lay on ground for 1/2 hour; Bridgetown (where it was 2 cm deep); Broomehill at day break, Williams, Greenbushes (snow fell throughout night and remained on the ground for 4 hours 7.5 cm deep), Wandering and Arthur River (heavy). There was even a report at Kurnalpi, northeast of Kalgoorlie
1900 Sep 10 Greenbushes, Narrogin, Darkan, Kojonup, Broomehill, Arthur River, Wandering Three inches (7.5 cm) of snow fell and remained on the ground for over four hours at Greenbushes. It appears to be one of the most extensive falls recorded in the state until that time
1899 Jul 12 Mt Barker, Kojonup, Broomehill, Balgarup, Jarrahdale, Mundaring, Gooseberry Hill Snow was reported
1898 Jul 28 Boorabbin Snow fell between 11am and midday and the weather was extremely cold
1898 Jun 12 Wagin Snow fell at Lime Lake and the weather was extremely cold
1897 Aug 11 Wagin Light snow fell and the weather was extremely cold
1897 Aug 10 Southern Cross Very light fall was recorded at 10am and melted as soon as it touched the ground. Heavier fall occurred later that day
1896 Southern Cross A brief snow fall occurred in October
1893 Sep 10 Porongorups Heavy fall of snow (5 cm deep) was recorded at Wattle Hill
1887 Jul 29 Chittering Snowstorm occurred lasting 3/4 hour. The flakes were described as about the size of a shilling (2 cm across)
1881 Jun 21 Porongorups Snow covered the Porongorup Range to a depth of several feet in places
1880 Aug 13 Manjimup Large snowflakes fell for ten minutes
1868 A snowfall was reported in the lower southwest
1846 Aug 11 Gordon River (near Cranbrook) The ground was covered by snow


Details of the Snowfalls of 1956 (June 26)

The snowfalls on June 26th 1956 were the most widespread experienced since European settlement, being reported from twenty official and nine unofficial weather stations, as well as numerous individuals. They ranged from Manjimup and Bridgetown in the south west to Wongan Hills and Cadoux in the north, and east to Ravensthorpe and Salmon Gums. Record or near-record cold was experienced in some places. Snow fell through much of the Darling Range, and was even seen in the Perth hills - one of only three occasions this has been known to happen. Despite the great extent, the snow was short lived, the heaviest falls being confined to the period from 5 to 7:30am and melting quickly. In the more northern and eastern areas the snowflakes melted upon touching the ground.

The summary below gives temperatures and snow depths in degrees Fahrenheit, and inches, respectively. These units of measurement were in use at the time, and are what was reported in the newspapers. Metric equivalents (degrees Celcius, centimetres) are shown in brackets.


Snow covered the ground to depths of 3 inches (7.6cm) at many places in the lower Great Southern district. The heaviest falls were near the Stirling Ranges, where up to 6 to 8 inches (15-20cm) covered paddocks near Borden, north of the ranges. Mr V Meir of Borden used a tractor to tow his family around their snow covered paddocks on a sleigh improvised from a sheet of iron. Snow covered the ground and roofs to provide a scene not witnessed in Borden by residents who had lived there 70 years, however it had all melted by 10am.

A West Australian newspaper staff photographer flew over the Stirling Ranges and reported snow lying thick on the hills, but showing signs of melting quickly. The Daily News reported snow 5 or 6 inches (13-15cm) deep on the ranges. There was still a light covering of snow on the lower foothills in the early afternoon, but by then no snow was visible on the flat ground outside of the ranges.

Children waiting for the school bus built a snowman outside the Porongorup post office. Remnants of the snowman could still be seen early in the afternoon. Some other children, also waiting for a school bus, built a snowman on a property west of Toolbrunup from the 3 inches (7.6cm) of snow on the ground.

The town of Mt Barker received about an inch (2.5cm) of snow, while Mt Barrow and some of the other peaks near Mt Barker were also temporarily covered.

Gnowangerup residents saw their first snow since 1920, when 2.5 inches (6.3cm) fell between 7:30 and 9am. Another 2 inches (5cm) fell between 9am and noon.

Snowballs were thrown at Holly siding in the Katanning district, where the ground was covered with snow one inch (2.5cm) thick as far as the eye could see.

At Tambellup, snow began falling at 7am. An hour later the snow was melted by rain, then at 10am it hailed.

Wagin received its first snowfall for more than 30 years, however the snowflakes melted as soon as they reached the ground. The air temperature at 8:15am was 34 F (1.1C).

Snow fell throughout the Kojonup district between 5 and 7:30 am. Residents woke to find their lawns coated by 0.75 to 1 inch (2-2.5cm) of snow; another report tells of a white blanket 2 inches (5cm) deep. It was the first snow recorded in the area, and the town's children lost no time in building snowmen and having snow fights. The minimum temperature of 32.5 F (0.9C) was a record. Snow also covered the ground at the nearby localities of Jingalup and Boscabel, residents being woken by their children who first noticed the white blanket at 5am.

Snow fell intermittently for more than an hour at Darkan, and was also reported during the morning at Arthur River.

A school bus driver in the Williams district, Mr L Gillett, reported ice half an inch (1.3cm) thick covering the windscreen, and snowflakes "floating like feathers" near the Culbin siding.


Manjimup - a night nurse at the Warren district hospital reported snow falling just after midnight. Other nurses went outside and were able to scrawl their names in the snow covering the ground. A similar fall at the same time was reported from Athol Ipsen at Perup, 12 miles (19km) to the east. A local resident saw snow in Manjimup's main street at 3am. Drivers of logging trucks near Deanmill reported a covering of snow "inches thick".

Boyup Brook received a little snow at 6:45am but it only lasted a few minutes and was followed by heavy rain. Areas to the north and east, however, reported falls lying up to 3 inches (7.6cm) deep on the roads.


The Collie district received snow for the first time in the memory of residents of 50 years. Night shift workers at the Collie power house noticed flakes falling but not settling at 5am, and pre-dawn snow also fell in the View St area on the north side of town. Just after dawn, sleet at Ewington was accompanied by fine flakes of snow. Snow fell at the Muja and Ewington open cuts, with a further fall at the latter at 11:30am. The heaviest fall was at Lyall's Mill where it was up to 2 inches (5cm) deep, and the remains of an early fall nearby on a hill on the Mumballup road could still be seen at 9am. Snow was also reported from Worsley and Bowelling.

Ironically, no snow was experienced in Wandering, the coldest town in the Darling Range.


In Wongan Hills, snow began falling at 9:30am. It snowed for 15 minutes but did not settle on the ground. This was the first snow seen in Wongan Hills since the only other fall there 23 years earlier, and is the most northerly snowfall experienced in WA.

Snow fell in the area surrounding Meckering, but not in the town itself. Children from outlying areas brought handfuls of snow to school on the school bus.


Mrs M Davey of Carilla saw two falls in her area, one at 8:15am and the heaviest at 9:15. She says "There were flakes as big as bantam eggs. It was a pretty sight to see it falling, but as it touched the ground it melted away. It did not last for long on the trees". A fine mist of snow in Karragullen and Roleystone melted upon reaching the ground. Snowflakes fell with the rain at Mundaring, Bickley and Walliston. Snow was also reported from Bartons Mill, Pickering Brook, Chidlow, and Mt Helena.

Snow fell at Bakers Hill at 7:30am, forming a thin layer on lawns.

Woorooloo experienced snow for the first time, falling for about 20 minutes at the sanitorium. It melted upon touching the ground, but blanketed the trees and remained there for a while.

Bus driver Terry McGrath reported that a blizzard stopped his bus for 5 minutes on the Toodyay Road near Nanamoolan. The snow "plastered itself all over the bus" and was piled "inches deep" on the road.

Perth experienced its lowest maximum temperature ever recorded: 47.8 F (8.8C). This "maximum" occurred at 2:15am, before dropping to a minimum of 39.4 F (4.1C) at 7:10am. During the daytime the temperature reached only 47.0 F (8.3C) at 2:40pm, and was only 40.5 F (4.7C) at noon. This combination of cold southerly winds and 40mm of rain (30 hours to 3pm) made this an exceptionally wintry day for Perth. No snow was reported from the Perth coastal plain, but conditions were not far from allowing snowflakes to reach near the ground.


New Norcia didn't see any snow, but an area just to the south of it did. Robert Swanell reported driving into a heavy snowstorm when approaching the town along the Great Northern Hwy about 9:15am. Large flakes fell for ten minutes over an area of 6 to 7 miles (9.6 to 11.2km), ending about 2 miles (3.2km) south of New Norcia. Nobody had heard of snow falling in the area before.


Snow was reported falling after 3pm at Salmon Gums, 60 miles (97km) south of Norseman. A white mantle covered buildings, housetops and vehicles, while children made snowballs. Kalgoorlie recorded its third coldest maximum temperature to that time of 50.3 F (10.2C). Snow was reported from Boorabbin, 100 miles (161km) west of Kalgoorlie, but this was seen by an aircraft pilot at an altitude of 2000 feet (610m), not on the ground. The following morning, motorists reported seeing a heavy fall of snow between Higginsville and Pioneer (between Norseman and Kalgoorlie).

The above information is a compilation of reports found in these newspapers:
The West Australian, June 27 1956
Daily News, June 26 & 27 1956
Blackwood Times, June 29 1956
Collie Mail, June 28 1956
The Gnowangerup Star, June 28 1956
Kalgoorlie Miner, June 27 & 28 1956

Details of the 1992 Snowfalls

The year of 1992 was memorable for its number of snow events and also their timing. Up to the middle of August the winter had been mild, then the weather changed. Two snow events in August were followed by one of the heaviest snowfalls for years in October (spring), then a surprise fall on November 19 which was the latest in the year that snow had been recorded in WA (just two weeks before summer). Snow was not confined to the Stirling Range either, with many low-lying towns seeing snow for the first time in decades.

August 16, 1992

The first snowfall of the year brought snow to the Stirling Range, commencing on the night of Saturday August 15 and continuing into the next day. Up to 5 cm was recorded on the summit of Bluff Knoll on the Sunday morning, with snow settling nearly halfway down the mountain at its greatest extent. Snow flurries were also seen in the car park at the base of Bluff Knoll but didn't settle there.

See Graham's own account of a snow chasing trip to Bluff Knoll on this day.- link opens in new window

Also the following scans:
Newspaper story (The West Australian, Aug 17) - link opens in new window
MSL pressure chart for Aug 16 - link opens in new window

August 28, 1992

Light snow fell on Bluff Knoll in the Stirling Range, and enough remained the following morning for children to build a "snowchild".

Mt Barker and nearby areas recorded two 10 minute bouts of light snow mixed with rain from 2:30 pm, with a 3 pm temperature of just 5°C. The maximum temperature of 7.5°C was Mt Barker's coldest August day in 35 years.

Ongerup recorded a record low August maximum temperature of 8.2°C, with reports of snow in the evening. There was also a report of snow from Katanning.

See also the following scans:
Newspaper story (The West Australian, Aug 29) - link opens in new window
Newspaper story (The Albany Advertiser, Sep 1) - link opens in new window

October 6, 1992

In the Stirling Range, snow on the summit of Bluff Knoll was estimated by ranger Allan Rose to be 20 cm deep, reputedly the heaviest snowfall since 1952. He estimated that 1000 or more people climbed the mountain to see the snow, and cars were lined up more than a kilometre from the car park - police were called in to help handle traffic to the small car park which normally sees less than a hundred visitors on a winter day. Further evening snow replenished the cover, which remained for two to three days and was a superb sight due to the cloud staying away.

Snow fell in the early morning at many low-lying inland towns including Katanning, Kojonup, Tambellup, Cranbrook, Borden, Gnowangerup and Arthur River. Katanning recorded a maximum temperature of only 7°C, while Albany on the coast recorded its lowest ever October maximum of 11.5°C.

At Broomehill, near Katanning, there were reports of snow 5 cm deep. Farmer Scott Witham said the snow began about 5:30 am and turned his farm into "a picture out of a storybook", with about 40 per cent of his crop flattened.

See also the following scans:
Newspaper story and accompanying picture (The West Australian, Oct 7) - link opens in new window
Newspaper story (The West Australian, Oct 8) - link opens in new window
MSL pressure chart for Oct 6 - link opens in new window

November 19, 1992

Snow-starved Western Australia's already above-average snow year was topped off with another record - the latest in the year that snow has been recorded. Only two other November snowfalls have ever been recorded in the State.

The heavier falls were on the Stirling Range and the Porongorup Range. Light falls were also recorded at lower lying inland towns of Greenbushes, Wagin, Darkan, Narrogin, Manjimup, Kojonup, Rocky Gully, Boyup Brook and Mt Barker. These falls were light and short lived, but the extent of the low-level snowfalls was especially remarkable due to the date being less than two weeks prior to the official start of summer.

See also the following scan:
Newspaper story (The West Australian, Nov 20 1992) - link opens in new window

The above information was extracted from reports in these newspapers:
The West Australian, Aug 17 1992
The Albany Advertiser, Aug 18 1992
The West Australian, Aug 29 1992
The Albany Advertiser, Sep 1 1992
The West Australian, Oct 7 1992
The West Australian, Oct 8 1992
The Albany Advertiser, Oct 8 1992
The West Australian, Nov 20 1992

Details of the 1998 Snowfalls

June 11, 1998

A light dusting of snow was reported on the summit of Bluff Knoll. The owners of the Stirling Range Retreat spotted snow in crevices and under rocky overhangs through binoculars, and climbers reported being hit by ice particles.

July 26, 1998

This was one of the heaviest falls of modern times, and also one of the longest lasting - snow was experienced on Bluff Knoll for three days.

Snowfalls commencing on the night of Saturday 25th had piled up to 20 cm of snow on Bluff Knoll by the following morning. Further snow, albeit a lesser amount, fell on the Sunday night. It became slushy and thawed within hours, but drifts on the back of bluff Knoll persisted into Tuesday.

Reports of the snow on TV on Sunday night led to crowds of people visiting the Stirling Range on Monday 27th. Rangers were busy, with over 200 cars and up to 1000 sightseers causing congestion in and around the Bluff Knoll car park. People came from as far as Kalgoorlie and Geraldton, and children were taken out of school so that they could see the unusual sight and enjoy the obligatory snowball fights. Further crowds were reported the next day, but by then little snow was left.

Snow was not reported from anywhere outside the Stirling Range. Maximum temperatures in Albany were 10.8 on Sunday 26 and 12.4 the following day.

The above information was extracted from reports in these newspapers:
The Albany Advertiser, June 12-14 1998
The Albany Advertiser, July 28 1998
The West Australian, July 29 1998
The Albany Advertiser, July 30 1998


Snow In Western Australia – part one

A special thank you to Graham Barker for generously granting permission to reproduce this article from his webpage: Graham's Paddock.

The images used in this article remain the property of the copyright holder and may not be used without prior written permission.

Mountain Snow

Mountain snowfalls are typically confined to the higher peaks of the Stirling Range in the state’s south. Situated about 90km inland from Albany, this is a range of rugged peaks rising abruptly from low-lying and mostly flat farmland. Five of the peaks rise to over 1000m, with the highest being Bluff Knoll at 1095m. Although not a great height as far as mountains go, in this region it is high enough to receive occasional light snowfalls in most years.

Sometimes, but far less often, snow may also fall on the Porongorup Range. This is a smaller, lower range (highest point 670m) between the Stirling Range and the coast. However, if snow falls on the Porongorups it will have also fallen in greater quantity on the Stirlings, and so Bluff Knoll is the focus for snow-seekers in WA.

Outside of the Stirling and Porongorup Ranges, the only other significant mountains in WA are too far north and of insufficient height to attract snow. Some areas in the southwest which have received rare snowfalls are hilly (eg Perth Hills and around Greenbushes), but these are generally considered as low level falls.

When It Falls

The earliest that snow has been recorded in WA is April 20th (1970), and the latest is November 19th (1992), however it is very improbable towards the ends of this range. Since WA was settled by Europeans (Albany in 1826, Perth in 1829) only three falls have ever been reported in November, and only four falls prior to June.

As the monthly distribution chart shows, the best likelihood of finding snow is in the July to September period. July and August are the coldest months of winter, and although September marks the start of spring, good wintry outbreaks can still occur and the mountain has cooled further. One of the heaviest falls in recent decades occurred in early October, so early spring should not be discounted.

From Bluff Knoll's summit facing east towards Ellen Peak - Image Credit: Graham Barker

When snow does fall, the best accumulation occurs overnight or in the morning when temperatures are lower. Snow showers may occur at any time of day, but even on the coldest days on the highest peaks maximum temperatures are above freezing, even if only just, and melting occurs.

How Often It Falls

Since 1965, reports of snowfalls known to the Bureau of Meteorology (as listed in the historical data page) suggest an average of just over one snow event per year, as shown on the chart below. Prior to this, reports become progressively less complete the further back in time one looks. Low level falls are much less frequent than those on the Stirling / Porongorup peaks, and appear to be decreasing with time. Whether this reflects climate change or not is hard to say, because of the probable incompleteness of the snow records.

Total number of known snow events in WA each year, from 1965 to 2003.
Each horizontal line indicates one day on which snow was reported in WA;
white asterisks (*) indicate snow at low levels as well as on the peaks.

It is not known with certainty how often it snows on the Stirling Ranges. Anecdotal evidence suggests snow may settle on Bluff Knoll two or three times per year on average, with the possibility of a couple more very light falls going unwitnessed or unreported. Instances of snowflakes briefly falling but not settling on the ground - such as in precipitation downdrafts over ground that isn’t cold enough - may occur more often.

Rime formations on Bluff Knoll - Image Credit: Graham Barker

The uncertainty is due to the lack of people in a position to clearly see the snow on the mountains. The Stirling Range sits in a large national park surrounded by thinly populated farmland; apart from the nearby ranger’s residence and caravan park (altitude only 218m, below the snowflakes), there simply aren’t many observers around. Light snow on the summits can be hard to spot from below and at a distance, even if not obscured by cloud, and when clear, a light dusting melts quickly. Falling snowflakes might only be seen by people climbing the summits, but few people do this in the inclement weather required for snow, especially outside of weekends and school holidays.

How Much Falls

Generally less than five centimeters accumulates on the Stirling Range summits, with only the upper portions of the mountains being affected.

Snow on Bluff Knoll summit, looking towards Ellen Peak. One or two light falls settle in most years, but snow of this depth (up to 5cm) is a once in 4 or 5 years event ... and it doesn't usually last the day, so you have to be quick! - Image Credit: Graham Barker

The biggest fall in recent times was in 1992 (October 6) when 20cm was recorded on Bluff Knoll, with snow on the ground all the way down to the car park (altitude approx 450m). Another fall of 20cm was reported in 1998 (July 26), and two falls of 10cm occurred in 1986 (July 15 and August 12). These four occasions were the only falls in the last twenty years to have exceeded 5cm. Prior to that, depth measurements are scarce, but falls described as "heavy" are equally uncommon. On the Stirling Ranges, anything more than a couple of centimeters is cause for celebration.

The Stirling Ranges are very rocky and thickly vegetated, so, other than photography, snowman building and snowball fights are the only snow activities that can be accommodated. Unfortunately skiing, snowboarding or tobogganing are not practical given the terrain and lack of snow depth. An unprecedented low-level snowfall covering the access roads would be needed for this.

How Long It Lasts

Thanks to the marginal conditions and small quantity, West Australian snow doesn’t persist for long. A typical light fall on the Stirling Range summits melts or is washed away within a day of it falling. The longest lasting known snowfall in WA occurred in 1923 when snow remained on the Stirlings for four days, and heavy falls in 1992 and 1998 lasted about two days, but these were rare exceptions. Even during WA’s most widespread snow event in 1956 (June 26), the Stirlings’ snow had all melted by the following morning.

Near Bluff Knoll's summit facing south - Image Credit: Graham Barker


Low Level Snow

Occasionally snow falls and perhaps briefly settles on the ground at low levels (from 100m to 350m altitude) in inland south-western Australia. Snow in the goldfields region (around 400m altitude) further to the northeast has also occurred but is more rare. Snow near the coast or near sea level is extremely rare.

A recent example occurred on August 22nd 2003 when 3cm of snow fell in Ongerup, 60km north of the Stirlings and only 286m above sea level. It also fell on surrounding areas including the towns of Gnowangerup and Jerramungup, but nowhere else in WA apart from the Stirling Range. Details of this event can be found at the Australian Weather News website.

The exceptional winter and spring of 1992 also brought snow (in four separate events) to Borden, Broomhill, Katanning, Wagin, Cranbrook, Mt Barker, Kojonup, Arthur River, Rocky Gully, Manjimup, Greenbushes, Boyup Brook and Darkan. Notably, the last of these snow events occurred in November, less than a fortnight before the official start of summer.

The most widespread snowfalls yet experienced in WA occurred one morning in 1956. Snow was observed in some Perth Hills suburbs, as far north as Wongan Hills, as far east as Salmon Gums, and a large number of other locations throughout the south west. Details of these and other WA snowfalls can be found in the historical data section.

The map below shows locations in Western Australia where snow has been reported (click on it to see a larger version, 337kb). Locations are marked by red squares, their size varying according to the number of times snow has been reported there (either 1, 2-3, 4-5 or 6+ times). Reported snowfalls vary from snowflakes melting upon touching the ground, through to snow settling and covering the ground - the map does not distinguish between type of snowfall (see the Historical Data pages for more details). Also be aware that snow has almost certainly fallen in places and at times not captured in the historical records.

In other words, use this map only as a general indication of where snow has occurred, not a definitive record.

Map showing distribution of low-level snowfalls in Western Australia

Low level snowfalls in WA share the following characteristics:

  • They are light, rarely reaching more than a few centimetres in depth. Many reports refer to snowflakes falling but melting upon touching the ground.
  • They are short-lived - even if snow settles on the ground it melts soon after, most commonly within half an hour.
  • They are very rare - in most years, low-level snowfalls do not occur at all. Any given area that receives one may not receive another for decades, or longer. For example, the 2003 Ongerup fall was the first significant low level snow in WA since 1992, and the first in Ongerup for 50 years.
  • They are often localised to small areas - it may snow in one town but not its neighbour. The 2003 Ongerup snowfall affected only that area, and not higher places further away.
  • They are virtually impossible to predict. Even in the unlikely event of favourable conditions, the localised nature of low level snowfalls makes it hard to guess which areas may get some and which will miss out.

Experiencing low-level WA snow isn’t something that can be planned or expected, so for anyone seeking snow in WA the only place where there is a realistic chance of finding any in a normal year is on the Stirling Ranges.

How Much Falls

The heaviest falls reported to have settled at low levels occurred on the morning of June 26, 1956. Up to 15-20cm covered paddocks near Borden, north of the Stirling Range. Some other Great Southern and Southwest district locations from Toolbrunup to Boyup Brook accumulated up to 7.5cm. Gnowangerup received 6.3cm, while 5cm was reported near Collie and Kojonup. None of these snow depths persisted very long.

Outside of that exceptional 1956 snow event, the most significant depth was 7.5cm at Greenbushes in 1900. Otherwise, falls are generally less than 2.5cm; any snow that settles and briefly covers the ground is a rare event.

How Long It Lasts

At low levels, the most durable snowfall occurred in the town of Greenbushes in 1900 when snow covered the ground for four hours. However, at low levels, snow rarely lasts more than half an hour before melting. This is due to temperatures at ground level seldom being low enough to sustain snow for long; even on the coldest of cold days, maximum temperatures rarely fail to reach 8°C in inland towns. Also, snowfalls are inevitably followed by rain, which quickly washes away what little cover there may have been. Many reports of snowfalls involve snow falling but melting upon contact with the ground.


Sprites, Blue Jets and Gigantic Jets

Sprites are large-scale electrical discharges that occur high above thunderstorm clouds, or cumulonimbus, giving rise to a quite varied range of visual shapes flickering in the night sky. They are triggered by the discharges of positive lightning between an underlying thundercloud and the ground. They are a type of electrical phenomena called transient luminous events (TLEs) that occur high in the atmosphere. They are rarely observed visually and not well understood. They can extend up to 90 km’s from the cloud top. Sprites are mostly red and usually last no more than a few seconds, and their shapes are described as resembling jellyfish, carrots, or columns. Because sprites are not very bright, they can only be seen at night. They are rarely seen with the human eye, so they are most often imaged with highly sensitive cameras.

They often occur in clusters within the atmosphere above the troposphere at an altitude range of 50–90 km (31–56 mi).

Diag. Credit: Abestrobi  CC BY-SA 3.0  via Wikimedia Commons

The following photo was taken in South East Queensland by Australian photographer and storm chaser Grant Rolph.

Sprite by Grant Rolph on 500px.comImage Credit: Grant Rolph, used with permission. You can see more of Grant’s weather photography here.

Elves are rapidly expanding disk-shaped regions of glowing that can be up to 460 km’s (300 miles) across. They last less than a thousandth of a second, and occur above areas of active cloud to ground lightning. Scientists believe elves result when an energetic electromagnetic pulse extends up into the ionosphere.



Blue Jets
Blue jets differ from sprites in that they project from the top of the cumulonimbus above a thunderstorm, typically in a narrow cone, to the lowest levels of the ionosphere 40 to 50 km (25 to 30 miles) above the earth. In addition, whereas red sprites tend to be associated with significant lightning strikes, blue jets do not appear to be directly triggered by lightning (they do, however, appear to relate to strong hail activity in thunderstorms). They are also brighter than sprites and, as implied by their name, are blue in colour. The colour is believed to be due to a set of blue and near-ultraviolet emission lines from neutral and ionized molecular nitrogen. Blue jets and gigantic jets are both forms of upper-atmospheric lightning or ionospheric lightning.

Photo Credit: Elka Liot & Muscapix Gregory Moulard

Gigantic Jets
Even rarer than blue jets, gigantic jets take place between clouds and the ionosphere where the electric potential is hundreds of kilovolts higher than earth’s surface. They are extremely powerful and while a typical lightning strike may travel less than ten kilometres, gigantic jets have been observed to shoot vertically up for more than 70 km.

Although their physics is not fully understood, researchers believe that gigantic jets could be a “missing link” in the Earth’s “global electric circuit” that helps maintain the potential difference of about 300,000 volts between the Earth’s surface and the ionosphere.

On August 13, 2016, photographer Phebe Pan caught a spectacularly clear wide-angle photo of a gigantic jet on a wide-angle lens while shooting Perseid meteors atop Shi Keng Kong peak in Guangdong province

Photo credit: Phebe Pan

The following photo was captured by my friend and fellow storm chaser Jeff Miles. Taken near a small town in the Pilbara region of Western Australia on Tuesday the 28th of March 2017, Jeff managed to capture a number of photos showing gigantic jets. You can check them out at his photography page here.

Photo credit: Jeff Miles / Snappermiles Photography . Used here with permission.

This article uses material from the Wikipedia 
article Upper-atmospheric Lightning which is released under 
the Creative Commons Attribution-Share-Alike License 3.0 and 
the NOAA National Severe Storms Laboratory with permission.



virga02Virga is an observable streak or shaft of precipitation that falls from a cloud but evaporates  before reaching the ground. At high altitudes the precipitation falls mainly as ice crystals before melting and finally evaporating; this is often due to compressional heating, because the air pressure increases closer to the ground.

 Image credit: ©Daniel Pardini – The setting sun lights up virga over the northern suburbs of Perth, Dec 2015.

Virga can cause varying weather effects, because as rain is changed from liquid to vapour, it removes heat from the air. In some instances, these pockets of colder air can descend rapidly, creating a dry microburst which can be extremely hazardous to aviation.

Virga also has a role in seeding storm cells whereby small particles from one cloud are blown into neighbouring supersaturated air and act as nucleation particles for the next thunderhead cloud to begin forming. When virga is occurring, you will often see precipitation on the rain radar, but the ground will be dry.

Virga from a decaying thunderstorm in the wheatbelt region of Western Australia.

This article is reproduced with permission and adapted under the Creative Commons Attribution-ShareAlike Licence.
Article source:



Pyrocumulus Clouds

A pyrocumulus cloud is produced by the intense heating of the air from the earths surface. The intense heat induces convection, which causes the air mass above the heat source to rise. As the air rises it expands and cools. If the water content in the rising air mass cools to the dew-point temperature in the atmosphere around it (the temperature at which condensation forms) then cloud formation occurs. This is the same mechanism that causes thunderstorms to form on a hot day.
Image credit: ©Amery Drage. Large pyrocumulus cloud from a fire in the Kalbarri National Park, 13th March 2014. The base of this fire is between 50-60km away from the photographer. Notice how the winds at a higher altitude are pushing the top of the cloud over to the right.

Jordan Cantelo 01Image credit: ©Jordan Cantelo. Arial view of the Boddington fire, Feb 2015.

Phenomena such as volcanic eruptions, forest fires, and occasionally industrial activities can form of this type of cloud. The detonation of a nuclear weapon in the atmosphere will also produce a pyrocumulus, in the form of a mushroom cloud. Condensation of the moisture already present in the atmosphere, as well as moisture evaporated from burnt vegetation or volcanic steam, occurs readily on the particles of ash in the cloud. In reality, a pyrocumulus cloud is made up of smoke, ash and condensed water vapour. This is why they always have a greyish to brownish colour, with the tops looking more cloud like than the smoke plume closer to the ground.
Brayden Marshall 01Image credit: ©Brayden Marshall. View of the Boddington fire from the Perth metro area. The base of this fire is approximately 120km away from the photographers location.

craig eccles 01Image credit: ©Craig Eccles. Small pyrocumulus cloud starting to form. Notice how the majority of the smoke is blown to the right but the hot air rising straight up is forming a cloud as water vapour condenses.

Pyrocumulus clouds often contain strong updrafts, which can cause strong wind gusts at the surface. This effect can make an already dangerous fire even more so. As the hot air mass rises, fresh air is pulled into the base. Even a relatively small fire can create it’s own in-draft, which further fuels the fires capacity to burn.  A large pyrocumulus, particularly one associated with a volcanic eruption, may also produce lightning. A pyrocumulus which produces lightning is usually called a pyrocumulonimbus cloud, but not all pyrocumulonimbus clouds produce lightning.

There have also been many recorded examples of pyrocumulonimbus clouds producing rain. There are even a few examples where pyrocumulonimbus clouds, caused by forest fires, actually quenched the fire that spawned them.

Sam Kaye 01Image credit: ©Sam Kaye – Sam posted this photo into PWL on 3rd Feb 2015. It shows the pyrocumulus cloud from the fire near Boddington. Notice how the smoke turns to cloud at about the same altitude and the conventional clouds.

This article is reproduced and adapted with permission under the Creative Commons Attribution-ShareAlike Licence.

Article source:



Just How Hot Is It?

Ever wondered why one day feels hotter or cooler than the next, even though the temperature is the same on both days? The answer is not as complicated as you might think. There is a formula for working out what the temperature of a given day might actually feel like.

The heat index (HI) is an index that combines air temperature and relative humidity in an attempt to determine the human-perceived equivalent temperature. In other words, how hot it feels.

The result is also known as the “felt air temperature” or “apparent temperature”. In Australia, the Bureau of Meteorology uses the term ‘feels like’.   For example, when the temperature is 32 °C with very high humidity, the heat index can be about 41 °C. That is why one day at 32 °C can ‘feel’ hotter or cooler than another day at 32 °C.

Screenshot at Jan 23 21-31-25
Screenshot from my personal weather station showing ‘feels like’ temperature. Note the high humidity.

But why? The human body normally cools itself by perspiration, or sweating. Heat is removed from the body by evaporation of that sweat. However, relative humidity reduces the evaporation rate because the higher vapour content of the surrounding air does not allow the maximum amount of evaporation from the body to occur. This results in a lower rate of heat removal from the body, hence the sensation of being overheated. However, it is important to note that this effect is subjective. How one person ‘feels’ might be completely different to another person.  The ‘feels like’ temperature measurement system has been based on subjective descriptions of how hot subjects feel for a given temperature and humidity. This results in a heat index that relates one combination of temperature and humidity to another.

This table is from the U.S. National Oceanic and Atmospheric Administration, and has been adapted to reflect temperature in degrees Celsius.

To find the ‘feels like’ temperature, look at the Heat Index chart above. For example, if the air temperature is 36°C and the relative humidity is 65%, the heat index—how hot it feels—is 51°C.

The table below highlights the potential effects of heat on the human body.
27–32 °C   Caution: fatigue is possible with prolonged exposure and activity. Continuing activity could result in heat cramps.
32–41 °C   Extreme caution: heat cramps and heat exhaustion are possible. Continuing activity could result in heat stroke.
41–54 °C   Danger: heat cramps and heat exhaustion are likely; heat stroke is probable with continued activity.
over 54 °C   Extreme danger: heat stroke is imminent.

As air temperature is usually measured in the shade, it is important to note that exposure to full sunshine can increase heat index values by up to 8 °C.

This article is reproduced with permission under the Creative Commons Attribution ShareAlike Licence.




Cirrus Vertebratus

A type of cirrus cloud resembling a spinal column or fish skeleton, hence the Latin vertebratus for vertebrae-like.
Image credit: Wikipedia, © used with permission under creative commons license.

These clouds form at very high altitudes (usually between 6000m and 14000m) and often indicate that stormy weather is coming. They usually start out as a smooth band of ice crystals that is then blown by crosswinds to create the fine streaks to either side of the central column. They can be straight or curved, like the one I saw today (in the image below, which is a very poor quality phone image, sorry)IMG_3161Image credit: ©PWL/Matt Fricker taken on 16 Nov 2014




Hail is created when small water droplets are caught in the updraft of a thunderstorm. These water droplets are lifted higher and higher into the thunderstorm until they freeze into ice. Once they become heavy enough, they start to fall. If the smaller hailstones get caught in the updraft again, they will collect more water on the way up.
As the stone gets higher and higher the water will freeze on the outside of the ice pellet, causing the hailstone to get bigger. Once the hail stone becomes heavy enough it will begin to fall. Depending on how many times this processes happens before the hailstone finally falls to the ground, will determine how big the hailstone will ultimately be.

In terms of diameter, the largest hailstone on record fell in Vivian, South Dakota, in the United States on 23 July 2010 (see image below). The stone measured 8 inches (20.32 cm) across, which is similar to the size of a small bowling ball. It likely was even larger when it fell, however, because it is believed to have melted somewhat before it was measured.


Hailstones can fall from a height of 9000 m (30,000 feet) and approach the earth at speeds of as much as 193 km/h (120 miles per hour).

A hail stone shape is circular at smaller sizes and becomes more irregular at larger sizes.  Hail is generally compared to common, everyday objects when reporting size. The following chart will give you a rough idea of the size of hail and the estimated updraft speed required to carry it high into the thunderstorm. When the updraft is no longer strong enough to support the weight of the hail, it will fall to the earth.

Object         Size        Updraft Speed
pea              6mm         39 km/h
marble         13mm       56 km/h
20c              27mm       79 km/h
50c              32 mm      87 km/h
golf ball        44mm       103 km/h
egg              50mm       111 km/h
tennis ball     64mm       124 km/h
baseball        70mm       130 km/h
grapefruit      100mm     158 km/h
softball          114mm     166 km/h

Giant hail stones are usually more irregular in shape. This is because irregularities of a smaller size hail stone are exacerbated as the hail stone gets bigger. Also, smaller hailstones can merge onto a bigger hailstone. When this happens the hail stone will have bulges and will have a larger diameter in certain directions.

Below are some of the recent images we have received from PWL chasers and followers.

Image© Daniel Pardini/PWL – Various sizes from 18th October 2014, Perth

Adam Delves_18oct2014
Image© Adam Delves/PWL – Single pea size hail from 18th October 2014, Perth

Image© Carmen Mallard/PWL – Boyup Brook, 22nd October 2014.

Image© Danica Justine/PWL – Large hailstone from Boyup Brook, 22nd October 2014

10723383_10205033548137855_1015474873_n  10728764_10205033548417862_858974483_n
Image© Blake Moore/PWL – Egg size hailstones from Boyup Brook, 22nd October 2014

To see more images like these, visit the Perth Weather Live Facebook page and like to receive regular updates, amazing photos and much more. You might like to check out these albums.



Sun Halo

A halo is a ring of light surrounding the sun or moon. Most halos appear as bright white rings but in some instances, the dispersion of light as it passes through ice crystals found in upper level cirrus clouds can cause a halo to have colour.

Sun Halo
Image Credit: ©Michael Beazley/PerthWeatherLive

Halos form when light from the sun or moon is refracted by ice crystals associated with thin, high-level clouds, like cirrostratus clouds, which are commonly found between 5-10kms in the troposphere.

There are several types of halos, but in this article we will look at the two most common ones seen in Western Australia.

The 22 degree Halo

A 22 degree halo is a ring of light 22 degrees from the sun (or moon) and is the most common type of halo observed and is formed by hexagonal ice crystals with diameters less than 20.5 micrometers (1mm = 1000 micrometers)

Light undergoes two refractions as it passes through an ice crystal and the amount of bending that occurs depends upon the ice crystal’s diameter.

A 22 degree halo develops when light enters one side of a columnar ice crystal and exits through another side. The light is refracted when it enters the ice crystal and once again when it leaves the ice crystal.

The two refractions bend the light by 22 degrees from its original direction, producing a ring of light observed at 22 degrees from the sun or moon.

The 46 degree Halo

A 46 degree halo is a ring of light observed 46 degrees from the sun or moon. Although they are less common than 22 degree halos, the process by which they form is similar.

What determines if a 46 degree halo or a 22 degree halo develops is the path of the light as it passes through hexagonal ice crystals. A 22 degree halo results from “in one side, out another side”, whereas a 46 degree halo results from “in one side, out the bottom“.

These ice crystals are hexagonal-shaped columns with diameters between 15 and 25 micrometers and have an appearance resembling tiny pencils.

A 46 degree halo develops when light enters one side of a columnar ice crystal and exits from either the top or bottom face of the crystal. The light is refracted twice as it passes through the ice crystal and the two refractions bend the light by 46 degrees from its original direction. This bending produces a ring of light observed at 46 degrees from the sun or moon.

Sometimes, you will notice that the inside of the halo appears darker. This is because light is deflected by the ice into a wide range of directions larger than 22°, but no light goes below that angle. This means that no light is added by the ice into the inside of the halo, at angles lower than 22°. Technically, the sky inside the halo is the same brightness as it would otherwise have been without the ice.

Sun Halo
Image Credit: ©Glenn Casey/PerthWeatherLive

Sun Halo
Image Credit: ©Mark Finley/PerthWeatherLive

Sun Halo
Image Credit: ©Craig Eccles/PerthWeatherLive

Moon Halo
Image Credit: ©Cameron Fisher/PerthWeatherLive

Moon Halo
Image Credit: ©Daniel Pardini/PerthWeatherLive

The text and diagrams in this article are used with permission 
under non-commercial rights from the University of Illinois WW2010 Project.
All photos are copyright and remain the property of named photographer.



Sun Rays

Sun Beams, God Rays, Cloud Breaks, Jacob’s Ladder, Ropes of Maui… these are just some of the names that sun rays are known by. But their proper name is crepuscular rays. A fairly common optical phenomenon, crepuscular rays appear when sunlight is blocked by objects, causing a visible contrast between sunlight and shadow. Clouds, mountains, building and even trees can cause this to happen.

The name comes from their frequent occurrences during the crepuscular hours (those around dawn and dusk), when the contrasts between light and dark are the most obvious. Crepuscular comes from the Latin word “crepusculum”, meaning twilight.

If the atmosphere contains particles that ‘capture’ or reflect the sun light (like very small water particles, smoke, dust and even airborne aerosols) this allows the sun light passing thought the atmosphere to become ‘visible’.
Crepuscular rays are usually red or yellow in appearance because particles in the air scatter short wavelength light (blue and green) much more strongly than longer wavelength (yellow and red) light. Also, because the light traveling through the atmosphere at this time of the day has to pass through as much as 40 times more air than it does when the sun is high in the sky, the filtering effect is more prominent. This process is known as Rayleigh scattering.

Image credit: ©Matt Fricker

As you can see in the images above, crepuscular rays seem to radiate outwards from behind the object. This is in fact an optical illusion due to our perspective as observers.


It is a similar effect to what you see when you stand next to railway tracks or power lines. As you look along the lines toward the horizon, they appear to get narrower.

Crepuscular rays are (almost) parallel. Interestingly, it is actually the shadows that get slightly narrower. This is due to an effect called umbra, and it is best explained in the following diagram.

The illustration on the left represents the situation we have on earth, with the sun as the source of light being much larger. This causes the darkest part of any shadow to gradually taper the further it gets away from the source. 
Image credit: unknown

Crepuscular rays can also form when the sun is high in the sky and light rays break through holes in the clouds.


Anti-Crepuscular Rays
Anti-crepuscular rays are similar to crepuscular rays, excepting that they are seen opposite the sun in the sky (the anti-solar point). Anti-crepuscular rays are most commonly seen at sunrise or sunset and are much less visible. This is due the amount of atmosphere that the sun light has to pass through.

Although anti-crepuscular rays appear to converge onto a point opposite the sun, the convergence is actually an illusion. The rays are in fact (almost) parallel, and the apparent convergence is due to the vanishing point at infinity.

To see anti-crepuscular rays, turn your back to the rising or setting sun and look towards the horizon.

Below are some great examples of anti-crepuscular rays captured by some of the PWL chasers.

Image credit: ©Cameron Fisher

Image credit: ©Jeff Miles – Near Esperance, Western Australia. See more of Jeff’s photography here.

Image credit: ©Grahame Kelaher – Grahame captured this image in Guam. See more of Grahame’s photography here.

I will add more images to this article as they become available.

**Parts of this article have been adapted and used with permission under Creative Commons Attribution ShareAlike License.