Australia Bushfire Smoke Now Warming the Lower Stratosphere? March 2020 Update

From Dr. Roy Spencer’s Blog

April 1st, 2020 by Roy W. Spencer, Ph. D.

Last month I noted how the global average lower stratospheric temperature had warmed considerably in recent months, especially in February, and tentatively attributed it to smoke from the Australian bushfires entering the lower stratosphere. You can read more there about my reasoning that the effect was unlikely to be due to the recent Taal volcanic eruption.

Here’s the March 2020 update, showing continued warming.


The effect cannot be as clearly seen in regional averages (e.g. tropics or Southern Hemisphere) because those regions routinely see large changes which are compensated for by changes of the opposite sign in other regions, due to strong adiabatic warming (sinking motion) or cooling (rising motion) in the statically stable stratosphere. Thus, global averages show the best signal of something new going on, even if that something new is only occurring in a specific region.

0 0 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
April 1, 2020 6:24 pm

Interesting to see the little known other side of the aerosol effect normally assumed to be backscatter cooling in the stratosphere but particulates are also able to absorb heat and cause warming in the lower atmosphere as for example in the so called Gottschalk Curve.

April 1, 2020 7:05 pm

I have two questions, not that I expect anyone to be able to answer them other than Dr Spencer. (If anyone knows him please ask him to share his thoughts)
1 The chart shown above covers the period 1978 to 2020. A problem with many charts related to climate change is that they are typically short duration – the author chooses a subset of available data rather than the whole set, because the sub-set supports his views and the full set does not. So, is there data going back beyond 1978, and if so, can we see it please?

2 In my simple way I look at the dataset and – apart from the three outliers – it appears to show a trend decline. Is that correct, and if so, what does it mean?

I have a horrible feeling that if the trend was up the climate activists would claim it was evidence of AGW, and even though it is down they will still say it is evidence of AGW for some arcae, non-testable reason.

Greg Goodman
Reply to  OldCynic
April 2, 2020 2:49 pm

There is tendancy for those doing “simplistic” analysis to try to model everything as a straight line “trend”. Please explain why you think that is a suitable model for a data set with two very clear spike events of know cause.

You seem to want put you thumb over the spikes and smooth the rest into a straight line even though even then it is clearly NOT a straight line.

This linear thinking is endemic in climatology because they already know where they want to end up. If you can make everything a straight line you can claim spurious “correlation” to CO2 forcing.

Since we know the cause of the spikes it should be obvious that the downward steps ( about 0.5 deg C at each time ) after each eruption were also caused by atmospheric changes caused by the eruptions.

Much of that is ozone loss , destroyed by sulphate aerosols. That is measured and pretty well accepted. Less ozone means lower stratosphere is more transparent , blocks less solar and warms less.

The period after Mt P is essentially flat for 30 years: no more event on that scale. Nothing to do with Montreal which was another fake UN crisis. They too try to pull the straight line trick, claim that was our CFCs and the magically flat bit is the Earth “healing” thanks to the UN. It was two events not a “trend”.

Work out the effect of a more transparent stratosphere and you are on your way to understanding the late 20th c. warming and why we got a troublesome “haitus” afterwards.

TLS is still in hiatus and one of the clearest indications of what is going on since it is so noise free, unlike the surface record which is plagued by weather.

Does that answer your question?

I wrote about this 8 years ago:

Reply to  OldCynic
April 2, 2020 3:42 pm

Dr Spencer does occasionally read this blog and reply to comments. But to comment on the date range:

Dr Spencer is primarily working with satellite data which has only been available for the timeframe given. In other words he has used the full data set. Satellite has a couple of benefits in that in provides the same measurement metric globally with no calculated interpolated or kriging, and can also provide information about different atmospheric layers well above ground level.

The satellite systems are not without weaknesses however, such as daily drift as the satellite orbits vary (measuring different places at the time on different days), altitude drift as the orbits decay and instrument decay over time vs replacement with new instruments.

However even with the known issues the better spatial coverage is enormously valuable.

As to the trends? I would be concerned that the existing data set is really still too short to conclude anything about a long term trend, but the short term perturbations are very interesting.

Reply to  OldCynic
April 4, 2020 12:20 pm

Dear OldCynic

That decreasing trend leveled out substantially in the late 1990’s and has had a much smaller gradient for the past 20 years. Until that became apparent, the AGWers especially John Crook et al used to crow incessantly about the decrease in stratospheric temperature being due to CO2-driven global warming. Now you never hear them bring up the stratosphere data and if challenged they’ll ‘explain’ the last 20 years by saying the ozone layer recovery has conveniently exactly cancelled out the CO2 effect.

April 1, 2020 7:14 pm

Where are all the other Australian bushfires of previous years?
These were not the only bushfires ever to have happened in Australia.

Furthermore, there are a huge number of fires all around the world, all the time.

The volcano seems a more plausible explanation to me – for the main spike at least.

Reply to  JCalvertN(UK)
April 1, 2020 7:23 pm

Good point, JCalvert. Well made.

Pat from Kerbob
Reply to  JCalvertN(UK)
April 1, 2020 9:05 pm

Wasn’t 74 or 75 far bigger for Bush fires in Oz
Much larger than this year?

Thought I saw a chart a month or two back on here that showed that

I don’t see anything unusual in the chart above for that time frame

Pat from kerbob
Reply to  Pat from Kerbob
April 1, 2020 9:07 pm

Never mind
Chart only starts at 79

Satellite data

Reply to  Pat from kerbob
April 1, 2020 9:45 pm

February 1983 – Ash Wednesday Bushfires

February 2009 – Black Saturday bushfires

Why don’t the Canadian or Californian Forest fires get a mention?

Patrick MJD
Reply to  Pat from kerbob
April 1, 2020 11:53 pm

Some of the really REALLY big fires, simply because no-one could fight them and they literally burnt out, were in the 19th century.

Reply to  JCalvertN(UK)
April 2, 2020 1:27 am


It was undoubtedly a big fire season if not overall the biggest, and it may have been especially smoky due to the type of materials the fire consumed.

Bearing in mind when Oz had some huge rains a few years ago it was supposed to have actually temporarily but noticeably reduced the amount of sea level rise, I wonder if Australia is in a particular geographic position to have a specific effect on climate related matters?

We have the same with the UK and CET which whilst not perfect has some sort of reasonably analogy to at least NH temperatures


Robert W. Turner
Reply to  JCalvertN(UK)
April 2, 2020 7:33 am

There are two ways for molecules to enter the stratosphere:
1. They are launched high enough from a volcanic eruption to overcome gravity and boundary layers – only the largest do this.
2. They are lifted there from a very powerful convective current/low pressure system – how most particles get into the stratosphere.

The conditions were just right for #2 apparently. Weather + the convection from the fires created pyrocumulonimbus clouds which reached into the stratosphere. The stratospheric winds sheer off the clouds and carry that material with them.

April 1, 2020 7:21 pm

An easy way to confirm this postulation is to monitor the next Aussie broad scale bushfires in oh, say 10 years time.
Because that’s about all the time it will take for the eucalyptus forests to re-grow to 2019 fuel levels.

And “those who ignore the lessons of history are sentenced to repeating them”

Bill Treuren
April 1, 2020 7:25 pm

Only smoke from fires near to people cause a problem. this may well be the linked but not by that. The fires some years ago 2003?? covered 3X the area but were central Australia could not see them.

Big spike!!

Reply to  Bill Treuren
April 1, 2020 9:32 pm

Much lower fuel load for central Australian fires so nowhere near as intense, smoke would not have risen as high

April 1, 2020 8:00 pm

There have been 21 volcanic eruptions since 2000 with an Volcanic Explosive Index (VEI) exceeding 4+ (Taal was also a VEI 4):

Since none of the other 21st-century VEI 4+ eruptions had any significant affect on lower stratosphere temps, it highly unlikely Taal was the exception.

There is excellent correlation of short-term lower stratospheric warming spikes following VEI 5+ eruptions, especially if the they are located close to the equator.

It’s also highly unlikely the large Australian brush fires caused any significant lower stratosphere warming because large Australian brush fires occur quite often and show little historic correlation.

What’s most likely occurring is we’re entering a 30+ year global cooling phase caused by a Grand Solar Minimum event that just started, and from the PDO and AMO moving towards their respective 30-year ocean cool cycles.

CAGW advocates hate to admit the affects of PDO/AMO warm/cool cycles on global temperature trends as it would mean most of the global warming from 1980 to the present was caused by PDO/AMO warm cycles and not CO2 forcing…

The next 5 years should be very interesting if global temps start to cool while CO2 levels continue to rise…

April 1, 2020 8:02 pm

So, if the stratosphere is warming, does that mean the lower troposphere is cooling?

I had always thought that, if the troposphere was a better insulator because of CO2, the stratosphere would cool as a result. I probably have it wrong but I don’t feel bad because our good buddy Gavin is also a bit confused. link

It seems to me that ‘they’ have to keep rejigging their atmospheric physics so CAGW can continue to work. Maybe I’m being a bit too cynical … maybe yes, maybe no …

Reply to  commieBob
April 2, 2020 2:40 am

This is carboncentric BS.

The long term cooling of the stratosphere is spuriously attributed to CO2 because they are obsessed with attributing everything to CO2.

If we look at the STEP cooling after the two eruptions we see that is is not a steady decline due CO2 blocking outgoing LWIR but a consequence of the aerosols.

A major factor is sulphate aerosols destroy ozone, secondly it seems that probably man made pollution of the stratosphere was flushed out by the natural processes which cleared the aerosols. The lower stratosphere became less opaque.

THAT was the cause of a least part of the late 20th c. warming that got everyone crapping themselves. The CO2 obsession is unscientif BS based on fitting linear “trends” to any and all datasets and then spuriously attributing to the similarly steady rise in CO2.

TLS shown here is the clearest indication this is a lie. We have a clear, noiseless signal which is clearly step like and NOT a linear trend.

The other thing this shows is that the Montreal protocol was based on a similar lie. Ozone was destroyed in two natural events , not a slow rise in CFCs.

J Mac
April 1, 2020 8:52 pm

If there were more occasions of large Aussie bush fires correlating with stratospheric warming, I would find this speculation more credible. But one data point does not establish a trend….

April 1, 2020 10:14 pm

Why do volcanic eruptions only ever occur when the temperature is already rising? Never when it’s trending down.

Izaak Walton
April 1, 2020 10:18 pm

I do find it very odd that the warming appears to only be apparent in the nothern hemisphere (since it is global and not apparent in either the southern hemisphere or the tropics). Plus all of europe experienced an abnormally warm winter. Perhaps the dastardly Russians stole all the smoke and used them to warm sibera
using heat pipes and Antony Watts suggested recently.

Keith Minto
April 1, 2020 10:26 pm

Anything to do with April 1 ?

“global averages show the best signal of something new going on, even if that something new is only occurring in a specific region.”……er, no, signals are regional, that’s what the climate debate is about.

Australia is in the SH, should show a SH signal.

I think we are being tested.

Joel O'Bryan
April 1, 2020 11:02 pm

What is happening in the Troposphere below 10 km?
If the Stratosphere is warming, the troposphere should be cooling. With a delay of about 6 months (seasonal).

April 1, 2020 11:56 pm

Anomalous lack of jet exhaust.

Pasi Autio
April 2, 2020 1:49 am

There is nothing extraordinary in this season’s bushfires in Australia. Quoted 14 Million hectares of burned forests are lower that normal (normal being more than 20 Million hectares per season).

Thus this idea of bushfires causing stratospheric warming is absolutely without merit.

Nothing has changed after I published this text. The season is still below average.

Reply to  Pasi Autio
April 3, 2020 6:39 am


As a resident of eastern Australia for 60 years, I have never seen a fire season like it. Both in the number of wild fires and the intensity. I have also never heard of smoke turning day into night and then travelling thousands of kilometers to dim the skies in New Zealand.

I have been looking at Modis satellite images for the last 14 years and again never witnessed smoke that dense emanating from Australia. I have little doubt that the pyrocumulus clouds seen on these images were pushing smoke into the stratosphere.

This was definitely not a below average fire season.

IMO the main culprit was a very positive Indian Ocean Dipole coupled with lax forest management.
As far as climate change goes, this might be a modest contributor.

Brett Keane
Reply to  dlb
April 3, 2020 10:41 pm

Dlb. Sorry, wrong. NZ gets Aussie smoke almost frequently, with lovely sunsets thank you. Brett Keane

April 2, 2020 2:32 am

Looking at the Australian Bureau of Meteorology data back in the 1860’s almost twice the area burnt this last season was burned in the 1850’s….may we ask where is the corresponding spike in temps which logically should have occurred , if these events were so significant??

Richard M
April 2, 2020 9:02 am

There was also an SSW event in Antarctica last September. Maybe it affected the air currents and drove more gases and particulates into the stratosphere than might otherwise occur.

April 2, 2020 9:50 am

Like Arte Johnson said, veeeery interesting.

Dave Fair
April 2, 2020 11:48 am

It seems a few commentators here are confusing stratospheric warming with tropospheric warming.

April 2, 2020 1:46 pm

Calculate the sum and dispersion (4 years on average for a VEI4) of the eruptions of the years from 2018 to 2020 and you will have the answer.
Fuego, Ambae, Anak Krakatau, Raikoke, Ulauwn, Manam and now Taal

Reply to  Al
April 2, 2020 2:55 pm

The answer to what? Maybe you could try actually saying what you trying to imply , then we can decide whether is it meaningful or just guff. You know, the testable hypothesis thing.

April 2, 2020 9:20 pm

There have been massive bushfires in Australia in the remote North West that burnt out areas the size of European Countries.

Claiming of a correlation with Australian Bushfires looks very weak to any Australian who has any knowledge of our bushfires and not just the media hype of ones that the smoke impacted on big city media.

Reply to  Richard
April 3, 2020 7:03 am

These “massive” bushfires the size of Europe are in the sparsely populated arid and semiarid lands in northern and western Australia. Much of this savanna country burns at the end of the dry season, often due to lightning strikes.

Compared to the recent forest fires in SE Australia, these savanna fires are of low intensity and their smoke would be unlikely to reach the stratosphere.

%d bloggers like this: