04. Vicious Cycles

A Study in Infra-Red

Part four –Vicious Cycles



What is out of the common is usually a guide rather than a hindrance.” Sherlock Holmes



A number of mechanisms are said to amplify the effect of global warming. These are known as positive feedback cycles. This feedback, as the name implies, requires an initial forcing, natural or otherwise, that precedes the feedback itself.


Let us look at the most relevant ones.


Humidity Cupidity


Positive Water Vapour Feedback


In a warming world, more water evaporates.


A warmer climate increases evaporation. It both sucks moisture from the ground, intensifying drought, and increases atmospheric humidity, which causes more rain to fall during extreme events.” James Hansen in Houston, 2007.


Climate expert sees drought, floods for Texas


As water vapour is the most potent greenhouse gas, this addition causes the temperature to go up even more. This is known as positive feedback.





This assumption is what enabled the IPCC to project an increase in world temperatures ranging up to 6°C as opposed to the 1.2°C that direct effects of CO2 alone would cause.

However, the increased precipitation also resulting from more water vapour in the atmosphere counteracts this. This is negative feedback.



Spencer on water vapor feedback


Spot the Hot spot

The IPCC predicted that a hot spot centred over the equator, where it is more moist, would arise due to warming at the surface (from whatever source) bringing about more evaporation and more moisture in the air. The more moisture in the air, the more latent heat is released. Warming aloft would be greater than warming at the surface leading to the predicted pattern - the hot spot.

The predicted hotspot has been conspicuous by its absence over the long term, as radiosonde data has revealed.


Missing hot spot, Radiosondes, weather balloons, climate models, global warming, water vapor.

Note the increase in temperature from the bottom to the top of the troposphere in the northern latitudes.

The actual pattern that has arisen is what one might expect from the disproportionate warming outlined in part 1.

Fig 1: Global map

This NASA GISS map and graph above show the trend in temperatures annually (Jan-Dec) from the period 1979 - 1999.

Note also from the radiosonde data covering the same period, the uniform and extreme cooling of the stratosphere which one would expect from CO2 and increased cirrus cloud cover trapping heat below, preventing it rising above the troposphere. There is also the additional factor of polar stratospheric clouds destroying ozone, which normally heats the stratosphere by trapping UV radiation.


But where is the water vapour?


Global Humidity


Global warming models predicted absolute humidity would increase with a warming atmosphere. Enough water vapour would evaporate at a rate that would also increase relative humidity despite warming temperatures. Has this been the case?





Global relative humidity, middle and upper atmosphere, from radiosonde data, NOAA Earth System Research Laboratory.


This graph shows that relative humidity has been declining for more than 60 years.


What is relative humidity?


Relative Humidity (RH) is specific measure of the amount of water vapour in the atmosphere expressed as a percentage of the maximum amount that the air could hold at the given temperature.

A warmer atmosphere can hold more water and so reduces relative humidity. More atmospheric water is required to increase relative humidity in a warmer atmosphere.


Absolute humidity is not rising rapidly enough for relative humidity to keep pace with warming temperatures.


In order to remove the noise of a warming atmosphere, let us now look at the trends in Atmospheric or Precipitable Water Vapour, a measure of the total amount of water vapour in the air, since 1988.




We see a decline in upper layer (around 18,000 to 30,000 feet) and middle layer (around 10,000 to 18,000 feet) water vapour  from 1995 to 2001. The near-surface layer (around surface to 10,000 feet) shows an increase in water vapour.





This chart, being a reflection of global atmospheric water vapour levels, reveals that levels in the middle and upper troposphere have decreased in the last 60 years. However, in the lower troposphere (blue line) they have increased. While the previous graph showed that at 700mb, the relative humidity declined, this graph shows that the total water vapour increased at that level.


The factor of a globally warming world leading to a reduction in relative humidity due to the atmosphere being able to hold more water vapour has somewhat masked the increase in water vapour in the lower atmosphere (layer surface to 700mb).


We still have a decline however, in the middle and upper layers in actual water vapour levels in addition to the decline in relative humidity.

So the uneven distribution of water, as a global phenomenon, occurs vertically as well as horizontally.


How do we reconcile this vertical distribution of water vapour, revealing an increase in the lower layers but a decrease in middle and upper layers, with projected models of CO2 induced water vapour feedback?


Rather than attempting to do that, let us look at another, related form of feedback.


Cloud Feedback


In addition to more water vapour and because of it, a warming atmosphere produces more clouds.

There are two competing forms of feedback here:


  • Negative feedback – more low clouds, like cumulus, are produced. Low clouds have an overall cooling influence in the region 60°N to 60°S, this being more pronounced the further one moves towards the equator.

  • Positive feedback – more high clouds, like cirrus, are produced. High clouds have an overall warming influence, this being more pronounced the further one moves towards the poles.




Skeptics of AGW claim that cloud feedback will be negative with more low clouds acting to cool and preserve the planet’s heat balance.

Proponents of AGW, citing more recent studies, claim that cloud feedback has been and will be strongly positive, with fewer low clouds. This is theorised to be due to heat transferred from the surface thinning the lower clouds out.


Recall from part 3 where it was revealed by satellite employing sophisticated Infra-Red instrumentation that higher clouds had increased globally in the region 60°N to 60°S over the period from 1985 to 2001 by 1.95% on average, per decade whilst the lower clouds had decreased by 1.7% per decade. An increase in high, warming clouds and a decrease in low, cooling clouds.





The geographical locations of changes in all-cloud and high-cloud frequency between the first and last 8 yr of this study

(1994–2001 minus 1985–92).


Trends in Global Cloud Cover in Two Decades of HIRS Observations



Assuming this shift in the ratio of lower to higher clouds has continued, the result would be significant warming globally, comparable to CO2.


A potential reason for this reduction in atmospheric water vapour, counter to the expected increase, could be aerosols. Normally, due to the presence of aerosols, it is unusual to find a relative humidity over 102% anywhere in the atmosphere. Aerosols lock up water and reduce both relative and absolute humidity. An increase in aerosol content at higher levels, particularly those that act as efficient ice nuclei, would result in a corresponding increase in high cloud that would extend from lower altitudes than would otherwise be the case. Conversely, a decrease in aerosol content at lower levels would result in a corresponding decrease in lower cloud formation.


This configuration of clouds, given a steady replenishment of aerosols at the upper levels, would remain constant, inhibiting the negative feedback of precipitation, and build up a stronger and more prolonged form of positive feedback than water vapor alone.


Now, the fact that water vapour has increased globally at the lower levels whilst lower clouds have decreased and water vapour has decreased at the upper layers whilst upper clouds have increased, suggests that aerosols are the driving force rather than water vapour alone.


This would make anthropogenically induced clouds a forcing in addition to a feedback.


Convective Cloud Feedback in the Arctic


The cloud configuration of more high clouds and fewer low clouds covering the globe from 60°N to 60°S, whilst accounting well for the vertical distribution of global humidity and an increase in global temperatures comparable to that induced by CO2, cannot by itself account for the uneven warming focused in the northern hemisphere and particularly the Arctic nor the related pattern revealed by radiosondes searching for the elusive hotspot.


However, when we observe the cloud cover over the Arctic itself, 60° - 90° N, a curious correlation emerges. We see that the trend since the 70’s has shown a significant increase in total cloud cover during all seasons with lower clouds appearing to be most responsible for this trend.


Lower clouds, which reflect more sunlight than the heat energy they trap, generally have a cooling influence in the region 60°N to 60°S where they have decreased, contributing to the overall warming. However, the further pole-wards the lower clouds occur, the greater the shift towards net warming until, over the Arctic, lower clouds have a warming effect overall.

This applies throughout the year, except, briefly, during the summer.


“On Greenland, which is covered in bright, light-reflecting snow, clouds primarily act to trap heat.”

Clouds, like blankets, trap heat and are melting the Greenland Ice Sheet



Arctic lower cloud in the region 60° - 90° N, having a warming influence, has increased, contributing significantly to the overall warming. When we take into account the fact that the Antarctic has not seen such an increase in cloud cover nor temperature, we can see how Arctic cloud cover has played a role in the disproportionate global warming biased towards the north pole.


“Overall, relationships between ice, temperature, and clouds indicate that cloud changes

in recent decades may enhance the warming of the Arctic and may be acting to accelerate the decline of Arctic sea ice. Emphasis mine



Interannual Variations of Arctic Cloud Types in Relation to Sea Ice




spatial distribution of trend in cloud cover


Spatial distribution of trends in cloud cover over twenty years. Provided by Axel J. Schweiger.


Climate Indicators - Clouds


It has been proposed that convective cloud feedback over the Arctic has played a role in accelerating Arctic sea ice loss.


Arctic Amplification


Arctic ice albedo loss, a feedback loop feeding on the others, and said to be the most urgent, is the extra heating caused by the melting of the Arctic snow and sea ice, estimated to be equal to 20 years of CO2 warming. This is due to the removal of albedo from the melting ice, leading to non-linear increases in temperature.


“A key factor is the Arctic sea ice, whose reflection of sunshine keeps the planet cool.  Remove the sea ice, and not only does the planet start to overheat, but the whole climate is suddenly changed.  The global weather systems, on whose predictability farmers rely, are dependent for their stability on there being a temperature gradient between tropics and the poles.  Remove the snow and ice at one pole, and the weather systems go awry and we have “global weirding”.

AMEG Strategic Plan

CO2 has been suggested as the initiator of this process. The idea is that CO2-initiated warming leads to sea ice reduction, which leads to increases in heat and moisture from the surface, leading to increased lower convective cloud cover and increased high altitude-moisture levels and presumably cirrus cloud cover, all of which trap heat and result in further sea ice loss - convective cloud feedback.


However, a study has shown that current levels of CO2 (400.16 ppm) are insufficient to have caused large reductions in winter sea ice through convective cloud feedback.

It was estimated that around 1120ppm of CO2 would be required to bring about convective cloud feedback and ocean heat transport feedback necessary for major Arctic sea ice loss.

Evidently something other than elevated CO2 levels must have kick-started the convective cloud feedback warming mechanism and the Arctic ice melt that we are seeing now, not in some 100 years in the future scenario.

Can a Convective Cloud Feedback Help to Eliminate Winter Sea Ice at High CO2 Concentrations?

Note that although Arctic ice loss is accelerating drastically at present, this process did not begin until around 1996.


PIOMAS model Arctic sea ice volume for autumn 1980–2014 (solid line) and spring 1981–2014 (dashed line). CryoSat-2 volume estimates (red stars) are plotted for 2010–2014.

Source: Nature Geoscience; Tilling et al. (2015).

The AMEG group claim that the reason for the weird weather we are now experiencing is the reduction in temperature gradient between the North Pole and the equator. This is, according to them, due to the non-linear albedo effect of the Arctic ice melt. However, it is clear from the NASA GISS map and graph below that the period from 1980 to 1996, when the Arctic ice had not begun its drastic downward decline, is marked by an extreme temperature increase in the Arctic circle around 66° to 90°N and also the mid-latitudes, relative to the equator and the Antarctic.