Posted by: erl happ | May 12, 2009

Climate Change ‘a la naturale’


Good science is plausible. It makes sense. Here is something from the UNIPCC that makes very little sense. It is from section1.4.6 of the most recent report accessible at:

“The mechanisms and predictive skill of ENSO are still under discussion. In particular, it is not clear how ENSO changes with, and perhaps interacts with, a changing climate.”

Has it occurred to those who have endorsed that statement that  ENSO cycles might actually be responsible for observed climate change?  Can we be sure that the thing that forces change on the inter-annual time scale is somehow inoperative on longer time scales? Of course not. We don’t even know what it is.

This post provides evidence that warming cycles in the tropics (ENSO in the Pacific manifestation) are directly responsible for the change that has occurred. The evidence is abundant and convincing.

The data reported here is referenced as:

Kalnay, E. and Coauthors, 1996: The NCEP/NCAR Reanalysis 40-year Project. Bull. Amer. Meteor. Soc., 77, 437-471.

This data set is available via the web at:

Like all data sets in climate studies it is the result of interpolation. Points of measurement  are well spread, sparse in parts, and we must have values for the spaces in-between. But in this study I am not concerned with minutiae. It is the big picture and the grosser type of change that is of interest. Used in this way the data set must be regarded as a truly magnificent resource, entirely suited to purpose.

All data is presented as a twelve month moving average centred on the seventh month so as to eliminate the influence of the seasons.

The first step in diagnosis of any ailment is to examine the patient’s symptoms. We must have answers to questions like: Where are the spots? Where does it hurt? What were you doing when the pain started?

The symptoms and some of the relevant physiology appear to include:

  1. Recent (since 1948) change in sea surface temperature in the tropics relates to change in inter-annual heating cycle minimum, rather than the maximum temperature.
  2. Low minima are associated with weak solar cycles. The sample is limited but this is certainly the case with cycle 20 and cycle 23, post 1998.
  3. After 1978 there is a consistent and unrelieved high level of evaporation from the tropical oceans. This is clearly associated with episodic inflation of temperature in the upper troposphere resulting in loss of cloud albedo allowing more solar energy to enter the ocean.
  4. The suns irradiance is invariable. The flux in cloud albedo and precipitation is the central dynamic driving change in surface temperature on all time schedules. In general an increase in albedo and precipitation accompanies warming of the tropical ocean but in extreme warming events, precipitation and albedo are slow to increase enabling fast surface temperature increase. This  occurs rarely. It manifested in 1972, 1983, 1997 and 2003. These are instances of gross failure of the Earth system to protect itself from rapid heating.
  5. Recently the Earth system has become much more successful in excluding energy from the sun. This is reflected in a decline of 100hpa temperature after the massive increase between 1976 and 1984 and return to situation where heating events are less frequent and less intense .
  6. Climate change is associated with change in energy accumulation in tropical waters and consequent strong change in high latitude sea surface temperature in winter. This is the signature of natural climate change in the tropics due to changing cloud cover. The period of the most recent of these cycles seems to be about thirty years.
  7. The tropics have cooled since 2005.  Sea surface temperature in both hemispheres has fallen, most particularly at high latitudes, in winter. There are reasons to expect this trend to continue.
  8. There is no discernible signature of greenhouse type warming in the tropical atmosphere where outgoing radiation is heaviest.The upper troposphere has been cooling since the mid 1980s. Warming in the lower troposphere is associated with the release of  latent heat at low cloud level, due to enhanced evaporation, due in turn to reduced cloud cover.


Tropical sea surface temperature since 1948

Figure 1 Sea surface temperature in the tropics

Figure 1. Sea surface temperature in the tropics

The fitted trend line in figure 1 exhibits an increase of 0.31°C over the sixty year period. If one assumes that the cause of the increase is monotonic and relentless, and fluctuations about the trend line are nothing more than meaningless ‘noise’ this may be an appropriate way to summarize the data. But, that assumption is insupportable.

Looking closely we note that the top temperature in the warming cycles has hardly increased at all. This suggests that the tropical ocean is near temperature saturated. In the tropics, water temperature seems limited to about 28-30°C just as the temperature of water in a kettle is limited to 100°C regardless of the energy applied. So, the most intense warming episodes of 1998 and 2006 are insignificantly warmer at their peak than the episodes of 1958 and 1970.  Manifestly it is the increase in the minimum in sea surface temperature that is responsible for the upward trend in figure 1. In 1978 alone, the minimum advanced by 0.3°C. However this degree of advance relates to the minimum of 1975-6. There is in fact very little advance on the minimum of 1948-50 until 1993.

The weak solar cycle 20 ran from August 1964 to March 1976. It is associated with the low minima of 1964, 72 and 76. Some may remember the global cooling scare of that time. Perhaps a weak solar cycle 24 might well return the tropics to the minima of 1948 or perhaps even the very low minima of the seventies. Viewed in these terms, the notion that change should be described in terms of a trend rate of increase can only be seen as naive. It is too early to say whether the increase in minima is permanent.

What actually lies behind the increase in the minimum sea surface temperature in the tropics?  We don’t have to look far. The answer is to be found in comparing the temperature at sea surface to that at low cloud level.

Temperature at sea surface and 850hPa between 20°N and 20°S latitude

Figure 2. Temperature at sea surface and 850hPa between 20°N and 20°S latitude

The temperature at 850hPa is 10°C less than the surface. Knowing the differential  and  the wet adiabatic lapse rate of 10°C per 1000 metres we see that the 850hPa pressure level is one kilometer in elevation.  So, it is quite close to the surface. Temperature at 850hPa reflects latent heat release as clouds form. So the changing temperature at 850hPa reflects the changing flux of solar energy into the ocean, the energy that evaporates water. Change in wind strength and relative humidity, both capable of affecting the rate of evaporation, should not materially affect the evaporation rate in this environment because the prevailing wind is light and humidity near saturation level. So, in the tropics temperature at 850hPa, broadly relates to energy received from sunlight. There is one further proviso. The temperature at 850hPa does not reflect all of the energy received because some is lost via transfer of tropical water to higher latitudes. Waters move from east to west, driven by the trade winds and are ultimately diverted pole-wards by the land masses. This is a leaky system.

The temperature increase at 850hPa is more than at the surface. In 1978 we observe an abrupt increase of 0.75°C in the minimum, more than double that at the surface. Minima vary little over the period 1948 to 1976 and once established at the higher level, are invariable.  The maxima in the warming events starting in 1969 and 1973 stand out as exceptional. These warming events are significant precursors to the step change in temperature that occurred in 1978.  So, the jump in temperatures at 850hPa in 1978, both minima and maxima, represents a step up in the energy supply to the tropical ocean. More sunlight.

After 1978 there is a consistent and unrelieved high level of evaporation from the tropical oceans.  This is particularly true of the period after 2002. The pertinent question is: Can the minimum temperature at 850hPa fall to the level that was reached most recently in the La Nina of 2000 and is it likely to do so some time soon? Common sense dictates that it will do so only if the flow of energy into the tropical oceans is reduced. We see evidence of this occurring in the prolonged La Nina’s of 2007-9. Perhaps the La Nina will renew after its pause in May 2009 and we will know sooner rather than later.

Temperature in the atmosphere in the tropics

The slice of the atmosphere between the equator and 10°S is where outgoing long wave radiation is greatest. It is therefore the place where the much touted greenhouse effect should manifest most strongly.  The outstanding change in sea surface temperature in the tropics is obviously the step change in 1978. It manifests strongly between 40°S latitude and 20°N latitude (see figure 7). If we see the thumbprint of change in 1978 in the atmosphere, we may have a clue about the origin of the recent surface warming.

Temperature between 700hPa and 100hPa between the Equator and 10°S latitude

Fig. 3 Temperature between 700hPa and 100hPa between the Equator and 10°S latitude

The three diagrams of figure 3 encompass an increasing vertical range from left to right.  The range is 20°C on the left, 35°C in the middle and 40°C on the right. The amplitude of temperature variation at 200hPa is actually double that at the surface. A true comparison is provided in figure 5. Plainly, upper troposphere temperature is driven in part by change in surface temperature but in larger part by some other factor.

The 1978 thumbprint is least noticeable at 500hPa and 400hPa. Upwards of 300hpa it becomes more noticeable than it is at the surface, increasing evident at higher elevations.

The 1978 climate shift is nowhere more obvious than at 100hPa. The temperature increase at 100hPa in 1978 is about 5°C. The average annual range of variation at 100hPa (between minima in May and maximum in August) is 4°C, and this variation relates to changing ozone excitation by the seasonal increase in outgoing long wave radiation in northern summer. The peaks in 100hPa temperature between 1978 and 1992 are logically due to either an increase in ozone content or an increase in long wave radiation emission from the Earth system, or a mixture of both. A gradual collapse of the Antarctic vortex from 1948 through to 1978 contributed to an ozone increase resulting in increased temperature in the stratosphere throughout the southern hemisphere. But the stratospheric temperature increase associated with that influence peaked in 1978, and has declined gradually ever since that time. So, the gradual reduction in 100hPa temperature after 1978 and after the secondary peak in 1991 is due to a slow decline in long wave emission and also ozone content. After a period of vulnerability that manifests strongly in 1978 the Earth system has become progressively more successful in excluding energy from the sun.

Energy flows

The relationship between outgoing long wave radiation (OLR) and tropical sea surface temperature (SST) is intriguing, on first sight even baffling. However, three modes of operation are discernible in figure 4.

Figure 4. Sea surface temperature and outgoing long wave radiation 0-10°S

Figure 4. Sea surface temperature and outgoing long wave radiation 0-10°S

  1. There is a sudden fall in OLR during the El Nino warming events (red rectangles) when the steepest rise in SST is experienced. For this to occur, precipitation (and therefore latent heat release that drives OLR and the cloud cover that would exclude sunlight) fails to increase as the ocean warms.
  2. In more moderate warming events, sea surface temperature and OLR simply increase together. In this instance an increase in precipitation occurs, releasing latent heat and promoting cloud cover, tending to slow the rise in temperature. This is the Earths thermostat working. Temperature gain is limited by condensation phenomena.
  3. Elsewhere (blue rectangles) we see periods where OLR increases while SST actually falls. In this instance an increase in precipitation is accompanied by increased cloud cover (what could be more natural when the atmosphere cools) excluding energy from the Earth system causing SST to fall.

These are fundamentally different modes.

Let’s look more closely at the first mode. During the strongest heating events, OLR falls as the ocean absorbs energy causing SST to rise strongly. The energy from the sun is relatively invariable. If the Earths albedo (via atmospheric cloud) changes to admit more energy, it will be absorbed by the ocean, causing surface temperature to rise. OLR fundamentally depends upon the rate of evaporation and latent heat release. The atmosphere warms fast, evaporation from the ocean lags behind and relative humidity falls. As a result, less cloud forms. More energy then reaches the ocean. But in the end evaporation must catch up or all the ocean would long since have completely evaporated. The limit to this process is the capacity of the atmosphere to hold water vapour without precipitating it. The evidence of past events suggests that this mode is limited to about a year. There is no better illustration than the El Nino of 1997-8. Such an event is followed by an orgy of precipitation as surely as night follows day but there is no gurantee that all the energy acquired by the ocean in the preceding heating event will be vented.

In mode two, encompassing the more moderate heating events, the energy gained from loss of albedo is partitioned to both increased sea surface temperature and outgoing long wave radiation via an increase in precipitation. This slows the increase in surface temperature. This process worked throughout the period of the massive increase in sea surface temperature centered on 1978. Even when the thermostat is working well a massive step change in ocean temperature is possible.

In mode three the Earth increases its radiation levels via release of latent heat of condensation drawing energy from the ocean via evaporation, yielding a reduction in sea surface temperature. For sea surface temperature to actually fall, the system must increase cloud albedo to exclude solar radiation. This might be described as an internally generated precipitation event. But, such an event has limits. When the atmosphere is sufficiently well dried and the source of moisture is cut off by declining surface temperature, cooling must falter. The best example of this occurrence is the La Nina of the year 2000.

The flux in cloud albedo and precipitation is therefore the dynamic driving change in surface temperature. The relationship between albedo and release of latent heat changes over time. The Earth admits energy at some times building surface temperature and at others it very capably sheds energy from the system. This dynamic is not the cause of these events because it has no autonomous existence, no pendulum action driven by the system itself. The infrequent nature of severe heating events, and the step changes tell us that. But, is this the dynamic behind the gain in temperature that began in 1978? Plainly yes, every solar cycle is marked by a different value of the Southern Oscillation Index. The SOI value is unrelated to sunspot numbers. It rises and falls on multi-decadal time scales following a pattern of changing El Nino or La Nina dominance. The SOI simply tracks the air pressure relationship that drives the trade winds across the Pacific. When the trades weaken, the ocean warms. The mechanism I have described elsewhere. It is the action of the sun changing the strength of the polar vortex which determines the ozone content and the reactivity of the upper troposphere to ultra violet B. This dance is called the QBO, long associated with change in temperature at the surface, the upper troposphere and in the stratosphere, according to today’s climate science driven by Kelvin Waves that are in turn driven by tropical convection. But that is putting the cart before the horse in an argument of hopeless circularity. It is the chosen viewpoint of those who view the Earth as an overpopulated spaceship hurtling towards disaster. These are men who would be gods.

Change in cloud albedo

The 200hPa pressure level is broadly representative of the upper troposphere between 400hPa and the tropopause at 100hPa. Figure 4 shows the relationship between temperature at the surface and  the upper troposphere. It is apparent that change in 200hPa temperature will change ice cloud density in the upper troposphere. This change affects the flux of solar energy into the ocean. A great disparity between 200hPa temperature and sea surface temperature was established between 1976 and 1983. That disparity has diminished over time. In the year 2000, and again in 2008-9 we see 200hpa temperature much more favourable to extensive ice cloud generation across tropical latitudes. From figure 4 we note a massive increase in OLR dating from about 2003. In 2008, the clear signature of energy exclusion appears with OLR increasing as surface temperature falls. The Earth is in energy exclusion mode. It cools.

Figure 5 Temperature at 200hPa compared to the surface both 0-10°S lat.

Figure 5 Temperature at 200hPa compared to the surface both 0-10°S lat.

The driver of 200hPa temperature

At this point let me simply point out that 200hPa temperature rises with ozone content in the upper troposphere. When the upper stratosphere begins to cool, convection there is weakened and ozone flows downwards from the zone of it’s creation into the lower stratosphere and upper troposphere between 30°N and 30°S. At this point the ocean begins to warm. I shall deal with the question of why the middle stratosphere periodically cools next post. At this point let me remark that it does so over a rotation of 27.1 months on the average. This is called the Quasi Biennial Oscillation.

Lessons from the past

Change in tropical sea surface temperature is the driver of climate change outside the tropics, particularly in winter. The oceans that comprise 70% of the surface of the Earth take up energy in the tropical portion and this energy raises water temperature at high latitudes in winter when the sun does not shine.

The variation in summer temperature by comparison with winter temperature in the northern hemisphere is shown in figures 5 and 6. The vertical scales are identical. Notice the much greater variability in winter temperature. The tropics are near temperature saturation. The flux in heating is expressed where the temperature is least. The decline in winter temperature from about 2005 signifies a cooling globe. That cooling is a direct result of sunlight exclusion  in the tropics.

Figure 6. Northern Hemisphere temperature in summer and winter

Figure 6. Northern Hemisphere temperature in summer and winter


The details of the temperature record by latitude

Now, let’s briefly examine temperature change by latitude in each hemisphere starting with the northern hemisphere. This is where the greenhouse argument begins to look really silly.

Figure 7 Sea surface temperature by latitude in the northern hemisphere

Figure 7 Sea surface temperature by latitude in the northern hemisphere

Temperature change in the northern hemisphere

  1. The Arctic is much warmer than Antarctica. Arctic temperature equates to temperature at about 70°south. In summer, when temperature rise above zero ice must melt. Polar bears must swim.
  2. There has been no change in the heating cycle peak temperature between 10° and 60°N between 1948 and 2009.
  3. Between 10°N and 70°N there is evidence of cooling up to 1978 and a simple return to 1948 temperature levels by 2009.
  4. There is an increase in SST of about 2°C at the highest latitudes especially north of 70°N starting in 1978. Tropical warming is associated with Arctic warming via oceanic energy transfer.
  5. Solar Cycle 20 cooling prior to 1978 shows up in the depth of minima reached at all latitudes between the Equator and 60°N. This is the result of change in energy intake in tropical waters.
  6. Warmer temperature at 80-90°N latitude during solar cycle 20 (1964 to 1976) reflects the impact of reduced tropical convection on Arctic temperature via weakening of the Brewer Dobson circulation and the Arctic vortex that brings very cold air to the surface from the mesosphere and upper stratosphere. So, tropical cooling results in polar warming via atmospheric effects. This acts in opposition to the ocean warming mechanism. Notice how jagged is the temperature line closest to the pole.
Fig. 8 Sea surface temperature by latitude in the Southern Hemsiphere

Fig. 8 Sea surface temperature by latitude in the Southern Hemsiphere

Temperature change in the southern hemsiphere

  1. Cooling at 80-90S following the 1978 climate shift represents the effect of enhanced tropical convection on Antarctic temperature via the Brewer Dobson circulation increasing the strength of the Antarctic vortex. Tropical warming causes Antarctic cooling. THe air over Antarctica is warmer than the frozen surface. The Penguins get to walk further and swim less in summer time because there is more ice. The sensible ones camp closer to the ocean.
  2. The temperature gain after 1978 is slightly smaller at high latitudes in Antarctica than in the Arctic.
  3. There is no temperature gain after 1978 between 40-70°S latitude.
  4. There is a strong increase in the heating cycle minima at 30-40S latitude after 1978.
  5. There is a collapse in the range of variation of temperature after 1978 in the 20-30°S latitude band.

Seasonal influences

There is a big difference in temperature gain between the summer and winter in high latitudes in both hemispheres.

Figure 10 Summer and winter temperature at high latitudes

Figure 10 Summer and winter temperature at high latitudes

In the northern hemisphere slight warming in Arctic summer since 2000 reflects the effect of equatorial cooling on the Brewer Dobson circulation and the Arctic vortex.

The step change to a higher range of temperature in winter in the Arctic after 1978 reflects heat gain in the tropical ocean and the increased flow of energy to the poles via ocean currents.

In Antarctica, the slight summer cooling  after 1978 reflects a stronger Brewer Dobson circulation due to enhanced convection in the tropics. This is an atmospheric effect. The atmospheric effect competes with an ocean circulation effect that raises temperature in winter. Antarctic is about 2.5°C warmer in winter in response to these competing influences.

The change in sea surface temperature since 1978 is characterized by a small increase in annual average temperature in the tropics and a marked increase in winter temperature at high latitudes. Mid latitude temperature remains unchanged. Increased warmth in frigid climates must be regarded as beneficial because of its effect on length of growing season. All species benefit.

One would expect any increase in temperature due to greenhouse forcing to be more uniformly distributed and to manifest in all seasons. The change in the composition of the atmosphere can not account for the observed pattern of temperature change in the ocean. The observed pattern of temperature change is consistent with a changing flux of solar energy into the tropical ocean. The change of average global temperature is due to ENSO.

The crusade aimed at controlling carbon dioxide emission is based on an erroneous diagnosis of the cause of ‘the problem’. But really, there is no problem. We can all relax.



  1. I recently examined the Pacific tropical SST 20N-20S and found a strong correlation with global temperatures:
    Your Figure 5 and comments related to that are interesting.

    • Alan,
      Thanks for your comment. I am aware of the material that you pointed out via the link and in fact I passed that address on to Carl Wolk last week. I think your observations are right on the money. If the IPCC agrees that tropical temperature trends are related to natural variation rather than Anthropogenic causes and the energy gain in the tropics explains global temperature variation, we and they have all the knowledge needed to throw out the AGW hypothesis. But are they? No, they are not really interested in the science of the matter at all. They are interested in careers and money.

      Unfortunately this matter will not be resolved in the peer reviewed journals. Nor will it be resolved in academia or by the major government funded institutions like Australia’s CSIRO or the BOM. We have got ourselves into a pretty pass.

      What you are doing with your site is a first class effort. And it is absolutely vital.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: