Figure 3 displays just such ‘sudden nonlinear transitions’ into and out of the Mediaeval Warm Period and the twin-troughed Little Ice Age.  At this time-scale, there is abundant evidence in similar vein.

deMenocal et al (2000) put it plainly enough:
 Whatever their ultimate cause, these millennial-scale Holocene SST variations  appear to have involved the entire North Atlantic basin, recurred with a ~1500  +/- 500 year period throughout glacial and interglacial intervals, were  accompanied by terrestrial climate changes, and involved large-scale ocean and  atmosphere reorganizations that were completed within decades or centuries,  perhaps less.  These climate perturbations continue to persist during “our time”.   The most recent of these, the LIA, ended in the late 19th century .....
and
 The Hole 658C SST record16 also supports the view that Holocene climate  variability has been increasing in recent millennia, with the LIA representing the  largest-amplitude event of the last 20 ky.

 6.2  Evidence from the past half-century
What of the past 50 years?  Here we have available a large, although still by no means comprehensive, record of physical and biological variations - some of which could well be linked to climate change.

  6.2.1  Who has heard of bioturbation - or the PDO index?
Two weeks before publication of the news and views item by Weaver and Zwiers (see Section 5.6, above), Nature of 21 September 2000 pre-delivered its rebuttal.  The paper by Stott et al (2000) was not supported by an In this issue flag, nor by a news and views explanatory piece.  In the week of its publication, those honours were reserved for a research paper entitled “Reduced calcification of marine plankton in response to increased atmospheric CO2“.

Unheralded it may have been, but nevertheless, the paper by Stott et al says it all.  This paper is also of personal interest, because it talks of that bane of oilfield production geologists - bioturbation.

______________________________________________________________________________________
16.  Corehole 658C samples seabed sediments at about 21 0N 19 0W, in 2300 metres of water off Cap Blanc Mauritania.  The proxy sea surface temperature record through the Little Ice Age, and back to the Last Glacial Maximum at about 20,000 years before the present, was reconstructed from variations in the assemblage of planktonic foraminifera recovered in the cored sediments.

The story goes something like this:
 * In the Santa Barbara, Santa Monica and Gulf of California basins, laminated sediments have spread systematically upward and outward over the past few centuries.
 * This observation is consistent with an increase in the upwelling of cold nutrient-rich water in the eastern Pacific and consequent increased biological productivity in waters off the west coast of North America.
 * A return to bioturbation since the late 1970s implies an increase in oxygen availability, because of a new limitation on the supply (‘rain’) of surface organic material to the seabed.  This change is attributed to a recent reduction in oceanic upwelling, and hence in nutrient availability and consequent surface productivity.

Stott et al find that:
 The temporal record suggests.... that there are factors operating on centennial  timescales that affect bottom-water oxygen levels ..... It is also now clear that  there are shorter-timescale environmental changes that also affect the North  Pacific marine environment, and that these are superimposed on the longer-scale  patterns of variability.  In particular, the mean climate state of the North Pacific  ..... has varied between a warm and a cold phase with a quasi-regular decadal  oscillation.  This decadal pattern has been referred to as the Pacific Decadal  Oscillation (PDO).
and
 The last clear PDO shift was coincident with a 1976-77 El Niño event.  Since that  time, the average sea surface temperatures within the southern California  Current during the months of upwelling (spring/summer) have increased by 1.5-3  0C.  Upwelling along the coast at the latitude of the Santa Monica basin during  the spring has also decreased.  The result of these changes has been a systematic  decline in marine fisheries in the southern California Current, which can be tied  to the reduced upwelling and lower primary productivity.

The lower graph in Figure 16 shows the PDO index (for May) during the upwelling season in the NE Pacific over the past century.

There is a concordance between the two tranches of positive PDO index (ie reduced upwelling of cold, deep water) at 1924-42 and from 1977 to the present, with the two main periods of observed warming at the surface in the Northern Hemisphere during the 20th century (see the upper graph in Figure 16, enlarged from Figure 15).

Particularly the latter (1976/77) change in sign of the PDO index marks a pronounced nonlinearity; and this change coincides with the jump observed in the global record of lower atmosphere temperatures, as portrayed in Figure 7.  These balloon measurements were taken mostly over land; and it appears that reduced upwelling of cold water in the NE Pacific is reflected well beyond the region.

6.2.2  Evidence from equatorial corals
Now, only three weeks after publication of the conceptually-dubious assessment by Allen et al (and its supporting commentary by Weaver and Zwiers) of the accuracy of model-based predictions of future global temperatures, Nature has published compelling evidence by Urban, Cole and Overpeck (2000) which demonstrates yet again the sudden nonlinear transition between climatic regimes in 1976/77.  This new record is from Maiana Atoll in the central Pacific.

The In this issue flag announces the article as follows:
 A new coral-based climate record provides a 155-year-long record of the El  Niño/Southern Oscillation (ENSO) system from the tropical Pacific.  The new  record gives a detailed view of climate change in the region since 1840.  A long- term trend from cooler/drier to warmer/wetter conditions occurs through gradual  change around 1915 and a sudden shift in 1976, and the period since 1976 is the  warmest and wettest on record.  The timescales of ENSO cycles varied even  before human influence was felt, possibly linked to these changes in the mean  climate.

No complaints here.  Nature has identified the prolonged, albeit discontinuous, global warming trend between about 1910 and 1945 (see Figure 1) in terms of its tropical Pacific manifestation.  Furthermore, and of crucial importance to the main thesis of this submission, it also recognises the renewed warming which began abruptly in 1976/77 and continues to the present.

Dunbar here provides the news and views commentary, from the point of view of the climate-modeller, saying:
 One factor missing from these models is decadal-scale climate variability, the  physics and cause of which are still unknown.  For example, during 1976-77  much of the northern and tropical Pacific warmed abruptly and stayed warm for  the next two decades, with large impacts on west-coast fisheries and long-term  average rainfall over parts of North America.
He continues:
 The Maiana record suggests that warming and possible freshening (caused by  greater rainfall) of the central tropical Pacific over the past 150 years was not  steady and incremental, but occurred mostly in two steps - a small shift in the  early twentieth century, and a larger shift around 1976-77.
and
 What does the coral record tell us about ENSO and global warming?  Although it  supports the notion that ENSO timescales respond to changes in the background  ocean temperatures, the simple fact that ENSO cycles as seen in the coral record  varied from three to more than ten years before the 1920s argues against a simple  human influence on modern El Niño variability.  Perhaps more important is the  observation that the warming and freshening of the central Pacific since the late  1970s is unique over the entire 155-year record.
 
Dunbar recognises the point I am making here.  The 1976/77 step-jump in equatorial Pacific sea-surface temperatures is a manifestation of the most striking single event in the record of 20th century climate-change.  He also recognises that the numerical climate models don’t and can’t take such nonlinearities into account.

I come now to the primary letters to nature research paper by Urban, Cole and Overpeck for which the quotes above serve as an introduction.  Figure 17 is reproduced from the Urban et al paper.

The top line in each panel of the Figure is the outcome of the study by these authors of oxygen isotope  variation in corals from an atoll at 1 N 173 E in the Pacific.  This oxygen-isotope record is a proxy for combined temperature/salinity change (and salinity relates to rainfall), and it is necessary to make assumptions if their separate influences are to be identified.

Nevertheless, the message is clear: there was a major climate change in the central Pacific at about 1976/77.  For comparison, the middle line in each panel displays the ENSO Index for the past 50 years; and the lower line is a generalised record of sea-surface temperatures further east (90 to 160 W) in the tropical Pacific for the 1856-1992 period.  Both these latter records contain the step-jump at 1976/77.

Urban et al put it as follows:
 Here we present a 155-year coral record from the central tropical Pacific that  provides new evidence of long-term Pacific climate variability, including  substantial decadal variability in the mid to late 19th century, and a trend  towards warmer, wetter conditions from 1840 to 1995.
and
 ..... this trend occurs in two stages, gradually in the early 20th century, and more  abruptly in 1976, with relatively stationary conditions elsewhere.

The 1976/77 event was particularly prominent at Maiana Atoll, it appears:
 Most of the Maiana coral ?18O trend occurred in the 1976 shift.  Instrumental  records indicate a warming at Maiana of 0.6 degrees C relative to the mean of the  previous 25 years.  This change is about twice the warming seen in the Niño 3.4  region further east .....

Later, the authors provide analysis which is particularly helpful to the theme of this submission.  They begin:
 Previous work has attributed differences between the recent spectral signature of  ENSO (1960-1998) and the average spectral signature (1858-1976) to the  influence of anthropogenic greenhouse forcing on post-1978 variability.   However, our results confirm that the spectrum of ENSO variability changes  substantially over multidecadal timescales before strong anthropogenic forcing.
and, crucially, continue
 The changes in time and frequency domains at Maiana do not correlate linearly  with global radiative forcings (solar, volcanic or greenhouse-gas  concentrations).  However, the Maiana record shift to warmer and wetter  conditions in 1976 coincides with the shift to warmer conditions that appears  clearly in hemispheric and global temperature indices.

Thus, Urban et al provide powerful support for my view that an abrupt re-ordering of oceanic heat transportation in 1976/77 was the main driver of the observed 20th century surface warming.

This extreme event is not the outcome of ‘strong anthropogenic forcing’.  Instead, it is an obvious example of a ‘sudden nonlinear transition between climatic regimes’ of the sort which is described by Weaver and Zwiers (see Section 5.6, above) as being ‘highly improbable’.  Such nonlinearities are specifically excluded from representation within the models used by IPCC to project future climate.

The 1976/77 “shift to warmer conditions that appears clearly in hemispheric and global temperature indices” is a natural not human-caused event.  But its cause is not solar variation or volcanic activity.  This extreme event is outside the ken of IPCC.

  6.2.3 Atlantic storminess and the destruction of meaning
The paper by Grevemeyer, Herber and Essen (2000) is a worry; and one wonders how it survived peer-review.  This exercise in wishful-thinking begins:
 Observations of the Earth’s near-surface temperature show a global increase  since 1901, occurring from 1925-44 and 1978-1997.  Over these periods global  temperature rose by 0.37 and 0.32 K, respectively.  The temperature change over  the past decade is unlikely to be entirely due to internal climate variability and  has been attributed to changes in the concentration of greenhouse gases caused  by human activity.

I see a mind-set here.

These authors relate microseisms recorded at the Hamburg seismological station to wave climate in the NE Atlantic; and they derive an index for the amplitude of secondary microseisms with periods of 6-8 seconds, as plotted in Figure 18(a), as a proxy for changes in the wave climate.  Implausibly, they then propose an explanation for their observations as follows:
 Recent simulations and analyses of the Earth’s temperature pattern exclude  purely natural forcing and attribute it largely to changes in the concentration of  greenhouse gases and aerosol loading due to human activity.  Therefore, it seems  reasonable to propose that greenhouse forcing affects the ocean’s wave climate  and hence coastal surf and storm surges along northern Europe’s coastlines,  which in turn produced the observed increase in microseisms.
 
The source of their problem is obvious - see Figure 18(a).  Quite rightly, they draw a horizontal line representing the 1954-77 years, ie the years leading up to the prominent step in their record.  If they had drawn a second horizontal line through the years following the step, all would have been well.

Doubtless, they would then have recognised the similarly-timed step in global atmospheric temperature shown in Figure 18(b).  Doubtless also, they would have recognised the coincident step in the Enso index and in equatorial Pacific sea-surface temperatures as illustrated in the lower graph of Figure 17.  Furthermore, they surely would have recognised the synchronous jump in the NE Pacific PDO index - see lower graph in Figure 16.

All this evidence in favour of an ocean-related nonlinearity as likely explanation for the change in NE Atlantic wave climate is ignored by Grevemeyer et al.

Instead, in a desperate attempt to sing the greenhouse song, they draw a sloping straight line between the beginning and end of their run of data - the broken line shown in Figure 18(a) - and call it a ‘trend’.

Can this be what we mean, when we speak of the ‘scientific tradition’?

 6.3  The mid-70s climatic event: implicating the oceans
All of the Atlantic, Indian and Pacific oceans show a nonlinearity in the record of heat content of their uppermost 300 metres in the mid 70s, as illustrated in Figure 19(a).  Indeed, the graph for the Atlantic shows a striking resemblance to the proxy for NE Atlantic wave climate in Figure 18(a) - yet more evidence that Grevemeyer et al are blinded by their preconceptions.

Furthermore, a crucial conclusion can now be drawn.  Figure 19(b) compares changes in upper-ocean heat content and lower-atmosphere temperature since the 1950s.  The only significant change in atmospheric temperature in the 43-year (1958-2000) record - the step at 1976/77 - is clearly a response to sudden contemporary changes in the oceans, and not a response to the continued growth in human-caused GHG emissions.

Therefore, greenhouse warming cannot be more than a minor contributor to observed surface warming in the second half of the 20th century.  Most of this warming is natural variability.

There is a second conclusion that flows from the science reviewed in this section.  The events of 1976/77 represent a “sudden nonlinear transition between climatic regimes” such as Weaver and Zwiers warn us would cause conclusions based on climate models to “fall apart” (see Section 5.6).  IPCC’s models cannot predict climate-change in a nonlinear world - such as the real world.

   You read it first here

© 2001  Bob Foster  Posted   9, April, 2001
www.globalwarming-news.com
Back to "Duel of the Hypotheses"  contents page
Back to Guests Page
Back to Front Page