It’s Only Natural

15 January 2021
It’s Only Natural - Featured image

This article from the Summer 2020 edition of the IPA Review is written by German chemistry professor and environmentalist, Fritz Vahrenholt.

Sometimes you have to go up in the air to better appreciate the big picture. From the helicopter perspective, many patterns and connections become clearer. This also applies to the climate system: 2,000 years ago, it was warm (Roman Warm Period), then it cooled down 1,500 years ago (cold period of the Migration Period), 1,000 years ago it was warm again (Medieval Warm Period), 500 years ago it was cold again, and today it is warm again. Can you see a pattern? The question should be rhetorical. In the current climate debate, however, a great many participants in the discussion seem to have problems recognising the regularity.

If modern global warming were essentially a continuation of a natural cycle—today, however, intensified by greenhouse gases—the situation and climate hazard would be completely different. Every thousand years, a warm phase; after five hundred years, a cold phase. If such a cycle were to occur, even if it were ‘only’ regionally pronounced, future generations would be better prepared.

The important news first: there is indeed a millennium cycle in the climate system that goes beyond the last 2,000 years and has been active for the past 10 millennia. A research group led by Gerard Bond of the Lamont-Doherty Earth Observatory in Palisades, New York, discovered it in 2001. The scientists had examined sediment cores from the North Atlantic and found layers of iceberg debris in them, which repeated themselves quasi-cyclically every 1,000-1,500 years. At these times, the icebergs made their way particularly far south to the core locations under investigation west of the British Isles. At other times, much further north, they were already melting and dropping their debris prematurely.

With the help of the iceberg debris layers, Gerard Bond and his colleagues were able to prove the existence of nine prominent cold phases in the last 12,000 years. The youngest cold phase corresponded to the Little Ice Age (1500 AD), the second youngest to the cold period of the Migration Period (500 AD). A 2003 study of Alaska provided a nice confirmation of the North Atlantic results. As if a metronome had set the pace, phases of abundant and sparse diatom growth alternated periodically throughout maritime history, with a cycle length of 1,000 years. In December 2003, Bond was awarded the Ewing Medal of the American Geophysical Union (AGU) for his work.


The Holocene ‘bond cycles’ have become an important reference, and the pioneering paper published in the prestigious journal Science 2001 has been cited 3,000 times in articles by other research groups. The cycles have been confirmed in many studies from all seven continents, in Europe, Africa, Asia, Oceania, North America, South America, and Antarctica.

IPCC reports downplay or ignore the role of millennium cycles.

The Millennium Cycle is not a new phenomenon that suddenly appeared out of nowhere 12,000 years ago. On the contrary, even during the last ice age (115,000-12,000 years before today) there were already pronounced climate cycles with a period of 1,500 years. The cycles took a very characteristic course. The warming phase was always very abrupt, with the annual average temperature rising by 6 to 10°C within only a few decades. The subsequent cooling phase, however, lasted many centuries. This sawtooth pattern of the temperature curve is the hallmark of these fluctuations. Some cycles are stronger, others weaker. The individual cycles, in turn, are grouped into bundles with longer-term cycles of a uniform trend. The cycles are presumably controlled by changes in the Gulf Stream. To what extent the strong millennium cycles of the last ice age are related to the weaker cycles of the last 12 millennia is still unclear. In any case, that the climate system is apparently in natural oscillation in a rhythm of 1,000-2,000 years is remarkable. It is unlikely that humans could have suddenly brought this long-periodic pulse of the climate system to a standstill.

Sediment Core Samples at Lamont-Doherty Earth Observatory. Source: K Kostel

Sediment Core Samples at Lamont-Doherty Earth Observatory.
Source: K Kostel


How does the Intergovernmental Panel on Climate Change (IPCC) deal with the climatic millennium cycles of the last 10,000 years, which could potentially become a ‘game changer’ once their functioning and driving forces are better understood? The IPCC was founded in 1988, more than a decade before the discovery of the millennium cycles. The first three IPCC climate status reports could not even have anticipated this significant phenomenon when they were written. The 4th IPCC report of 2007 deals with the cycle in a subchapter, but quotes critics at the same time, so in the end no great importance is attached to the phenomenon. Attention in the 5th IPCC report of 2013 to millennium cycles is even shorter. In the chapter ‘Palaeoclimate’ the topic is completely omitted, while in the chapter ‘Clouds and Aerosols’ there is only a scant reference to the Bond paper. Publications of other research groups on the millennium cycles are also largely searched in vain in the 5th IPCC report. It can be assumed the 6th IPCC report, which will be published in 2021, will also exclude the topic.

The IPCC’s notorious lack of interest in this topic—which is currently being actively researched by a large number of climate scientists—leaves one perplexed. Would it not be of great importance to thoroughly investigate this long-rhythmic heartbeat of the climate system in order to integrate it into climate models or, after careful consideration, discard it? The silence of the IPCC on this topic could also be due to the fact that the issue is too uncomfortable to be discussed ‘in a big round’ in climate reports in a transparent and open-ended manner. Could it be that the decision was made too early and there is simply no room for further climate mechanisms? The IPCC special report on the 1.5-degree target for 2018 now assumes a 100 per cent anthropogenic contribution to global warming over the last 150 years. Clearly, the natural millennium climate cycle would challenge the IPCC view that there is only one cause. More case studies on the bond cycles are still being published every month, but they have no chance of being considered by the IPCC.

Why not investigate the climate system’s heartbeat?

A basic problem in research on climatic millennium cycles is the often-limited data density of the studies and uncertainties in radiocarbon dating. When studying a climate development over 10,000 years and a resolution of two data points per century, 200 samples from a sediment core or another climate archive would have to be examined. Due to the usually low-resolution data, only long-term trends can usually be determined, whereas higher-frequency climate changes often remain undetected. Even if high-resolution data series have been generated, the temporal dating of the identified climatic events is often only possible with plus/minus 100 years of inaccuracy. If the peak of a warm phase is only 200 years, the same warm phase can appear in different studies as two different warm events, offset by up to 200 years, simply because of age uncertainty. A purely mathematical-statistical summary of the individual data will therefore always lead to a temporal ‘smearing’ of the respective heat event and an underestimation of the real heat level. The climate archives therefore show us a rather blurred picture in the past with fewer ups and downs, fewer temperature increases and decreases. For this reason, the current warming—drawing on modern temperature measurements—appears as a unique and unprecedented development.


There is no question that most of our planet has warmed up in the last 140 years. However, in some places temperatures have remained stable or even decreased. These include a huge marine area in the North Atlantic Ocean south of Greenland and Iceland, covering some two million square kilometres, which cooled significantly during the industrial phase. On climate maps, the region appears as a large bruise, which is why experts refer to it as a “warming hole”. Many regions of the Southern Ocean around Antarctica have also cooled down. No statistically significant warming trend has been observed in East Antarctica—which makes up the largest part of Antarctica—and on the Antarctic Peninsula (see also the IPA’s Fact Sheet #8, No Evidence of Warming at Mawson Antarctica,

Large parts of the North Pacific also have remained without warming over the past 140 years. Another significant warming hole is in the south-east of the United States, where the climate has cooled in several states over the last 100 years. Temperatures here oscillated between colder and warmer phases, still not significantly exceeding the warmth of the 1930s and 1940s.

Looking only at the last three to five decades, the cooling in the North Atlantic, Southern Ocean and North Pacific are particularly striking. Temperatures in East Antarctica have hardly changed since the 1950s, and in some cases areas have even cooled in recent decades.

Leopard Seal lazing in Antarctica. Source: Wikimedia/Godot13

Leopard Seal lazing in Antarctica.
Source: Wikimedia/Godot13

“No statistically significant warming trend has been observed in East Antarctica.”
Fritz Vahrenholt


In the first decade-and-a-half of the new millennium, global warming came to a virtual standstill. The average global temperature stagnated, which is why the period 2000-2014 is also known as an ‘hiatus’. Numerous research groups confirmed and analysed this unexpected phenomenon. None of the current climate models predicted the warming pause. In a 2013 interview the Hamburg climate modeller Jochem Marotzke called the hiatus “an unusual, surprising and not yet understood phenomenon”.

More than a dozen hypotheses were put forward, but there was also a minority of scientists who did not want to acknowledge the lack of warming and actively wrote against it. With the help of subsequent changes in temperature data, alternative regional averaging and clever selection of temporal start and end points, they tried to show the warming was continuing after all. Respected climate researchers resisted these attempts and called on their colleagues to recognise the hiatus. Ultimately, whether the warming has come to a complete standstill (‘hiatus’) or has merely been slowed down considerably (‘slow down’) does not matter. The decisive factor is that the climate models expected a warming of 0.2°C per decade, which did not occur.

Remarkably, politicians prevailed over science.

The ‘hiatus’ posed a major challenge for the authors of the 5th IPCC Climate Status Report. For some inexplicable reason, the term does not appear in the summary for policymakers. According to Der Spiegel there were at the time tough negotiations as delegates of the German Federal Government in particular—against the resistance of many researchers—apparently wanted to keep the warming of the previous 15 years out of the summary of the new IPCC report. At best, a ‘slowing down of the temperature rise’ was to be admitted, which was ‘scientifically irrelevant’. In the end the politicians prevailed, which is remarkable for a scientific report.

In mid-2014, a strong El Niño set in the Pacific region, the strongest in the entire measuring history. El Niño is an abrupt oceanographic-meteorological change in the currents in the equatorial Pacific, which occurs every 2-7 years and usually causes the global mean temperature to shoot up. This is exactly what happened this time. El Niño 2014-16 ended the global warming pause and caused temperatures to rise again. The peak of the extraordinary heat surge was reached in 2016. Since then, temperatures have been falling again, but at a higher level than during the hiatus phase at the beginning of the millennium.

What exactly triggered the hiatus continues to be the subject of controversial debate. The most probable explanation is the temporary slip of the Pacific Decadal Oscillation (PDO) ocean cycle into the negative phase, which had already led to pauses in warming and cooling in the past. Scientists from the Hamburg Max-Planck-Institute for Meteorology gave the public little hope in 2017 that the mechanisms could ever be clarified.


Time and again we hear in the media that the climate is warming up faster than ever before. What is the truth of this assertion? We are trying to build a bridge between the modern and pre-industrial climate change. The global temperature has risen by about 1.0°C over the last 150 years, which corresponds to an average warming rate of 0.07°C per decade. However, the warming was mainly concentrated in three temperature spurts: from 1860 to 1880, 1910 to 1940, and 1975 to 1998. The rate of temperature increase of the three episodes was similar and was about 0.15°C per decade. Between the warming phases, the climate cooled down slightly or stagnated.

Have there really been no phases in the last few millennia in which similar global warming rates have been achieved? Since the search horizon is in the pre-industrial era, we unfortunately do not have any direct measurement data and therefore have to rely on ‘proxy data’ from geological reconstructions based on sediment cores, cave dripstones and other natural climate archives. The biggest problem: the pre-industrial data are not available from everywhere and are not always easy to interpret. On closer examination, it quickly becomes clear that the above-mentioned assertion is difficult to substantiate. In general, geologically determined warming rates are underestimated due to a variety of problems. The paleothermometer methods used do not work without errors, which means some temperature change remains undocumented. Also, the sampling density and thus data resolution cannot be increased arbitrarily with many methods, which leads to data gaps. The radiocarbon method is subject to systematic errors which, in extreme cases, can cover the entire period of the industrial phase of 100-150 years. The warming spurts therefore appear with a time lag, which artificially reduces the statistically determined warming rate.

The incomparability of modern temperature measurement data with geologically determined temperature-proxy data has already become clear in connection with a global Holocene temperature curve published in 2013. The web platform Klimaretter reported at the time on the study:

Global temperatures are currently rising faster than ever before since the end of the last ice age. This is the result of a study in which a team of scientists around the geologist Shaun Marcott from the Oregon State University in Corvallis reconstructed the temperatures of the past 11,300 years.

However, shortly after the publication of the work, a controversial debate developed, as the low-resolution geological temperature curve was supplemented by yearly measurement data from the last 150 years, which shot up like a rocket at the end of the temperature series. A common presentation of the two data types was, however, not allowed at all, as data of different resolution are not statistically comparable. Rapid pre-industrial warming phases would not appear at all in the data set due to the lack of resolution of only 120 years. This was obviously overlooked in the review of the article in Science. Incidentally, the manuscript had previously been submitted to Nature, where it was probably rejected.

Interestingly, the modern measurement data were still missing in Marcott’s doctoral thesis from 2011, which formed the basis of the young researcher’s Science publication. Why the modern values were finally added in the publication remains unclear.

Did Marcott’s doctoral supervisor Peter Clark’s involvement as lead author of the 5th Climate Status Report and his interest in drawing the most dramatic conclusions possible play a role? In any case, the Science paper was already cited in the IPCC report in September 2013, although the Marcott paper was not officially published until 8 March 2013. If the paper had been published only one week later, it would not have been used in the report, as the so-called ‘literature cut-off’ would have been exceeded on 15 March 2013.

On 31 March 2013, three weeks after the paper was published, Shaun Marcott and his co-authors finally admitted in a blog article at that the temperatures of the 20th century in the published curve are not statistically robust.


The Heidelberg paleoclimatologist Augusto Mangini affirmed in a 2007 FAZ article there was intensive climate change even in pre-industrial times:

[It is] wrong to claim that the current warming is happening much faster than earlier warmings. The fact is that over the last ten thousand years there have been significant global and, above all, equally rapid climate changes that have had a very strong impact on human beings.

As global pre-industrial temperature curves for the last 2,000 and 10,000 years are not yet reliable, we can initially only rely on individual reconstructions. The first example is the warming in the transition from the cold period of the migration period to the medieval warm period. In the Eifel mountain range in western Germany and eastern Belgium, temperatures rose by 4°C within 400 years, which corresponds to a rate of 1°C per century—almost exactly the same rate as today. However, this only applies to the smoothed temperature development. If the unfiltered original data is used, temperature jumps of up to 7°C within a few decades can be detected.

Today’s rate of warming is not unique.

The situation is similar in the Spannagel Cave in the Austrian Central Alps. There, temperatures rose by 2°C between 750 and 850 A.D. Rapid warming phases also occurred on other continents at that time. In the lakes of the East African Rift Valley, for example, temperatures rose by 1.5°C within only 100 years. A Chilean lake in Patagonia even warmed by 4°C within a century. There are many other examples from all seven continents.


The study of pre-industrial phases of ‘abrupt climate change’ even constitutes a sub-discipline of its own in paleoclimatology. Strong temperature fluctuations occurred, for example, during the last 10,000 years in the course of the climatic millennium cycles. In the high-resolution Vostok ice core from the East Antarctic, for example, a large number of strong warming and cooling phases have been documented.

During the last ice age the climate made particularly strong and rapid temperature jumps. Within the framework of a 1500-year cycle, temperatures rose by 6 to 10°C within only a few decades. The subsequent cooling phases then dragged on for many centuries. The sawtooth pattern in the temperature curve is the hallmark of these fluctuations, which were named the Dansgard and Oeschger cycles after their Danish and Swiss discoverers. Two dozen of these events have been recorded in the Atlantic. Some of the cycles are stronger, others weaker. The individual cycles, in turn, are grouped into bundles with longer-term cycles of a uniform trend.

Today’s rate of warming is therefore by no means as unique as often claimed, neither on the scale of the last 1,000 years nor in the context of the last 100,000 years. It is only through geological inaccuracies in the proxy data that the pre-industrial warming phases are “washed out” in the global temperature reconstructions synthesised from many individual data.

Before any robust statements can be made here, hundreds of new pre-industrial data series would have to be generated. At the moment, large parts of the interior of Africa and South America are virtually unexplored from a palaeoclimate perspective. The uncomfortable truth is that we still do not have global pre-industrial temperature curves that would be suitable for high-resolution detailed comparisons.

Fritz Vahrenholt

Fritz Vahrenholt is a German chemistry professor, former energy industry executive, politician and pioneering environmentalist.

Support the IPA

If you liked what you read, consider supporting the IPA. We are entirely funded by individual supporters like you. You can become an IPA member and/or make a tax-deductible donation.