Solar variability and the Earth’s climate

By Andy May

According to Javier and the IPCC total solar radiation output varies little, less than 0.1%. This is only 0.7 to 1.4 Watts/m2 and converted to incident radiation striking the Earth this reduces to about 0.05 to 0.4 W/m^2.  The latter values can be compared to an IPCC anthropogenic effect estimate of 2.3 Watts/m2. They believe it has a small effect on the Earth’s climate. Others, like Abdussamatov, think solar output is more variable, perhaps varying 3 Watts/m2 (their Figure 3). Other variable stars, similar to the sun, seem to have 3% dimming in their minima, which is certainly significant. Both of the latter two examples are larger than the IPCC estimate of man’s influence. We don’t want to get any further into this debate here other than to note the IPCC may be significantly underestimating the effect of solar and ocean cycles in their models. The key point is we don’t know what drives the Earth’s climate. There are a bewildering number of natural and man-made factors that influence it.

While variations in total solar irradiance (TSI) may be small, there is clear evidence that Earth/solar cycles affect our climate. This is discussed in detail by two well referenced posts by Javier here and here. While measured TSI variations are small, the solar UV (ultraviolet) output varies by up to 10%, this affects ozone heating in the stratosphere which may have an influence on the troposphere. The varying UV radiation from the sun and other solar impacts on climate are discussed by Dr. Isaac Held and others at an NRC workshop here.

An interesting quote from the NRC (National Research Council) workshop in 2013:

“In recent years, researchers have considered the possibility that the sun plays a role in global warming. After all, the sun is the main source of heat for our planet.”

They follow this by explaining that solar influence is regional, how exactly does that work? The sun is 109 times larger in diameter than the Earth and 93,000,000 miles away, how can its influence be regional? The Pacific Ocean covers almost one third of the Earth’s surface and 68% of the landmass is in the northern hemisphere; so changes in the surface that the solar radiation hits are bound to cause uneven warming in the short (hundreds or thousands of years) term. As they correctly note, solar changes and regional surface variations cause changes in precipitation and in air circulation. Uneven warming can be expected to do this. However, an uneven warming effect does not disprove solar-caused global warming. It just means global warming of a heterogeneous surface cannot occur evenly everywhere instantaneously. The main means of heat distribution are through water phase changes, that is evaporation, circulation and precipitation. The adjustment of the Earth’s surface to a change in solar activity takes a long time, thus we have long term ocean cycles like the 1,500-year cycle.

The effects of irregularities in the Earth’s orbit

The largest climatic effects appear to be related to long term changes in the Earth’s orbit. These orbital changes occur roughly in cycles of about 413,000, 100,000, and 41,000 and 21,000 years. They are probably, at least part of, the cause of the glacial periods of the Pleistocene geological epoch. The 41,000-year cycle is a change in the Earth’s axial tilt or it’s obliquity. Short term changes (geologically short, that is only thousands of years) are probably related to obliquity and orbital precession (the 26,000-year cycle). Probably obliquity has a larger influence on our climate than precession. Both appear to play a role in initiating and ending major periods of glaciation. The seasons change more dramatically when the tilt is high (24.5°) than when the tilt is low (22.1°). The current tilt is intermediate at 23.5° and decreasing rapidly. Precession controls the distance from the sun during the seasons. Right now the sun is closest to the Earth in the northern hemisphere winter, this moderates the northern winters and makes the southern hemisphere winters more severe.

Below (Figure 1) is a plot of orbital eccentricity, obliquity and precession from 110,000 years ago to 60,000 years from now. The plot was made using a calculator based on Lasker et al.’s algorithm at Colorado State University.

Figure 1

The last glacial period is shaded in blue and the present day is shown with the heavy vertical line. For reference the last glacial maximum (LGM) and the Younger Dryas cool period (YD) are marked. The bottom graph is the computed mean daily insolation at 65°N on the summer solstice. Because most of the land mass is currently in the northern hemisphere this is a key latitude for initiating a glacial period as well as for ending one. It is easier to accumulate long-lived ice on land than on water. The last glacial period began when insolation was headed toward a low of 440 Watts/m2 at 65°N. The last glacial maximum was reached when insolation was 460 Watts/m2. The highest insolation, over 540 Watts/m2, occurred early in the glacial period. By then a lot of ice had accumulated and presumably increased the northern hemisphere albedo enough to keep the ice from melting.

The important points to observe in Figure 1 are that today the obliquity is falling rapidly. Falling obliquity nearly always coincides with cooling temperatures. There is only one exception in the last million years at the end of the Younger Dryas. But, total insolation was quite high and rising at the time. The other key point is that solar insolation at the critical 65°N latitude varies a remarkable 100 Watts/m2! This is over 50 times the IPCC’s estimate of the effect of anthropogenic carbon dioxide and 44 times the total estimated anthropogenic effect.

Javier presents the following illustration (Figure 2) showing the relationship of obliquity to climate in our recent past:

Figure 2

So, the overall natural cooling trend we have observed for the last 5,500+ years is mostly caused by declining obliquity. The decline in temperatures is modified by shorter climate cycles. These shorter cycles are weaker than the orbital cycles, but strong enough to be detected. In Figure 2, the purple line is obliquity, the blue boxes represent periods of glacial advance in various parts of the world and the red curve is Bond’s ice raft debris hematite-stained grain curve (inverted). The black curve is the Marcott, et al. global temperature anomaly adjusted so the difference between the LIA and the Holocene Thermal Optimum is 1.2 degrees to fit paleontological evidence (see Javier’s appendix and here for more on Marcott). To see the present orbital situation compared to the starting point for the last glacial period see Javier’s Figure 17. The long cooling trend from 5,500 BP to the present day is sometimes called the “Holocene temperature conundrum” because it is the opposite of what would be expected when greenhouse gas concentrations are rising. This is discussed in Liu, et al. and graphed here from Knownuthing’s bucket and shown in Figure 3 below. The red curve is CO2 concentration and the blue is methane. The green curve is an ensemble of three computer models (CCSM3, FAMOUS, and LOVECLIM) of global temperature based primarily on the CO2 and methane curves. The discrepancy between the computer model results and the Marcott, et al. reconstruction is obvious.

Figure 3

Other important solar cycles

Javier notes:

Frequency analysis of solar variability during the Holocene identifies several cycles (McCracken et al., 2013), with the most important being the 11.4-yr Schwabe cycle, the 87-yr Gleissberg cycle, the 208-yr de Vries cycle, the ~ 1000-yr Eddy cycle, and the ~ 2400-yr cycle. Even longer cycles can be identified from 10-Berilium (10Be) records in ice cores, like a 9600-yr cycle (Sánchez-Sesma, 2015). Comparison of climate and solar variability records leads to the important observation that the length of the cycle correlates with the amplitude of the climate effect observed and in general the longer the cycle the more profound [its] effect … on climate.”

The post on Professor Curry’s website mostly discusses the 2450 year Bray cycle also called the Hallstatt cycle. Estimates of the length of this cycle vary from 2100 years to 2500 years. Since the estimates are based on 14C dates, this variability is to be expected. The best 14C dates are only good to +-100 years or so, and they can be much further off. The cause of the Bray cycle is unknown, but by process of elimination it is likely to be related to solar cycles. Scafetta, et al. suggest it is due to the orbits of Jupiter, Saturn, Uranus and Neptune. Geoff Sharp suggests that the overall cycle is 4627 years divided into two severe cold periods at roughly 2100 years and 2500 years. Specifically, Geoff Sharp has shown that all grand minima happen when Jupiter, Uranus and Neptune are together with Saturn opposite. These are attractive ideas, but the climate cycles have imprecise periods and tying them to specific solar cycles, with a specific cause has yet to be done.

Whatever the cause of the Bray cycle, historical records show that it has a measurable effect on climate. Javier points out that the little ice age (LIA) occurs at a Bray cycle low. Bray cycle lows correspond with grand solar minimums which are clusters of solar minima, such as those observed in the LIA. The Bray cycle lows in the Holocene are marked in gray in Figure 2.

There are two other important climate cycles, the 1,500-year oceanic cycle and the 1,000 year long solar Eddy cycle. The 1,500-year oceanic cycle is not directly related to solar cycles as discussed here. The 1,000-year Eddy cycle is directly related to a solar cycle and shows up clearly in all records.

The earliest Bray minimum (B-5) occurs during the recovery from the Younger Dryas period 10,300 years ago. This corresponds with Bond event 7. The event is clearly seen in Petit et al.’s Antarctic temperature reconstruction, but it is only a change in slope on the Alley, 2004 Greenland reconstruction. None the less, it is a major ice raft anomaly in the North Atlantic. Evidence of colder temperatures in this period are seen in Norway, Germany, California, and Tibet. At this time the religious monument at Gobekli Tepe (southern Turkey) was deliberately and mysteriously buried. The city wall around Jericho was first built at this time.

The second Bray minimum (B-4) occurs about 7,700 years ago. It corresponds with Bond event 5a and occurs about 500 years after the dramatic 8,200-year BP event. The 8,200-year event is related to the Eddy cycle and the 1,500 year oceanic cycle, but not related to the Bray cycle. The B-4 event is a long slow cooling event that does not end until 7,100 BP (in this post BP means before 1950). This event coincides with the beginning of the Ubaid period. This period also sees the end of the European hunter-gatherer culture and the rise of agriculture.

The B-3 event marks the beginning of a long period of cooling that lasts until the depths of the LIA. The peak insolation (see Figure 1) occurs about this time and falls after. This is the end of the Holocene Climatic Optimum and the beginning of the Holocene Neoglacial period. From this point on precession moves perihelion (Earth closest to the sun) toward the northern hemisphere winter and orbital obliquity falls. This period coincides with Bond event 4. By this time the Sahara desert has mostly formed, replacing the lush savannah that existed during the Holocene Climatic Optimum. Numerous glacial advances around the world show that B-3 (sometimes called the 5.2 kyr event) was strong and took place all over the world.

The B-2 event coincides with the Homer grand solar minimum about 2800 BP and Bond event 2a. This occurs during the collapse of the Minoan civilization and during the Greek Dark Age. A great drought started in the Black Sea area around 1177 BC and this drove the “Sea Peoples” to invade Greece and Egypt. This initiated the Greek Dark Age, ended the Minoan civilization and the Mediterranean and European Bronze Ages. The 3.2 kyr event, when the megadrought began, is not associated with the Bray cycle and may have been caused by the long term ocean cycle or the Eddy cycle or both.

The B-1 event is the little ice age (LIA). It coincides with the Wolff, Sporer, Maunder and Dalton cluster of grand solar minima and with Bond event zero. As with all of these events placing a starting date is difficult. Javier places the start of the LIA at 1258 AD, others place the start after 1500 AD. Either way this is a long period of colder weather that reached its coldest between 1600 and 1800 AD. The LIA is unusual. It was very cold relative to other cold periods and it lasted almost 600 years, longer than any of the other cold periods. Because it started late in a long period of cooling (see Figure 2) it would be expected to be colder than earlier cold periods, it started from a colder point. As to the length, it began and ended with periods of significant volcanism. These could be responsible for extending the period. It also occurred at a confluence of the Bray cycle, the 1,500-year oceanic cycle and the Eddy cycle.


The IPCC bases its conclusion that man has caused most of the warming in the late 20th century solely on two assumptions. The first is that the only natural causes of warming or cooling are TSI (total solar irradiance) and volcanism. Further, they assume the variability of TSI is very small and the climatic effect on the Earth is instantaneous and evenly distributed. We can see from the references above and here and here that this assumption is weak. The second assumption is that the warming from 1951 to 2010 is mostly due to man, see figure 10.1 here. This assumption is also dubious since the warming from 1910 to 1944 is very similar as shown here. How can one claim that the warming from 1910 to 1944 is natural and the warming from 1951 to 2010 is man-made? Further, as shown above, many natural climate cycles (both oceanic and solar) are much longer than 59 years. The IPCC calculation of man’s influence on climate was not based on data, it was computed from the difference between two climate model runs. One model used TSI, volcanism and the IPCC estimates of man’s influence and one was based only on TSI and volcanism. The “Holocene temperature conundrum” casts serious doubt on the climate model results.

So, given that many natural climate cycles are much longer than 59 years and poorly understood; how can we have confidence in the IPCC calculation of man’s influence? We are not suggesting that man has no influence on climate, but we do not believe that man has caused most of the recent warming.

A key take-away is that solar variability and the Earth’s orbit can have a large effect on global climate. But, the conditions on the Earth at the time of the solar change coupled with the uneven distribution of oceans, ice and land on the surface cause the impact of any solar change to be distributed unevenly. This delays the global impact on temperature and causes what we observe as long term oceanic cycles. These long-term cycles are not properly accounted for in the climate models.

We could argue with some of Javier’s points or conclusions, but he has provided a very good overview of natural climate cycles. These cycles are in the available literature piecemeal, but his well referenced and well organized posts are an excellent summary. English is not Javier’s first language and we need to look past this, but his research and content are first rate.

September 27, 2016

Leave a Reply