Tag: Solar forcing

From Andy May Petrophysicist

By Andy May

The IPCC references global mean temperature increases to the 1850-1900 mean (IPCC, 2021, Ch. 2&4), and radiative forcing increases to 1750 (IPCC, 2021, Ch 7), then uses the two numbers to help them model warming since the pre-industrial period. According to Pages 2K the world warmed 0.1 to 0.2°C (see figure 1) between 1750 and 1850-1990 (Pages2K, 2019). AR6 (p 150) estimates a change of 0.1°C (-0.1 to 0.3°C):

“This assessed change in temperature before 1850–1900 is not included in the AR6 assessment of global warming to date, to ensure consistency with previous IPCC assessment reports, and because of the lower confidence in the estimate. There was likely a net anthropogenic forcing of 0.0–0.3 W/m² in 1850–1900 relative to 1750 (medium confidence), with radiative forcing from increases in atmospheric greenhouse gas concentrations being partially offset by anthropogenic aerosol emissions and land-use change. Net radiative forcing from solar and volcanic activity is estimated to be smaller than ±0.1 W/m².” (IPCC, 2021, p. 150)

The key assumption in this AR6 quote, unsupported in my view, is that natural climatic forcing is negligible, or varies randomly over shorter climatic periods than 100 years.

Evidence for Solar Forcing

There is considerable evidence for natural solar driven climate oscillations longer than 100 years (Stefani et al., 2024) & (Scafetta & Bianchini, 2022). Figure 1 shows the Pages 2K temperature reconstruction (Pages2K, 2019) since 1600AD, this is one of the temperature reconstructions cited and used in AR6 (IPCC, 2021, p. 61).

The AR6 “temperature 0” and “radiative forcing 0” are not from the same point in time and both are within, or just after, the Little Ice Age, a period of rapid natural climate change. In this section I will discuss the AR6 assertion that natural solar forcing is negligible.

Problems with tree-ring data

The Pages 2K estimate of the global warming between 1750 and 1850-1900 is possible given the data available, but, as AR6 notes, the data is sparse and of poor quality before 1850. The data is mostly tree-ring data (see figure 4 here), but its accuracy, at least since ~1950 and probably sooner, is affected by increasing CO₂ concentrations and possibly other anthropogenic factors (Briffa et al., 1998b). Since tree-ring records are calibrated in the modern era, increasing CO₂ creates calibration difficulties. Additional CO₂ accelerates tree growth, independently of temperature and precipitation, this is especially true in the colder, dryer regions preferred for tree ring analysis (Briffa et al., 1998b).

Briffa, et al. emphasize that increasing CO₂, whether from anthropogenic or natural sources, using standard tree ring paleotemperature techniques, can:

“… result in systematic overestimation of past temperatures, particularly in regions where the loss of low-frequency temperature sensitivity in tree growth is greatest (eastern Siberia and eastern North America).” (Briffa et al., 1998b).

Is TSI an adequate measure of the Sun’s climatic effect?

Another problem is that AR6 only considers TSI (total solar irradiance) when evaluating solar forcing, when it has been shown that other solar changes influence our climate (Lean J. , 2017), (Lean & Rind, 1998), (Haigh J. , 2011), (Hoyt & Schatten, 1997). According to Judith Lean, Earth’s climate system somehow amplifies changes in solar output, because the effects (Earth’s surface temperature changes) are larger than can be accounted for from the changes in TSI forcing alone. Exactly how this works is not fully understood due to “ambiguous observations, dubious correlations, and lack of plausible mechanisms.” (Lean J. , 2017).

One long-term TSI reconstruction is from Steinhilber et al. The data I used for figure 2 is from Steinhilber et al. 2012, but also see Steinhilber et al. 2009. AR6 does not directly use Steinhilber’s reconstructions, they use SATIRE-T and NRLTSI2, but Steinhilber’s reconstruction is similar. I prefer Steinhilber’s cosmogenic isotope reconstruction because it shows more detail in periods of low solar activity. The problem with SATIRE-T and NRLTSI2 is that they rely too much on sunspots and zero sunspots does not equate to zero solar activity.

AR6 does not use Steinhilber’s reconstruction because they require annual resolution and Steinhilber’s reconstruction only produces 22-year averages. The sunspot-based magnetic flux models used in AR6 correlate better with modern satellite TSI measurements, but unfortunately due to problems with the satellite datasets, the correlations are ambiguous and open to interpretation, as discussed in more detail here (see especially figure 5 in the linked post). Some sunspot-based reconstructions show up to 5 W/m² difference between 1700 and the present day. The satellite record of total solar irradiance is short and fraught with problems (see especially figure 7 in the linked post).

AR6 explicitly ignores:

• Solar magnetic field strength doubled from the beginning of the Modern Solar Maximum around 1930 to the peak of solar cycle 19 (~1958), the strongest solar cycle in recorded history. The solar magnetic field then leveled off until about 1990, followed by a slow gradual decline (Lockwood, 2003).
• Solar wind became more geoeffective (efficiency of transferring energy into Earth’s magnetic field) during the Modern Solar Maximum. This led to enhanced coupling with Earth’s magnetosphere.
• The Heliosphere has become better at reducing cosmic ray flux.
• Solar UV variability is much more variable than TSI.
• Stratospheric ozone shows a 1-3% increase at solar maximum.
• The Modern Solar Maximum (~1930–2000) is the longest in over 600 years.

As Joanna Haigh writes:

“Understanding the role of variability in solar activity is essential for the interpretation of past climate and prediction of the future. In the past, although much had been written on relationships between sunspot numbers and the weather, the topic of solar influences on climate was often disregarded by meteorologists.” (Haigh J. , 2011).

As an example, the temperature changes in the upper troposphere and lower stratosphere (~20 km) are three times higher than at the surface. The main point that Joanna Haigh and Judith Lean make is that the IPCC is oversimplifying the impact of solar variability on climate by only considering TSI. AR6’s estimate of solar forcing is not a measure of the total solar influence on the climate, as Judith Lean writes:

“Because solar radiation impinges primarily at low latitudes, the increased radiant energy alters equator-to-pole thermal gradients, initiating dynamical responses that produce regions of both warming and cooling at mid to high latitudes. Because solar energy deposition depends on altitude as a result of height-dependent atmospheric absorption, changing solar radiation establishes vertical thermal gradients that further alter dynamical motions within the Earth system.” (Lean J. , 2017).

As Christopher Scotese has shown, the equator-to-pole temperature gradient is fundamental to average global surface temperature over geological time (Scotese et al., 2021). A steep gradient means a colder planet and a shallow gradient means a warmer one (see Scotese’s figure 7 for more details).

The Little Ice Age

The deepest and most widespread cold period during the Holocene is roughly 1645-1715AD. After this cold period, glaciers reached their greatest Holocene extent in Europe, Iceland, Alaska and New Zealand. The period contains the historically coldest winters in Europe and North America and follows five consecutive solar grand minima, including the longest solar grand minimum of the past 6,000 years.

There were also numerous large volcanic eruptions during the period, Parker (Philippines, 1641), Serua (Indonesia, 1693), Fuji (Japan, 1707), plus three others that have been identified in sediments and ice with unknown locations. In addition to these, there were probably three other eruptions in the western Pacific in 1654 (Japan), 1673 (Indonesia), and 1674 (Indonesia). The Little Ice Age is identified in figure 3 and compared to high-quality temperature proxies from Indonesia, Greenland, Antarctica, and the logarithm of CO₂ from the Dome C ice record.

The key point of figure 3 is that the Little Ice Age is the coldest period in the entire 12,000-year Holocene Epoch, at least in Greenland, Indonesia, the North Pacific, and in the New Zealand area. It is also cooler in Antarctica, but not as much. For more details on the proxies see here and here. Thus, the AR6 implication that this period (the “pre-industrial”) is somehow preferred to today’s temperature or is some sort of climatic baseline is disingenuous. Further, how can a pre-industrial climatic baseline be composed of a baseline temperature and radiative forcing that are separated by over 100 years and within the climatically unstable Little Ice Age (Parker G. , 2012)?

Antarctic CO₂

Figure 3 also shows that the logarithm of atmospheric CO₂ concentration in Antarctica does not correlate well with temperature in Greenland or Indonesia, although it does visually correlate reasonably well with the Antarctic Dome C ice core temperature proxy. Regional temperatures naturally capture regional amplification effects, but since CO₂ is a well-mixed gas it is supposed to have a global effect. The trends in figure 3 cast considerable doubt on this idea. AR6 relies very heavily on the Dome C CO₂ record (Jouzel et al., 2007) & (Lüthi et al., 2008) for their studies, but it only seems to support a CO₂-temperature correlation for Antarctica, not a global correlation.

The Little Ice Age Glacial Maximum

It was quite cold between 1809 and 1820, at least partly due to the famous Tambora eruption in 1815. But by 1850 the worst of the cold was gone, and the world began to warm. Thus, it is difficult to compare warming from 1850-1900 and radiative forcing since 1750, the climate (or more accurately global mean surface temperature) was quite variable between 1750 and 1900 as shown in figure 1. Table 1 shows glacier losses since the Little Ice Age glacial maximum (~1750) in 1900 and 2020.

Table 1. Glacier loss by region in 1900 and 2020.

I had Microsoft Copilot generate an illustration of the glacier positions in New Zealand, Alaska, and in the European Alps at important times. It is shown in figure 4.

Glacier extent is a very sensitive indicator of climate change, in the sense of average surface temperature change, in the glacier’s region (Bray, 1968) & (Zumbühl & Nussbaumer, 2018). It is well understood from historical records and paleo-glacial-moraine data that the maximum glacier extent in the three regions shown in figure 1 occurred between 1700 and 1800, with a maximum reached around 1750. This is just after the historically coldest part of the Little Ice Age, ~1640-~1715 when one-third of the world’s population died (Parker, 2008) & (Parker G. , 2012). There is a delay between the glacial maximum and the cold that caused it due to the time necessary for the snow to compact, form the glacial ice, and move to its ultimate maximum point.

The data sources for the New Zealand illustrations in figure 4 are the NIWA Climate Database (CliFlo) climate snow and ice records (link), the NIWA Snow and Ice Network (SIN) high-elevation glacier/snow monitoring group (link), and the NIWA Glacier Inventory and terminal position update group (Part of the new Climate Present and Past group: link).

The Alaska data is from the USGS (link and link). More details are available in Connor, et al. (Connor et al., 2009), who write:

“Neoglacial climate and landscape dynamism created difficult but endurable environmental conditions for the Huna Tlingit people living there. Choosing to cope with environmental hardship was perhaps preferable to the more severely deteriorating conditions outside of the Bay as well as conflicts with competing groups. The central portion of the outwash plain persisted until it was overridden by ice moving into Icy Strait between AD 1724—1794. This final ice advance was very abrupt after a prolonged still-stand, evicting the Huna Tlingit from their Glacier Bay homeland.” (Connor et al., 2009)

Thus, the maximum ice advance in southern Alaska was around 1750. Additional sources were the NOAA NSIDC data access portal (link) and NCEI (link).

For the European Alps the data are from WGMS (link, link, and link) and swisstopo (link, link, and link). In 1644 the Mer de Glace reached its maximum extent after swallowing three villages in the mountain valley. The terrified and homeless villagers asked Bishop Jean‑Amédée de Milliet to perform an exorcism ritual at Chamonix, France to halt the advancing ice in 1645. The exorcism’s effect was temporary however, in 1821 the glacier reached a second maximum, which was nearly as far down the mountain as the first in 1644. The glacier’s ultimate retreat finally began in the mid-1850s (Zumbühl & Nussbaumer, 2018).

As we can see in table 2, 20-40% of the loss from that maximum occurred by 1900, the end of the temperature baseline period. Thus, it is hard to relate radiative forcing change, as seen in glacial records, since 1750 to temperature changes in 1850-1900. Another complication is that the last of five consecutive grand solar minima was in 1680 and the beginning of the longest grand solar maxima in 600 years was in about 1930 (Usoskin, 2017).

Summary

From this discussion, it should be clear that the Little Ice Age is not a good climatic baseline from the standpoint of humans or human civilization. It was a time of war, famine, drought, plague, and misery in Europe, North America, the Middle East, Japan, China, and probably elsewhere (Parker G. , 2012). The climate of the Little Ice Age is not a climate any sane person would want to return to. This baseline was chosen by the IPCC out of scientific convenience because it had the best available data and not for any meaningful purpose. It is madness to imply that the “pre-industrial” or the Little Ice Age is an ideal climate.

So, what is the optimal global temperature for humankind? This period must be a time when agriculture was productive, civilizations expanded, ecosystems were stable, extreme cold was rare, and glaciers were not advancing into farmland and villages. Historically, when were these times? Refer to figure 3 and here and we see the best times for humans were the Holocene Climatic Optimum (~8000BC to 4200BC), the Medieval Warm Period (~800AD to 1250AD), and the Modern Warm Period (~1905 to present). These were the times that civilizations grew and prospered. Table 2 lists the average temperature in these periods from all three curves in figure 3. The average shown for the Modern Warm Period is mostly from the Pages2K instrumental data after it was moved to the same reference period used for the proxies.

The values in table 2 are the change in temperature, during the Holocene Climatic Optimum (HCO), the Medieval Warm Period (MWP), and the Modern Warm Period relative to a 1961-1990 mean (top table) and the “pre-industrial” (1850-1900). In both tables, the Modern Warm Period falls in between the MWP and the HCO.

The values in table 2 are very rough (perhaps ± 0.2° to 0.3°C or so), but directionally correct. The MWP was probably about the same temperature as the late 20^th century, and the HCO was probably warmer than today. This is important because humans and human civilization thrived during these warm periods and suffered during the Little Ice Age or pre-industrial period, see here and here for a discussion and comparison. The best human-centric climate baseline is not the Little Ice Age, it is the climate seen in the HCO, MWP, and today.

Download the bibliography here.