[Norwegian version here, published by Climate Realists of Norway (Klimarealistene)]
According to a study from last year, it’s not clear whether it’s human activity or changes in solar activity that has contributed more to global warming since pre-industrial times.
With current knowledge, we simply cannot know, according to the study, which, among other things, summarizes knowledge and theories about the Sun’s influence on Earth’s temperature changes.
The study’s title is “How much has the Sun influenced Northern Hemisphere temperature trends? An ongoing debate”. The study should be referenced as Ronan Connolly et al 2021, but for simplicity, I’ll just call it Connolly 2021. There are 22 co-authors in addition to Ronan Connolly, but only two (Willie Soon and Michael Connolly) contributed to the first draft. The study is 60 pages plus references, but it’s a relatively easy read – and very interesting.
The study’s conclusion is in sharp contrast to the Intergovernmental Panel on Climate Change (IPCC), which believes it is extremely likely that human activity has caused most of the warming in the last 100 years or so. 1)
According to the IPCC, the Sun has had almost no impact on changes in global temperature since at least 1950. In other words, IPCC believes the Sun’s radiation has varied very little in this period.
Scientists agree that the Sun’s radiation varies relatively little (around 0.1%) throughout one solar cycle (which is about 11 years). However, they disagree about how much the Sun’s radiation varies on longer timescales.
There are also theories proposing that variations in the radiation the Earth receives from the Sun can be amplified by indirect effects, meaning that parameters other than the amount of solar energy that reaches the Earth can have an impact on Earth’s temperature.
A theory promoted by Henrik Svensmark, among others, is that increased solar activity leads to a stronger magnetic field around the Sun, which causes fewer cosmic rays to reach the Earth’s atmosphere. According to theory, this then leads to less low level clouds and thus a warmer climate.
That increased solar activity and a stronger magnetic field around the Sun cause fewer cosmic rays to reach Earth’s atmosphere is uncontroversial. However, IPCC does not believe that this in turn causes less low clouds and higher temperatures. 2)
As a curiosity: If Svensmark’s theory is correct, then the solar system’s position in the Milky Way may actually affect Earth’s climate. If our solar system is in a region with a lot of cosmic rays (for example in one of the galaxy’s spiral arms), we get more low clouds and a colder climate. If we are in an area with less cosmic rays (for example outside of the spiral arms), we get less low clouds, and it gets warmer than it would otherwise be. The position of the solar system in our galaxy may thus affect the timing of ice ages on Earth (Shaviv 2002).
Uncertainty in solar irradiance measurements
To be able to accurately measure solar irradiance (the amount of solar energy that reaches Earth), we need to measure from above the atmosphere. This has only been possible since 1978, when the first satellites were launched. Unfortunately, we don’t have continuous high-quality measurements of solar irradiance for the entire period from 1978 until today due to the 1986 Space Shuttle Challenger disaster, which led to the postponement of new satellite launches. The first generation of satellites with accurate sensors for measuring solar irradiance was retired in 1989. The next generation was launched in 1991, more than two years later.
In the two years from 1989 to 1991, satellites with less accurate sensors provided measurements. Two different interpretations of these data (ACRIM and PMOD) give qualitatively different results: If PMOD is correct, then solar irradiance has decreased slightly in the entire period from about 1980 until today (each consecutive solar minimum is lower than the one before). On the other hand, if ACRIM is correct, then solar irradiance increased from 1980 to 2000 before starting to decrease:
The IPCC believes the PMOD interpretation is correct and concludes that solar irradiance has decreased somewhat since about 1950, while global temperatures have increased. They thus conclude that greenhouse gases, in particular CO2, have caused most or all of the warming since 1950.
Connolly 2021 emphasizes that there’s disagreement among scientists as to whether ACRIM or PMOD is more appropriate and refers to research on both sides of the controversy.
(That the IPCC favors PMOD is maybe not surprising, considering that Judith Lean, who actually helped create the PMOD interpretation, was lead author of chapter 2.7.1 on Solar Variability in IPCC’s 4th Assessment Report.) 3)
ACRIM vs PMOD
Let’s briefly take a closer look at ACRIM and PMOD. ACRIM1, ACRIM2 and ACRIM3 are the ACRIM project’s three satellites. They’re considered to be very accurate, but new satellites must be calibrated against older ones to get comparable values. Unadjusted solar irradiance values for ACRIM1 were approximately 6 W/m2 higher than for ACRIM2:
Two satellites, Nimbus7/ERB and ERBS/ERBE, 4) measured solar irradiance during the ACRIM-gap. Unfortunately, and as we can see in the figure above, the two satellites show different trends during the ACRIM-gap: According to ERB, solar irradiance trended up, but according to ERBE it was trending down.
Scafetta et al. 2019 argue that ERB is more reliable than ERBE since ERBE probably experienced excess degradation of its sensors during its “first exposure to the high UV radiation levels characteristic of solar activity maxima” during the ACRIM-gap. The ERB satellite had experienced this 11 years earlier, and a further strong degradation was considered unlikely. 5)
PMOD, on the contrary, argue that ERBE is more reliable. They have adjusted data for both ACRIM1, ERB and ACRIM2 to conform them to predictions of proxy models for solar irradiance. Fröhlich & Lean 1998 justify the adjustments based, among other things, on sensor degradation, but both Richard Willson and Douglas Hoyt, principal investigators for ACRIM and ERB, respectively, believe that the PMOD adjustments are unwarranted. Willson has even said the adjustments are incompatible with the scientific method:
The TSI [Total Solar Irradiance] proxy models, such as Lean’s, are not competitive in accuracy or precision with even the worst satellite TSI observations. To ‘adjust’ satellite data to agree with such models is incompatible with the scientific method.
But we don’t need to conclude here that the ACRIM interpretation is better than PMOD or vice versa. IPCC has chosen to rely on PMOD, and we can at least take note that there’s no scientific consensus on that being correct.
NASA
(I didn’t include this section about NASA in the Norwegian version of the article, but I thought it could be interesting to some of you.)
NASA has a web site, climate.nasa.gov, which is dedicated to climate change. The assumption seems to be that the IPCC is correct. Under Causes (of climate change) they have included an image showing solar variation and average surface temperature, where we can see that Earth’s temperature has increased while (total) solar irradiance (TSI) has decreased since around 1980.
NASA doesn’t discuss the ACRIM vs PMOD controversy, but they assume that PMOD is correct. This isn’t explicitly stated anymore, but previously, the TSI source was listed as “SATIRE-T2 + PMOD”. According to the Wayback Machine, the text ” + PMOD” was removed around November 2021 (old version | new version). The removal of PMOD from the TSI source is the only change in the image.
So, if you see someone referring to this image from NASA, you now know that there are scientist who believe that the TSI/solar irradiance graph is inaccurate. Actually, there’s also some uncertainty in the temperature graph. I’ll come back to that shortly.
SATIRE-T2, by the way, is a proxy model for solar variation.
High or low variability?
In order to be able to estimate solar irradiance prior to the satellite era and potentially very far back in time, the datasets for satellite-measured solar irradiance are used as a starting point. By comparing the measured solar irradiance with how the number and size of sunspots have changed, scientists actually find that the two agree well. It’s thus possible to translate from sunspots to solar irradiance a few hundred years back in time. There are also other methods for estimating solar irradiance, allowing scientists to estimate solar irradiance considerably further back in time.
Unfortunately, we’ve seen that there’s disagreement about which dataset is more reliable when it comes to satellite-measured solar irradiance. If PMOD is correct, solar irradiance seems to have varied relatively little over time. However, if ACRIM is correct, solar irradiance seems to have varied significantly more.
Solar irradiance datasets that go further back than the satellite era can be divided into two categories, high variability datasets and low variability datasets. Datasets based on the ACRIM interpretation are mainly high variability, while those based on PMOD are low variability.
The IPCC writes in their latest report from 2021 that the change in solar irradiance was between 0.7 and 2.7 W/m2 in the period from about 1680 (Maunder Minimum, 1645-1715) to about 1975 (second half of the 20th century). According to Judith Curry, this range (0.7-2.7 W/m2) includes both high and low variability datasets.
Despite IPCC having included datasets with high variability in this range, they still recommend that the climate models (which produce estimates of future global temperatures, among other things) should use two datasets with low variability (Matthes et al. 2017).
The effect of this is that the climate models predict higher temperatures in the future than they would have done if datasets with high variability had also been used. This is because low variability datasets indicate that the Sun’s had little impact on temperature changes in recent decades, and thus that greenhouse gases have had a bigger impact. If greenhouse gases have had a big impact in the past, they’ll likely also have a big impact in the future. So if low variability is correct, we can expect a relatively large temperature increase in the future due to more greenhouse gases in the atmosphere. If, on the other hand, high variability is correct, the effect of greenhouse gases has been smaller in recent decades than the IPCC believes and will probably be smaller in the future as well.
Uncertainty in Earth’s surface temperature trend
More and more areas are being developed by humans. This means that weather stations previously located in rural areas suddenly may find themselves in proximity to urban or semi-urban areas. Areas developed by humans are generally warmer than remote areas. This means that weather stations that were previously remote can start to show higher temperatures – not because of global warming, but simply because they’re now closer to urban areas. This effect is called the urban heat island effect.
In order to – as accurately as possible – determine Earth’s temperature changes over time, scientists attempt to correct for the urban heat island effect. The correction process is called statistical homogenization.
Connolly 2021 has compared temperature data for remote weather stations with temperature data for all stations. By exclusively considering stations that are still remote, they found that Earth’s temperature increase has been substantially lower than assumed by the IPCC.
So it is possible that the homogenization process does not correct enough for higher temperatures around previously remote weather stations.
According to Connolly 2021, the standard estimate is that there’s been a temperature increase of 0.86℃ per century over land in the period 1841-2018 in the Northern Hemisphere. But when considering temperature data from remote stations only, Connolly 2021 finds that the temperature increase has been slightly less than half of the standard estimate, 0.41℃ per 100 years.
How much has the Sun influenced Earth’s surface temperature? 80 different answers
Connolly 2021 considered 16 plausible datasets for solar irradiance from the 19th century (or earlier) until today. Eight of the datasets have low variability, and eight have high variability. They then combined each of the solar irradiance datasets with the following five temperature datasets for the Northern Hemisphere:
Assuming, among other things, that there’s a linear relationship (which might not necessarily be the case) between solar irradiance and the associated temperature changes on Earth, this gives a total of 5 x 16 = 80 possible answers to how much the Sun has influenced temperature changes in the Northern Hemisphere since the 19th century.
The results vary from 0 to 100%. On the one extreme, the Sun may have had no role in temperature changes since the 19th century (this is what the IPCC believes). On the other extreme, the Sun may have been the cause of almost all changes in average temperature since the 19th century:
The blue bars in the above two figures show measured temperature increase per 100 years, the yellow bars show the Sun’s contribution to this and the gray bars show the contribution from humans (primarily greenhouse gases). 6)
Connolly 2021 concludes that we cannot know how much of the temperature changes has been caused by the Sun and how much has been caused by humans, and that the debate is still ongoing:
In the title of this paper, we asked “How much has the Sun influenced Northern Hemisphere temperature trends?” However, it should now be apparent that, despite the confidence with which many studies claim to have answered this question, it has not yet been satisfactorily answered. Given the many valid dissenting scientific opinions that remain on these issues, we argue that recent attempts to force an apparent scientific consensus (including the IPCC reports) on these scientific debates are premature and ultimately unhelpful for scientific progress. We hope that the analysis in this paper will encourage and stimulate further analysis and discussion. In the meantime, the debate is ongoing.
Edit: An article criticizing Connolley 2021 has been published on RealClimate. I haven’t looked closely at the criticism.
Footnotes:
1) IPCC states in their 5th assessment report from 2013 that: “It is extremely likely that human influence has been the dominant cause of the observed warming since the mid-20th century.” In their most recent assessment report from 2021 (Summary for Policymakers), the message is that human activity accounts for all net warming since the late 19th century:
The likely range of total human-caused global surface temperature increase from 1850–1900 to 2010–2019 is 0.8°C to 1.3°C, with a best estimate of 1.07°C. It is likely that well-mixed [greenhouse gases] contributed a warming of 1.0°C to 2.0°C, other human drivers (principally aerosols) contributed a cooling of 0.0°C to 0.8°C, natural drivers changed global surface temperature by –0.1°C to +0.1°C, and internal variability changed it by –0.2°C to +0.2°C.
2) The IPCC only includes four paragraphs in their most recent assessment report on how (galactic) cosmic rays (chapter 7.3.4.5) may affect surface temperatures on Earth. They’re skeptical of Svensmark’s theory and say there’s a high probability that cosmic rays have negligible effect:
There is high confidence that [Galactic Cosmic Rays] contribute a negligible [Effective Radiative Forcing] over the period 1750 to 2019.
Svensmark, on the other hand, has published (with three co-authors) a new study after the IPCC’s working group I published their report. In the study, they look at how explosions on the Sun (Forbush Decrease events) have affected, among other things, low clouds via changes in cosmic rays. They found a reduction in low clouds that was strongest 5-7 days after the explosion. The corresponding increase in solar irradiance reached about 2 W/m2 for the strongest explosions.
3) Judith Lean also said in 2003 that one reason she wanted to look at the solar irradiance data was that she didn’t want people to take the ACRIM data as an excuse to do nothing about greenhouse gas emissions:
The fact that some people could use Willson’s [ACRIM dataset] results as an excuse to do nothing about greenhouse gas emissions is one reason we felt we needed to look at the data ourselves.
Since so much is riding on whether current climate change is natural or human-driven, it’s important that people hear that many in the scientific community don’t believe there is any significant long-term increase in solar output during the last 20 years.
But it should be mentioned that Judith Lean, too, recognizes that there’s uncertainty in how much solar irradiance has varied on longer timescales. However, she believes that, in any case, the resulting temperature change has been very small:
It remains uncertain whether there are long-term changes in solar irradiance on multidecadal time scales other than due to the varying amplitude of the 11-year cycle. If so the magnitude of the additional change is expected to be comparable to that observed during the solar activity cycle. Were the Sun’s activity to become anomalously low, declining during the next century to levels of the Maunder Minimum (from 1645 to 1715), the expected global surface temperature cooling is less than a few tenths °C.
A change this small from a Maunder Minimum level assumes there are no solar indirect effects.
4) According to Willson 2014, the ERB and ERBE satellites have less accurate sensors than the ACRIM satellites:
The traceability of ERB and ERBE results are degraded, relative to the TSI monitors, by: (1) the absence of dedicated solar pointing, (2) brief and infrequent data acquisition opportunities; (3) inability to calibrate sensor degradation and (4) infrequent electrical self-calibration.
5) From Scafetta et al. 2019:
ACRIM contends that the Nimbus7/ERB TSI upward trend during the ACRIM-gap period is more likely correct than the ERBS/ERBE TSI downward trend because it agrees with the solar activity-TSI variability paradigm established by the ACRIM1 experiment [references]. In fact, the downward ERBE trend is at variance with the paradigm and was caused by well-documented degradation of its sensors that were experiencing their first exposure to the high UV radiation levels characteristic of solar activity maxima. High UV exposure is known to cause excess degradation of TSI sensors in other satellite results [references]. Nimbus7/ERB experienced a similar excess degradation 11 years earlier during the TSI maximum of solar cycle 21 [references] and would have already reached its asymptotic degradation limit during the ACRIM Gap. Clearly, the Nimbus7/ERB TSI record is the most reliable choice to relate the ACRIM1 and ACRIM2 records across the ACRIM-Gap.
6) The sum of the human and solar contribution doesn’t always add up to exactly 100% of the observed trend. Connolly 2021 states that if the sum is greater than 100%, this means that either the solar or the human contribution (or both) is exaggerated. If the sum is less than 100%, there may have been other contributing factors.
2