18 - A planet-sized magnetic heat pump
The sun shines cold yet the ground grows hot
The previous post documented that Earth’s magnetic field intensity varies with changes in solar wind and SPE strength, and that these variations are mostly anti-phased to geothermal heat flux and surface temperature variations. Many may therefore be wondering if these relationships are reflected in recent solar, geomagnetic and temperature data. Wonder no more.
Historical Solar Activity
Solar activity at its most elemental level is driven by the Sun's 11-year solar (Schwabe) cycle[1]. The magnetic field of the Sun reverses itself after each such cycle and therefore returns to its original state after 2 cycles (the 22 year Hale cycle). A similar yet possibly unrelated duality is also observed in the longer-period solar cycles, such as the 200 / 400-500 year Grand solar cycles, as well as the 2400 / 4800 year Bray cycles (see below).
Four major (and one minor) Grand cycle minima - Oort, Wolf, Spörer and Maunder (and Dalton) - have very likely occurred during the last 1000 years[1,2]. During a solar minimum the Sun’s sunspot, solar flare, SPE and solar wind activity diminish[1-3]. Such minima - according to the previous post’s integrated geomagnetic-geothermal-surface temperature model (IGGT model) - should therefore represent periods of prolonged Outer Core cooling and increasing geomagnetic intensity. This is confirmed by the graph below: over the last 1000 years all periods of significant geomagnetic intensity increases concur with major solar minima.
Under the IGGT model these periods of increasing geomagnetic intensity should correspond to periods of Outer Core cooling, lower geothermal heat flux and lower surface temperatures. This is confirmed by the graph below from Lockwood et al.,[4]
the best estimate of the variation in the average air surface temperature in the northern hemisphere since AD850. This is a combination by Masson-Delmotte et al. (2013) of 18 separate reconstructions and was presented in the fifth assessment report (AR5) of the International Panel on Climate Change (IPCC). Each of these reconstructions employs multiple temperature proxies, including data from boreholes, corals and sclerosponges, ice cores, insect numbers, instrumental data, pollens, lake levels, loess (wind-blown silt), glacier extents, plant macrofossils, diatoms, molluscs, foraminifera, dinoflagellates, ostracods, heavy minerals, grain-size, trace elements speleothems, dendrochronology and historical records (recorded freeze/thaw dates, harvest yields and dates, etc)
The IGGT model correctly predicts key historical temperature lows: periods of lower sunspot/solar wind activity concur with periods of temperature decline/lower temperatures. A look even further back (below) confirms this remarkable correlation: a temperature decrease/GI increase until 250 AD, a temperature increase/GI decrease until 750 AD, a possible mismatch between 750-900 AD, the correlation as above until 1750, and a temperature increase/GI decrease until present.
Note that while the Lockwood et al. graph carries the IPCC’s seal of approval, Loehle’s reconstruction does not, although many objections may be overstated. The main bone of contention is the lack of an anthropogenic warming “hockey stick” increase in the post-industrial age, so most objections are likely irrelevant to this discussion.
Also note that solar insolation differences are almost certainly not responsible for e.g. the temperature decrease during the “Little Ice Age”/Maunder minimum as the solar insolation at higher Northern Hemisphere latitudes when compared to present was less than ~1 W/m2 lower during the winter months, while up to ~1 W/m2 higher during the summer months
Geomagnetic variations, climate change and societal collapse
One of the “most severe climatic events of the Holocene epoch” was the so-called 4.2 ka event. According to Wikipedia:
Starting around 2200 BC, it probably lasted the entire 22nd century BC. It has been hypothesised to have caused the collapse of the Old Kingdom in Egypt, the Akkadian Empire in Mesopotamia, and the Liangzhu culture in the lower Yangtze River area. The drought may also have initiated the collapse of the Indus Valley Civilisation, with some of its population moving southeastward to follow the movement of their desired habitat, as well as the migration of Indo-European-speaking people into India
A quick comparison to the Icelandic temperature proxies reveals that geomagnetic intensity was mostly anti-phased to surface temperature proxies over the 2000-8000 b2k interval: the 7 ka temperature maximum concurred with the geomagnetic intensity minimum. A spectral analysis of the geomagnetic data (presented in the next post) confirms a strong 2400/4800 year cyclicity (Bray/Hallstadt solar cycles), while the de-trended data show pronounced ~1000/2000 year cyclicity (Eddy solar cycles), as well as 200/400 year cyclicity, confirming solar and oribital forcings are driving geomagnetic intensity variations.
A more in-depth comparison between Icelandic temperature proxies and geomagnetic intensity variations reveals that most of the sharp declines in temperature, including the 4.2 ka event, occurred during periods when the rate of change of geomagnetic intensity was maximum (mid upward curve). In other words, they occur when the rate of change in incident SPE and solar wind energy (solar maxima to minima) is largest, and therefore the rate of increase in geomagnetic intensity and decrease in geothermal heat flux is largest. The 3.2 ka Late Bronze Age collapse and the 5.5-5.0 ka cooling event[5] also concur with increasing geomagnetic intensity events. But aren’t these dramatic climate change events caused by solar insolation variations?
Small solar irradiation variations do not cause climate change
Over the past 140 years total solar irradiation has only varied by minor amounts on the order of 0.1 W.m-2 - or 0.1% - over its 11-year solar cycles.
For example, Earth’s global mean surface temperatures (GMST) did not significantly change over the 1965 - 1976 solar cycle: -0.11 °C in 1965 versus -0.1 °C in 1976. Total Solar Irradiance anomalies (TSI) started at -0.5 W.m-2 in 1965, rose to +0.3 W.m-2 in 1969, and ended the cycle at -0.3 W.m-2 in 1976. The least-squares GMST interpolation (Q65_76) peaks roughly 2-3 years (1971-1972) after the TSI anomalies peak (in 1969), mainly due to Earth’s greenhouse effect delaying the atmospheric energy increase’s (thermal radiation) emission to space. The graph is fairly typical of Earth’s steady-state behavior: small solar insolation increases cause small increases in surface-emitted thermal radiation that in turn are partially captured by Earth’s atmospheric greenhouse gasses that in turn delay their energy emission to space by 2-3 years, thereby causing a small, temporary surface temperature increase on the order of 0.1-0.2 ºC. These small solar insolation variations therefore do not cause systemic changes in GMST, a conclusion also reached by the IPCC[6].
But geothermal heat flux variations are too small to cause temperature change!
This is a popular misconception. The cartoon below is fairly typical.
The big yellow box above represents “incoming solar radiation” or solar insolation, which is by far the largest energy source heating the Earth. It, and Earth’s atmospheric greenhouse effect, are the main reasons the Earth’s average temperature is a comfortable 15 ºC. When dealing with climate change however, for example surface temperature fluctuations on the order of 1 ºC, the absolute size of large yellow box becomes irrelevant, as it is only the variations in energy inputs - typically on the order of 1 W.m-2 - that are important. Even so, there is apparently an order of magnitude difference between e.g. the interpreted anthropogenic climate change forcing of 1.6 W.m-2 and the average (change in) geothermal heat flux of ~0.1 W.m-2. The following sections demonstrates that geothermal heat forcing punches far above its weight by acting as a catalyst. For example, when an 0.1 W.m-2 geothermal heat flux melts a sea ice cover (albedo=0.9) the amount of energy absorbed by an open ocean (albedo=0.07) from 400 W.m-2 solar insolation increases by 332 W.m-2
Geothermal heat flux variations change the glacial and interglacial steady-states
Earth’s atmosphere can only accumulate a fraction of the amount of heat energy accumulated in its oceans[7]. When oceans accumulate or lose heat they can become important drivers of climate change, as variations in their thermal radiation directly heat or cool the atmosphere and its weather systems. Unlike ocean surface heat fluxes all geothermal heat energy must be transferred from the seafloor through the entire ocean column before being radiated to space via the atmosphere.
Consider the glacial steady-state in the graph above. A low geothermal heat flux can co-exist with a land or sea ice cover if the ice can successfully transfer and radiate the supplied heat from below the ice cover to the cold atmosphere. If it transfers less then heat accumulates and ice melts, especially during the summer months when atmospheric temperatures are warmer: the ice thickness is inversely proportional to the absolute geothermal power density. Increasing this heat flux by even small amounts will cause the ice cover to thin and possibly melt, as is currently happening to the Thwaites Glacier. A reduction in geothermal heat flux will cause ice covers to thicken by growing from below, as was happening to the Ross Glaciers in 2004. Note that a change in surface power flux, e.g. a 1.6 W.m-2 increase in radiative forcing, has relatively little effect, as the ice sheet’s high albedo and large negative thermal radiative forcing severely limit its melting capacity. Relatively small geothermal heat flux increases are therefore far more effective at melting land or sea ice cover and thereby reducing surface albedo.
The interglacial steady-state effects a different balance. Solar irradiation energy is absorbed and radiated back to the atmosphere, which in the steady-state case will have the same or slightly lower temperature as the ocean surface. The surface-absorbed thermal energy is also radiated and convected towards the deeper, cooler water column, causing a thermocline. Studies[11] indicate a relatively small geothermal heat flux on the order of 50 W.m-2 can cause:
overall warming of bottom waters by about 0.4°C, decreasing the stability of the water column and enhancing the formation rates of North Atlantic Deep Water and Antarctic Bottom Water by 1.5 Sv (10%) and 3 Sv (33%), respectively.
So small increases in seafloor geothermal heat power are magnified multifold at the ocean’s surface. Global ocean surface temperatures increased by about ~1 ºC (~0.2 ºC / decade) while deeper ocean columns temperatures increased by ~0.2 ºC (~0.04 ºC / decade) between 1960-2010 [8; graph above]: the thermocline shift could partially be due a relatively minor increase in seafloor geothermal heat flux. This could also plausibly explain why most of the 2023 record high temperatures were overlying geothermal heat anomalies (graph above).
In addition, small increases in ocean bottom temperatures due to minor geothermal heat increases have disproportionately large effects on ocean currents, for example the Atlantic meridional overturning circulation, a large horizontal advector of heat (see future posts).
And then there’s volcanos, too
A catastrophic increases in local geothermal heat flux can cause major local temperature disruptions, as e.g. in the case of the Hunga Tonga volcano on the map above. A 2001 sonar survey [9] revealed two previously undiscovered active volcanos on the Arctic Ocean Gakkel ridge. The Gakkel Ridge has historically been one of the slowest spreading ridges on Earth, so the discovery of active volcanos came as a great surprise to the oceanographers, causing them to significantly modify their models to reflect this major increase in geothermal heat flow. While the number and size of the Gakkel Ridge volcanos is uncertain, a quick back-of-the-envelope demonstrates their 1995-2020 heating potential. The Kilauea, Hawaii volcano produces an average volume of roughly 0.1 km3/y of 1000 °C lava[10]. Each gram of lava releases 400 cal when cooled to 0 °C, so two 0.1 km3 volumes of lava, roughly corresponding to 6.1014 g, can raise the temperature of 2400 km3 of sea water by 0.1 °C per year, or 60000 km3 over the 1995-2020 period. The paragraphs above illustrate the effect is magnified multifold at the ocean’s surface, so the two volcanos have potentially raised ocean surface temperatures by significant amounts over a fairly sizeable area.
The volcanoes’ significance however also lies in their existence: they demonstrate an unexpectedly hot Arctic Ocean floor caused by a recent, dramatic increase in regional geothermal heat, similar to the increased geothermal heat fluxes currently melting the Thwaites and Pine Island Glaciers. And more importantly: similar to the increases in geothermal heat flux that triggered the Antarctic sea ice cover melting that caused the end of the Last Glacial Period.
Summary
Solar cyclicity causes changes in sun spots, SPE’s and solar wind intensity, but causes little variation in solar insolation intensity. Geomagnetic intensity has historically varied with solar cyclicity, and its variations are mostly anti-phased with geothermal heat and surface temperature variations. Geothermal heat flux variations are a plausible climate change forcing as small variations can cause significant changes in Earth surface albedo, ocean heat accumulation, and ocean current strength. The next posts will document how a geothermal heat increase effectively caused the first interglacial climate change domino to fall.
References:
[1] Moussas, X. et al., 2005, Solar Cycles: A Tutorial. Advances in Space Research
[2] Eddy J., 1976, The Maunder Minimum. Science, 192, 1189-1202
[3] Wu, C. et al., 2018, Solar activity over nine millennia: A consistent multi-proxy reconstruction. Astronomy & Astrophysics, 615, 10.1051/0004-6361/20173
[4] Lockwood, M. et al., 2017, Frost fairs, sunspots and the Little Ice Age. Astronomy & Geophysics., 58, 2.17-2.23. 10.1093/astrogeo/atx057.
[5] Geirsdóttir, Á.et al., 2019, The onset of neoglaciation in Iceland and the 4.2 ka event, Clim. Past, 15, 25–40, https://doi.org/10.5194/cp-15-25-2019
[6] Bindoff, N.L. et al., 2013: Detection and Attribution of Climate Change: from Global to Regional. In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.
[7] Levitus, S. et al., 2000, Warming of the World Ocean. Science, 287, 2225-2229. 10.1126/science.287.5461.2225.
[8] Rhein, M.et al., 2013, Observations: Ocean. In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press.
[9] Report on East Gakkel Ridge at 85°E (Undersea Features), 2009, Bulletin of the Global Volcanism Network, 34, no. 5
[10] Verhoogen, J., 1980, Energetics of the Earth. National Acad. of Sciences Collection, doi: 10.17226/9579
[11] Hofmann, M. and Morales Maqueda, M., 2009, Geothermal heat flux and its influence on the oceanic abyssal circulation and radiocarbon distribution, Geophys. Res. Lett., 36, L03603, doi:10.1029/2008GL036078.