As intimated in my earlier post [1], the above borrows again from Michael Lewis' The Big Short, with Michael Burry again in the spotlight. What follows is both an affirmation of Burry's acumen and a warning about another, previously almost unnoticed negative aspect of climate change, one that is exacerbated by the continued use of fossil fuels and nuclear fission in power generation. What is happening is that ground moisture is being lost, as manifested by a contribution to sea level rise, and there is no reason to think the process will be reversed. My main source for this post is an article by Benjamis von Brackel in the Süddeutsche Zeitung [2] which in turn is based on an article in Science by Ki-Weon Seo and his colleagues [3].
It has been known for some time that climate change is causing glaciers and sea ice to melt, and increased evaporation and transpiration water losses from the soil and plants, but the ability to actually measure soil moisture and ground water losses has emerged only recently, principally with the advent of satellite gravity missions, in particular the Gravity Recovery and Climate Experiment (GRACE) and its follow-on, starting in 2002.
In case you are wondering how a gravity sensor detects changes in ground water, allow me to introduce an explanation. A satellite in orbit experiences a gravitational attraction proportional to the mass of the planet it orbits, and inversely proportional to the square of the distance between its center of mass and that of the planet. If the planet was for example a perfect sphere of uniform density, its center of mass would be at its geometric center. Since the earth is not a perfect sphere and is not of uniform density, its center of mass is not located at its geometric center, and therefore the satellite will experience a varying gravitational attraction, which can be used to test the accuracy of models of the earth's mass distribution. In particular, if water is lost from land to the sea, the land mass decreases and the sea level increases, which results in a detectable change in the orbit of the satellite.
The redistribution of mass has yet another effect, which was used by Seo and his colleagues to extend the period over which soil moisture and ground water loss to the seas could be measured, and also to help test and refine models of those losses due to climate change. Their major contribution was to tie redistribution of ground water and soil moisture to motion of the earth's axis of rotation, i.e., the positions of the poles.
To explain that, we start with a toy top, which is a body that is symmetric about its axis of rotation. If one sets a top in motion, its axis of rotation will wobble to the degree that the top was not oriented perfectly vertically when set in motion. This wobbling is called precession. Even if the earth were a perfect, uniform sphere it would precess, because its axis is not perpendicular to the plane of its orbit around the sun and because the moon's orbit is also out of plane. These introduce periodically varying torques, resulting in precession. Likewise, a change in the planet's mass distribution introduces a non-periodic perturbation leading to an added (or subtracted) motion of the axis.
By combining GRACE measurements with other measurements of soil moisture and ground water loss and taking into account glacier and sea ice melting, together with the change in sea level due to thermal expansion of sea water, together with astronomical data on the motion of the earth's poles, Seo's team were able to reconcile differences between climate model predictions and precision satellite radar altimeter measurements of sea level over the years. For example, model estimates predicted a 1.3 mm increase in sea level due to the thermal expansion of the oceans, and a 1.8 mm increase due to melting of glaciers and ice caps, over the period 2003 – 2015. But actual satellite altimeter measurements produced a result of 3.5 mm, implying 0.4 mm resulting from soil moisture and ground water losses. The Seo, et al. calculations based on polar motion produced the same result: 0.4 mm increase in sea level due to soil moisture and ground water loss to the oceans.
Having confirmed that the analysis of polar motion can be used to accurately portray the planetary mass distribution, Seo and is team extended the analysis by using data from satellite altimeters and polar motion only, thereby extending the study back to 1979 and forward to the present. They found that the increase in sea level due to ground water and soil moisture loss over the period 1979 – 2016 amounts to more than 1 cm. Now, a centimeter doesn't seem like much, but that increase over the area of the earth's oceans and seas corresponds to 4 Trillion metric tons of water, which equates to slightly more than the volumes of Lakes Erie and Huron combined.
What's more, the rate of loss of soil moisture and ground water to the oceans is increasing. This is shown for the worldwide average in the figure below. Note that the plot Is normalized to the 1979 – 1999 average, that is, it shows the losses or gains relative to average over that period.
The losses are not uniformly distributed, and some regions have actually seen increases. Per Figure S-4 of [3], the largest contiguous areas experiencing large (> 50 mm) losses over the period 2003 – 2021 were in the US midwest. Argentina and Brazil in South America, Equatorial Africa, East Central to Southeastern Europe stretching to the Urals, Southern Siberia and Northern China.
When considering soil moisture or ground water loss, the easiest visualization probably involves moisture take up by plants and transpiration through their leaves, or alternatively dew or puddles evaporating. There are human causes as well, and one we are inclined not to think about is in conjunction with energy production. Power plants that produce electricity by boiling water to produce steam, which in turn drives turbines, require cooling of the working fluid, whether the water is boiled using fossil fuel or nuclear energy. Some plants are deliberately situated on rivers or other large bodies of water, and use that water for cooling. Others employ massive cooling towers, relying on evaporation to do the job. (If you've ever seen one, you will notice the massive quantities of “smoke” billowing out of the tower. But it's not smoke, but rather water droplets or steam.) Such plants evaporate about 1.8 liters of water for every kilowatt hour of electricity produced. [4] Hydroelectric power can also result in evaporation, when the water driving the generators is accumulated in a lake behind a dam. The presence of the dam greatly magnifies the surface area of the river, thereby facilitating a commensurate rate of evaporation. Naturally, solar and wind power consume no water whatsoever in producing electricity, though like virtually everything else, some water is used in their manufacture.
Coming full circle, it is clear that the global response to the climate crisis has been inadequate, and recent political developments in the historic leader in greenhouse gas emissions and current 2nd worst offender - the US – promise to roll back even what little has been achieved here. So once again, Michael Burry is right, this time by going long on water. [5] [6]
Notes
[1] https://stephenschiff.substack.com/p/profiting-from-calamity
[2] von Brackel, B. Der Wüstenplanet, Süddeutsche Zeitung, 31 März 2025, Nr. 75 s 12
[3] Seo, K-W., et al., Abrupt sea level rise and Earth's gradual pole shift reveal permanent hydrological regime changes in the 21st century, Science, 27 Mar 2025, Vol 387, Issue 6741, pp. 1408 – 1413. Full disclosure: The article is paywalled. An English language abstract and supplementary material are obtainable at no cost via https://science.org/doi/10.1126/science/adq6529
[4] https://visualizingenergy.org/what-methods-of-electricity-generation-use-the-most-water/
[5] https://revealnews.org/blog/what-comes-after-the-big-short-going-long-on-water/
[6] Nothing in this post is intended as, nor should be considered to be, financial or investment advice.