There are millions of square kilometers of below sea level dry land, mostly in desert areas. Reducing the mean temperature of the earth by 2 °C requires increasing cloud cover of stratus clouds by about 3% of the total area of the earth (assuming 80% reflection, typical of stratus), roughly 3.36 million square kilometers. I understand there are issues with how much evaporation would become cloud cover (and that cloud cover varies by type of cloud), nevertheless it seems that increased evaporation would increase cloud cover more or less in proportion to the distribution of clouds and most clouds are stratus, thus providing a cooling effect.


3 Answers 3


Barring the technical feasibility of what you are proposing, there's a few points to consider which make your proposal problematic.

  1. Water vapor is a greenhouse gas. If you increase evaporation, you increase GHGs in the atmosphere.
  2. The albedo of water is very low, so if you increase surface area of water on Earth, you decrease the albedo
  3. Clouds are temporary, not a long-term solution to Earth's albedo.
  4. The best natural source of high albedo on Earth is snow/ice, not clouds.
  5. Unless you are using renewable energy sources, pumping large volumes of water continuously will require fuel usage, which will increase GHGs in the atmosphere.
  6. In order to address the issue of global temperature increase, we need to stop burning fuel as a globe and start sequestering carbon. Efforts to increase albedo artificially are surely of little value while humans still emit 10 gigatons of CO2e per year.
  • $\begingroup$ I appreciate your comments. Let me reply to some of them. The albedo of water is low, however the energy content of water is very high so evaporating water absorbs a lot of energy. The surface temperature of the planet is what we need to lower, so lowering the surface temperature of extremely hot deserts will lower the temperature of the atmosphere above that area significantly mitigating the effects of the lower albedo. We don't need to pump water if we flood below ocean level areas, we just need to (essentially) dig ditches (with gates to regulate flow). Stratus clouds can be 80% reflective. $\endgroup$ Commented Oct 2, 2021 at 13:24
  • 9
    $\begingroup$ @MichaelLampel While the desert surface might transfer its heat energy into evaporating the water, this energy is not "lost". The water will at some point emit this energy again when it condensates, thereby releasing it to its surroundings. Unless energy is radiated into space, it stays in Earth's system, so you would be doing the global equivalent of trying to cool your home by leaving your fridge open. $\endgroup$ Commented Oct 2, 2021 at 14:22
  • 2
    $\begingroup$ which btw, won't cool your house down because there is a radiator on the back of the fridge that heats up the house (radiating out the heat that is absorbed by the cooling elements) at the same time that the cooling elements on the inside of the fridge try to cool it down $\endgroup$
    – minseong
    Commented Oct 2, 2021 at 19:48
  • $\begingroup$ So evaporating water takes a lot of energy. Condensing water releases that energy, which happens in the atmosphere. As pointed out half of that energy (the half angled toward the horizon up to straight up) will go out to space, thus releasing half of the absorbed energy back out. This is in addition to reflecting out radiation (if thick enough stratus) once the cloud is formed. Seems like a winning idea to me. $\endgroup$ Commented Oct 3, 2021 at 15:45

Clouds form when moist air rises in a convection cell - cloud coverage area is largely controlled by the size of the convection cells, not moisture content.

When temperature differences arise between different patches of the Earth, the air over the warmer part expands, creates low pressure, and pulls in moist air at the bottom as it rises. As the altitude increases the pressure drops and that reduces the temperature. When it gets high and cold enough the water falls out as rain. More moisture means heavier rain, starting at a lower altitude. But the size of the cloud is a function of the convection cell size, which depends on the pattern of temperature differences.

To get more area covered by clouds, you would have to change the balance between warm areas and cold areas, so that the warmth was less intense and more geographically diffuse and the cold bits more intense and more geographically concentrated. The average temperature increasing or decreasing doesn't change the differences. Even the differences getting more or less intense doesn't change the sizes of the patches where it is higher or lower than the average. It's a third-order effect.

Cloud physics is very complicated and not fully understood, and it's effect on the global energy balance likewise. Clouds form in warm areas, in the process cutting off the sunlight and cooling the land beneath them. They act a bit like a thermostat, smoothing out temperature variations - although not in the sense of aiming for a specific temperature. This results in feedback loops with various time delays - local warming causes more clouds causes local cooling causes reduced clouds. Their effect on the global energy balance depends on their altitude. Low clouds are made of warm water and radiate lots of infrared to space, high clouds are cold and radiate much less. They have different effects during the day and the night - at night they cause less infrared to be radiated to space, being at a higher, cooler altitude, causing net warming. So what effect they have depends on what time of day they form. And there are feedback interactions with the temperature of the oceans which cause a massive time lag compared to the land which warms and cools very quickly, the effect of mountains and terrain and vegetation, different amounts of sunlight at different latitudes and seasons, and so on. The effect on the climate of clouds at the equator is very different to the effect of clouds in Antarctica during the long darkness of winter.

Feedback cycles with time delays result in oscillations, and an oscillator forced by regular pushes from an independent driver (like summer/winter) tends to result in the phenomenon of 'chaos', as pioneered by the meteorologist Edward Lorenz. Even with perfectly understood deterministic rules, the behaviour can be irregular, complex, apparently 'random'; a fractal structure with variation on every time scale.

In addition, the effect on global mean temperature is a cumulative process. If one supposed the net cloud cover over a year was a random number, centred on a fixed value but varying slightly from year to year, then the amount of heat in the system would be pushed up and down from the previous year's value by a random amount - a classic 'random walk'. If you plot a sequence of independent random numbers, you get a uniform fuzzy cloud of points. If you plot the cumulative sum of the same numbers, you get a rough line that wanders up and down like a mountain range. Over short periods, you get what appear to be rising and falling 'trends', just from random numbers. It's very hard to distinguish this sort of thing from deterministic drivers; different researchers have come to radically different conclusions based on the same data. When the process driving it is not a nice, independent, mathematically well-behaved random number generator, but a chaotic process that can hop between different meta-stable states at long intervals, the issue is even worse. The natural background level varies by relatively large amounts over long time periods. Clouds add precisely this sort of noise to the climate.

There isn't any global shortage of water surface to evaporate from - two thirds of the Earth's surface is water. And it's not generally any cloudier over the sea. Clouds are driven by temperature differences causing air to rise or fall convectively. Clouds rising cool, and drop the excess moisture as rain. More moisture results in thicker cloud and more rain, but not necessarily a greater area covered. And the effects of cloud on climate are not so simple and straightforward, or easily understood.

If you really want to reduce the albedo of the Earth, you'd be better off painting the ground white. Clouds are a highly unreliable way of doing it.

  • $\begingroup$ Thank you for your detailed answer. I agree that clouds do not have very good models - I modeled clouds for a bit while working at JPL. Let me make a couple of points, first clouds do preferentially come from evaporated sea surface water, wherever they might coalesce. Second in lieu of a better model, stratus clouds will form preferentially as they are most common especially over sea surface. Think of this - if the Mediterranean were desert wouldn't it most likely increase mean surface temperature? $\endgroup$ Commented Oct 3, 2021 at 15:40
  • $\begingroup$ To the point of no shortage of water; I am suggesting that flooding areas in the Sahara or Arabian deserts (or the Salton Basin in CA) will provide a very large (several percent) increase in evaporated water - this is potentially a very big deal. After all 2C is only ~0.7% of the mean temperature of the earth, but it is looking like a catastrophic change for the climate and ecosystem. I contend that doing that is environmentally more friendly than any other large scale intervention in the climate - carbon capture, seeding the atmosphere with other chemicals, or anything else. $\endgroup$ Commented Oct 3, 2021 at 16:18

We agree that additional evaporation enhances energy transport from the surface to the atmosphere and intensifies the hydrological cycle and cloud formation, and that some of the most serious climate change issues such as:

  • Rapidly expanding deserts with forest and species extinction
  • Accelerated rising sea level and earth temperature
  • In many places sinking groundwater levels, drainage of the regions and continents
  • Periods of drought with temperature records and heavy rain events with flooding

…are all linked to the presence or absence of water !

You can read here what a slightly modified concept for reducing sea level rise through water and rain retention and for cooling the Earth's temperature through additional cloud albedo could look like: https://climate-protection-hardware.webnode.com/english/

It differs in:

  • that it is not sea water, that is discharged and evaporated, but clean spring, river or rain water, which not only guarantees additional evaporation rates, but also improves groundwater aquifers, soil moisture, photosynthesis and CO2 absorption.
  • that the water does not evaporate on an uninhabited, barren desert surface, but where water scarcity is already putting the global population, agriculture and nature under enormous pressure.
  • that the concept is not based on a single, expensive large-scale project, but results in millions of small-scale, cheap and quickly implementable measures.

In principle, it is a simple, worldwide request to politics, agriculture, industry, but also to private individuals, wherever/whenever possible, to create extensive water reserves and also to save water, in order to generously use it in plant growth, evaporation, clouds and "water cooling" during periods of drought in spring and summer.

The global strategy starts with rain barrels with overflow on unsealed areas...and explains many other private and municipal measures, which you can find on the website.

There you will also find some links to scientific papers, which should support this holistic, alternative climate protection strategy.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.