This article is part of a series
Oxygen gas is difficult to measure
In case you are arriving here for the first time, I’ve been investigating how our atmosphere may be running low on oxygen, which you might call a global cooling gas. As you know, for over a century humanity has been burning carbon-based fossil fuels (C) plus atmospheric oxygen (O2) to produce energy. This process creates emission byproducts of CO2 and other greenhouse gases and poisons, which are killing the planet’s ecosystem and ability to photosynthesize more oxygen.
Update: I did start with 10 reasons why global oxygen measurements may be wrong. Now I’m up to 16 reasons.
Table of Contents
Is oxygen the missing global cooling gas?
There is a lot of evidence that climate change is being driven by greenhouse gas emissions, which warm the atmosphere and cause volatile weather — and I do NOT argue that. What I am saying is that climate change is also being driven by a lack of oxygen. So on the one hand, we have an increase of carbon dioxide and other pollutants and on the other hand, we have a decrease in molecular oxygen (O2), the kind we breathe. So, we can reasonably assume that the weather would be at least twice as bad as predicted if only accounting for greenhouse gases. And, it does seem like everyone agrees the weather is worse than predicted.
But how much oxygen are we missing?
In a previous article, I demonstrated how to calculate the mass of oxygen in the atmosphere. Calculating the accepted number for how much oxygen is in the air is easy. The challenge is that atmospheric science is complicated, and the numbers are estimates based on current theories and knowledge. In this article, I explore ten reasons why the accepted amount of oxygen in the atmosphere might not be accurate, including my fishbowl theory. And, even more problematic than an error in measurement are hidden sources of oxygen that offset measurements, including warming oceans, melting glaciers, and new bonus reason #11, CO2 fertilization.
I hope these ideas inspire some further investigation.
Okay, let’s get started with a recap of how oxygen is measured by scientists.
How is oxygen measured?
Before I list my reasons why I think they are errors in the measurement, let’s look at how oxygen is actually measured. [You may have already read this section in my previous article. How much oxygen is in the atmosphere?]
It’s complicated. For starters, oxygen is measured either by concentration (percentage) or by partial pressure. The percentage of oxygen in our atmosphere is theoretically the same everywhere (except at extreme altitudes) but the partial pressure changes. So, at high altitudes, the percentage is the same, but there are fewer molecules. So, it gets harder to breathe.
The Scripps O2 Program is the organization that collects air samples from 9 locations worldwide and measures changes in atmospheric oxygen concentration. (Per their data, oxygen is going down step by step, matching the rise in carbon dioxide.) Two points interest me:
One) Per their website: “Measuring the changes in O2 is challenging because the changes are so small. The data reported on this website are measured using a technique developed by the project leader (R. Keeling) as part of his Ph.D. thesis back in the 1980s. The technique involves detecting changes in the refractive index of air via a very precise measurement method known as interferometry. The data are reported as changes in the O2/N2 ratio in ‘per meg’ units.”
Two) Then, they compare this ratio to a sample of air they collected in the mid-1980s.
Okay, wow! So, the reference sample of oxygen is only about 35 years old. This is well after the industrial revolution when fossil fuels began being burned. What’s more surprising is that the Scripps Institute only measures a change in the ratio between oxygen and nitrogen, not a change in the actual amount of oxygen.
This raises so many questions:
- How did they measure the total mass of oxygen? (Their measure is almost the same as mine and presumably also 35 years old.)
- Was this method accurate in the 1980s?
- Where does nitrogen come from or go? (If this has changed, then the ratio is off.)
- And, not only are water vapor and pollutants subtracted from the equation but so are three of the top five gases, argon, carbon dioxide and neon.
- And many more.
Actually, I had so many questions that I now list them below as possible errors in measurement.
Update: Coincidentally, I found another paper by the same scientist above. It reports that the ocean is currently losing about 1.5−3.1 gigatons (1.5−3.1 billion tons) of oxygen each year (Keeling and Garcia, 2002; Schmidtko et al., 2017). So, my first question is: where does the oxygen go? If it is evaporating into the air, then it is offsetting the loss and the measurements are wrong. Perhaps we can assume the melting glaciers are releasing an equal amount of oxygen into the atmosphere.
A hiccup in the data
Okay, I was so surprised at my accuracy in calculating the mass of oxygen in the atmosphere that I had to consult a real research scientist. They didn’t seem surprised. They said, “This is how science works. Doing science is like following a recipe. Anyone can duplicate the process and arrive at the same answer.” They also said that hundreds if not thousands of scientists came before me to help make this process very easy, like Newton and Avogrado. That being said, if one theory, one scientist or one piece of data collected is wrong, our recipe fails. The good news is that science is meant to be scrutinized. If a theory breaks, that is an opportunity to see how the world really works.

At this point, I should admit that I have a bias — let’s call it a theory. I believe that the real driving force behind climate change is NOT carbon dioxide emissions but oxygen depletion. So, as part of my investigation, I’m looking to see if there is some measurement error. Is there less oxygen than believed?
So far, the science has held up to my scrutiny. But I have found some potential gaps.
Take a close look at this graph. It shows a decline in oxygen that seems to match an increase in carbon dioxide. Actually, oxygen is declining about ten times faster. Given the formula for combustion of a fossil fuel (CH4 + 2 O2 → CO2 + 2 H2O), oxygen should be declining only twice as fast. Of course, plants absorb carbon dioxide, but theoretically, they also produce oxygen. So, these graphs give me my first clue that something is off.
Possible errors in measurement
Below I will give a brief introduction to each idea. Perhaps in the future, I will investigate each idea thoroughly. However, I have learned that one can only go so far with the existing scientific literature. At some point, these ideas may require novel, in-the-field research studies.
Oversimplification
My first idea is circumstantial evidence. As I mentioned, it was just too easy to calculate the mass of oxygen in the atmosphere. I used nothing more than high-school-level geometry, chemistry and physics. Think of all the complicating factors that should have made this calculation far more complex: calculating the average landmass (Is this done by satellite measures or theory?), accounting for seasonal variations in water vapor and temperature, differences in pressure (wind), the oblate spheroid shape of the Earth, and many more factors, some that I’m guessing are still unknown to science. My ideas below are more complicating factors. I don’t know if they are new, but I did think of them myself or brainstorm them with friends, like the Fishbowl Theory.
Lack of measurement stations
There are measuring stations for carbon emissions all over the world, including the top of mountains, in the middle of the ocean and in Antarctica. However, Scripps, the leading institute on oxygen research only has 9 measuring stations. These all seem like nice fresh places on the ocean shores. I think we need some in the middle of deserts and the top of mountains. (Oxygen is heavier so a decline would be measured up here first. See my fishbowl theory below.) I’d also like to see a measuring station in the smog-filled valley of Mexico City.
- Alert, NWT, Canada
- Cold Bay, Alaska
- Cape Kumukahi, Hawaii
- La Jolla Pier, California
- Mauna Loa Observatory,Hawaii
- American Samoa
- Cape Grim, Australia
- Palmer Station, Antarctica
- South Pole
Nitrogen bias
As we discussed above, the rise or fall of oxygen isn’t measured directly; it’s measured compared to nitrogen as if the amount of nitrogen in the atmosphere never changes. But is this true? The study of the nitrogen cycle is another complex subject. The short story is that nitrogen levels in the atmosphere do fluctuate. It seeps out of the land and volcanoes release more. Nitrogen is crucial to life, which both consumes and excretes it. Like oxygen, measuring nitrogen is difficult. But it is possible for the ratio of nitrogen to oxygen to stay the same while the density goes down. Though I doubt this is a significant problem if at all, I do think it is an area for further explanation. For example, how much nitrogen gas does the biosphere produce versus consume? And if the biosphere is failing does this mean less nitrogen and less oxygen?
By the way, when measuring oxygen compared to nitrogen, all other atmospheric gases are excluded, like argon, carbon dioxide and neon. This accounts for about 1.1% of the atmosphere. If any of these change then the practical amount of available oxygen available changes.
The water vapor problem
In this chart of the five most common gases in our atmosphere, what should have ranked as third is water vapor. According to many sources online, the percentage of water vapor in the atmosphere varies from 0.2% to 4%. There is so much water vapor in the air that if it all rained down, it would raise the ocean levels by 1.5 inches!
Measurements of the atmosphere are done with “dry air.” The water is condensed or frozen out of the sample before measurement. Now I understand that scientists are using this method to achieve consistent numbers despite the weather, but what does this mean under practical circumstances? Does water vapor displace oxygen or nitrogen? On a very humid day, does that mean there is only 17% percent oxygen? Do internal combustion engines perform worse? Does it get harder to breathe?
Weight versus volume
As I mentioned, we calculated the mass of oxygen in the atmosphere, not volume. The volume of oxygen in a given parcel of air is 20.946%, but its weight is 23.14%. Arguably this is the same amount of oxygen any way you look at it. But practically, it is very different. For example, I rode my bicycle to Mount Everest Base Camp. It got very hard to breathe. It was the same percentage of oxygen but in very thin air. It takes a lot of acclimatization to survive at this altitude and some never can. The thin air is an example of Boyle’s law: as the pressure decreases, the volume of air increases. So, on the ground level, the percentage may remain the same, but the air could be getting thinner or thicker due to unknown factors. For example, wouldn’t global warming cause the air to expand and get thinner?
Saying the same thing in a different manner: The standard measure of oxygen by percentage does not count the number of molecules of oxygen, which varies tremendously due to pressure; in other words, though measurements may still indicate an atmospheric concentration of 20.95%, if the pressure were half as much there would only be half as many molecules of oxygen. Reduction in pressure occurs with an increase in altitude or a change in the composition of the atmosphere, such as a storm system (water vapor is lighter than air), or pollution (ground-level ozone), even a reduction of oxygen itself would cause a change.
Change in atmosphere composition
Similar to the point above. If there is more carbon dioxide or less oxygen or both, this changes the density of the air being measured. And, changes in pressure affect other systems. For instance, the ocean may absorb or outgas oxygen.
Air pollution
I think air pollution, which is also subtracted (or at least, discounted) from the measures, has much the same effect as water vapor. So, imagine a smoggy and a humid day combined. The oxygen levels would drop even further. Now imagine a smoggy, humid day while living at a high altitude. Now imagine it got really hot! You can understand how the practical amount of breathable oxygen can get quite low.

Atmosphere shape and size
This also doesn’t seem to be a problem, but worth mentioning. The Earth is not round and the atmosphere less so. Our math above assumed the Earth and atmosphere are perfect spheres and that the pressure around the world is constant, but it is a much more complicated place. However, it appears to all average out. If the pressure drops in one area, gases will flow into the hole. We call this weather. And, as you can see in the illustration, the atmosphere (bottom of the troposphere) is shorter at the poles and taller at the equator. It would seem that the pressure is much greater at the equator, but this air is warm and moist, thus less dense. And if the air does change, it will circulate. It doesn’t seem to be a problem; still, I think there may be areas left to study. Consider the next three points.
Doldrums and dead zones
Pictured in the illustration are the doldrums (Inter-Tropical Convergence Zone), where sailboats can get stuck in the windless waters. There are many zones that may act as walls, similar to those heaters that blast hot air in front of a cold door.
Also, given that there are high and low-pressure systems, and deserts (no oxygen production), and forests (oxygen-producing), and urban areas (burning oxygen), there have to be high and low areas of oxygen, much like the dead zones in the sea. Do scientists measure these invisible dead zones as they float around? Do planes fly through them? Do the Himalayas block the jet streams and create a pocket of dead air?)
I’m reminded of the classic book, On the Beach by Nevil Shute. In this nuclear holocaust, the superpowers in the Northern Hemisphere have destroyed themselves. However, the weather patterns have temporarily delayed the spread of the deadly radiation. Earth’s last survivors in Melbourne, Australia, await their fate.
This article recognizes the problem of oxygen depletion in the ocean, but only as far as it is related to ecosystems like coral reefs and fisheries. Ocean scientists call for global tracking of oxygen loss that causes dead zones. Phytoplankton in the ocean produces most of our breathable oxygen, and there is evidence that phytoplankton are dying in the warming waters. This affects all land animals and, even, plants.

Inversion layers
Similar to the above is the idea that inversion layers may be trapping pockets of oxygen-depleted air, much luck inversion layers trap smog. Also, if you do an image search for “inversion layers in the atmosphere”, you will see that the temperature and density of the atmosphere is not a smooth gradient from bottom to top. The troposphere is capped by the stratosphere. The two mix very slowly. So, I would predict that if the oxygen is dropping that the ceiling of the troposphere is also dropping. That leads me to my next point:
Fishbowl Theory
Okay, this is my favorite idea and what seems like the biggest possible error when measuring the mass of oxygen in the atmosphere. It’s super simple to understand, but first, a little background.
Per this article about air composition and molecular weight, “Air is usually modeled as a uniform (no variation or fluctuation) gas with properties averaged from the individual components.” That means if you put a bunch of different gases in a box and close the lid, all the gases will mix evenly because all the molecules are in constant motion. But, of course, the Earth is not a box. It is more like an open fishbowl with temperature and gravity diminishing towards the top.
So imagine the water inside the fishbowl is nitrogen; now we pour in some oxygen, which is heavier (see the chart above), and it falls to the bottom like sand. You can even imagine that as oxygen gets burned and turned into other pollutions these heavier molecules are like rocks that fall to the bottom. In other words, we should expect a greater density of oxygen at ground level. Evidence of this stratification of layers is very obvious. You see it almost every day — clouds. Water vapor is lighter than both nitrogen and oxygen. (If it wasn’t, the Earth would be a very different place.)
This subject requires a lot more research. I haven’t found much yet to disprove it. In fact, the same article we just quoted also says, “The composition of air is unchanged until [an] elevation of approximately 10.000 m.” Unfortunately, there is no mention of what the composition of the atmosphere is above this altitude (10 kilometers).
And this article about the exosphere lends further proof. It says, ”The exosphere layer is mainly composed of extremely low densities of hydrogen, helium and several heavier molecules including nitrogen, oxygen and carbon dioxide closer to the exobase. The atoms and molecules are so far apart that they can travel hundreds of kilometers without colliding with one another. Thus, the exosphere no longer behaves like a gas, and the particles constantly escape into space.”
The other layers of the atmosphere also have slightly different compositions, but their weight pressing down on the surface of the Earth is still part of the formula above.
Oxygen depletion in the ocean has been researched a lot more. Above, we see an example of dead zones in the ocean. We can see the levels are affected by weather patterns, continents and the depth of the ocean. Measuring oxygen levels at the surface would yield much different results. I think that this graph models what we would expect to see in the atmosphere in different locations different altitudes and different altitudes.

The Ox family
Known as the Ox family, are the different forms of oxygen easily measured as different?
- O, aka: atomic oxygen, monooxygen, photooxygen
- O2, molecular oxygen, diatomic oxygen, dioxygen, breathable oxygen,
- O3, ozone, trioxygen
- Triplet oxygen
Measuring non-breathable forms of oxygen
I am doubtful this is a problem, but I will mention it anyway. When scientists measure oxygen, what kinds of oxygen are they measuring? As mentioned above, there are different forms of oxygen. And there are a lot of gases that have oxygen as part of their makeup: carbon dioxide (CO2) and carbon monoxide (CO) and water vapor (H2O).
Factors offsetting measurement
Even more problematic than an error in measurement are hidden sources of oxygen that offset measurements. These sources may be temporarily replenishing supplies of oxygen giving us the false impression that everything is fine. Oceans warming, glaciers melting and CO2 fertilization may herald a sudden and dramatic drop in our atmospheric oxygen levels.
Oceans warming

Imagine opening a can of soda on a hot day and leaving it in the sun. What happens? The dissolved carbon dioxide bubbles out, and the soda goes flat. In a similar way, this is what is happening to our oceans. As a liquid warms up, it can’t hold as many dissolved gasses. So, as the oceans warm due to global warming, they release not only their dissolved carbon dioxide but also their dissolved oxygen.
If the oxygen is evaporating into the air, then it is offsetting the loss of oxygen by combustion engines. Maybe the measurement isn’t wrong as much as it isn’t complete. The oxygen levels in the atmosphere may appear stable for now, but what happens when the oceans reach a new equilibrium and stop venting the stored oxygen reserves?
Our oceans are boiling away
Update 2: Our oceans are hotter than ever! A new study published in the academic journal Advances in Atmospheric Sciences revealed that the ocean had absorbed 14 zettajoules of energy in the last year, or 14,000,000,000,000,000,000,000 joules of energy!
More than 90% of global warming heat eventually ends up in the ocean. Warmer oceans can store energy that offsets global warming, but it also causes changes in weather patterns, higher sea levels, glacial melt, disruptions in ocean currents, and — from our POV — it causes the saturated oxygen to boil away.
Update: Coincidentally, I found another paper by the same scientist above from Scripps Institute. It reports that the ocean is currently losing about 1.5−3.1 gigatons (1.5−3.1 billion tons) of oxygen each year (Keeling and Garcia, 2002; Schmidtko et al., 2017).
Glaciers melting
Likewise, we can assume the melting glaciers are releasing their stored oxygen into the atmosphere. Ancient glaciers have both air bubbles trapped inside them, and dissolved gases trapped in the frozen gas. When glaciers melt, they release these gases back into the atmosphere.
CO2 fertilization
Some big news in the oxygen world happened yesterday. And I think made a big discovery as a result. I think it explains why there is more oxygen in the atmosphere than expected. Carbon dioxide levels have been going up for a long time, and oxygen is being measured to go down. However, the decline in oxygen should be much steeper than observed. In part because more oxygen is consumed than carbon dioxide emitted. This concept of “CO2 fertilization” may explain why. And, it may herald a future dramatic drop in oxygen levels.
A new study was reported in the esteemed scientific journal Nature. A constraint on historic growth in global photosynthesis due to increasing CO2. This research study reports an increase in photosynthesis by 11.85 ± 1.4%. This increase is due to the increased carbon emissions in our atmosphere. From the point of view of this research study, more carbon dioxide means more plant growth which means more carbon dioxide absorbed. So, ironically, mankind’s effect on the atmosphere is offset by about 1/3. Conversely, from my point of view, increased plant growth means an increased release of oxygen.
Here is the missing link and the big problem! When humanity hits the ceiling on photosynthesis, yet continues to burn fossil fuels, oxygen may go down dramatically. In other words, there is an upper limit to plant growth due to land area and sunlight, so eventually, plant growth will stabilize and no longer produce excess oxygen. So, in the near future oxygen levels could suddenly diminish.
The article says this same thing but in relation to carbon dioxide.
“We don’t know what the future will hold as far as how plants will continue to respond to increasing carbon dioxide,” he said. “We expect it will saturate at some point, but we don’t know when or to what degree. At that point land sinks will have a much lower capacity to offset our emissions. And land sinks are currently the only nature-based solution that we have in our toolkit to combat climate change.”
Trevor Keenan, Berkeley Lab scientist and lead author of the study.
Expanding and/or densifying atmosphere
This is an idea that may explain why our weather is so volatile.
I need to a lot more research, but here is the basic concept. When a barrel of oil is burned inside internal combustion engines, it is a liquid going in and an extremely hot oxide gas coming out. When a liquid becomes a gas it expands greatly. In the case of water, one liter of liquid water becomes 1700 liters of steam at standard pressure and 100°C. I’m not sure what the conversion is for a barrel of oil, but it is obviously a lot. Now multiply that by 100 million barrels of oil burned per day. Wow!
Per this article on Livescience.com, “The lowest part of Earth’s atmosphere has been rising by 164 feet (50 meters) per decade since 1980.” It also may be expanding just due to the temperature increase of global warming.
Not only is that atmosphere expanding, it is also getting denser. Many of the oxides produced from burning oil, with exceptions like water, are actually denser and heavier than air.
So, adding up all these factors (expanding atmosphere, denser atmosphere, hotter atmosphere, more carbon dioxide and less oxygen), it is no wonder climate change is becoming an unavoidable problem.
Summary
We discussed ten reasons why there might be less oxygen than believed. This is part of my series exploring the theory that it is actually oxygen depletion, NOT carbon dioxide emissions, that is the major factor in climate change. I hope that I raised a few areas worth further exploration. I think figuring this all out could take a long time! But humanity has no time to lose.
If you can think of any more reasons, or you find any flaws in my thinking, please leave a comment below. Thanks.
Footnotes
See links above, plus the following:
This article explains the one above in layman’s terms. New Research Shows Plants Are Photosynthesizing More in Response to More CO2 in the Atmosphere.