The Keeling Curve

Atmospheric Carbon Dioxide Variations at the Mauna Loa Observatory, Hawaii; By Charles D. Keeling, Robert B. Bacastow, Arnold E. Bainbridge, Carl A. Ekdahl, J.R., Peter R. Guenther, & Lee S. Waterman, Scripps Institution of Oceanography, University of California at San Diego, La Jolla, California, U.S.A & John F. S. Chin, National Oceanic & Atmospheric Administration, Mauna Loa Observatory, Hawaii (January 27th, 1976). Keeling 1976


Many of the pivotal papers I can foresee myself dissecting here will be loaded with a plethora of content that could never be covered in one fell swoop. I hope that as I do this longer that these summaries will contain less tangental material and become more concise. However, it would be a disservice to exclude topics that contribute to the big picture. Eventually I will find a balance.

This paper was sourced from CarbonBrief, a speciality website dedicated to, “covering the latest developments in climate science, climate policy, and energy policy.” Their article The most influential climate change papers of all time reached out to climate experts and scientific leaders to get their opinion on the matter. A common metric used to measure a paper’s success is to see how many times other papers cite or reference it in their own work. However, in this instance they asked for subjective opinions and compiled the responses here: Who Said What? They selected to discuss the papers that were referenced most often.


The Mauna Loa Observatory, located on ‘The Big Island’ of Hawaii, collected atmospheric COsamples uninterrupted for fourteen years. Collaborators compiled the data and analyzed it using a variety of methods to isolate atmospheric changes from local influences to ultimately generate the Keeling Curve. The results show a seasonal oscillation in CO2 levels, as well as a long term trending increase as a consequence of fossil fuel combustion.

The observatory sits high above the island of Hawaii, far removed from industrial activities and anthropogenic influences. Additionally, much of the area surrounding the base of the active volcano is covered in forest reserve, relegating human development to the edges of the island. Even still, the first half of the paper is spent trying to address and correct for local impacts.

The term ‘diurnal‘ can be used to describe a variety of behaviors or outcomes that fluctuate or occur on a daily schedule. For example, organisms living in deeper sections of the ocean perform diurnal vertical migrations, whereby they travel from the depths to the surface every day around dawn to feed under the cover of night only to return at dawn to avoid predation. The slope of Mauna Loa produces a similar diurnal fluctuation in CO2 concentrations. At night, low humidity air blows downslope, which produces steady COreadings. However, occasional and sporadic spikes in COlevels are seen. They attribute this to volcanic venting from a location ~4km upslope. During the day, an upwind draft brings in high humidity air rich in CO2 that peaks around midday. Around 6:00 P.M. the COconcentrations are lowest and humidity is highest.

Steady, accurate nightly readings can be attributed to atmospheric mixing and minimal upstream contamination. Just like food coloring dripped into water, the longer the two are allowed to mingle the more diffuse the coloring becomes until it eventually occupies the whole glass and evenly distributes itself. Air blow down from above is an accurate representation of global levels because it’s sourced from a homogenous air column high above the planet. On the other hand, air samples being blow up from below are subject to a plethora of different influences. The most obvious of these are human sources like industry, agriculture, and combustion. The less obvious are those expansive forest ‘buffers’.

Let’s take a trip down memory lane to review energy production and acquisition:

Photosynthesis – 6CO2 + 6H2O + Light → C6H12O6 + 6O2

Respiration – C6H12O6 + 6O2 → 6CO2 + 6H2O + ATP (Energy)

Plants perform photosynthesis and respiration simultaneously during the day, but cease to photosynthesize when the sun sets. Respiration continues into the night because, just like humans, plants need energy around the clock. Stomata are minuscule openings on the underside of every leaf that number in the thousands and help regulate COuptake as well as water loss – transpiration. Since photosynthesis is discontinued at night, the stomata close to prevent water loss, which could lead to the plant drying out – desiccation. However, this traps internal COfrom escaping until the morning. Without photosynthesis to actively consume atmospheric, local, or internal CO2, it begins to accumulate. So, when dawn finally arrives the stomata open en masse and a wave of CO2 blows up the volcano to the observatory. Additionally, anthropogenic sources can be clearly identified in the data series.

Midday peaks can be attributed to automobile combustion such as when the observatory held a three day ‘open house’. Likewise, the researchers observed long term trends in the frequency of COpeaks above a predetermined threshold. These peaks correlated strongly with the completion of a nearby paved road. Counter to this, the frequency dropped when the observatory instituted a locked gate at the access road in March 1971.

Mauna Loa CloseMauna Loa Far

‘Reference gas calibration’ and ‘Secular increase in CO2‘ are two sections that can be wholly ignored. Unless you feel like inducing a headache, these sections are primarily reserved for experts. The former outlines how the “dual detector infrared analyzer” – the instrument used to measure atmospheric COlevels – remained accurate by calibrating it with reference gases. These gases included pure nitrogen, mixtures of 20, 40, and 60% oxygen in nitrogen, and argon. The latter discusses two methods used to, “separate the annual increase of about 0.8 ppm from the 6 ppm seasonal variation found in the Mauna Loa record” (547). From here it gets complicated fast. They use a plethora of statistic terms and tests like power series, Fourier harmonics, moving averages, and cubic trend functions. These steps of the process are extremely important, but also a matter of trust more than anything. Even with my science background, albeit still relatively undeveloped in the grand scheme, I quickly become lost. However, we shouldn’t disparage the paper or its findings because we can comprehend it. A hallmark of the scientific method is reproducibility and it’s what lets me rest easy. Reproducibility is the ability of a researcher to perform an experiment, their own or another’s, and verify the original outcome. You can expect that the more a new finding shatters previously held beliefs, the more independent verifications there will be. It is because of this that I don’t feel the need to completely comprehend both sections mentioned above.

Although reproducibility should be a cornerstone of experimentation, it isn’t always that way. The website Nature discusses the results of their survey: 1,500 Scientists Lift the Lid on Reproducibility, which I recommend you peruse even if it’s only for the video and infographics.

Keeling Curve

Above you can see the famous Keeling Curve in all its glory, but why the sawtooth pattern? Tim Lueker, a research scientist at the Scripps COResearch Group, explains in one sentence, “Springtime comes in May in Siberia.” You can read his concise explanation or mine if you prefer, or you can just consider him a riddler.

The abscission of tree leaves in the fall leads to the microbial decomposition of leaf litter through winter. The microbes respire and produce CO2, which elevates atmospheric levels. It’s not until May when springtime returns that plants regenerate their lost foliage and begin drawing down COfrom the atmosphere once again. This is why COconcentrations peak in May. Lueker goes onto explain that spring and summer in the Northern Hemisphere doesn’t cancel out fall and winter in the Southern Hemisphere for two main reasons. Mixing within a hemisphere is much faster than between hemispheres. Additionally, the Northern Hemisphere contains a much larger proportion of Earth’s land mass, including Russia’s expansive Siberian wilderness. As a result, the CO2 readings will naturally swing further at Mauna Loa, located in the Northern Hemisphere. The paper describes the phenomenon here, “Statistical analysis of daily and monthly averages of the original continuous data establish a seasonal oscillation described by annual and semiannual harmonic oscillations about a long term trend” (550).

I want to close by quickly discussing the greenhouse effect and how climate differs from weather. Greenhouses work by allowing light energy to pass through glass and hit an internal opaque surface, like plants, potters, or the floor, whereby it transforms into heat. A portion of the incoming energy escapes depending on its wavelength, but, overall, the system is efficient enough to keep a greenhouse warm in the winter and foster life (Hydroponics Simplified). The Australian Government’s Department of the Environment & Energy has an excellent graphic to demonstrate how this phenomenon works on our atmosphere. It’s important to remember that the greenhouse effect is vital to support life on Earth. However, anthropogenic factors are exaggerating it more than intended by artificially introducing more greenhouse gases than normal. These gases include methane, water vapor, ozone, nitrous oxide, CFCs, and of course carbon dioxide. There are certainly other reports dedicated to the dissemination of each gases’ contribution to the problem, but the focus is usually CObecause of its leading role. The EPA provides a breakdown of each greenhouse gas – points of origin, global warming potential, emission rates, etc.

‘Weather’ is short-term, eradicate, and more or less unpredictable while ‘climate’ is a longer-term average based on energy balances and is almost entirely foreseeable. Neil deGrasse Tyson and his dog illuminate the distinction. I admittedly don’t keep up with politics as much as I should, but Senator Jim Inhofe’s attempt to disprove global warming by bringing a snowball to the Senate floor is perhaps my favorite (The Washington Post).

If you want to understand the nuances between climate change and global warming, see the The Climate Reality Project website. I particularly like the paragraph that discusses our outdated use of the phrase, “inadvertent climate modification.”

Keeling Curve Today
The Keeling Curve Today


“The air at Mauna Loa Observatory may be slightly influenced by local processes which cannot be expunged from the record, but the observed long term trend of rising CO2 appears clearly to be in response to increasing amounts of industrial CO2 in the air on a global scale” [my italics] (550).

Global atmospheric concentrations of COare increasing and the Keeling Curve produced by the Mauna Loa Observatory, Scripps, and NOAA proves that. The seasonal oscillations are a result of the natural change in plant coverage in the fall/winter and spring/summer months, but the long-term rise is a direct result of anthropogenic fossil fuel emissions. Scientists around the world are working hard to determine how this will impact wildlife and how to curb the curve through different carbon sequestration methods and alternative energy sources (A New Way to Remove COFrom the Atmosphere).

If any of this is confusing or ambiguous, if you think I have poorly explained a topic, or if you think I have altogether incorrectly analyzed or presented material, please let me know. This is new to me, so I’m learning as well. Again, I highly encourage everyone to read the 1976 paper themselves, but maybe skip the two sections I mentioned early.

Additional Resources:

Climate Change Debate: Last Week Tonight with John Oliver

Quantifying the Consensus on Anthropogenic Global Warming in the Scientific Literature – Skeptical Science – CO2 has Been Higher in the Past

6 Claims Made by Climate Change Skeptics – And How to Respond

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s