Earth has experienced dramatic natural warming and cooling cycles throughout its history, driven by shifts in its orbit, volcanic eruptions, and changes in solar output. But the warming observed over the past century does not match the fingerprint of any of those natural drivers. Formal climate attribution studies find that natural forces contributed negligibly to the warming recorded between 2010 and 2019, while human-caused greenhouse gas emissions accounted for essentially all of it. The question is worth taking seriously, because the natural cycles are real and well-documented, and understanding how they work makes it easier to see why scientists are so confident the current trend is something different.
Earth’s Orbital Cycles and the Ice Ages
The most powerful natural climate cycles operate over tens of thousands of years. Slow wobbles and stretches in Earth’s orbit around the Sun change how much sunlight reaches different parts of the planet at different times of year. These shifts come in three main flavors: a roughly 100,000-year cycle in how elliptical the orbit is, a roughly 42,000-year cycle in the tilt of Earth’s axis, and a roughly 23,000-year cycle in the wobble of that axis. Together, they pace the advance and retreat of ice sheets. A landmark 1976 study of deep-sea sediment cores found that about 50 percent of climatic variance over the past several hundred thousand years was concentrated at the 100,000-year period, with another 25 percent at 42,000 years and 10 percent at 23,000 years, each matching the corresponding orbital period.1PubMed. Variations in the Earth’s Orbit: Pacemaker of the Ice Ages Later work confirmed that these climatic rhythms are phase-locked to the orbital changes, meaning the climate shifts happen when the orbital geometry predicts they should.2Reviews of Geophysics. Milankovitch Theory and climate
These orbital cycles are the reason Earth has spent the last few million years swinging between ice ages and warmer interglacial periods like the one we live in now. They are genuine, well-understood natural warming cycles. But they operate on timescales of millennia, not decades. The current interglacial period, the Holocene, has been relatively stable for about 10,000 years. Earth’s orbital configuration today would, if anything, be nudging the planet slowly toward cooler conditions over the next several thousand years, not producing the rapid warming observed since the mid-twentieth century.
What About the Sun?
If the Sun were getting brighter, that would be a natural explanation for warming. And solar output does vary, mainly on an 11-year cycle. An early reconstruction of solar irradiance since 1610 suggested that changes in solar brightness may have contributed about half of the surface warming observed between 1860 and 1970.3Geophysical Research Letters. Reconstruction of solar irradiance since 1610: Implications for climate change That finding was influential for a time and is still sometimes cited by people arguing that the Sun drives modern warming.
The problem is that since the late 1970s, satellites have been directly measuring the Sun’s energy output, and it has not increased. On multi-decade averages, total solar irradiance has stayed within a narrow band, implying a temperature effect of less than one hundredth of a degree Celsius over the past 30 years.4Journal of Space Weather and Space Climate. Changes in the Total Solar Irradiance and climatic effects Earth’s temperature, meanwhile, has climbed steeply over the same period. The Sun and global temperature were moving in the same direction for much of the early twentieth century, but they decoupled after about 1980. One analysis projects that solar irradiance may contribute a small additional increment to future warming, on the order of a few percent of the total trend, but that is a modest supplement to greenhouse gas forcing, not a replacement for it.5Advances in Space Research. Solar irradiance, climatic indicators and climate change – An empirical analysis
Volcanoes and Ocean Patterns
Volcanic eruptions can temporarily cool the planet by injecting reflective particles into the upper atmosphere. The 1991 eruption of Mount Pinatubo, for example, produced measurable global cooling that lasted a couple of years.6Earth-Science Reviews. Evaluating the relationship between climate change and volcanism A string of smaller eruptions between 2000 and 2013 may have contributed a modest cooling of a few hundredths of a degree, partially masking the warming trend for a few years, though they were not enough to halt it.7Atmospheric Science Letters. The impact of volcanic eruptions in the period 2000–2013 on global mean temperature trends evaluated in the HadGEM2‐ES climate model Volcanoes are a real natural climate force, but they cool rather than warm, and their effects fade within a few years.
Ocean circulation patterns like El Niño and La Niña redistribute heat between the ocean and the atmosphere and can shift global temperatures up or down by a fraction of a degree from year to year. Some researchers have argued that multidecadal oscillations in the Pacific and Atlantic oceans could account for much of the observed warming trend.8Energy & Environment. Multidecadal Tendencies in ENSO and Global Temperatures Related to Multidecadal Oscillations These patterns are real and they do create bumps and dips in the temperature record, which is why individual years or even individual decades can look warmer or cooler than the long-term trend would suggest. But ocean oscillations move heat around within the climate system; they do not add new energy to it. They can make a warming trend look faster or slower in the short run without being the underlying cause.
The Medieval Warm Period and the Little Ice Age
Two episodes from the past thousand years come up constantly in debates about natural climate variability. The Medieval Warm Period, roughly 800 to 1300 AD, saw temperatures warm enough for Viking settlements in Greenland and expanded agriculture in parts of Europe. The Little Ice Age, roughly 1400 to 1900 AD, brought harsh winters, advancing glaciers, and crop failures. Paleoclimate records from Chesapeake Bay sediments show temperature shifts of about two to four degrees Celsius during these periods, possibly linked to changes in Atlantic Ocean circulation.9Global and Planetary Change. Medieval Warm Period, Little Ice Age and 20th century temperature variability from Chesapeake Bay
These events are sometimes invoked to argue that today’s warming is just another swing in a natural rhythm. But a closer look complicates that story. The Medieval Warm Period was not globally uniform; different regions warmed at different times, and the peak warmth in any single region was generally below what we see today across the globe simultaneously. The Little Ice Age, meanwhile, now appears to have been driven in large part by a cluster of major volcanic eruptions. Recent modeling shows that frequent strong volcanism during the Little Ice Age overwhelmed the temperature patterns normally associated with Atlantic Ocean variability, making the cooling look more like a volcanically forced event than a pure ocean cycle.10Geophysical Research Letters. Volcanic Forcing Overwhelmed AMV’s Influence on Eurasian Winter Temperatures During the Little Ice Age Neither event was a simple, unexplained oscillation. Both had identifiable physical causes, and those causes are not operating in the same way today.
How We Know the Carbon Is From Fossil Fuels
One of the strongest pieces of evidence that modern warming is human-caused comes not from temperature records but from the chemistry of the atmosphere itself. Carbon atoms come in different isotopes, and fossil fuels have a distinctive isotopic signature. Plants preferentially absorb lighter carbon-12 over heavier carbon-13 during photosynthesis, so plant-derived material, including the fossil fuels formed from ancient plants, is depleted in carbon-13. Fossil fuels are also completely devoid of carbon-14, which decays over thousands of years. When you burn coal, oil, or gas and release that carbon into the atmosphere, the ratio of carbon-13 to carbon-12 in atmospheric CO₂ drops, and so does the ratio of carbon-14.11PubMed Central. Changes to Carbon Isotopes in Atmospheric CO(2) Over the Industrial Era and Into the Future
Both of those signatures have been measured. Ocean surface water in the Pacific showed a decline in carbon-13 of dissolved carbon between 1970 and 1990, consistent with the uptake of fossil-fuel-derived CO₂ from the atmosphere.12PubMed. Oceanic Uptake of Fossil Fuel CO2: Carbon-13 Evidence This isotopic fingerprint rules out natural sources like volcanoes or ocean outgassing as the main contributors to rising CO₂. Volcanic carbon, for instance, does not have the same depleted carbon-13 signal. The atmosphere is filling up with carbon that has the specific chemical stamp of burned ancient organic material.
Satellite Observations of the Greenhouse Effect in Action
Beyond chemistry, satellites have directly observed the greenhouse effect strengthening in real time. Continuous measurements of outgoing longwave radiation from 2003 to 2021 show decreases in the energy escaping to space at the specific wavelengths absorbed by CO₂, methane, and nitrous oxide. At the same time, energy escaping through atmospheric “windows” where those gases do not absorb has increased, along with water vapor absorption bands. Radiative transfer simulations confirm that the observed changes match what you would expect from the measured increases in greenhouse gas concentrations.13Geophysical Research Letters. Greenhouse Gas Forcing and Climate Feedback Signatures Identified in Hyperspectral Infrared Satellite Observations In plain terms, scientists can watch the planet’s energy budget being altered in exactly the way greenhouse gas physics predicts.
The Atmospheric Fingerprint That Only Greenhouse Gases Produce
If the Sun were causing the warming, you would expect every layer of the atmosphere to warm. More solar energy coming in means more heat everywhere. What scientists actually observe is the opposite pattern in the upper atmosphere: the troposphere, the lower layer where weather happens, is warming, while the stratosphere above it is cooling. This was predicted in 1967 as a signature specific to greenhouse gas warming, and it has now been confirmed by decades of weather balloon and satellite temperature data.14PubMed Central. Exceptional stratospheric contribution to human fingerprints on atmospheric temperature The cooling gets more pronounced higher in the stratosphere, exactly as theory predicts.15AGU Advances. Modeled and Observed Stratospheric Temperature Changes: Implications for Fingerprint Studies
The logic behind this is straightforward. Greenhouse gases trap outgoing heat in the lower atmosphere, warming it. But that same trapping means less heat reaches the stratosphere, so it cools. An increase in solar output would warm both layers. The fact that the stratosphere is cooling while the surface warms is about as close to a smoking gun as climate science gets. No natural mechanism proposed to date replicates this pattern.
Another telling pattern involves the daily temperature range. Over the late twentieth century, nighttime temperatures over land rose faster than daytime temperatures, narrowing the gap between daily highs and lows.16Communications Earth & Environment. Historical diurnal temperature range trends constrain future climate projections This is consistent with an enhanced greenhouse effect, which acts around the clock, warming nights more than days. Solar forcing, by contrast, would primarily affect daytime temperatures. Early work on this asymmetry noted that the cause was uncertain and could involve changes in cloud cover, but the pattern itself has been robustly observed across the Northern Hemisphere.17Geophysical Research Letters. Global warming: Evidence for asymmetric diurnal temperature change
Formal Attribution Studies
Putting all of this together, climate scientists run formal attribution analyses that compare observed warming against what models predict from natural forces alone and from human forces alone. The most direct finding: anthropogenic forcings caused between 0.9 and 1.3 degrees Celsius of warming in global mean temperature from 2010 to 2019, relative to 1850–1900. The observed warming over that period was about 1.1 degrees. Natural forcings, including solar and volcanic influences, contributed negligibly.18Nature Climate Change. Constraining human contributions to observed warming since the pre-industrial period The fact that the anthropogenic contribution slightly exceeds the observed warming reflects the offsetting cooling effect of human-produced aerosols, which mask some of the greenhouse gas warming.
Ice Cores and the CO₂ Lag Argument
A common counterargument goes like this: ice cores show that in past climate cycles, temperature rose before CO₂ did, so CO₂ must be a consequence of warming, not a cause. The underlying observation is real. High-resolution Antarctic ice cores show that CO₂ concentrations rose by 80 to 100 parts per million roughly 600 years after the onset of warming during the last three deglaciations.19PubMed. Ice core records of atmospheric CO2 around the last three glacial terminations The Vostok ice core, which extends back 420,000 years, confirms that CO₂ and temperature track each other closely through multiple glacial cycles, and that current atmospheric CO₂ levels are unprecedented in that entire record.20Nature. Climate and atmospheric history of the past 420,000 years from the Vostok ice core, Antarctica
But the lag does not mean what the counterargument implies. In natural glacial cycles, the initial warming trigger is orbital. Small orbital changes warm the planet slightly, which causes the oceans to release CO₂ (cold water holds more dissolved gas than warm water). That released CO₂ then amplifies the warming, which releases more CO₂, in a feedback loop. CO₂ acts as both a response and an amplifier. The fact that it can be nudged upward by warming does not mean it cannot also cause warming when added independently, which is exactly what fossil fuel combustion does. Today, the initial trigger is not orbital. Humans are adding CO₂ directly, and the physics of how it traps heat works the same regardless of where the carbon came from.
Deep-Time Parallels
For a sense of what happens when massive amounts of carbon enter the atmosphere quickly through natural means, the Paleocene-Eocene Thermal Maximum, or PETM, about 56 million years ago, is the closest geological analogue. Global temperatures spiked by several degrees over a few thousand years. Recent isotopic work points to volcanism associated with the North Atlantic Igneous Province as the main carbon source, releasing more than 10,000 petagrams of carbon.21Nature. Very large release of mostly volcanic carbon during the Palaeocene–Eocene Thermal Maximum Climate modeling suggests that pre-PETM CO₂ levels were already elevated, and the additional volcanic carbon pushed the system into extreme warming.22Paleoceanography. The Paleocene‐Eocene Thermal Maximum: How much carbon is enough?
What makes the PETM relevant is not just the warming itself but the rate. Even so, the carbon release during the PETM played out over thousands of years. Some evidence suggests that just before the main event, a rapid pulse of carbon release caused ocean surface warming of at least 2 degrees Celsius and a measurable drop in ocean pH, in a pattern that researchers note is more similar in rate to modern anthropogenic emissions than the main PETM event was.23PubMed Central. Surface ocean warming and acidification driven by rapid carbon release precedes Paleocene-Eocene Thermal Maximum The current rate of carbon release exceeds anything in the geological record that scientists have been able to resolve. That is not a comforting comparison.
The Cosmic Ray Hypothesis
One alternative natural explanation that gained attention in the 2000s proposed that cosmic rays from deep space seed cloud formation. The idea was that during periods of low solar activity, more cosmic rays reach Earth, creating more clouds, which reflect sunlight and cool the planet. If true, this would mean the Sun indirectly controls climate through a mechanism not captured in standard solar irradiance measurements. It was a creative hypothesis, and it motivated real experimental work.
The evidence has not been kind to it. A study examining the relationship between galactic cosmic rays and low cloud cover found that over midlatitude Eurasia, the data actually showed the opposite of what the cosmic ray theory predicts: cloud cover decreased when cosmic rays increased, arguing against the idea that cosmic ray ionization enhances cloud formation.24Scientific Reports. The influence of solar-modulated regional circulations and galactic cosmic rays on global cloud distribution The hypothesis has not been definitively buried, but it has failed to produce the kind of robust observational support that would make it a serious competitor to greenhouse gas forcing.
Feedback Loops and Why They Matter
Whether warming starts naturally or through human emissions, once it gets going, the same set of feedback loops kicks in. Warming in the Arctic melts sea ice, exposing darker ocean water that absorbs more sunlight, which causes more warming.25One Earth. Many risky feedback loops amplify the need for climate action Permafrost soils in high latitudes contain enormous amounts of organic carbon. As they thaw, microbes break down that carbon and release CO₂ and methane, adding more greenhouse gas to the atmosphere. One modeling study projected that permafrost thaw and related changes could increase Arctic methane emissions substantially, though the exact magnitude remains uncertain.26PubMed Central. Permafrost carbon-climate feedbacks accelerate global warming The timing and scale of permafrost carbon release are among the bigger unknowns in climate science, but the direction of the feedback is clear: warming unlocks more carbon, which drives more warming.27Nature. Climate change and the permafrost carbon feedback
These feedbacks are the same ones that amplified orbital warming during past ice age cycles. The difference is that in those cycles, the initial nudge was tiny and the feedbacks unfolded over thousands of years. Today, the initial push is large and fast, and the feedbacks are being activated on top of it. Earth’s energy imbalance, the gap between incoming and outgoing energy, has been confirmed by measurements of increasing ocean heat content.28PubMed. Earth’s energy imbalance: confirmation and implications The ocean has absorbed the bulk of the extra energy, acting as a buffer that temporarily slows surface warming but stores heat that will influence the climate for centuries.29PubMed Central. Improved estimates of ocean heat content from 1960 to 2015
Where the Uncertainty Actually Lives
The science is not uncertain about whether human activity is driving modern warming. That part is settled about as firmly as anything in Earth science gets. Where genuine uncertainty persists is in how much warming to expect going forward, because that depends on future emissions choices and on the precise strength of feedbacks that are difficult to pin down. Projections for warming through the end of the twenty-first century range from about 1.0 to 3.7 degrees Celsius depending on emission scenarios, based on the ensemble results of current climate models. The spread in that range comes largely from how models handle clouds, aerosol interactions, and carbon cycle feedbacks, not from any remaining doubt about the basic physics of greenhouse gas warming.
Permafrost thaw is a good example. Scientists know it will release carbon. They know it will amplify warming. But how much, and how fast, depends on temperature thresholds and microbial processes that are hard to model at global scales. Similarly, the behavior of ice sheets in Greenland and West Antarctica could accelerate sea level rise beyond current central estimates, but the timing of potential rapid ice loss is genuinely uncertain. These are the frontiers of the science, and they matter enormously for planning. They do not, however, reopen the question of whether the warming itself is natural.