Climate change feedbacks are natural processes that affect how much global temperatures will rise based on the amount of greenhouse gas emissions. Positive feedbacks increase global warming, while negative feedbacks reduce it. Feedbacks influence both the amount of greenhouse gases in the atmosphere and the amount of temperature change that occurs. While emissions are the main cause of climate change, feedbacks work together to determine how sensitive the climate is to those emissions.
Although the overall effect of feedbacks is negative, this effect is becoming weaker as greenhouse gas emissions increase. This means that warming is slower than it would be without feedbacks, but warming will speed up if emissions continue at current levels. Net feedbacks will remain mostly negative because the Earth emits more heat as it warms, an effect that is much stronger than other individual feedbacks. Therefore, human-caused climate change alone cannot lead to a runaway greenhouse effect.
Feedbacks can be divided into physical feedbacks and biological feedbacks. Physical feedbacks include reduced surface reflectivity (due to less snow and ice) and more water vapor in the atmosphere. Water vapor is a strong greenhouse gas and also affects cloud formation and atmospheric temperatures. Biological feedbacks are mainly linked to how quickly plants take in CO₂ as part of the carbon cycle. The carbon cycle absorbs more than half of all CO₂ emissions each year into plants and the ocean. Over time, the ability of these carbon sinks to absorb CO₂ will decrease as they become full and as higher temperatures cause effects like droughts and wildfires.
Scientists estimate the strength and relationships of feedbacks using global climate models, which are adjusted based on real-world data when possible. Some feedbacks quickly affect how sensitive the climate is to warming, while feedbacks from ice sheets take centuries to fully develop. Feedbacks can also create local differences, such as faster warming near the poles due to reduced snow and ice. While basic feedback relationships are well understood, there is uncertainty in some areas, especially about how clouds affect feedbacks. Uncertainty in the carbon cycle comes from the large amounts of CO₂ absorbed by plants and released when plants burn or decay. For example, melting permafrost releases both CO₂ and methane in ways that are hard to predict. Climate change models use these feedbacks to estimate how Earth will respond to greenhouse gas emissions over time, including how feedbacks will change as the planet warms.
Definition and terminology
The Planck response is the extra heat that objects release when they become warmer. Whether this is considered a climate feedback depends on the situation. In climate science, the Planck response is often seen as a natural part of warming that happens independently of other feedbacks, such as those related to radiation or the carbon cycle. However, the Planck response is still included when scientists calculate how sensitive the climate is to changes.
A feedback that increases the size of an initial change is called a positive feedback, while a feedback that decreases the size of an initial change is called a negative feedback. Climate feedbacks are always linked to global warming, so positive feedbacks make warming stronger, and negative feedbacks make warming weaker. The words "positive" and "negative" only describe how the feedback affects changes, not whether the feedback is good or bad.
The initial change that starts a feedback can come from outside the climate system, such as human activities or natural events, or it can happen naturally within the climate system. External forcing refers to changes caused by something outside the climate system, such as human actions (like greenhouse gas emissions or changes in land use) or natural events (like volcanic eruptions). These external influences can push the climate system toward warming or cooling.
Physical feedbacks
The Planck response is the most basic way the Earth's climate system reacts to changes in temperature. According to the Stefan–Boltzmann law, as the temperature of a black body increases, it emits more infrared radiation, with the amount of radiation increasing in proportion to the fourth power of its absolute temperature. This means that as the Earth warms, it sends more heat back into space. This reaction is a strong stabilizing force and is sometimes called the "no-feedback response" because it depends only on temperature and not on other factors. While Earth’s emissivity is less than that of a perfect black body, the concept of black body radiation is still useful for studying changes in the Earth’s outgoing radiation.
The Planck response, as observed in real-world data or climate models, is calculated using the Stefan–Boltzmann equation. The expected strength of this response is approximately −3.8 W/m²·K. However, some climate models show slightly weaker responses due to factors like the stratosphere’s properties, which were not fully accounted for in earlier models.
Other factors that affect Earth’s outgoing radiation, such as its "grey body" properties, are usually included in climate models as separate feedbacks. As analysis methods improve, estimates of the Planck response from models, observations, and black body theory are expected to become more consistent.
The Clausius–Clapeyron relation explains that warmer air can hold more water vapor. As the atmosphere warms, the amount of water vapor increases. This is sometimes called the specific humidity feedback because relative humidity remains nearly constant over oceans but decreases over land, where warming is faster. Since water vapor is a greenhouse gas, more water vapor in the air traps more heat, leading to further warming. This creates a cycle that continues until other cooling effects balance it. Observations from satellites estimate this feedback’s strength at 1.85 ± 0.32 W/m²·K, which is close to model estimates of 1.77 ± 0.20 W/m²·K. This feedback nearly doubles the warming caused by increases in carbon dioxide alone.
The lapse rate is how quickly temperature decreases with height in the atmosphere. It is related to how radiation interacts with the atmosphere. The lapse rate feedback is generally a cooling effect, but in polar regions, it can act as a warming effect. This happens because the surface warms faster than the air above it, reducing the atmosphere’s ability to release heat into space.
The greenhouse effect depends on how quickly temperature decreases with height. Global warming reduces this rate, weakening the greenhouse effect and creating a negative lapse rate feedback.
Albedo measures how much sunlight a surface reflects. Surfaces with high albedo, like ice and snow, reflect more sunlight and absorb less heat, cooling the planet. When ice melts, darker surfaces like land or open water replace it, absorbing more heat and causing further melting. This cycle is known as the ice-albedo feedback. This feedback is a major reason for faster warming in the Arctic, which has warmed nearly four times faster than the global average since 1979. In contrast, Antarctica’s ice sheet remains stable due to its high elevation, resulting in less warming.
As of 2021, the total surface feedback strength is estimated at 0.35 W/m²·K (with a range of 0.10 to 0.60 W/m²·K). Declining Arctic sea ice between 1979 and 2011 contributed 0.21 W/m²·K of radiative forcing, equivalent to a quarter of the warming caused by CO₂ emissions over the same period. Changes in sea ice cover between 1992 and 2018 are estimated to have contributed 10% of all human-caused greenhouse gas emissions. The strength of the ice-albedo feedback depends on the rate of ice loss, with models suggesting it will peak around 2100 before declining.
If Arctic sea ice disappears completely during summer, as predicted by some climate models, it could raise global temperatures by 0.19 °C (0.34 °F), with a range of 0.16–0.21 °C. Regional temperatures in the Arctic could rise by over 1.5 °C (2.7 °F). These estimates include effects like changes in lapse rate, water vapor, and clouds, and they do not add extra warming beyond what models already predict.
Clouds affect the climate in two ways: from below, they trap heat and warm the surface; from above, they reflect sunlight and cool the planet. Low clouds are very reflective and cause strong cooling, while high clouds are thin and trap heat, leading to warming. Overall, clouds have a cooling effect. However, climate change is expected to change cloud patterns, reducing their cooling effect and increasing global warming. While some regions may see cooling from clouds, the overall effect is a warming one.
As of 2021, cloud feedback strength is estimated at 0.42 W/m²·K (with a range of −0.10 to 0.94 W/m²·K). This wide range is due to the difficulty of observing certain cloud types, especially over oceans, which limits the accuracy of climate models.
Biogeophysical and biogeochemical feedbacks
There are both positive and negative climate feedbacks connected to Earth's carbon cycle. Negative feedbacks are large and play an important role in studies about climate inertia or how climate changes over time. These feedbacks are often considered less affected by temperature changes and may be studied separately or ignored in research that focuses on how sensitive the climate is to changes. Since the IPCC Fourth Assessment Report (AR4) in 2007, global warming predictions have included carbon cycle feedbacks. Although scientists understood these feedbacks less clearly at the time, their knowledge has improved since then. Positive feedbacks include more frequent and severe wildfires, major losses in tropical rainforests due to fires, drying, and tree loss in other areas.
The Amazon rainforest is a well-known example because of its size and importance. Damage to the rainforest from climate change is made worse by ongoing deforestation. Together, these two threats could turn much or all of the rainforest into a savannah-like area, though this would likely require a temperature rise of about 3.5 °C (6.3 °F).
Currently, land and ocean carbon sinks absorb about half of all emissions. Their ability to absorb emissions will change in the future. If emissions decrease, these sinks will absorb a larger share of remaining emissions, up to three-quarters, but the total amount absorbed will be less than today. If emissions increase, the total amount absorbed will grow, but the share absorbed could drop to one-third by the end of the 21st century. If emissions remain very high after the 21st century, carbon sinks may eventually be overwhelmed. The ocean’s ability to absorb carbon would decrease further, and land ecosystems could become a net source of carbon. In some cases, strong carbon dioxide removal efforts might cause land and ocean sinks to become net sources for several decades.
According to Le Chatelier's principle, the Earth's carbon cycle will adjust in response to human-caused CO₂ emissions. The ocean is the main force behind this, absorbing CO₂ through a process called the solubility pump. Today, the ocean absorbs about one-third of current emissions, but over centuries, most (about 75%) of human-caused CO₂ will dissolve in the ocean. However, how quickly the ocean will absorb CO₂ in the future is uncertain and may be affected by ocean warming and changes in ocean circulation. Scientists believe the strength of the global carbon sink depends largely on the Southern Ocean, especially its overturning circulation.
Over long periods, chemical weathering removes CO₂ from the atmosphere. With current global warming, weathering is increasing, showing strong connections between climate and Earth’s surface. Biosequestration also captures and stores CO₂ through biological processes. For example, ocean organisms form shells that remove CO₂ from the ocean over thousands to hundreds of thousands of years.
The growth of plants and phytoplankton increases as more CO₂ fuels their photosynthesis, a process called the CO₂ fertilization effect. Plants also use less water as CO₂ levels rise because they lose less moisture through their leaves. However, increased droughts in some regions can still limit plant growth, and extreme warming can harm plants. By the 21st century, plants are expected to grow more in high-latitude areas near the poles but less in tropical regions. Scientists are not certain whether tropical ecosystems will store more carbon than they do now, but they are confident that land carbon sinks will remain a net carbon absorber.
Global warming may change the release of gases like methane, nitrous oxide, and dimethyl sulfide. Methane from wetlands and nitrous oxide from land and oceans are known positive feedbacks. For example, warming may shift microbial communities in freshwater ecosystems to produce more methane and less carbon dioxide. Changes in vegetation, such as larch trees being replaced by spruce trees in some sub-arctic forests, can also affect Earth’s reflectivity, or albedo. Larch trees shed their needles in winter, allowing more snow to cover them, which reduces warming compared to spruce trees that keep their dark needles year-round.
Changes in emissions of compounds like sea salt, dimethyl sulfide, dust, ozone, and biogenic volatile organic compounds are expected to have mostly negative effects. As of 2021, these non-CO₂ feedbacks are believed to largely cancel each other out, but scientists have low confidence in this. Combined feedbacks could range up to 0.25 W/m² per degree Celsius in either direction.
Permafrost is not included in earlier estimates because it is difficult to model and its role depends strongly on time and warming levels. Instead, it is treated as a separate process that contributes to near-term warming. Studies suggest that total emissions from permafrost thaw will be smaller than human-caused emissions but still significant. The IPCC Sixth Assessment Report estimates that permafrost could release 14–175 billion tonnes of carbon dioxide equivalent per 1 °C (1.8 °F) of warming. For comparison, annual human-caused CO₂ emissions in 2019 were about 40 billion tonnes. A 2022 study found that if global warming is limited to 2 °C (3.6 °F), permafrost emissions over the 21st century would be similar to Russia’s 2019 emissions. Under a scenario where warming stays below 3 °C (5.4 °F), permafrost emissions would match Western Europe or the United States’ 2019 emissions. In the worst-case scenario, emissions could reach levels similar to China’s 2019 emissions.
Few studies directly link permafrost impacts to warming. A 2018 study estimated that limiting warming to 2 °C (3.6 °F) would add about 0.09 °C (0.16 °F) to global temperatures by 2100. A 2022 review suggested that each 1 °C (1.8 °F) of warming could add 0.04 °C (0.072 °F) and 0.11 °C (0.20 °F) from abrupt thaw by 2100 and 2300, respectively. At 4 °C (7.2 °F) of warming, widespread and sudden permafrost collapse could occur, adding 0.2–0.4 °C (0.36–0.72 °F) to global temperatures.
Long-term feedbacks
The Earth has two large ice sheets: the Greenland ice sheet and the Antarctic ice sheet. These ice sheets cover the world's largest island and an entire continent. On average, both ice sheets are about 2 km (1 mi) thick. Because they are so large, their response to warming takes thousands of years and happens in two stages.
The first stage involves the effect of melting ice on ocean currents. When ice melts, it adds fresh water to the ocean. This fresh water makes it harder for the top layer of ocean water to sink, which disrupts the movement of oxygen, nutrients, and heat between different layers of the ocean. This process can slow down warming, but research on this effect is limited. A second, longer-term effect happens when ice sheets change their reflectivity, or albedo, in response to long-term temperature changes. If warming continues, this change would increase warming further.
If the Greenland ice sheet completely melts, it is estimated to add about 0.13 °C (0.23 °F) to global warming. The loss of the West Antarctic ice sheet is expected to add about 0.05 °C (0.090 °F), while the loss of the East Antarctic ice sheet could add about 0.6 °C (1.1 °F). The melting of the Greenland ice sheet would also raise temperatures in the Arctic by 0.5 °C (0.90 °F) to 3 °C (5.4 °F). In Antarctica, temperatures could rise by 1 °C (1.8 °F) after the loss of the West Antarctic ice sheet and 2 °C (3.6 °F) after the loss of the East Antarctic ice sheet.
These estimates assume that global warming stays at an average of 1.5 °C (2.7 °F). The impact of ice loss on warming depends on the level of global warming. At lower warming levels, like those in the 2020s, the effect of ice loss may be greater. However, if warming increases further, the effect may become smaller. If long-term warming reaches about 1.5 °C (2.7 °F), the Greenland and West Antarctic ice sheets are likely to melt completely. The East Antarctic ice sheet would not be at risk of melting entirely until global warming reaches very high levels, such as 5–10 °C (9.0–18.0 °F).
Methane hydrates, also called methane clathrates, are frozen structures where methane is trapped inside a water crystal. These structures are found beneath ocean sediments, usually about 1,100 m (3,600 ft) below the ocean surface. In 2008, scientists were worried that methane hydrates in shallow Arctic areas, such as the East Siberian Arctic Shelf, might release large amounts of methane quickly, possibly causing a rapid rise in global temperatures. However, current research shows that methane hydrates react slowly to warming, and it is difficult for methane to reach the atmosphere after it is released from the seafloor. Because of this, no significant impact on global temperatures from methane hydrates is expected this century. Some studies suggest that over thousands of years, hydrate release could cause a warming of 0.4–0.5 °C (0.72–0.90 °F).
Forcing-feedback formulation of climate sensitivity
Earth is an energy system where long-term temperature changes depend on the global energy imbalance (EEI). This imbalance is calculated by comparing absorbed solar radiation (ASR) and outgoing longwave radiation (OLR) at the top of Earth's atmosphere. When EEI is positive, Earth is warming; when it is negative, Earth is cooling; and when it is close to zero, there is no significant warming or cooling. The ASR and OLR values include many temperature-related properties and complex interactions that affect how the system behaves.
To understand how the system responds to changes near a stable equilibrium, scientists study small changes in EEI, represented by the symbol Δ. These changes are often caused by radiative forcing (ΔF), which can be natural or human-made. The system's response to these changes—either returning to stability or moving further away—is called feedbacks, represented by λΔT.
A feedback is a process that affects how energy moves within the system, while a forcing is an action that changes the system's energy balance, based on thermodynamic principles.
Feedbacks can be described using a simplified value called λ and the temperature change ΔT. This is because all parts of λ (which are assumed to act independently and add up) depend on temperature in different ways, as defined for thermodynamic systems.
Important feedbacks that influence EEI include:
– wv = water vapor,
– c = clouds,
– a = surface albedo (how much sunlight Earth reflects),
– cc = carbon cycle,
– p = Planck response (how Earth emits heat based on temperature),
– lr = lapse rate (how temperature changes with altitude).
These values represent global averages. The temperature (T) is often measured at Earth's surface because it directly affects humans and other life.
The Planck response is a strong temperature-dependent effect. It is sometimes separated from other feedbacks to focus on the relative strength of other components, called feedback gains (g_i). For example, the water vapor feedback gain (g_wv) is approximately 0.5.
In modern climate models, the simplified approach has limited use. However, it helps compare the strength of different feedbacks. When the net feedback remains negative and the system reaches a new stable state (ΔEEI = 0) after some time, scientists can estimate how sensitive Earth's climate is to a forcing.
Implications for climate policy
Uncertainty about how climate change affects the environment can influence decisions about how to reduce greenhouse gas emissions. For example, uncertainty about how the carbon cycle changes may impact goals for lowering emissions. These goals are often based on aiming to keep greenhouse gas levels in the atmosphere at a certain level or limiting how much the planet warms. Both of these goals require knowing how the carbon cycle will change in the future.
If predictions about future changes in the carbon cycle are incorrect, then goals for keeping greenhouse gas levels or limiting warming may not be reached. For instance, if models do not accurately predict how much carbon will be released into the atmosphere from effects like melting permafrost, they might also underestimate how much emissions need to be reduced to meet a goal for greenhouse gas levels or temperature limits.