Causes of climate change
The causes of climate change comprise natural and anthropogenic factors that disrupt Earth's radiative energy balance, thereby altering global surface temperatures and atmospheric circulation patterns over timescales ranging from decades to millennia. Natural drivers include variations in solar irradiance, which have fluctuated by approximately 0.1% over the 20th century, major volcanic eruptions that temporarily cool the planet via stratospheric sulfate aerosols, and orbital forcings such as eccentricity, obliquity, and precession that operate on 20,000–100,000-year cycles.[1][2] Anthropogenic influences primarily stem from elevated concentrations of greenhouse gases—carbon dioxide surpassing 420 ppm, methane exceeding 1,900 ppb—resulting from fossil fuel combustion, cement production, deforestation, and livestock emissions, alongside aerosol effects from industrial pollution that partially offset warming.[3][4] Empirical data from ice cores, satellite measurements, and instrumental records document a global temperature increase of about 1.1°C since the late 19th century, with attribution analyses apportioning most post-1950 warming to human-induced greenhouse gas forcings after accounting for natural variability.[5] However, early 20th-century warming aligns more closely with solar and oceanic oscillations, and discrepancies persist between observed trends and climate model simulations, which often overestimate recent warming rates when natural forcings are included.[6][7] Key controversies center on the magnitude of equilibrium climate sensitivity—the expected long-term temperature response to doubled atmospheric CO2—which empirical assessments place in a broad range of 1.5–4.5°C, with some instrumentally derived estimates favoring the lower end and highlighting amplified natural feedbacks like solar or cloud responses over model-derived values.[8][7] Institutions such as the IPCC emphasize high-confidence anthropogenic dominance, yet critiques note systemic overreliance on tuned models that underperform in replicating paleoclimate variability or mid-century cooling phases influenced by aerosols and ocean dynamics, underscoring uncertainties in causal attribution amid biased funding and publication incentives in academic climate research.[9][7]Fundamentals of Climate Dynamics
Radiative Forcing and Energy Balance
The Earth's climate system achieves thermal equilibrium when the energy absorbed from incoming shortwave solar radiation balances the outgoing longwave infrared radiation emitted to space. At the top of the atmosphere, the global average incoming solar flux is approximately 340 W/m², with about 100 W/m² reflected back to space by clouds, aerosols, atmospheric gases, and the surface, leaving roughly 240 W/m² absorbed by the Earth-atmosphere system. This absorbed energy is re-emitted as longwave radiation, maintaining a stable average surface temperature of about 288 K (15°C).[10][11][12] The natural greenhouse effect, primarily from water vapor, carbon dioxide, and methane, intercepts a portion of this outgoing longwave radiation, absorbing and re-emitting it downward, which reduces the net flux to space by about 150 W/m² compared to a planet without an atmosphere. This trapping effect raises the surface temperature by approximately 33°C above the effective radiating temperature of 255 K that would prevail absent such absorption. Without this baseline forcing, Earth would be uninhabitably cold.[13][14] Radiative forcing measures the perturbation to this energy balance induced by changes in atmospheric composition, solar input, surface albedo, or other factors, expressed as the net change in radiative flux (in W/m²) at the tropopause. Instantaneous radiative forcing assumes fixed temperatures except in the stratosphere, capturing the direct flux change from the perturbation. A positive forcing creates a temporary energy surplus, prompting system warming that increases outgoing radiation until balance is restored, while negative forcing induces cooling.[15][16] Effective radiative forcing (ERF) refines this by including rapid adjustments, such as alterations in cloud cover, water vapor distribution, or lapse rate, that occur in response to the initial perturbation but before substantial global temperature shifts. ERF provides a more accurate indicator of the eventual temperature response, as it accounts for these near-immediate atmospheric and surface responses. Observational data from satellite measurements indicate Earth's current energy imbalance—effectively the net ERF after feedbacks—stands at about 0.9 W/m² averaged over 2005–2019, having roughly doubled from earlier decades due to accumulating forcings. This surplus drives heat accumulation primarily in the oceans.[17][18]Role of Natural Cycles in Baseline Variability
Natural climate variability arises primarily from internal oscillations within the Earth's climate system, including interactions between the atmosphere and oceans, as well as external drivers like periodic changes in solar output. These processes generate fluctuations in global mean surface temperature (GMST) and other metrics on timescales from years to centuries, establishing a baseline range of variability that predates significant anthropogenic influences. For instance, the El Niño-Southern Oscillation (ENSO) cycle, with a typical periodicity of 2–7 years, drives interannual GMST anomalies of approximately 0.1–0.2 °C during strong events, as sea surface temperatures in the equatorial Pacific redistribute heat globally via atmospheric teleconnections.[19][20] Similarly, the 11-year solar cycle modulates total solar irradiance by about 1 W/m² at the top of the atmosphere, correlating with GMST variations of roughly 0.1 °C, as evidenced in 150 years of global sea surface temperature records.[21] Longer-term modes, such as the Pacific Decadal Oscillation (PDO) with 20–30 year phases and the Atlantic Multidecadal Oscillation (AMO) with 60–80 year cycles, further contribute to baseline multidecadal variability. The PDO influences Pacific Ocean temperatures and modulates ENSO activity, leading to GMST shifts of about 0.1–0.2 °C across its positive and negative phases, while the AMO drives North Atlantic sea surface temperature differences of 0.3–0.4 °C between phases, with downstream effects on global circulation patterns.[22][23] Analyses of instrumental records from 1880 to 2017 indicate that such internal modes, including AMO and PDO analogs, account for roughly 30% of observed multi-decadal GMST changes, underscoring their role in setting the envelope of natural fluctuations against which trend detection occurs.[23] These cycles demonstrate that climate exhibits inherent dynamism, with variability often rivaling or exceeding forced signals on decadal scales, as seen in non-monotonic 20th-century warming patterns even after low-pass filtering.[24] Attribution studies must therefore isolate anthropogenic forcings from this baseline by employing statistical methods, such as optimal fingerprinting, that account for mode-specific patterns; failure to do so risks conflating internal noise with external drivers.[25] Empirical reconstructions from proxies like tree rings and sediments confirm that pre-industrial variability aligns with these mechanisms, providing context for assessing whether modern excursions fall within or beyond historical norms.[26]Natural Causes
Solar Activity and Irradiance Changes
Solar activity encompasses phenomena such as sunspots, faculae, and coronal mass ejections, which vary cyclically, primarily on an 11-year Schwabe cycle driven by the Sun's dynamo. These variations modulate total solar irradiance (TSI), the integral solar energy flux at Earth's orbit, with cycle-amplitude fluctuations of approximately 1 W/m² (0.1% of the mean TSI value of 1361 W/m²).[27] Longer-term modulations occur over decades to centuries, reconstructed via proxies including sunspot records, ¹⁰Be and ¹⁴C isotopes in ice cores and tree rings.[28] Satellite radiometry since 1978, via instruments like those on SORCE and ACRIM series, confirms TSI's cyclic behavior without a net positive trend; post-1980, TSI exhibited a slight decline amid the modern grand maximum of solar activity ending around 2008-2009.[29][30] This contrasts with global surface temperature increases of about 0.7°C since 1970, yielding an inverse correlation that precludes solar irradiance as the primary driver of late-20th- and 21st-century warming.[31][32] Proxy-based TSI reconstructions indicate solar forcing contributed modestly to early-20th-century warming (roughly 0.1-0.2 W/m² net increase from 1900-1950), but forcing turned neutral or negative thereafter as activity waned, accounting for at most 7% of century-scale temperature change and negligible post-1980 influence.[28] The peak-to-peak radiative forcing from an 11-year cycle is ~0.17 W/m², dwarfed by anthropogenic forcings exceeding 2 W/m².[33] Empirical temperature responses to solar cycles yield sensitivities of 0.08-0.18 K per W/m², consistent with direct radiative effects but insufficient to explain observed trends absent amplification.[33] Proposals for indirect solar influences—such as ultraviolet-driven stratospheric heating altering jet streams or cosmic-ray modulation of cloud nucleation—suggest potential amplification, with some multi-proxy models estimating total solar activity impacts up to twice direct TSI forcing.[34] However, these mechanisms lack robust empirical validation and fail to reconcile the post-1980 solar-temperature divergence, as evidenced by unchanged low-cloud trends despite varying galactic cosmic rays.[34] Causal attribution thus assigns solar variability a minor role in multidecadal climate shifts, subordinate to greenhouse gas accumulation in explaining contemporary disequilibrium.[6]Orbital Forcing (Milankovitch Cycles)
Orbital forcing arises from periodic variations in Earth's orbital parameters, which alter the distribution and intensity of incoming solar radiation, or insolation, across latitudes and seasons. These changes, quantified as variations in radiative forcing measured in watts per square meter (W/m²), influence global climate primarily over tens to hundreds of thousands of years through feedbacks involving ice sheets, ocean circulation, and atmospheric composition. The theory, developed by Milutin Milankovitch in the early 20th century, posits that such orbital perturbations initiate glacial-interglacial transitions by modulating summer insolation in the Northern Hemisphere high latitudes, where large ice sheets form or melt. Empirical evidence from ice cores, marine sediments, and speleothems confirms strong spectral peaks in paleoclimate records matching these cycles, particularly a dominant ~100,000-year periodicity in the late Pleistocene.[35] The three primary Milankovitch cycles are eccentricity, obliquity, and precession. Eccentricity describes the deviation of Earth's orbit from a perfect circle, varying between 0.005 and 0.058 over cycles of approximately 100,000 years (with a longer ~413,000-year modulation), which modulates the Earth-Sun distance and thus the annual insolation contrast between perihelion and aphelion by up to 20-30%. Obliquity refers to the tilt of Earth's rotational axis, oscillating between 22.1° and 24.5° with a ~41,000-year period; greater tilt enhances seasonal extremes, increasing high-latitude summer insolation by up to 13% relative to lower latitudes and promoting ice melt during interglacials. Precession involves the precessional wobble of Earth's axis, combining ~19,000- and ~23,000-year cycles into an average ~21,000-year period, which shifts the timing of seasons relative to orbital position, amplifying or dampening insolation peaks when perihelion aligns with Northern Hemisphere summer. Combined, these yield insolation variations of up to 25% at 65°N during summer, sufficient to trigger albedo feedbacks where reduced ice cover lowers reflectivity and amplifies warming.[36][37][35] Paleoclimate records demonstrate orbital forcing's causal role in Quaternary climate oscillations, with benthic oxygen isotope ratios from ocean sediments showing ~41,000-year obliquity dominance until ~1 million years ago, shifting to ~100,000-year eccentricity-paced cycles amid falling atmospheric CO₂ and expanding ice volumes. This transition, termed the Mid-Pleistocene Transition, reflects nonlinear amplifications via ice sheet dynamics and carbon cycle feedbacks, where low summer insolation favors snow persistence and glacial buildup. Model simulations and proxy data indicate orbital insolation changes of ~20-50 W/m² at key latitudes drove ~4-7°C global temperature swings between glacials and interglacials, with CO₂ rising ~80-100 ppm during deglaciations as a response and amplifier rather than primary driver. Such cycles have modulated climate for hundreds of millions of years, evident in older sedimentary records, underscoring their long-term pacing over shorter solar or volcanic forcings.[38][39] In the current Holocene interglacial, which began ~11,700 years ago following the Last Glacial Maximum, orbital forcing trends toward declining Northern Hemisphere summer insolation, with eccentricity decreasing and precession shifting perihelion away from boreal summer, conditions historically conducive to glacial inception within ~1,500-50,000 years absent other influences. Total solar input varies minimally (~0.1-0.2%), but distributional shifts yield a net negative radiative forcing of ~0.5-1 W/m² globally over the past century from orbital parameters alone, insufficient to account for observed 20th-century warming rates exceeding 0.1°C per decade. Climate models incorporating orbital data confirm this forcing is dwarfed by anthropogenic greenhouse gas increases, which impose positive forcings of ~2-3 W/m², rendering Milankovitch effects negligible on centennial scales.[40][35]Volcanic Aerosols and Eruptions
Volcanic eruptions release sulfur dioxide (SO₂) into the stratosphere, where it oxidizes to form sulfuric acid aerosols that scatter incoming solar radiation, thereby reducing the amount of sunlight reaching Earth's surface and inducing a temporary global cooling effect.[41] These aerosols typically persist for 1–3 years before settling out, limiting their influence to short-term climate perturbations rather than sustained trends.[41] The magnitude of cooling depends on eruption size, plume height, latitude, and season, with tropical eruptions producing more widespread and prolonged effects due to stratospheric circulation patterns.[42] The 1815 eruption of Mount Tambora in Indonesia, a VEI-7 event, ejected approximately 100–150 teragrams of SO₂, leading to a sulfate veil that lowered global temperatures by about 0.4–0.7°C in 1816, with regional drops exceeding 1°C in some areas and contributing to the "Year Without a Summer" characterized by crop failures and famines in Europe and North America.[43] Model simulations indicate this cooling reduced global precipitation by roughly 3–5%, amplifying agricultural disruptions.[43] Similarly, the 1991 Mount Pinatubo eruption in the Philippines injected around 20 teragrams of SO₂, resulting in a peak negative radiative forcing of approximately -3 W/m² and a global surface temperature decline of 0.4–0.5°C sustained for 1–2 years.[44][45] In terms of radiative forcing, volcanic aerosols contribute a net negative effective radiative forcing (ERF) that fluctuates with eruption frequency but averages near zero over decadal scales, as assessed in IPCC AR6, with individual large events exerting forcings of -1 to -5 W/m² for months to years.[42] Unlike persistent greenhouse gases, this forcing is transient and does not accumulate, serving primarily to introduce interannual variability that can temporarily mask underlying trends.[42] Empirical reconstructions show that volcanic activity has been subdued since the mid-20th century, with no major stratospheric injections comparable to Pinatubo, contributing negligibly to the observed multi-decadal warming.[46] While subaerial volcanoes emit CO₂—estimated at 0.13–0.44 gigatons annually, dwarfed by anthropogenic sources—their net climatic role remains cooling-dominant in the short term due to aerosol dominance over trace gas releases.[46] Recent events like the 2022 Hunga Tonga–Hunga Ha'apai eruption introduced stratospheric water vapor, which has a warming influence via enhanced radiative absorption, but its overall forcing (+0.05 to +0.1 W/m²) is minor and short-lived compared to aerosol cooling from prior eruptions.[47] Long-term geological records indicate volcanic forcing has driven past climate shifts, such as cooling episodes in the Holocene, but current low eruptivity underscores that natural volcanic variability does not explain the rapid 20th–21st century temperature rise, which aligns more closely with sustained positive forcings from other agents.[48] Attribution studies confirm volcanoes as a modulator of decadal fluctuations rather than a primary driver of directional change.[42]Ocean-Atmosphere Oscillations
Ocean-atmosphere oscillations encompass coupled modes of variability between oceanic and atmospheric circulation, manifesting on interannual to multidecadal timescales and driving regional to global climate fluctuations through heat redistribution and teleconnections. These patterns, including the El Niño-Southern Oscillation (ENSO), Pacific Decadal Oscillation (PDO), and Atlantic Multidecadal Oscillation (AMO), arise from internal dynamics such as wind-driven upwelling, thermocline adjustments, and basin-scale sea surface temperature (SST) anomalies, independent of external forcings like solar or volcanic activity.[49][50] Empirical reconstructions from SST observations since the mid-19th century reveal these oscillations as superimposed on baseline climate states, contributing to decadal-scale temperature variances of 0.1–0.3°C globally.[51][52] The ENSO, centered in the tropical Pacific, operates on 2–7 year cycles, with El Niño phases characterized by weakened trade winds, suppressed equatorial upwelling, and eastward propagation of warm SST anomalies exceeding 0.5°C, which enhance global tropospheric temperatures via atmospheric convection shifts and reduced heat release to deeper oceans. La Niña counterparts involve cooler eastern Pacific SSTs and intensified trades, yielding opposite cooling effects. Observations link strong El Niño events, such as those in 1997–1998 and 2015–2016, to temporary global surface temperature rises of approximately 0.1–0.15°C above trend lines, as seen in the 2023 spike of 0.29 ± 0.04 K from 2022 baselines driven by concurrent ENSO onset amid elevated baseline warmth.[53][54] These impacts extend via teleconnections, altering jet streams and precipitation, though ENSO amplitude shows no statistically significant long-term trend in instrumental records spanning 1850–2020.[55][56] On decadal scales, the PDO features an El Niño-like SST dipole in the North Pacific north of 20°N, with positive phases (e.g., 1925–1946, 1977–1998) correlating with cooler central tropical Pacific waters and enhanced Aleutian Low pressure, influencing North American drought frequency and modulating ENSO teleconnections to extratropics. Instrumental indices derived from 1900–present SST data indicate PDO shifts explain up to 40% of North Pacific decadal temperature variance, with negative phases (e.g., post-2000) linked to intensified marine heatwaves in the Northeast Pacific via altered stratification.[49][57] Similarly, the AMO involves basin-wide North Atlantic SST fluctuations on 60–80 year periods, with positive phases (e.g., since ~1995) associated with weakened meridional overturning and radiative feedbacks amplifying hemispheric warmth by 0.1–0.2°C.[50][58] AMO-positive intervals correlate with Sahel rainfall increases and U.S. Southeast droughts, while interconnections, such as AMO modulation of ENSO–North Atlantic Oscillation links, introduce nonstationarities in teleconnection strengths.[51][59] Despite widespread use in climate analyses, the oscillatory persistence of PDO and AMO faces scrutiny; statistical decompositions of 1850–2018 SST fields suggest these modes may reflect stochastic noise or external forcing responses rather than autonomous internal cycles, with no detectable multidecadal signal in preindustrial simulations or noise-filtered observations.[52][60] Nonetheless, their empirical signatures in reanalysis data (e.g., ERA5, 1950–2020) confirm roles in modulating interannual-to-decadal climate signals, such as PDO's contribution to Pacific salmon productivity declines during cool phases and AMO's influence on Eurasian heat extremes via thermodynamic pathways.[61][62] These oscillations thus represent intrinsic variability that can mask or accentuate underlying trends but lack the sustained forcing needed for centennial-scale change.[63]Anthropogenic Influences
Greenhouse Gas Emissions
Anthropogenic emissions of greenhouse gases have elevated atmospheric concentrations beyond pre-industrial levels, primarily through combustion of fossil fuels, land-use changes, and industrial processes. Carbon dioxide (CO₂) dominates, comprising 74.5% of total global greenhouse gas emissions in 2024, followed by methane (CH₄) at 17.9% and nitrous oxide (N₂O) at 4%.[64] These emissions, measured in CO₂-equivalent terms, totaled approximately 53 GtCO₂eq in 2023 excluding land-use changes.[65] CO₂ emissions from fossil fuels and cement production reached a record 37.4 Gt in 2024, up 0.8% from the prior year, driven largely by coal (44%), oil (32%), and natural gas (22%) combustion.[66][67] Energy-related activities, including electricity generation and transportation, account for the majority, with industrial processes like cement manufacturing contributing about 8% of global CO₂ or 2.8 Gt annually.[68] Deforestation and land-use changes add further CO₂ releases, though estimates vary; cumulative anthropogenic CO₂ from 1750 to 2022 totals around 2,500 Gt, with fossil sources predominant since the mid-20th century.[69] Atmospheric CO₂, measured continuously at Mauna Loa Observatory since 1958, averaged 424.61 ppm in 2024, up from 315 ppm at the record's start and 280 ppm pre-industrially, with the annual growth rate accelerating to 2.5 ppm in recent years.[70][71] Methane emissions, with a global warming potential 28–34 times that of CO₂ over 100 years, stem mainly from agriculture (38% of anthropogenic total), including enteric fermentation in livestock (28%) and rice cultivation, alongside energy sector leaks (35%) and waste decomposition in landfills (20%).[72][73] Globally, anthropogenic CH₄ reached about 380 Mt in recent estimates, with agriculture as the largest source due to manure management and anaerobic processes.[74] Incomplete biomass combustion, including agricultural waste, adds nearly 18 Mt annually.[74] Nitrous oxide, with a warming potential 265–298 times CO₂, arises predominantly from agriculture, which drives nearly three-quarters of human emissions through synthetic fertilizers, manure, and soil management practices.[75] Nitrogen inputs to soils stimulate microbial N₂O production, accounting for over 60% of global anthropogenic sources, with U.S. agricultural soils alone responsible for 69% of domestic N₂O.[76][77] Fluorinated gases, though minor in volume (<1% of total GHG), have high potency from industrial uses like refrigeration and semiconductors.[78]| Greenhouse Gas | Share of Total Emissions (2024) | Primary Anthropogenic Sources |
|---|---|---|
| CO₂ | 74.5% | Fossil fuel combustion, cement production[64] |
| CH₄ | 17.9% | Agriculture, energy fugitives, waste[64] |
| N₂O | 4% | Agricultural fertilizers and manure[64] |
| Fluorinated | <1% | Industrial processes[78] |