Niall Mac Dowell, Centre for Environmental Policy, 911½ñÈÕºÚÁÏ, SW7 2AZ
Given the contemporary focus on climate change and net zero, one could be forgiven for thinking that humanity had only recently become aware of this area. The reality is that we have been studying this space since the late 18th century. Arguably, our understanding of climate science started with de Saussure’s heliothermometer experiments in the 1770s[1].
In the 1830s, building on Fourier’s earlier work[2,3], Pouillet[4,5] was one of the first to theorise that certain constituents of earth’s atmosphere would trap heat – specifically water vapour and carbon dioxide. In the 1850s, Foote and Tyndall independently confirmed this. Foote demonstrated that a glass cylinder filled with CO2 heated up much more than one filled with air, whilst Tyndall measured the infrared absorption capabilities of the atmospheric gases and confirmed that both water vapour and carbon dioxide were primarily responsible for the “greenhouse effect”3.
Towards the end of the 19th century, in 1896, Svante Arhenius published his seminal work that presented the first quantitative climate model that incorporated atmospheric CO2 concentration, infrared radiative absorption and surface temperature[6]. Remarkably, given the relative simplicity of his model (solved by hand, using paper and pencil), his calculations were in the correct order of magnitude – he estimated that a doubling of CO2 concentration would lead to 4-6C of warming (this is now understood to be closer to 2.5 – 3.5C). Importantly, Arrhenius explicitly recognised that fossil fuel use was increasing atmospheric CO2 concentrations, and that this would gradually warm the climate.
Whilst theoretical work continued in the 20th century, warming effects started to become observable, with Callendar being the first to recognise in 1938[7] and again in 1961[8] that a warming trend was already underway. Then the baton was passed to Charles David Keeling who established the Mauna Loa Observatory CO2 record in 1958[9]. The Mauna Loa Observatory[10] continues to make measurements of the earth’s atmosphere to this day.
In 1965, US President Lyndon B Johnson requested an assessment from the President’s Science Advisory Committee (PSAC) on pollution and planetary impacts which included a chapter on atmospheric carbon dioxide[11]. When endorsing the PSAC report, President Johnson explicitly recognised the anthropogenic and harmful nature of climate change.
Climate change entered mainstream diplomacy in 1989 when Prime Minister Margaret Thatcher raised the issue at both the G7 and the UN General Assembly, arguing that the science was sufficiently clear to justify action.
Figure 1: Historical milestones in the scientific and political understanding of anthropogenic climate change, shown against annual global COâ‚‚ emissions
The United Nations Framework Convention on Climate Change (UNFCCC) was adopted at the 1992 Rio Earth Summit, establishing, inter alia, the annual Conference of the Parties (COP) as the decision-making mechanism. The Intergovernmental Panel on Climate Change (IPCC), established in 1988, became the scientific foundation for this process, with each Assessment Report (AR) shaping subsequent negotiations. A major landmark was the Kyoto Protocol, an international treaty under the UNFCCC that committed industrialised countries to legally binding targets for reducing greenhouse gas (GHG) emissions, aiming to combat climate change by assigning greater responsibility to developed nations. Adopted in 1997 but not ratified until 2005, it established emissions trading and mechanisms like the Clean Development Mechanism (CDM) but ultimately highlighted the need for broader, more inclusive agreements. Arguably this process culminated at the 21st COP in Paris, 2015 where 195 countries adopted the Paris Agreement to “to hold global temperature increase to well below 2°C above pre-industrial levels and pursue efforts to limit it to 1.5°C above pre-industrial levels”[12].
At the time of de Saussure’s experiments, anthropogenic emissions of CO2 were approximately 14 Mtpa. At the time of the Paris Agreement, this had increased to 35,400 Mtpa, and in 2024, it was 38,600 Mtpa[13].
Following the Paris Agreement, the IPCC spent two years working to understand the implications of the agreement. When they reported in 2018[14], two things became clear: the importance of transitioning to a net zero greenhouse gas emission economy, and the need to remove CO2 from the atmosphere at the gigatonne scale.
The ensuing years saw a great many countries and organisations making pledges to transition to net zero. In 2019, the UK was the first major economy to make this a legal commitment and by 2023, 89% of global emissions were covered by a net zero pledge of one kind or another. At the time of writing, the USA has exited the Paris Agreement for the second time, reducing this to 77%[15].
The US withdrawal from the Paris Agreement caused significant consternation, and was referred to as a “major disappointment” by the United Nations[16]. However, emission reductions in the US have still been projected to continue, even if some of the policy support is withdrawn[17]. This makes sense when viewed through a macro lens.
Figure 2: U.S. COâ‚‚ emissions by year, coloured by party. While the Oval Office shapes national climate posture, the trajectory of emissions is driven primarily by energy-market dynamics, technology, economic cycles, and state-level policy — not by partisan shifts alone
As illustrated in Figure 2, historically, US emissions rose and fell largely due to structural forces — energy markets, technology shifts, economic cycles, and state-level policy — not the party of the sitting President. US emissions rose more or less consistently from 1850 until the 2005 – 2008 period, when they peaked and began a relatively steep decline primarily owing to a combination of fuel switching from coal to gas, reduced cost of renewable energy, and improved efficiencies. Other factors, such as the 2008 recession and state level policies and regulations contributed to a much lesser extent.
Similarly in the UK and the EU, emissions peaked in 1972 and 1990 respectively and have since subsequently declined by approximately 50% and 40% respectively.
In the UK, the initial wave of decarbonisation was driven by the coal to gas switch, the rise of the services economy, and the initial deployment of renewable energy. Similarly, in the EU key drivers for emissions reduction were the post-Soviet industrial collapse, increase in energy efficiency, and expansion of renewable energy.
Across the US, EU, and UK, the major ‘easy-to-abate’ components of decarbonisation have already been captured—principally deindustrialisation, coal-to-gas switching, efficiency gains, and the large-scale deployment of renewable electricity. As these structural wedges become exhausted, we expect emissions to continue declining, but at a slower pace, as each region enters the ‘hard-to-abate’ phase involving industry, heat, transport, and large-scale infrastructure build-out.
Future progress will also be shaped by systemic constraints such as transmission capacity, grid congestion, permitting delays, and the availability of firm low-carbon power. Although the US, EU, and UK have followed broadly similar decarbonisation pathways to date, their trajectories are likely to diverge over time due to differences in industrial structure, energy systems, policy frameworks, and investment dynamics.
To close this section, we must recognise that the UK, EU, and United States have reduced territorial emissions in large part through the long process of deindustrialisation involving significant carbon exporting. But the modern economy remains irreducibly material. Steel, cement, ammonia, plastics, and other industrial staples are not optional inputs; they are the foundations of urbanisation, food production, transport, and infrastructure[18]. Their global output has not fallen. It has increased several-fold since 1965[19], and production has shifted to regions where energy is cheaper, labour is abundant, and environmental constraints are less stringent[20]. China absorbed most of this shift after 1990. In the past decade, parts of this activity have begun relocating again—to South and Southeast Asia, and increasingly to Africa[21].
Industrial emissions do not decline simply because production moves across borders. They rise with population growth, higher per-capita consumption, expanding urban areas, and the construction needs of developing economies. Between now and mid-century, the Global South will add billions of people, build hundreds of new cities, and increase its demand for cement, steel, fertilisers, and manufactured goods by multiples, not percentages. These processes are energy-intensive and difficult to decarbonise. As a result, even if advanced economies continue to reduce their territorial emissions, global industrial COâ‚‚ output is likely to keep increasing or shifting geographically. The OECD may have peaked; the world’s material metabolism has not.
[1] Kete, K. “The Alpine Enlightenment: Horace-Bénédict de Saussure and Nature's Sensorium”. University of Chicago Press, 2024. Chicago Scholarship Online, 2025.
[2] Fourier, J. B. J. (1824). Remarques générales sur les températures du globe terrestre et des espaces planétaires, Annales de Chimie et de Physique, 27: 136–167
[3] Fleming, James R, 'Joseph Fourier’s Theory of Terrestrial Temperatures', Historical Perspectives on Climate Change (New York, 1998; online edn, Oxford Academic, 12 Nov. 2020),
[5] Pouillet, C. S. M. (1837). “Mémoire sur la chaleur solaire, sur les pouvoirs rayonnants et absorbants de l’air atmosphérique, et sur la température de l’espace.” Comptes Rendus de l’Académie des Sciences 4, 24–65.
[6] Arrhenius, S. (1896). “On the Influence of Carbonic Acid in the Air upon the Temperature of the Ground.”
Philosophical Magazine, 41: 237–276.
[7] Callendar, G. S. (1938). “The Artificial Production of Carbon Dioxide and Its Influence on Temperature.” Quarterly Journal of the Royal Meteorological Society, 64(275): 223–240.
[8] Callendar, G. S. (1961). “Temperature Fluctuations and Trends over the Earth.” Quarterly Journal of the Royal Meteorological Society, 87: 1–12
[9] Keeling, C. D. (1960). “The Concentration and Isotopic Abundances of Carbon Dioxide in the Atmosphere.” Tellus, 12(2): 200–203.
[11] PSAC (1965). Restoring the Quality of Our Environment.
[15] https://climateactiontracker.org/global/cat-net-zero-target-evaluations/
[18] IEA The Future of Petrochemicals 2018
[19] For each of steel, cement, and plastics, production has increased by a factor of ~ 4, 9.5, and 27 respectively since 1965. For context, primary energy consumption has increased by a factor of ~ 3.5. Data from
[21] UN IDO Industrial Development 2022