case study report on global warming

• Climate Change - A Global Issue
• Dag Hammarskjöld Library
• Research Guides

Major Reports

• A Global Issue
• At the United Nations
• Books & Journals
• Consulting the Experts
• Keeping up to date
• Data & Statistics
• AR6 - 6th IPCC Assessment Report / Intergovernmental Panel on Climate Change The main activity of the IPCC is to, at regular intervals, provide Assessment Reports of the state of knowledge on climate change. The IPCC is now in its sixth assessment cycle, in which it is producing the Sixth Assessment Report (AR6) with contributions by its three Working Groups and a Synthesis Report, three Special Reports, and a refinement to its latest Methodology Report.

• Global Landscape of Climate Finance 2023 / Climate Policy Initiative Date: 2023 Provides information about which sources and financial instruments are driving investments, and how much climate finance is flowing globally. The report aims to provide an updated picture on how, where, and from whom finance is flowing toward low-carbon and climate-resilient actions globally, and to improve understanding of how public and private sources of finance interact.

This site contains links and references to third-party databases, web sites, books and articles. It does not imply the endorsement of the content by the United Nations.

• << Previous: At the United Nations
• Next: Books & Journals >>
• Last Updated: Jan 5, 2024 5:23 PM
• URL: https://research.un.org/en/climate-change

UN climate report: It’s ‘now or never’ to limit global warming to 1.5 degrees

Facebook Twitter Print Email

A new flagship UN report on climate change out Monday indicating that harmful carbon emissions from 2010-2019 have never been higher in human history, is proof that the world is on a “fast track” to disaster, António Guterres has warned , with scientists arguing that it’s ‘now or never’ to limit global warming to 1.5 degrees.

Reacting to the latest findings of the Intergovernmental Panel on Climate Change ( IPCC ), the UN Secretary-General insisted that unless governments everywhere reassess their energy policies, the world will be uninhabitable.

#LIVE NOW the press conference to present the #IPCC’s latest #ClimateReport, #ClimateChange 2022: Mitigation of Climate Change, the Working Group III contribution to the Sixth Assessment Report. Including a Q&amp;A session with registered media. https://t.co/iIl81zXev7 IPCC IPCC_CH

His comments reflected the IPCC’s insistence that all countries must reduce their fossil fuel use substantially, extend access to electricity, improve energy efficiency and increase the use of alternative fuels, such as hydrogen.

Unless action is taken soon, some major cities will be under water, Mr. Guterres said in a video message, which also forecast “unprecedented heatwaves, terrifying storms, widespread water shortages and the extinction of a million species of plants and animals”.

Horror story

The UN chief added: “This is not fiction or exaggeration. It is what science tells us will result from our current energy policies. We are on a pathway to global warming of more than double the 1.5-degree (Celsius, or 2.7-degrees Fahreinheit) limit ” that was agreed in Paris in 2015.

Providing the scientific proof to back up that damning assessment, the IPCC report – written by hundreds of leading scientists and agreed by 195 countries - noted that greenhouse gas emissions generated by human activity, have increased since 2010 “across all major sectors globally”.

In an op-ed article penned for the Washington Post, Mr. Guterres described the latest IPCC report as "a litany of broken climate promises ", which revealed a "yawning gap between climate pledges, and reality."

He wrote that high-emitting governments and corporations, were not just turning a blind eye, "they are adding fuel to the flames by continuing to invest in climate-choking industries. Scientists warn that we are already perilously close to tipping points that could lead to cascading and irreversible climate effects."

Urban issue

An increasing share of emissions can be attributed to towns and cities , the report’s authors continued, adding just as worryingly, that emissions reductions clawed back in the last decade or so “have been less than emissions increases, from rising global activity levels in industry, energy supply, transport, agriculture and buildings”.

Striking a more positive note - and insisting that it is still possible to halve emissions by 2030 - the IPCC urged governments to ramp up action to curb emissions.

The UN body also welcomed the significant decrease in the cost of renewable energy sources since 2010, by as much as 85 per cent for solar and wind energy, and batteries.

You might also like

• ‘Lifeline’ of renewable energy can steer world out of climate crisis
• 8 reasons not to give up hope - and take climate action

Encouraging climate action

“We are at a crossroads. The decisions we make now can secure a liveable future,” said IPCC Chair Hoesung Lee. “ I am encouraged by climate action being taken in many countries . There are policies, regulations and market instruments that are proving effective. If these are scaled up and applied more widely and equitably, they can support deep emissions reductions and stimulate innovation.”

To limit global warming to around 1.5C (2.7°F), the IPCC report insisted that global greenhouse gas emissions would have to peak “before 2025 at the latest, and be reduced by 43 per cent by 2030”.

Methane would also need to be reduced by about a third, the report’s authors continued, adding that even if this was achieved, it was “almost inevitable that we will temporarily exceed this temperature threshold”, although the world “could  return to below it by the end of the century”.

Now or never

“ It’s now or never, if we want to limit global warming to 1.5°C (2.7°F); without immediate and deep emissions reductions across all sectors, it will be impossible ,” said Jim Skea, Co-Chair of IPCC Working Group III, which released the latest report.

Global temperatures will stabilise when carbon dioxide emissions reach net zero. For 1.5C (2.7F), this means achieving net zero carbon dioxide emissions globally in the early 2050s; for 2C (3.6°F), it is in the early 2070s, the IPCC report states.

“This assessment shows that limiting warming to around 2C (3.6F) still requires global greenhouse gas emissions to peak before 2025 at the latest, and be reduced by a quarter by 2030.”

Policy base

A great deal of importance is attached to IPCC assessments because they provide governments with scientific information that they can use to develop climate policies.

They also play a key role in international negotiations to tackle climate change.

Among the sustainable and emissions-busting solutions that are available to governments, the IPCC report emphasised that rethinking how cities and other urban areas function in future could help significantly in mitigating the worst effects of climate change.

“These (reductions) can be achieved through lower energy consumption (such as by creating compact, walkable cities), electrification of transport in combination with low-emission energy sources, and enhanced carbon uptake and storage using nature,” the report suggested. “There are options for established, rapidly growing and new cities,” it said.

Echoing that message, IPCC Working Group III Co-Chair, Priyadarshi Shukla, insisted that “the right policies, infrastructure and technology…to enable changes to our lifestyles and behaviour, can result in a 40 to 70 per cent reduction in greenhouse gas emissions by 2050. “The evidence also shows that these lifestyle changes can improve our health and wellbeing.”

• climate action
• Intergovernmental Panel on Climate Change (IPCC)

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 21 March 2024

Global warming and heat extremes to enhance inflationary pressures

• Maximilian Kotz   ORCID: orcid.org/0000-0003-2564-5043 1 , 2 ,
• Friderike Kuik 3 ,
• Eliza Lis 3 &
• Christiane Nickel 3  

Communications Earth & Environment volume  5 , Article number:  116 ( 2024 ) Cite this article

7926 Accesses

1364 Altmetric

Metrics details

• Climate-change impacts

Climate impacts on economic productivity indicate that climate change may threaten price stability. Here we apply fixed-effects regressions to over 27,000 observations of monthly consumer price indices worldwide to quantify the impacts of climate conditions on inflation. Higher temperatures increase food and headline inflation persistently over 12 months in both higher- and lower-income countries. Effects vary across seasons and regions depending on climatic norms, with further impacts from daily temperature variability and extreme precipitation. Evaluating these results under temperature increases projected for 2035 implies upwards pressures on food and headline inflation of 0.92-3.23 and 0.32-1.18 percentage-points per-year respectively on average globally (uncertainty range across emission scenarios, climate models and empirical specifications). Pressures are largest at low latitudes and show strong seasonality at high latitudes, peaking in summer. Finally, the 2022 extreme summer heat increased food inflation in Europe by 0.43-0.93 percentage-points which warming projected for 2035 would amplify by 30-50%.

Similar content being viewed by others

Global supply chains amplify economic costs of future extreme heat risk

Yida Sun, Shupeng Zhu, … Dabo Guan

Warning of a forthcoming collapse of the Atlantic meridional overturning circulation

Peter Ditlevsen & Susanne Ditlevsen

The impact of Russia–Ukraine war on crude oil prices: an EMC framework

Qi Zhang, Yi Hu, … Shouyang Wang

Introduction

The effects of climate change on the economy are becoming increasingly well understood. Key progress has been made using empirical methods to demonstrate impacts on labour productivity 1 , agricultural output 2 , 3 , 4 , energy demand 5 , 6 , and human health 7 , 8 from historical weather fluctuations. The resulting consequences for macroeconomic production have also been quantified empirically, with non-linear impacts of average temperature 9 , 10 , 11 , temperature variability 12 , and various aspects of precipitation 13 on aggregate economic output identified in historical data. The future changes in weather conditions expected due to greenhouse gas emissions imply considerable welfare losses when evaluated through both these micro- 3 , 14 , 15 and macroeconomic impact channels 10 , 11 , 16 , 17 .

Despite these advances, weather impacts on inflation and, in particular, the implications for inflation risks under future climate change, remain understudied. Advancing this understanding is crucial to a comprehensive assessment of climate change risk because rising or unstable prices threaten economic 18 , 19 and human welfare 20 , 21 as well as political stability 22 . The 2021/2022 cost of living crisis provides an example of such implications, with estimates by the United Nations having suggested that an additional 71 million people may have fallen into poverty due to rapidly rising prices 23 . Moreover, the potential for climate change to impact inflation dynamics is of increasingly high-relevance for the conduct of monetary policy and for central banks’ ability to deliver on their price stability mandate in the future 24 , 25 , 26 . A comprehensive assessment of climatic risks on inflation is therefore an important element in guiding the mitigation and adaptation efforts of governments, as well as informing monetary policy concerning the risks posed by climate change.

Previous work in this area has used historical weather fluctuations to identify impacts on inflation from changes in average temperatures 27 , 28 , 29 , 30 , temperature variability 30 , as well as from annual precipitation 31 . However, assessments of the implications of these historical impacts under future climate change are lacking. Here, we provide a comprehensive assessment of the historical impacts on inflation from fluctuations in a wide range of weather conditions, while flexibly accounting for the heterogeneity of their impacts across seasons and regions given different baseline climatic and socio-economic conditions. Moreover, by combining our results with projections from physical climate models we are able to assess the implications of these impacts under the weather conditions projected with future climate change.

We combine measures of national exposure to different weather conditions, based on high-resolution data from the European Centre for Medium-range Weather Forecast Reanalysis version 5 (ERA5) 32 , with a dataset of monthly price indices for different aggregates of goods and services across 121 countries of the developed and developing world over the period 1996-2021 (see Supplementary Table  S1 for summary statistics) 33 . As well as providing over 27,000 observations, the availability of monthly price indices allows a detailed assessment of the temporal dynamics of the response of inflation to weather shocks and the heterogeneity of such effects across seasons. Our empirical framework quantifies the plausibly causal effects of fluctuations in historical weather conditions on national, month-on-month inflation rates (measured as the change in the logarithm of consumer price indices (CPI)) by exploiting within-country variation using fixed-effects panel regression models. Country-fixed effects account for unobserved differences between regions such as baseline climate and inflation rates, while the use of year fixed effects accounts for contemporaneous global shocks to both variables such as El Nino events or global recessions. We further include country-month fixed effects to account for country-specific seasonality – a crucial step given the strong seasonal cycle in both monthly inflation and weather data. Furthermore, our baseline specification accounts for country-specific time trends to avoid spurious correlations arising from common trends. Consequently, our framework accounts for a wide variety of un-observed confounders, and our results stem from the deviations of weather conditions from their national and seasonal patterns which cannot be accounted for by global shocks or country-specific trends. Combined with the exogenous nature of weather fluctuations, these methodological choices strengthen confidence in a causal interpretation of our results 34 .

Temperature increases cause nonlinear, persistent increases in food and headline inflation

We find a rich response of inflation in different price aggregates to fluctuations in a variety of weather conditions (see Supplementary Fig.  S1 , Tables  S2 and S3 ). The strongest and most consistent signal arises from fluctuations in average monthly temperatures (Fig.  1 and Supplementary Fig.  S1a & f ). Impacts are strongest in the food price component (Fig.  1b and Supplementary Fig.  S1f ), indicative of a supply-side productivity shock given the considerable evidence for impacts on agricultural production from temperature 2 , 4 and other weather fluctuations (Fig.  1a ) 35 , 36 , 37 . Although larger in food prices, these impacts also translate into considerable effects on headline inflation (Fig.  1c ). We find limited evidence for impacts on other price sub-components asides from weak evidence in the electricity sector (Supplementary Figs.  S1 & S2 ).

a A schematic outline of the mechanisms via which temperature shocks may impact inflation via agricultural productivity and food prices. The results of fixed-effects panel regressions from over 27,000 observations of monthly price indices and weather fluctuations worldwide over the period 1996-2021 demonstrate persistent impacts on food ( b ) and headline ( c ) prices from a one-off increase in monthly average temperature. Lines indicate the cumulative marginal effects of a one-off 1 C increase in monthly temperature on month-on-month inflation rates, evaluated at different baseline temperatures (colour) reflecting the non-linearity of the response by baseline temperatures which differ across both seasons and regions (see methods for a specific explanation of the estimation of these marginal effects from the regression models). Error bars show the 95% confidence intervals having clustered standard errors by country. Full regression results are shown in Tables  S2 & S3. Icons are obtained from Flaticon ( https://www.flaticon.com/ ) using work from Febrian Hidayat, Vectors Tank and Freepik.

The response to average temperature is strongly non-linear, such that increases in hotter months and regions cause larger inflationary impacts (Fig.  1 ). Consequently, increases in average temperatures at high latitudes cause upwards inflationary pressures when occurring in the hottest month of the year, opposing downward pressures when occurring in colder months. By contrast, increases in average temperatures at lower latitudes cause upwards inflationary pressures all year round (Supplementary Fig.  S3 ). These heterogeneities arise from the dependence of the impacts on baseline temperatures in the empirical model expressed through an interaction term (see methods), rather than explicit dependence on season or latitude, a distinguishing feature from previous work 28 . By using lagged weather variables, we further find that the impacts of a 1 C increase in monthly temperature on the price level persist across the entire 12 months following the initial shock (Fig.  1 ), causing a cumulative effect on food inflation of 0.17 percentage points over the following year (when occurring in country-months with a temperature of 25 C, under our central specification shown in column 1 of Supplementary Tables  S4 & S5 and Fig.  1a ). That is, the initial spike in inflation is not offset by a decline in prices over the following year.

The response of inflation to other weather variables

In addition to the impacts arising from average temperature changes, we also assess impacts from daily temperature variability (the standard deviation of daily temperatures within each month, see methods). We find significant upwards pressures on food and headline inflation from increased variability (Supplementary Tables  S2 & S3 , Fig.  S1b & g ), which depend on the magnitude of the seasonal temperature cycle, with larger impacts at lower latitudes where the seasonal cycle is less pronounced (Supplementary Fig.  S3c ). This reflects the same patterns of vulnerability as that identified to the impacts of daily variability on economic growth 12 . Impacts from variability persist over twelve months, although with increasingly large errors (Supplementary Fig.  S1b & g ).

With regards to precipitation, we assess exposure to monthly extremes using the Standardised Precipitation Evapotranspiration Index (SPEI, see methods for further details). Excess wet conditions cause upwards impacts on food and headline prices which persist over twelve months, independent of baseline climate conditions (Supplementary Table  S2 & S3 , Fig.  S1c & h ). Excess dry conditions have some significant upwards impacts when coinciding with hot months or regions, but these are generally less persistent or significant (Supplementary Fig.  S1i ). These results are qualitatively consistent under different SPEI timescales and thresholds (see Supplementary Fig.  S4 ). We further consider the impacts of daily precipitation extremes (defined as population exposure to the grid-cell level relative exceedance of the 99 th percentile, see methods Eq.  1 for further details) to assess potential heavy-precipitation impacts arising over shorter timescales such as flooding 13 . Statistically significant upwards pressures on headline inflation can be identified in hot months in the first month following the shock, but these impacts appear not to persist with insignificant cumulative impacts at further time-horizons (Supplementary Tables  S2 & S3 , Fig.  S1e & j ).

Robustness of the impact of temperature on inflation

In general, we find the strongest and most significant historical weather impacts on inflation from changes in average temperature. These effects are robust to a number of tests and alternative specifications, an overview of which is shown in Supplementary Tables  S4 and S5 (the results of the robustness tests for all weather variables can be found in Supplementary Figs.  S5 – 9 ). Such tests include using a dynamic panel specification to account for auto-correlations in inflation, for example, associated with inflation developments through the business cycle, using Driscoll-Kraay errors to account for cross-sectionally correlated errors, and including explicit controls for changes in monetary policy frameworks (Supplementary Tables  S4 & S5 Columns 2-4, Figs.  S5 & S6a–j ) 38 . Moreover, we conduct tests in which we split our estimates based on national income (estimated from World Bank GDP and population data), as well as when normalising inflation data by its historical volatility. Doing so we find the effects of temperature increases to be consistent across both higher- and lower-income countries and when accounting for different historical inflation volatility (Supplementary Tables  S4 & S5 Columns 5 & 6, Figs.  S7 & S8 ).

The fact that fluctuations of average temperature cause equivalent impacts on inflation in high- and low-income countries suggests that historical adaptation to temperature increases via socioeconomic development has been limited. To further test whether adaptation can be seen in the historical period we alternatively define temperature shocks with respect to a moving average over the past 30 years rather than a static 1990–2021 average. This choice would reflect the fact that agents may adjust their expectations as long-term climatic conditions change. Empirical results using these temperature shocks provide very similar response functions (Fig.  S6k–t ). Moreover, Information Criteria do not provide strong evidence for using shocks defined in either way. Empirical models of impacts on headline inflation produce Bayesian Information Criteria (BIC) and Akaike Information Criteria (AIC) values of −146636 and −163551.5 when using shocks defined with a moving baseline, compared to −146638.8 and −163554.3 when defined with a fixed baseline. For impacts on food inflation the BIC and AIC show values of −105528.5 and −122443.7 for moving baseline shocks and −105526.7 and −122442 for shocks with a constant baseline. We interpret this as a lack of evidence for significant adjustment in the historical period.

In further tests we use an alternative price index dataset from the World Bank 39 , finding a qualitatively and quantitatively consistent response of food inflation (Supplementary Fig.  S9 ). Estimates for headline inflation differ notably when using World Bank data, most likely due to the inclusion of imputed rents in the World Bank data which may bias estimates away from the effects on widely consumed goods (see Methods for further discussion).

Following the principle of parsimony, and since the magnitude of its impacts lie in the middle range of the other specifications, we use the simple specification of column 1 in Supplementary Tables  S4 & S5 as our baseline for the rest of the paper. We nevertheless continue to discuss the robustness of our results to this choice and present a range of uncertainty arising from this choice of baseline empirical specification (see methods).

Future warming to amplify pressures on inflation

The empirical evidence for the historical impacts of weather shocks on inflation suggests that the ongoing warming and intensification of weather extremes and variability due to anthropogenic greenhouse gas emissions 40 may have consequences for future inflation. To assess these consequences, we evaluate the empirical responses identified above for temperatures under projected future climate conditions. Future projections are taken from an ensemble of 21 bias-adjusted climate models from the Coupled Model Intercomparison Project Phase 6 (CMIP-6) under different emission forcing scenarios from the Shared Socioeconomic Pathways (SSP) 41 (see methods for further details and comparison to the forcing scenarios considered by the Network for Greening the Financial System 42 ). We focus on the role of average temperature due to the persistence of its impacts across income groups and price aggregates, as well as due to the stronger response of average temperatures to greenhouse gas forcing compared to other weather variables.

We consider the impacts we estimate using future climate projections as the effects of future weather conditions on inflation, which would occur (i) in the absence of historically un-precedented adaptation via socioeconomic development or adjustment to warmer climatic conditions, (ii) without a targeted monetary policy response, and (iii) abstracting from any possible interactions with macroeconomic developments. (i) Though we do not introduce explicit models of future adaptation, which could mitigate impacts from future climate change we note that several robustness tests account for historical adaptations, which may have evolved via socio-economic development or prolonged exposure to different climate conditions, or through adjustment to warmer climatic conditions (see above). The results suggest that historical adaptation to temperature increases through socio-economic development and adjustment to warmer climatic conditions has been limited. (ii) A monetary policy reaction aimed at limiting persistent impacts on inflation from long-term changes in average weather conditions is plausible. However, central banks usually pursue a medium-term orientation with respect to their price stability objective, which allows them to be patient when confronted with temporary shocks, such as the weather shocks that we identify historically (Fig.  1 ). (iii) We do not aim to forecast inflation or provide scenarios for it, which would require a range of assumptions on socioeconomic and macroeconomic developments as well as a suite of structural models (as for example, used in the context of the scenarios of the Network for Greening the Financial System 42 ). Rather, we provide an assessment of the potential future exogenous pressure on inflation from future climate conditions, based on the causal relationships inferred with the empirical models, and assuming other socioeconomic factors such as demographic developments and changes in the consumption basket remain constant (principle of ceteris paribus). As such, these results should provide helpful guidance on the likely magnitude and range of exogenous pressures to which society will be exposed and to which monetary policy may have to respond (see also the discussion section).

We find that the temperature conditions projected for 2035 under future warming imply upwards inflationary pressures across all of the world (Fig.  2a, b ). In the global average, these effects constitute persistent upwards pressures on food inflation of 1.49±0.45 or 1.79±0.54 percentage-points per year (p.p.p.y.) respectively in a best- (SSP126) or worst-case (SSP585) emission scenario (uncertainty indicating the standard deviation across climate model projections). Pressures on headline inflation are approximately half as large, 0.76±0.23 or 0.91±0.28 p.p.p.y. under a best- or worst-case emission scenario (Fig.  2a, c ). These results are qualitatively robust across empirical specifications, although impacts vary quantitatively dependent on this choice (shown in Supplementary Tables  S4 & S5 row 5 and Figs.  S10 – 12 ). Combining the uncertainty arising from empirical specification, emission scenario and range of climate model projections (see methods) results in a range of potential pressures on food inflation of 0.92-3.23 p.p.p.y. by 2035, and of 0.32-1.18 p.p.p.y. for headline inflation, on average across the world. These results therefore provide robust evidence that projected global warming would cause persistent upward exogenous pressures on inflation of considerable magnitudes already during the next few decades, independent of future emission trajectories and assuming ceteris paribus.

Maps of the pressure on annual national inflation in the food ( a ) and headline ( b ) price aggregates from the average weather conditions expected by 2035 under a high-emission scenario (SSP585) as estimated from the projections of CMIP-6 climate models. The annual pressure on inflation aggregated across world regions (population weighted), at different time periods under both a low (SSP126) and high (SSP585) emission scenario for food ( c ) and headline ( d ) price aggregates. Point estimates show the average, and error bars the standard deviation, of impacts as projected across the ensemble of 21 CMIP-6 climate models. Impacts are estimated accounting only for increasing average temperatures using the baseline empirical specification shown in column 1 of Supplementary Tables  S4 & S5 . Estimates reflect the exogenous pressure on inflation arising from future weather conditions in the absence of historically un-precedented adaptation, policy response, and abstracting from any possible interactions with macroeconomic developments (see text for discussion). Data on national administrative boundaries are obtained from the GADM database version 3.6 ( https://gadm.org/ ).

Exogenous pressures on inflation from projected future temperature conditions are generally larger in the global south, with the largest pressures found across Africa and South America robustly across specifications (Fig.  2 & Supplementary Figs.  S10 – 12 ). This occurs despite projected warming being greater at higher latitudes (Supplementary Fig.  S13 ). This indicates that the heterogenous vulnerabilities to temperature increases due to different baseline temperature levels (as encoded in our empirical model shown in Eq.  3 ) outweigh heterogeneity in projected warming. Nevertheless, the magnitudes of pressures on inflation are also already considerable by 2035 across advanced economies, in the range of 1-2% on food inflation in North America and Europe under our baseline specification.

Beyond 2035 the magnitude of estimated pressures on inflation diverges strongly across emission scenarios (Fig.  2c, d ), suggesting that decisive mitigation of greenhouse gases could substantially reduce them. By 2060, there is a strong and robust difference in the average global pressures on food inflation between the highest and lowest emission scenarios: 2.1 p.p.p.y. in our central estimate with a range of 1.6-3.8 across empirical specifications and climate models (Row 6 of Supplementary Tables  S4 & S5 ). Under a best-case emission scenario, exogenous pressures on inflation are only marginally larger in 2060 than in 2035, but a worst-case emission scenario would cause pressures on food inflation exceeding 4 p.p.p.y. across large parts of the world (Fig.  2c , Supplementary Figs.  S10 – 12c ).

Although the empirical evidence indicates that adaptation to temperature shocks has been limited historically, we explore the potential of adaptation via adjustment to changing temperatures to reduce these future impacts. We do so by using empirical models in which temperature shocks are defined relative to a 30-year moving average rather than a constant baseline (Fig.  S6k–t ), and by evaluating potential impacts using future temperatures defined in this way. This method indicates that adaptation via adjustment could substantially reduce future impacts (Supplementary Fig.  S14 ). In particular, in a low emission-scenario most impacts could be removed by adjustment once global temperatures stabilise (Supplementary Fig.  S14c, d ). However, in scenarios of un-mitigated warming, persistent impacts of considerable size remain despite introducing adjustment of this type which has not been observed historically (Supplementary Fig.  S14 ).

The seasonality of pressures on inflation from future warming

The use of monthly CPI data allows us to further assess how the estimated pressures on inflation from future temperature conditions under projected climate change are distributed across the year. Concerning food inflation, these impacts are fairly constant across seasons at low latitudes but vary considerably across seasons in Northern mid-latitudes (20-40 N) where they can be more than twice as large in summer compared to winter (Fig.  3a ). At the highest latitudes (>40 N) upwards pressure in summer contrasts downwards pressure in winter. This seasonal and spatial heterogeneity is robust across empirical specifications (Supplementary Figs.  S15 – S17 ), although accounting for different historical baseline inflation volatilities (column 6 of Supplementary Tables  S4 & S5 ) introduces additional noise (Supplementary Fig.  S17 ). Moreover, similar patterns are observed for headline inflation (Supplementary Figs.  S18 – S21 ).

a The pressures on monthly food inflation averaged across latitudinal bands estimated from the temperature conditions expected by 2035 under a high-emission scenario, as projected on average across the ensemble of CMIP-6 climate models. Impacts are estimated accounting only for increasing average temperatures. b The percentage change in the seasonal variability of food inflation under the pressures from future temperature conditions, estimated as the change in the standard deviation of the seasonal inflation cycle. c – f Country-specific examples of the pressures on the seasonal cycle of food inflation for the United States, Germany, Colombia and Kenya. Black curves show the historical average month-on-month percentage change in the food consumer price index (CPI) with blue error bars indicating the standard deviation across the years of the historical period (1996–2021). Red curves show this historical average plus the pressures estimated from the future weather conditions under projected warming, with the error bars indicating the standard deviation of projections across climate models. Estimates reflect the exogenous pressure on inflation arising from future weather conditions in the absence of historically un-precedented adaptation or policy response (see text for discussion). Data on national administrative boundaries are obtained from the GADM database version 3.6 ( https://gadm.org/ ).

These heterogeneities arise from the dependence of the impacts on baseline temperatures as outlined in the empirical model (Eq.  3 ), rather than an explicit dependence on season or latitude. Large seasonal cycles of temperature at higher latitudes lead to stronger upward pressures in summer contrasting weak or downward pressures in winter, whereas less variable baseline temperatures throughout the year at low-latitudes result in fairly constant impacts across seasons. Projected temperature increases are typically stronger in winter than in summer in Northern mid-to-high latitudes (with the exception of Europe, Supplementary Fig.  S13 ) indicating that most of the seasonality observed at high latitudes in Fig.  3 results from the distribution of baseline temperatures across seasons rather than differential warming between seasons (except in Europe, where more rapid warming in summer also contributes to these patterns).

These seasonally heterogenous pressures would cause alterations to the usual seasonal cycle of food inflation, resulting in an amplification of seasonal variability across most of the global south and the USA, and reductions in seasonal variability across most of Europe (excluding Spain) and the higher northern latitudes (Fig.  3b & Supplementary Fig.  S15 – 21b ). A reduction in seasonal variability arises when the strongest upwards pressures occur in months with historically lower inflation rates, as compared to other months (as shown in the case of Germany in Fig.  3d ).

Amplified impacts from unpredictable heat extremes

In addition to shifting average conditions, climate change is also altering the intensity and frequency of unpredictable hot extremes which may pose additional short-term risks to inflation. The summer heat extreme in Europe in 2022 is a prominent example in which combined heat and drought had wide-spread impacts on agricultural and economic activity. These effects likely added to inflationary pressures in Europe, but the magnitude of their contribution has so far been difficult to assess, particularly in the context of other pressures from the Russian invasion of Ukraine and the aftermath of the Covid-19 pandemic. Combining our empirical results with estimates of monthly temperatures in June, July and August of 2022 (from the ERA5 reanalysis of historical observations), we estimate that the anomalous heat over these three months alone caused a cumulative annual impact of 0.67 percentage-points (0.43–0.93 across empirical specifications) on food inflation and 0.34 percentage-points (0.18–0.41) on headline inflation in Europe, with larger impacts across Southern Europe (Fig.  4a , see Supplementary Figs.  S22 – S24 for results using other empirical specifications).

The cumulative annual impacts on food ( a ) and headline ( b ) inflation from the observed temperatures of June, July and August of 2022 across Europe. ( c , d ) Regionally aggregated (using a population weighting) impacts from the historical 2022 summer temperatures, as well as those impacts which would result from an equivalent summer if amplified by future warming as projected by CMIP-6 climate models (see methods) under future emission scenarios specified by the SSPs. Point estimates and error bars show the mean and standard deviation of impacts across climate models. Data on national administrative boundaries are obtained from the GADM database version 3.6 ( https://gadm.org/ ).

Future climate change will amplify the magnitude of such heat extremes, thereby also amplifying their potential impact on inflation. To assess such effects, we make use of the fact that climate change will alter the distribution of future summer temperatures predominantly by shifting their mean 43 , 44 . We therefore add the future summer warming projected to occur from 2022 onwards in the CMIP-6 projections to the historical temperatures realised in 2022, and re-evaluate their impact using our empirical response functions (see methods for further details). This approach suggests that if amplified by future warming, an equivalent extreme summer (i.e., in the upper tail of the shifted temperature distribution) would – ceteris paribus - cause impacts on food inflation in Europe of 1.0 percentage-points (0.6–1.6, uncertainty range across climate models and empirical specifications) in 2035 under a high-emission scenario, or of 0.9 percentage-points (0.5–1.4) under a low-emission scenario (Fig.  4c ). These constitute an amplification of the impacts of extreme heat on inflation in Europe by 30–50% due to climate change already by 2035. By 2060, the amplification of such extreme impacts would diverge under different emission scenarios, remaining at 1.1 percentage-points (0.6–1.8) under the most optimistic scenario compared to 1.8 percentage-points (1.0–3.2) under the most pessimistic scenario of emission mitigation, an amplification of nearly 200%. These results highlight the short-term risks to inflation posed by unpredictable heat extremes which are already occurring under present climatic conditions, and which will be amplified by future warming.

This work has identified a number of weather variables with significant historical impacts on headline and food inflation globally (Supplementary Fig.  S1 ), but limitations persist in providing a comprehensive relationship between weather conditions and inflation. For example, the fact that we do not find such significant or consistent impacts of precipitation changes on food prices may be surprising given the clear sensitivity of agricultural productivity to precipitation 36 . However, precipitation changes exhibit a higher spatial variability than temperature 34 and the use of national-level data may therefore be a limiting factor in our ability to accurately detect such effects should they exist. The development of consistent datasets of consumer prices at higher spatial resolutions, such as for sub-national regions may reduce these issues to the extent that local prices reflect local production. To the extent that local prices reflect imported production, assessments of spill-overs via trade 45 or pressures arising through global commodity prices may provide further interesting insights.

Second, our empirical results refer predominantly to food and headline inflation, whereas we find a limited response of other price aggregates to weather changes. However, the strong response of electricity demand to temperature 5 , 6 suggests that impacts on electricity prices are plausible. Indeed, we find that electricity prices show some consistent and persistent response to temperature increases (Supplementary Fig.  S1k ), but with much larger uncertainty which precludes statements of significance at conventional levels. Lesser data availability for this more detailed price aggregate as well as complex and heterogeneous electricity price-setting practices may contribute to these large errors. However, as electricity supply is increasingly met with renewable sources, the price sensitivity to weather may change. A detailed analysis of electricity and other price aggregates may be a fruitful avenue of future work.

Compared to previous literature, our empirical results are similar to those of Mukherjee 29 in identifying persistent impacts of temperature increases in both developed and developing countries. Moreover, they are similar to those of Faccia et al. 28 in identifying that temperature increases in hot seasons cause the largest and most significant upwards pressures on food inflation. Our approach is qualitatively different to Faccia et al. 28 in that it models the heterogeneity across seasons and regions using interactions of the temperature shocks with baseline temperatures rather than assessing shocks in specifically defined seasons. Faccia et al. find contemporaneous impacts of 0.38%-points on food inflation from a 1.5 C quarterly summer temperature increase. Evaluating the regression coefficients pertaining to average temperatures in our central model (shown in Column 1 of Supplementary Table  S2 ) at the baseline temperature observed in our dataset on the three hottest months of the year on average across countries (23.5 C), and given a 1.5 C temperature increase, indicates an impact of 0.17%-points. Given that our data are monthly, we must further consider a temperature shock, which persists across all three months of a quarter to compare to Faccia et al., implying an impact of 0.49%-points which is closely consistent. The slightly larger estimates we obtain may result from the use of more granular climate data (monthly vs quarterly) which likely limits attenuation of the impact signal.

The implications of our empirical results under future temperature conditions are considerable regarding societal welfare in general and price stability in particular: our results suggest that climate change is likely to alter inflation seasonality, increase inflation volatility, inflation heterogeneity and place persistent pressures on inflation levels.

In our empirical results we find upward pressures on food and headline inflation from higher-than-normal temperatures, especially when occurring in hot months and countries. This implies short-term rises in inflation from exceptionally hot periods such as that experienced in Europe in the summer of 2022 (Fig.  4a, b ). With the intensity of hot extremes and their impacts on inflation being amplified with continuing climatic change (Fig.  4c, d ), while being unpredictable in the medium- to longer-term, this relationship is set to increase inflation volatility. This in turn may pose challenges to inflation forecasting and monetary policy, likely increasing the difficulty of identifying temporary supply shocks and disentangling them from more persistent drivers.

We find that the inflationary impact of temperature shocks depends on the baseline climatic conditions. At the same time, future climate change implies different warming levels depending on the season and latitude. Taken together, this implies that temperature shocks under future climate change would both amplify inflation heterogeneity (Fig.  2 ) and alter the seasonality of inflation within individual countries (Fig.  3b ). Inflation heterogeneity poses challenges in monetary union areas such as the euro area, where larger inflationary pressures from climate change in southern Europe (Fig.  2 & 4 ) may increase inflation differentials, making the calibration of a single monetary policy more difficult 46 . Moreover, heterogeneous effects on inflation within an economic union such as the EU could exacerbate pre-existing welfare discrepancies, which can fuel anti-EU sentiment 47 . In addition, an altered inflation seasonality could pose additional challenges to inflation forecasting, which may however be (partially) mitigated through the development of weather-dependent forecasts for production 48 and inflation 49 .

Finally, evaluating our empirical results under future temperature conditions suggests that – ceteris paribus - persistent upward pressures on annual food inflation of 1-3 percentage-points per-year could result from temperatures projected for 2035 (Fig.  2c ). In addition, we test for adaptation via socio-economic development (Supplementary Fig.  S8 , Table  S4 & S5 column 5), prolonged exposure to higher temperatures (Fig.  1 and all other empirical specifications), and adjustment to gradual warming (Supplementary Fig.  S6k–t ), with results suggesting that these forms of historical adaptation have been very limited (see earlier discussion). It should however be noted that our estimates assume constancy in other factors which may be important for future developments of inflation such as general macroeconomic developments and structural changes in the economy. Our estimates should therefore be understood as the likely exogenous pressure on inflation from future climate conditions based on the causal relationships inferred from the empirical models, in the absence of unprecedented adaptation. More persistent upward inflationary pressures from increasing temperatures under a changing climate would have important implications for monetary policy, as it would render the identification of drivers of inflation more difficult when relying on traditionally used models, and also risk the de-anchoring of inflation expectations. As a result, central banks may need to make monetary policy decisions also in response to weather and climate shocks, as in such a situation weather and climate shocks can no longer be considered temporary. Moreover, persistent upward pressures on inflation may have adverse effects on purchasing power, often with regressive distributional effects and potential impacts on social cohesion 50 , as well as inefficiency costs due to nominal rigidities and adverse interactions with taxation 50 . Overall, these results strongly highlight the importance for central banks and macroeconomic modelling in general to consider future climate change in their macroeconomic assessment and forecasting tools.

Future adaptation to climate change through unprecedented technological changes – which we do not explicitly model - offers an opportunity to limit pressures on inflation in a changing climate. For example, planned adoption of space cooling could limit heat stress impacts on labour productivity and crop switching could limit agricultural productivity losses, two major channels of impacts with potential relevance to inflation. Exploring the possibility for historically un-precedented adaptation to reduce impacts via adjustment to changes in long-term climate conditions indicates that it could do so substantially (Supplementary Fig.  S14 ). However, without considerable mitigation of greenhouse gas emissions pressures on inflation would remain persistent and sizeable, even when accounting for such adaptation which goes beyond what has been observed historically. The efficacy and opportunity costs of the necessary investments in these adaptations also remain largely unknown and therefore present an important avenue for further research on the scope to limit the risks to inflation from a warming climate and intensifying heat extremes.

Inflation data

Data on national-level inflation of different price aggregates are obtained from a dataset developed by reference 33 (see the Supplementary Methods for further information). The data used here constitute monthly, non-seasonally adjusted prices at different levels of aggregation. Data are available for 121 countries with varying temporal coverage from 1996-2021. The countries included cover most of the developed world (minus Australia and New Zealand where monthly data are not available), as well as large parts of the developing world. Coverage across South America and Africa is good, but large gaps exist in South East Asia where detailed information on price aggregates at monthly timescales are not available. Month-on-month inflation rates are used as the main dependent variable, estimated as the first difference in the logarithm of consumer price indices (CPI).

In a robustness test conducted in Fig.  S9 , we alternatively use monthly inflation data from the World Bank cross-country database on inflation 39 . Differences in the aggregation procedures exist and are documented extensively in reference 33 . Two important differences are the inclusion of imputed rents in some headline inflation indices in the World Bank data and differences in the aggregation of food (see Supplementary Methods for further details). We use the data compiled by reference 33 as our main specification because the inclusion of imputed rents may bias estimates away from the impacts on widely consumed goods. In those countries where imputed rents are incorporated, they typically have a large weight, but there are many indices that do not incorporate them, notably including all European countries using the Harmonised Index of Consumer Prices. We find that the impacts on food inflation from mean temperature are qualitatively and quantitatively consistent when using World Bank data, Fig.  S9 . The response of headline inflation differs considerably, likely due to the inconsistent inclusion of imputed rents in headline inflation in the World Bank data.

Climate data

The primary source of climate data for this study is the ERA-5 reanalysis of historical observations 32 . ERA-5 combines satellite and in-situ observations with state-of-the-art assimilation and modelling techniques to provide estimates of climate variables with global coverage and at 6-hourly resolution. Daily 2 m air temperature and surface precipitation rates for the years 1990-2021 are used as well as monthly average temperature for the months of June, July and August in 2022 for use in Fig.  4 . All data from ERA-5 is obtained on a regular 0.25-by-0.25-degree grid for the years 1990-2021. For the estimates of SPEI, we follow the literature 51 in using monthly mean temperature and monthly precipitation totals from the Climate Research Unit (CRU) TS v4.05 for the years 1901-2021. This data is obtained at the same resolution and on the same grid as ERA-5.

Weather variables

Monthly, m, averages, \({\bar{T}}_{x,m}\) , and standard deviations, \({\widetilde{T}}_{x,m}\) , of daily ERA-5 temperatures are calculated at the grid cell, \(x\) , level. Moreover, the relative exceedance of certain high precipitation thresholds, \({T}_{x}\) , are calculated according to

where \({P}_{x,d}\) are daily precipitation totals, \(H\) the Heavide step function and \({D}_{m}\) the number of days in a given month. Following reference 13 , we use the 99th percentile of the distribution of historical daily rainfall to set thresholds locally (1990-2021).

Standardised Precipitation Evapotranspiration Indices (SPEI) are calculated following the methods of reference 51 , applying their publicly available code to monthly temperature and precipitation data from CRU TS v4.05. The SPEI calculation is based on a physical model of moisture balance and considers contributions to dry or wet conditions from both temperature and precipitation. It is a widely used tool to flexibly compare dry and wet conditions across countries. Moreover, its flexible estimation over different timescales allows exploration of different impact-relevant timescales. We estimate SPEI at one, two-, three-, six- and twelve-month timescales to flexibly assess the impacts of shocks across these timescales.

Spatial aggregation

We use gridded population estimates from the History database of the Global Environment (HYDE) 52 to estimate national-level exposure to changes in these climate variables. The data are provided at 0.25-by-0.25-degree resolution by the Inter-Sectoral Impact Model Intercomparison Project (ISIMIP). Monthly average temperature, temperature variability and the measure of daily precipitation extremes are aggregated to the national level using a population weighted average. In this weighting we also account for the proportion of grid-cells falling within a given administrative unit, estimated by evenly distributing 100 points within each grid-cell and estimating the proportion which fall within the given administrative unit. Given these weightings, \({w}_{x,n},\) for all \({{{{{{\boldsymbol{N}}}}}}}_{{{{{{\boldsymbol{x}}}}}}}\) grid-cells falling at least partially within the administrative boundary of a country, \(c\) , this weighted average reads:

for monthly average temperature for example. The equivalent procedure applies for temperature variability and the measure of daily precipitation extremes.

For SPEI, we first estimate binary variables indicating whether a grid cell is experiencing conditions which exceed a certain level of dry or wet (indicated as SPEI  <  and SPEI >, respectively in Eq.  3 ). We choose thresholds of one, one-point-five, two and two-point-five deviations to flexibly assess the exceedance of different thresholds. We then apply the same spatial aggregation procedure as outlined in Eq.  2 to estimate the proportion of population exposed to these excessively wet or dry conditions. By calculating the grid-cell level exceedance of certain SPEI thresholds and aggregating the proportion of national population exposed to these excess wet or dry conditions, we aim to limit the issue of spatially averaging over opposing effects. This is particularly relevant for precipitation given its larger spatial variability 34 .

The magnitude of the temperature seasonal cycle, \({\hat{T}}_{c}\) is estimated as the difference between the maximum and minimum national monthly temperatures within a given year, which is then averaged over the historical period (1990-2021) before use in the regression models. Deviations of average temperature, \(d{\bar{{{{T}}}}}_{{{{c}}}{{{,}}}{{{m}}}}\) , and temperature variability, d \({\widetilde{T}}_{c,m}\) , from their historical average (1990-2021) over the same calendar month are also calculated for use as dependent variables. This choice is intended to reflect the impact of deviations of monthly climate conditions from their historical seasonal patterns, following the intuition that the economy is well adapted to historical weather patterns from which deviations are a source of potential impacts. We note that the use of country-month fixed effects in the empirical model (see next section), results in an equivalent differencing process for the other independent variables.

Empirical framework for estimation of causal effects

The combination of price and climate data results in 27,340 country-month observations across 121 countries. We then apply fixed-effects panel regression models to identify the causal effects of changes in weather variables on national level inflation. Month-on-month national level inflation rates, \({dlCP}{I}_{c,m}\) are the dependent variable. Deviations of average temperature are included with an interaction with the average temperature level, whereas deviations of temperature variability are included with an interaction with the magnitude of the seasonal temperature cycle, following the heterogeneities identified in previous studies on the impacts of climate conditions on growth 12 . The interaction with the average temperature level introduces a heterogeneity in the response to temperature shocks across both geographical locations and seasons, based on the prevailing temperature of those regions and seasons. This choice follows previous literature which finds larger impacts on inflation in hotter seasons 28 , and larger impacts on different economic factors in hotter regions 1 , 2 , 10 , 11 . Daily precipitation extremes are included with an interaction with the monthly average temperature level (having also tested alternative interactions with the monthly share of annual precipitation). Both positive (excess wet) and negative (excess dry) SPEI threshold exceedance are used. For the former we find no significant effect of interactions with the monthly average temperature level or the monthly share of annual precipitation and as a result include no interactions in our main specification. For the latter we find a significant effect of an interaction with monthly average temperature level and therefore include this in our main specification. We include all weather variables simultaneously to ensure that any effects we identify occur independently of one another and are therefore additive 12 , 34 . Each weather variable is included with 11 lags in addition to the contemporaneous term, to assess the delayed effects of monthly weather shocks over the course of the following year and in particular whether they are recovered or persist over this time frame.

Our baseline specification includes country, \({\mu }_{c}\) , date, \({\eta }_{t},\) and country-month, \({\pi }_{c,m},\) fixed effects, in addition to country specific linear time-trends, \({\gamma }_{c}y\) . Country fixed effects account for unobserved differences between regions such as baseline climate and inflation rates, while the use of date fixed effects accounts for contemporaneous global shocks to both variables such as El Nino events or global recessions. The inclusion of country-month fixed effects accounts for country specific seasonality – a crucial step given the strong seasonal cycle in both monthly inflation and weather data. This constitutes an additionally conservative step by ignoring inflation impacts which could repeatedly occur seasonally due to seasonal weather patterns. This ensures that our results only estimate the impacts of deviations from normal seasonal conditions. Finally, our baseline specification accounts for country specific time trends to avoid spurious correlations arising from common trends. This is important given the presence of strong warming trends in the historical period which could cause spurious correlations to inflation changes. Interestingly, we find that accounting for these linear trends enhances the magnitude of estimated effects, suggesting that it indeed assists in removing estimation biases. Estimates without linear time trends are nevertheless qualitatively and quantitatively similar. The regression model of the baseline specification then reads:

where \(t\) is the date in terms of a given year and month and \({\varepsilon }_{c,t}\) is the country-date residual error. Note that here \(t\) refers to the date i.e., the month of a specific year, whereas m refers to all general occurrences of a particular month, and y refers to the particular year. In our baseline specification, errors are clustered by country. Coefficients \(\alpha\) and \(\beta\) describe the common impact across countries and months of a 1-unit increase in each independent variable on month-on-month inflation rates and are shown in Tables  S2 and S3 .

In alternative robustness tests we estimate a dynamic model in which we also include 11 lags of the inflation rates, \({dlCP}{I}_{c,t}\) to account for serial correlations due to for example business cycles (Supplementary Tables  S4 & S5 Column 2, Fig.  S5 ), account for cross-sectionally correlated and heteroskedastic errors using Driscoll Kraay errors 38 (Supplementary Tables  S4 & S5 Column 3, Fig.  S5 ), and test an inflation database provided by the World Bank (Supplementary Tables  S4 & S5 Column 7 Fig.  S9 ). Moreover, in an additional robustness test we include controls for transitions in monetary policy using data from the Comprehensive Monetary Policy Framework project 53 . We use the data in its most granular form (32 classifications of monetary policy frameworks) introducing dummy variables in the regression for each potential framework. Our results are qualitatively and quantitatively robust to these additional controls (Supplementary Tables  S4 & S5 Column 4, Fig.  S6a–j ). Furthermore, we also estimate models in which we include additional interactions of each climate variable with a binary term indicating whether a given country has above or below median national income per capita (based on world bank estimates of GDP and population, see Supplementary Tables  S4 & S5 Column 5, Fig.  S8 ), and also when normalising monthly inflation rates by their interannual standard deviation to account for differing baseline inflation volatilities (see Supplementary Tables  S4 & S5 Column 6, Fig.  S7 ). These robustness tests of our main results are summarised in Supplementary Tables  S3 and S4 of the supplementary information.

We do not include further controls for variables which affect inflation such as employment and economic output for a number of reasons. First, important aspects of their effects which are linked to business cycles are already accounted for by the use of a dynamic panel with lagged inflation as independent variables, as used in similar contexts with global panels 28 . Second, such data is not available at the monthly resolution used here for most countries in our panel. Third, including such control variables could only alter our results if they were correlated with the weather variables. Given the strong a-priori assumption of exogeneity between weather and these variables, the presence of such correlations would indicate that these control variables are themselves impacted by the weather variables. Therefore, any alteration to our results when including these variables would not alter our main interpretation but rather indicate that these variables (employment/output) are a mediating variable through which weather impacts inflation. While such mechanistic insights may be interesting, due to data availability they are beyond the scope of our manuscript which primarily aims to understand the overall impacts of climate variables on inflation.

Cumulative marginal effects

In Fig.  1 and a number of supplementary figures we display the results of the empirical models by plotting the cumulative marginal effects of each climate variable. These cumulative marginal effects reflect the theoretical cumulative impact on prices from a 1-unit climate shock. These effects are estimated by summing the lagged coefficients (shown in Eq. ( 3 )) which are relevant to a particular climate variable. Moreover, because of the use of interaction terms, the coefficient pertaining to the interaction term must be multiplied by a chosen value of the moderating variable of the interaction. For example, in Fig.  1 the cumulative marginal effects, \({ME}\) , of average temperature are plotted, having been calculated as follows:

Calculating the cumulative marginal effects therefore requires an evaluation of the coefficients at a particular baseline temperature, and their summation over the different lags. In Fig.  1 , these marginal effects are plotted when evaluating the above summation over different numbers of lags, from 0 to 11 months after the initial shock. We show results having evaluated Eq. ( 4 ) at the temperatures observed at the lower and upper quartiles and median of the distribution of country-month temperatures present in our data. We conduct equivalent procedures for estimating and plotting the cumulative marginal effects for the other climate variables in the Supplementary Figs.  S1 , S2 & S4 – S9 .

Climate model data

Daily 2-m temperature and precipitation totals are taken from 21 climate models participating in CMIP-6 under the most pessimistic (SSP-RCP8.5, referred to as SSP585 in the main text) and most optimistic (SSP-RCP2.6, referred to as SSP126) greenhouse gas emission scenario from 2015-2100. SSP126 provides approximately equivalent emission forcing as the orderly and dis-orderly transition scenarios provided by the Network for Greening the Financial System (NGFS), with an average end-of-century global temperature change of 1.7 C. SSP585 provides stronger emission forcing (4.9 C end-of-century global temperature change) than the hot-house world scenario from NGFS (3.2 C end-of-century temperature change). While considered by some as un-realistic 54 , RCP8.5 tracks recent emissions well and is arguably likely to provide a good estimation of emission forcing up until mid-century based on current (2020) policies 55 . The data have been bias-adjusted and statistically downscaled to a common half-degree grid to reflect the historical distribution of daily temperature and precipitation of the W5E5 dataset (WATCH Forcing Data Methodology applied to the ERA5 data) using the trend-preserving method developed by ISIMIP 56 , 57 .

Estimating impacts from projected future warming

We evaluate the hypothetical impact on inflation which future weather conditions under projected climate change would cause given our empirical models. We note the important distinction that these are not projections of future inflation, but simply an evaluation of this particular mechanism via which climate conditions effect inflation under future conditions. Important factors including demographic developments, changes in the consumption basket, and fiscal and monetary policies are purposefully held fixed (although we note that our empirical results are strongly robust to changes in the regime of monetary policy, Supplementary Fig.  S6a–j ; also see the discussion included in the main part of the manuscript).

To do so, we evaluate the first terms pertaining to monthly average temperatures of Eq.  3 under future temperature conditions. That is, we calculate future monthly average national temperatures from the CMIP-6 models (using a population weighting equivalent to that used with the historical data), \({\bar{T}}_{c,y > 2020,m}\) , as well as their deviations from the 1990-2021 average, \(\Delta {\bar{T}}_{c,y > 2020,m}\) , and then apply these to the first terms of Eq.  3 to calculate the impacts on inflation:

Note that in practice, \(m-L\) , may need to refer to a month in the preceding year. These impacts are then averaged over 30-year periods centered on the future period in question (usually 2035 or 2060). These impacts at the country-month level are then either summed over the year to provide estimates of annual impacts on inflation as in Fig.  2 or presented at the monthly level as in Fig.  3 . This procedure is conducted separately for each of the 21 climate models in the CMIP-6 ensemble, from which the mean and standard deviation are presented as central estimates and errors.

In Fig.  4 we assess the impacts of the 2022 extreme summer heat in Europe using ERA-5 estimates of monthly temperatures in June, July and August. In this case, the total impacts on inflation from those three summer months are estimated using the temperature levels in those months, \({\bar{T}}_{c,m}\) , their deviation from the historical (1990-2021) average, \(\Delta {\bar{T}}_{c,m}\) , and the relevant terms from Eq.  3 pertaining to average temperature impacts:

As such, these impacts reflect the estimated effects on net inflation from June 2022 to August 2023 from the three months of temperature in June, July and August of 2022.

We further assess how the impacts from such extremes could be amplified under future warming. To do so, we evaluate the future warming occurring between 2022 and 2035 or 2060 in each summer month in each country (using the difference between 30-year averages of temperature centered on 2022 and 2035 or 2060 in each climate model and emission scenario). This additional month-specific warming is then added to the historically observed 2022 summer temperatures, and the impacts on inflation evaluated as before using Eq.  6 . This approach assumes that future warming will shift the mean of the distribution of possible summer temperatures and does not account for the potential role of changing temperature variability in altering the intensity of future temperature extremes. However, evidence for a role of temperature variability in enhancing extremes at monthly time-scales is limited 43 , 44 .

When presenting estimated inflationary impacts under projected future climate, we also present country-level impacts aggregated to larger spatial regions. In Figs.  2 and 4 we do so using a population weighted average (using World Bank estimates of national level population in 2017) to reflect the human exposure to future inflationary pressures. In Fig.  3a and S3d we take binned averages across latitudinal zones to convey the relationship between latitude and the seasonality of inflation response and impacts. Countries are considered part of a latitudinal zone if their centroid falls within the zone’s boundaries, and in this context, we use an average without population weighting to reflect the nature of the relationship between latitude and impacts rather than to reflect the average human exposure to impacts.

Uncertainty in estimated impacts from future weather conditions under projected warming arises from a combination of factors, including the choice of empirical specification, the range of climate model projections, as well as future emission scenarios if their differences are not explicitly compared. In Figs.  2 – 4 we show projection estimates for a particular empirical specification, showing the range of projections across climate models and emission scenarios visually. In the text, we discuss the robustness of these figures to the use of different empirical specifications (results of which are shown in the Supplementary Information Figs. S10 – S24 ), and report estimates of projected impacts with an uncertainty range accounting for these contributing factors. This uncertainty range spans the lowest projection across the empirical specifications shown in columns 1-7 of Supplementary Tables  S4 & S5 (and emission scenario unless explicitly comparing their differences) combined with the lower range of climate model projections (the mean minus one standard deviation of impacts across climate models), and the largest projection across empirical specifications combined with the higher range of climate model projections (the mean plus one standard deviation of impacts across climate models). This framework provides a transparent assessment of uncertainty across a range of factors.

Data availability

ERA-5 climate data are publicly available at https://www.ecmwf.int/en/forecasts/datasets/reanalysis-datasets/era5 , and raw data is available for 10 of the bias-adjusted climate models from the ISIMIP repository at: https://data.isimip.org/ . Raw data on price indices was taken from a forthcoming publicly available dataset developed by ref. 33 . All processed climate data and anonymised inflation data (until publication of the inflation data-set by ref. 33 ) necessary for replication of our analysis is publicly available at: https://doi.org/10.5281/zenodo.10183679 . Source data required only for reproducing the main figures is also available at the same repository.

Code availability

All code used for the replication of our analysis is publicly available at: https://doi.org/10.5281/zenodo.10183679 .

Dasgupta, S. et al. Effects of climate change on combined labour productivity and supply: an empirical, multi-model study. Lancet Planet. Health 5 , e455–e465 (2021).

Article   PubMed   Google Scholar  

de Lima, C. Z. et al. Heat stress on agricultural workers exacerbates crop impacts of climate change. Environ. Res. Lett. 16 , 044020 (2021).

Article   ADS   Google Scholar  

Moore, F. C., Baldos, U., Hertel, T. & Diaz, D. New science of climate change impacts on agriculture implies higher social cost of carbon. Nat. Commun. 8 , 1–9 (2017).

Article   Google Scholar  

Moore, F. C., Baldos, U. L. C. & Hertel, T. Economic impacts of climate change on agriculture: a comparison of process-based and statistical yield models. Environ. Res. Lett. 12 , 065008 (2017).

Auffhammer, M., Baylis, P. & Hausman, C. H. Climate change is projected to have severe impacts on the frequency and intensity of peak electricity demand across the United States. Proc. Natl. Acad. Sci. 114 , 1886–1891 (2017).

Article   CAS   PubMed   PubMed Central   ADS   Google Scholar  

Wenz, L., Levermann, A. & Auffhammer, M. North–south polarization of European electricity consumption under future warming. Proc. Natl. Acad. Sci. 114 , E7910–E7918 (2017).

Song, X. et al. Impact of ambient temperature on morbidity and mortality: an overview of reviews. Sci. Total Environ. 586 , 241–254 (2017).

Article   CAS   PubMed   ADS   Google Scholar  

Guo, Y. et al. Temperature variability and mortality: a multi-country study. Environ. Health Perspect. 124 , 1554–1559 (2016).

Article   PubMed   PubMed Central   Google Scholar  

Dell, M., Jones, B. F. & Olken, B. A. Temperature shocks and economic growth: Evidence from the last half century. Am. Econ. J. Macroecon. 4 , 66–95 (2012).

Burke, M., Hsiang, S. M. & Miguel, E. Global non-linear effect of temperature on economic production. Nature 527 , 235–239 (2015).

Kalkuhl, M. & Wenz, L. The impact of climate conditions on economic production. Evidence from a global panel of regions. J. Environ. Econ. Manag. 103 , 102360 (2020).

Kotz, M., Wenz, L., Stechemesser, A., Kalkuhl, M. & Levermann, A. Day-to-day temperature variability reduces economic growth. Nat. Clim. Change 11 , 319–325 (2021).

Kotz, M., Levermann, A. & Wenz, L. The effect of rainfall changes on economic production. Nature 601 , 223–227 (2022).

Bressler, R. D. The mortality cost of carbon. Nat. Commun. 12 , 1–12 (2021).

Rode, A. et al. Estimating a social cost of carbon for global energy consumption. Nature 598 , 308–314 (2021).

Burke, M., Davis, W. M. & Diffenbaugh, N. S. Large potential reduction in economic damages under UN mitigation targets. Nature 557 , 549–553 (2018).

Moore, F. C. & Diaz, D. B. Temperature impacts on economic growth warrant stringent mitigation policy. Nat. Clim. Change 5 , 127–131 (2015).

Dotsey, M. & Ireland, P. The welfare cost of inflation in general equilibrium. J. Monet. Econ. 37 , 29–47 (1996).

Chu, A. C. & Lai, C.-C. Money and the welfare cost of inflation in an R&D growth model. J. Money Credit Bank 45 , 233–249 (2013).

Gazdar, H. & Mallah, H. B. Inflation and food security in Pakistan: Impact and coping strategies. IDS Bull 44 , 31–37 (2013).

Nord, M., Coleman-Jensen, A. & Gregory, C. Prevalence of US food insecurity is related to changes in unemployment, inflation, and the price of food . (2014).

Abaidoo, R. & Agyapong, E. K. Commodity price volatility, inflation uncertainty and political stability. Int. Rev. Econ . 1–31 (2022).

Molina, G. G., Montoya-Aguirre, M., and Ortiz-Juarez, E. Addressing the cost-of-living crisis in developing countries: Poverty and vulnerability projections and policy responses. U. N. High-Level Polit. Forum Sustain. Dev. HLPF (2022).

Bolton, P. et al. The green swan. BIS Books (2020).

Boneva, L., Ferrucci, G. & Mongelli, F. P. To be or not to be “green”: how can monetary policy react to climate change? ECB Occas. Pap . (2021).

Drudi, F. et al. Climate change and monetary policy in the euro area. (2021).

Parker, M. The impact of disasters on inflation. Econ. Disasters Clim. Change 2 , 21–48 (2018).

Faccia, D., Parker, M. & Stracca, L. Feeling the heat: extreme temperatures and price stability. ECB Work. Pap. Ser . (2021).

Mukherjee, K. & Ouattara, B. Climate and monetary policy: do temperature shocks lead to inflationary pressures? Clim. Change 167 , 1–21 (2021).

Ciccarelli, M., Kuik, F. & Hernández, C. M. The asymmetric effects of weather shocks on euro area inflation. ECB Work. Pap. Ser. , (2023).

Moessner, R. Effects of Precipitation on Food Consumer Price Inflation. (2022).

Hersbach, H. et al. The ERA5 global reanalysis. Q. J. R. Meteorol. Soc . 146 , 1999–2049 (2020).

Osbat, C., Parker, M. & Franceschi, E. Navigating the global inflation surge: Rising tide or choppy wake? ECB Mimeo (2024).

Auffhammer, M., Hsiang, S. M., Schlenker, W. & Sobel, A. Using weather data and climate model output in economic analyses of climate change. Rev. Environ. Econ. Policy (2020).

Wheeler, T. R., Craufurd, P. Q., Ellis, R. H., Porter, J. R. & Prasad, P. V. Temperature variability and the yield of annual crops. Agric. Ecosyst. Environ. 82 , 159–167 (2000).

Liang, X.-Z. et al. Determining climate effects on US total agricultural productivity. Proc. Natl. Acad. Sci. 114 , E2285–E2292 (2017).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Damania, R. The economics of water scarcity and variability. Oxf. Rev. Econ. Policy 36 , 24–44 (2020).

Driscoll, J. C. & Kraay, A. C. Consistent covariance matrix estimation with spatially dependent panel data. Rev. Econ. Stat. 80 , 549–560 (1998).

Ha, J., Kose, M. A. & Ohnsorge, F. One-stop source: A global database of inflation. J. Int. Money Finance 102896 (2023).

Arias, P. et al. Climate Change 2021: The Physical Science Basis. Contribution of Working Group14 I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change; Technical Summary. (2021).

Riahi, K. et al. The shared socioeconomic pathways and their energy, land use, and greenhouse gas emissions implications: an overview. Glob. Environ. Change 42 , 153–168 (2017).

NGFS. NGFS Climate Scenarios for Central Banks and Supervisors. (2021).

Lenton, T. M., Dakos, V., Bathiany, S. & Scheffer, M. Observed trends in the magnitude and persistence of monthly temperature variability. Sci. Rep. 7 , 1–10 (2017).

Article   CAS   Google Scholar  

van der Wiel, K. & Bintanja, R. Contribution of climatic changes in mean and variability to monthly temperature and precipitation extremes. Commun. Earth Environ. 2 , 1–11 (2021).

Kuhla, K., Willner, S. N., Otto, C., Wenz, L. & Levermann, A. Future heat stress to reduce people’s purchasing power. PloS One 16 , e0251210 (2021).

Consolo, A., Koester, G., Nickel, C., Porqueddu, M. & Smets, F. The Need for an Inflation Buffer in the Ecb’s Price Stability Objective–the Role of Nominal Rigidities and Inflation Differentials. ECB Occas. Pap . (2021).

Kuhn, T., Van Elsas, E., Hakhverdian, A. & van der Brug, W. An ever wider gap in an ever closer union: Rising inequalities and euroscepticism in 12 West European democracies, 1975–2009. Socio-Econ. Rev. 14 , 27–45 (2016).

Iizumi, T. et al. Prediction of seasonal climate-induced variations in global food production. Nat. Clim. Change 3 , 904–908 (2013).

Vidal-Quadras Costa, I., Saez Moreno, M., Kuik, F. & Osbat, C. Forecasting food inflation with climate and commodities data using a non linear model. forthcoming .

Cecioni, M. et al. The Ecb’s Price Stability Framework: Past Experience, and Current and Future Challenges. (2021).

Vicente-Serrano, S. M., Beguería, S. & López-Moreno, J. I. A multiscalar drought index sensitive to global warming: the standardized precipitation evapotranspiration index. J. Clim. 23 , 1696–1718 (2010).

Klein Goldewijk, K., Beusen, A., Van Drecht, G. & De Vos, M. The HYDE 3.1 spatially explicit database of human-induced global land-use change over the past 12,000 years. Glob. Ecol. Biogeogr. 20 , 73–86 (2011).

Cobham, D. Monetary policy frameworks since Bretton Woods, across the world and its regions. World Its Reg. Sept . 12 2022 (2022).

Hausfather, Z. & Peters, G. P. Emissions–the ‘business as usual’story is misleading. Nature 577 , 618–620 (2020).

Schwalm, C. R., Glendon, S. & Duffy, P. B. RCP8. 5 tracks cumulative CO2 emissions. Proc. Natl. Acad. Sci. 117 , 19656–19657 (2020).

Cucchi, M. et al. WFDE5: bias-adjusted ERA5 reanalysis data for impact studies. Earth Syst. Sci. Data 12 , 2097–2120 (2020).

Lange, S. Trend-preserving bias adjustment and statistical downscaling with ISIMIP3BASD (v1. 0). Geosci. Model Dev. 12 , 3055–3070 (2019).

Download references

Acknowledgements

We gratefully acknowledge the work of Miles Parker and Chiara Osbat in collecting the data on consumer price indices and providing it to us prior to its publication. We also thank them both for insightful feedback on early versions of the manuscript. MK gratefully acknowledges funding from the Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH on behalf of the Government of the Federal Republic of Germany and Federal Ministry for Economic Cooperation and Development (BMZ). Part of the work for this paper was also conducted during a joint project procured by the European Central Bank. We note that the views expressed are those of the authors and should not be reported as representing the views of the European Central Bank (ECB) or the Eurosystem.

Author information

Authors and affiliations.

Research Department IV Complexity Science, Potsdam Institute for Climate Impact Research, Potsdam, Germany

Maximilian Kotz

Institute of Physics, Potsdam University, Potsdam, Germany

European Central Bank, Frankfurt, Germany

Friderike Kuik, Eliza Lis & Christiane Nickel

You can also search for this author in PubMed   Google Scholar

Contributions

M.K. designed the weather variables, empirical approach, the use of physical climate models, conducted all analyses, produced all figures and lead the writing of the manuscript. F.K. proposed the collaboration, contributed to the design of the weather variables and the empirical approach as well as the interpretation of the results and writing of the manuscript. E.L. contributed to the design of the weather variables and the empirical approach as well as the interpretation of the results and writing of the manuscript. C.N. gave feedback on results and contributed to the writing of the manuscript.

Corresponding authors

Correspondence to Maximilian Kotz or Friderike Kuik .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Peer review

Peer review information.

Communications Earth & Environment thanks Simon Dikau and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editors: Alessandro Rubino and Clare Davis. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary information, peer review file, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Kotz, M., Kuik, F., Lis, E. et al. Global warming and heat extremes to enhance inflationary pressures. Commun Earth Environ 5 , 116 (2024). https://doi.org/10.1038/s43247-023-01173-x

Download citation

Received : 27 March 2023

Accepted : 08 December 2023

Published : 21 March 2024

DOI : https://doi.org/10.1038/s43247-023-01173-x

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List

Climate change effects on desert ecosystems: A case study on the keystone species of the Namib Desert Welwitschia mirabilis

Pierluigi bombi.

1 Institute of Research on Terrestrial Ecosystems, National Research Council, Monterotondo, Rome, Italy

Daniele Salvi

2 Department of Health, Life and Environmental Sciences, University of L’Aquila, Coppito, L’Aquila, Italy

Titus Shuuya

3 Gobabeb Namib Research Institute, Walvis Bay, Namibia

Leonardo Vignoli

4 Department of Science, University of Roma Tre, Rome, Italy

Theo Wassenaar

5 Department of Agriculture and Natural Resources Sciences, Namibia University of Science and Technology, Windhoek, Namibia

Associated Data

Data cannot be shared publicly because they regard endangered populations of a species that is potentially sensitive to illegal collections. Data are available from the CNR-IRET Institutional Data Access (contact via ti.rnc.teri@aireterges ) for researchers who meet the criteria for access to confidential data.

Deserts have been predicted to be one of the most responsive ecosystems to global climate change. In this study, we examine the spatial and demographic response of a keystone endemic plant of the Namib Desert ( Welwitschia mirabilis ), for which displacement and reduction of suitable climate has been foreseen under future conditions. The main aim is to assess the association between ongoing climate change and geographical patterns of welwitschia health, reproductive status, and size. We collected data on welwitschia distribution, health condition, reproductive status, and plant size in northern Namibia. We used ecological niche models to predict the expected geographic shift of suitability under climate change scenarios. For each variable, we compared our field measurements with the expected ongoing change in climate suitability. Finally, we tested the presence of simple geographical gradients in the observed patterns. The historically realized thermal niche of welwitschia will be almost completely unavailable in the next 30 years in northern Namibia. Expected reductions of climatic suitability in our study sites were strongly associated with indicators of negative population conditions, namely lower plant health, reduced recruitment and increased adult mortality. Population condition does not follow simple latitudinal or altitudinal gradients. The observed pattern of population traits is consistent with climate change trends and projections. This makes welwitschia a suitable bioindicator (i.e. a ‘sentinel’) for climate change effect in the Namib Desert ecosystems. Our spatially explicit approach, combining suitability modeling with geographic combinations of population conditions measured in the field, could be extensively adopted to identify sentinel species, and detect population responses to climate change in other regions and ecosystems.

Introduction

Climate change is one of the strongest threats for ecosystems worldwide. Variations in the density of species, range changes, and extinction events have been documented at local and global level [ 1 – 3 ]. Furthermore, changes in species diversity, ecosystem functioning, and service provision are expected for the future as a consequence of climatic pressures on natural populations [ 4 – 6 ]. In Africa, deep impacts by climate change have been forecasted for animals [ 7 – 9 ], plants [ 10 – 12 ], and biodiversity in general [ 4 , 13 ].

Desert ecosystems are predicted to be one of the most vulnerable ecosystems to global climate change [ 14 – 16 ]. Rising temperature, decreasing rainfall, and increasing atmospheric CO 2 , are expected to strongly affect the structure and function of desert ecosystems [ 15 , 17 ]. Desert-adapted species are vulnerable to climate change [ 18 ] and among them endemic plant species are particularly susceptible to the loss of suitable habitat [ 19 ]. The negative effect of climate change on desert plants has been demonstrated worldwide [ 20 ].

In the arid regions of southern Africa, projections of climate change impacts on species persistence indicate a high vulnerability of endemic plant diversity to climate change [ 19 , 21 ]. For example, climate-linked increases of mortality have been observed for the quiver tree ( Aloidendron dichotomum Klopper & Gideon 2013; [ 22 ], and a potential decrease of climatic suitability was recently pointed out by Bombi [ 23 ] for welwitschia (common name for Welwitschia mirabilis Hooker 1863). Welwitschia mirabilis is regarded as a living fossil, representing an ancient lineage of gymnosperm plants, and it is recognized as a symbol of the Namib Desert biodiversity. This species has a peculiar morphology, being a long-living dwarf tree with only two leaves growing throughout its entire life [ 24 ]. This is also a keystone species for the Namib desert ecosystems, where it provides food, water, and refuge for many animal species, including mammals, reptiles, and insects [ 25 , 26 ]. Welwitschia mirabilis is endemic to the central and northern Namib Desert, ranging between the Kuiseb River in Namibia and the Nicolau River, north of Namibe, in Angola [ 27 , 28 ]. In this area, welwitschia plants occur in four separated sub-ranges, three in western Namibia ( Fig 1 ) [ 29 ] and one in south-western Angola. Bombi [ 23 ] showed that populations living in the three Namibian subranges have experienced and will in the future face rather different climatic conditions. The same study predicted a significant reduction of climatic suitability in the northernmost Namibian subrange (which lies in a transition zone between desert and arid savanna) under current climate change scenarios. In particular, the ongoing rise of temperature can drive the local climate out of the realized niche for the northern populations, thus increasing their extinction risk [ 23 ]. Although these findings were potentially important for conservation planning, the study was based on low spatial resolution data (2.5 arcmin) available at the national scale, thus limiting its utility for targeting individual populations.

In the main map, the black polygon indicates the study area, the red polygons show the known species distribution, and the blue polygons represent the boundaries of the observed extent of occurrence in Northern Namibia. In the inset map the red circle indicates the general location of the study area in Africa. Map tiles by Stamen Design, under CC BY 3.0.

Many species respond to climate change by changing their distribution range [ 30 – 32 ]. These changes have been generally described as poleward and/or upward movements to track suitable temperature conditions along latitudinal and altitudinal gradients [ 33 – 35 ]. However, in many cases documented geographic patterns of response are complex and do not align with simple latitudinal and altitudinal shifts [ 36 ]. Indeed, the assumption of simple, uni-directional distribution shifts does not account for intricate interactions among temperature, precipitation, and species-specific tolerances and can lead to substantial underestimation of the effect of climate change on species distributions [ 37 ]. To overcome these drawbacks, one promising approach to quantify possible range changes is based on the comparison of the species-specific spatial pattern of climatic suitability variation (the expected responses), generated by predictive models, with the pattern of appropriate metrics of population conditions measured in the field (the observed responses) [ 38 ]. This approach can increase our ability to identify the effect of climate change on species dynamics.

The main aim of this study was to determine whether the observed geographic combination of population condition (plant health, reproductive status, and size) of welwitschia in northwestern Namibia is associated with ongoing climate change. Secondly, we tested if the same pattern follows a latitudinal or altitudinal gradient in agreement with the assumption of a poleward or upward range shift. More specifically, we wanted to first validate the projections of potential impacts of climate change on W . mirabilis with field-based data. Then, we wanted to assess whether the simple assumption of a poleward/upward range shift is suitable for detecting climate change effects. To do this, we compared the geographic combination of population conditions, measured in the field, with the expected pattern of response, estimated by ecological niche models. If climate change is affecting welwitschia populations, we expected the worst population condition in sites where climatic suitability is decreasing and the best where suitability is increasing. Moreover, if a poleward/upward range shift is the major response to climate change, we could expect a latitudinal or altitudinal trend in the observed patterns of response. Since potential divergent responses to climate change by intraspecific lineages have been observed before [ 39 ] and different realized niches were described for each distinct Namibian subrange [ 23 ], we focused on populations in the northern subrange and considered them as an independent ecological unit, with its own climatic niche and with its (sub)specific expected response. By testing our main and secondary hypotheses, we hope to inform the long-term conservation of W . mirabilis and further contribute to the scientific debate on the climate change impacts on biodiversity.

Materials and methods

Field data collection.

During May 2019, we carried out a field expedition in the northernmost Namibian subrange of W . mirabilis , as defined by the ’Digital Atlas of Namibia’ [ 29 ], in order to obtain information relevant for the species conservation. This study was authorized by the National Commission on Research, Science and Technology of Namibia (Research Permit RCIV00032018) and performed in public lands, managed by the Orupembe, Sanitatas, and Okondjombo Communal Conservancies. During the expedition, we spent 10 full days, in a team of six persons, searching for welwitschia plants across the northernmost Namibian subrange by (1) driving at low speed along the available tracks (more than 330 km) while recording the presence of plants in a ~30 m wide transect on each side of the vehicle, and (2) walking across potentially suitable habitats (more than 65 km) in ~60 m wide transects on each side, in both valley bottoms and hill slopes. Doing this, we explored comprehensively more than 65 km 2 (330 km x 30 m x 2 sides + 65 km x 60 m x 2 sides x 6 persons). The starting points and spatial extent of our walking transects were informed by the knowledge of our local team members, who have an intimate knowledge of the area. We are confident that the combination of local knowledge and systematic transects extending beyond the known range have allowed us to establish the extent and characteristics of the majority of this sub-range.

We collected detailed data on four categories of plant traits: plant location, health condition, reproductive status, and plant size. More specifically, we recorded the precise coordinates (using a handheld GPS) (1), the sex (2), and the presence/absence of cones (3) for almost all the individual plants we observed (just a few, unreachable plants were excluded). The health condition (4), the stem diameters (minimum (5) and maximum (6) along the two main axes of the stem), and the mean leaf length (along the curved trajectory of the leaves) (7) were measured in sites with a sufficient number of plants (> 25 for variable 4 and > 35 for variables 5–7) for reducing the effect of chance on categorical and numerical variables. In sites with more than 60 plants, we considered a random subset of ~60 plants. We ranked health condition on a four-point scale (dead, poor, average, good) based on leaf color (see S1 Fig ). Although this is a relatively coarse scale, the brightness of the green color and the ratio of red/brown to green together are a remarkably consistent and accurate indicator of good health condition as measured by photosynthesis efficiency [ 40 ]. The green color of the leaf is associated with the chlorophyll content and the photosynthetic efficiency of the tissues [ 41 , 42 ], which is influenced by environmental stress [ 43 , 44 ]. An estimate of health condition such as the above is both a direct reflection of the environmental (including climatic) stress that the plant experiences and an index of the likelihood that its resistance to parasites might be compromised [ 45 , 46 ]. We expected that changes in local climate will be visible in its leaf color as a quick proxy of plant health.

Observed pattern of response

For each welwitschia stand (defined a posteriori , through a GIS-based analysis, as a group of plants separated from the other groups by a distance larger than the intra-group mean distance), we calculated three categories of synthetic indicators of population response (derived from plant health, reproductive status, and size) from the field-measured data. For each stand, we calculated the proportion of plants that were dead or in poor, average, and good condition to the total number of plants in the stand. We also calculated the reproductive status (the proportion of plants in the stand that had cones) and the plant size (average stem major axis, stem minor axis, and leaf length). The presence/absence of cones were used as a proxy of population recruitment potential instead of other, more common methods (e.g. plant size) in order to gather a trend of the last few years (after 2000) in plants with an extraordinary low growth rate.

In order to test a previously proposed gradient in plant size, health condition, and reproductive status from hill slopes and valley bottoms [ 27 ], we compared these variables for plants growing in the drainage systems with those in steeper locations.

Expected pattern of response

We used a spatially explicit approach based on ecological niche modeling to estimate the geographic combination of plant response expected as a consequence of climate change. To do this, we defined our study area as a bounding box three times larger than the latitudinal and longitudinal extent of the previously known subrange of welwitschia in northern Namibia [ 29 ]. Inside this study area, we fitted models on 1000 pseudo-presence/absence points by using historical (1950–2000) climate data from the WorldClim databank, version 1.3 [ 47 ] at the spatial resolution of 30 arcsec (about 1 km) in the R -based [ 48 ] biomod2 Package [ 49 ]. In order to control the model-associated uncertainty, we adopted an ensemble forecasting approach [ 50 ]. In particular, we used Generalized Linear Models [ 51 ], Generalized Additive Models [ 52 ], Generalized Boosting Models [ 53 ], Classification Tree Analyses [ 54 ], Artificial Neural Network [ 55 ], and Random Forest [ 56 ] methods, which are widely used and recognized as robust methods.

Pseudo-presence/absence points were randomly generated across the study area and classified as presence or absence points based on their position inside or outside the species extent of occurrence, generated as a minimum convex polygon from our detailed distribution data. Doing this, we overcame the problems related to the small number of real sites of presence and to the spatial autocorrelation due to the non-homogeneous distribution of the real sites [ 57 ]. Multicollinearity among predictors was reduced by discarding those with variance inflation factor higher than five [ 58 ]. Thus, four variables were retained for modeling (i.e. Temperature Seasonality, Mean Temperature of the Driest 3-months period, Precipitation Seasonality, Precipitation of the Warmest 3-months period) (See S2 Fig ). Each independent model was projected into the study area under historical climatic conditions and three-fold cross-validated by calculating the true skill statistic (TSS) [ 59 ]. Finally, we generated a single consensus model of historical suitability. More specifically, we used the six individual models to predict a continuous value of suitability (between 0 and 1), so the final consensus suitability is the TSS-weighted average of six values for each 30 arcsec pixel. Suitability ranges between 0 (no suitability) and 1 (perfect suitability). Future climate suitability was predicted by projecting the models into future climatic conditions across the study area. We used Global Climate Models (GCMs) elaborated by 19 research centers within the Coordinated Modelling Intercomparison Project Phase 5 (CMIP5), which represented the basis for the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC-5thAR) [ 60 ]. GCMs were elaborated for the four Representative Concentration Pathways (RCP2.6, RCP4.5, RCP6.0, and RCP8.5) for 2050 (average for 2041–2060) and downscaled at the same spatial resolution as the historical climate data using WorldClim as baseline climate [ 61 ].All the 59 available scenarios of future (2050) climate from CMIP5 were utilized for projecting the models. We compared the historical climate data and the future climatic scenarios by calculating the Multivariate Environmental Similarity Surfaces (MESS) [ 62 ] in order to estimate the reliability of predictions. The MESS calculation estimates the extent to which the predictor variables in the study area are similar to the conditions experienced by the species in the presence sites. Negative MESS values indicate areas where at least one variable is outside the range of the experienced conditions and thus where model predictions can be less robust [ 62 ].

Suitability variation over time was calculated as the difference between future and historical suitability (Suitability variation = future suitability–historical suitability) and assigned to the observed plant stands on the basis of their location. Suitability variation ranges between -1 (perfectly historically suitable and perfectly unsuitable in the future) and 1 (perfectly historically unsuitable and perfectly suitable in the future). Following our approach, the predicted suitability variation represents an ongoing process, ranging between 2000 and 2050, and the population data collected in the field represent a snapshot approximately in the middle of this process.

Association between observed and expected patterns

Test of model-based pattern.

For each variable, we tested the linkage between the observed and the expected patterns of responses at the stand level (i.e. measured synthetic indicators vs modelled suitability variation) by adopting a null-model approach [ 63 – 65 ]. First, we quantified the observed correlation between measured values and expected suitability variation in the same sites ( r obs ) by calculating the Pearson r. Second, we generated in R (as all the other analyses) 30,000 random permutations of the measured values and calculated the simulated correlation with the expected suitability variation for each permutation ( r sim ). Third, we calculated the probability of the null hypothesis that the observed correlation was drawn at random from the distribution of the simulated correlations [ 66 ]. Finally, in order to control the familywise error rate due to multiple comparisons, we corrected our p values adopting the approach proposed by Benjamini & Hochberg [ 67 ]. These corrected p values ( p corr ) measure the level to which the suitability variation (corresponding to the expected response) due to climate change explains the actual responses observed in the different stands.

In addition, we tested whether the observed responses follow a general and simple geographic combination. We tested the hypothesis of a latitudinal (equator-to-pole) or altitudinal (low-to-high elevation) range shift, as if often assumed for detecting climate change impacts on species distributions. To do this, we adopted the same approach used for testing the linkage between the observed and the model-based expected patterns of responses. We contrasted each measured variable with the stand latitudes and altitudes. We quantified the correlation between the measured variable and the latitude/altitude, calculated the probability that the observed correlation comes randomly from the simulated correlations after 30,000 random permutations, and corrected our p values with the Benjamini & Hochberg [ 67 ] approach. With this approach, we assumed that populations at higher latitude/elevation should be in better general conditions (i.e., lower proportion of plants in poor conditions and of dead plants and higher proportion of plants in good conditions and of plants with cones) if climate change effect is following a simple geographic gradient. As a result, we obtained an estimation of the extent to which climate change effects can be explained as a simple geographic gradient.

Overall, we recorded 1330 plants within the known distribution of W . mirabilis in northern Namibia. These plants are clustered in 12 distinct stands, which are scattered across the central part of the known range at elevations between 806 and 991 m above sea level. On the basis of our field effort and the expert knowledge of our local team members, we are confident that the recorded/observed individuals, and their resulting extent of occurrence, represent the majority of plants in this northern-Namibian sub-range. The area of each recorded stand varied from 2000–825,000 m 2 (for a total surface of about 1.5 km 2 ) and the number of plants per stand varied between four and about 400. The extent of occurrence (estimated as the minimum convex hull) of welwitschia in the area covers about 215 km 2 and the inter-stand distance varied from 1.8 to 30 km. Even though we searched extensively throughout the study area, we defined a markedly smaller extent of occurrence than the distribution map previously published for the species in northern Namibia [ 29 ], which was based on very general information.

The available climatic models revealed that the temperature historically available in the current extent of occurrence of welwitschia in northern Namibia is becoming almost completely unavailable in the same area ( Fig 2A ). In particular, annual mean temperature within the stands is rising about 1.5–2.5°C, with strong variations among the different CMIP5 scenarios. In contrast, the total annual precipitation is likely remaining relatively stable ( Fig 2B ), with small reductions or increases forecasted by different scenarios.

Density plot of historical data (in green) and expected future values (in orange) for annual mean temperature (A) and annual precipitation (B) in the welwitschia extent of occurrence. Density plots use kernel density estimates to show the probability density function of the variables.

Observed variation of plant parameters

The most common class of health condition was ‘average’, with 50% of all the plants and a range between 32% and 74% across individual stands being found in this status ( Fig 3B ). Plants in ‘poor’ condition were 32% (range: 11–50%), but only 10% of all plants were in a ‘good’ condition (range: 0–30%) ( Fig 3A and 3C respectively). Seven percent of all plants were dead (range: 0–30%) and 56% (range: 10–90%) had cones ( Fig 3E and 3D respectively). Not all individuals could be sexed, but among those that were, 56% were males, with a sex ratio (males/females) ranging between 0.6–1.7 across stands ( Fig 3F ). Stem major axis and stem minor axis were highly variable, ranging from 2 to 100 cm (18.8 ± 14.1 cm; range: 10–33 cm) and from 0.3 and 55 cm (10.3 ± 9.8 cm; range 4.6–22 cm), respectively ( Fig 3G and 3H respectively). Leaf length varied from almost 0 cm (completely browsed plants) up to 93 cm (18.7 ± 13.4; range 11–40 cm) ( Fig 3I ) (see S1 Table ).

Density plots of measured values for: proportion of plants in poor conditions (A), proportion of plants in average conditions (B), proportion of plants in good conditions (C), proportion of plants with cones (D), proportion of dead plants (E), sex ratio (F), stem major axis (G), stem minor axis (H), and leaf length (I). Density plots use kernel density estimates to show the probability density function of the variables.

The previously proposed difference between plants growing in valley bottoms and on hill slope (Kers, 1967) was not verified. Indeed, most of the tested variables were not significantly different for the two groups of plants (stem minor axis: Mann-Whitney U = 10773, p = 0.49; leaf length: U = 10364, p = 0.1968; health condition: Mann-Whitney U = 189, p = 0.32; reproductive status: χ 2 = 0.05, p = 0.82). The plant stem major axis only was different ( U = 13120, p = 0.02) but in the opposite direction respect to the proposed pattern, with plants on slopes bigger than plants on valleys (mean: 14 ± 7.85 cm and 11 ± 8.48 cm respectively).

Expected pattern of species response

The strongest reduction of climatic suitability is expected in the eastern half of the extent of occurrence of welwitschia, as well as in some areas extending further north and south to the extent of occurrence ( Fig 4 ). On the other hand, our models predict an increase in climatic suitability to the northwest of the current extent of occurrence ( Fig 4 ). The MESS analysis indicated that all the pixels in the species extent of occurrence have positive values (minimum: 1.10, mean: 7.19 ± 2.47), suggesting that geographical extrapolation on future predictions are reliable (see S3 Fig ). As a result, all the recorded stands are expected to be facing suitability reductions in the period 2000–2050, with variability between almost no reduction and complete reduction.

Expected suitability variation from climate change (red and green shades indicate negative and positive variations respectively) calculated as the difference between future and historical suitability. Black polygon indicates the study area, and the blue polygon represents the boundaries of the observed extent of occurrence in Northern Namibia.

Observed vs expected patterns

Stronger predicted reductions of climatic suitability in the stand sites are associated with lower plant health condition, fewer plants with cones, and an increased number of dead plants. More specifically, the proportion of plants in poor condition in each stand increases with the reduction of suitability ( Fig 5A ). In contrast, the proportion of plants in average and good condition decreases as suitability variation decreases ( Fig 5B and 5C ). The proportion of plants with cones (i.e. a proxy of the potential population recruitment) is lower in stands where stronger reductions of climatic suitability are expected ( Fig 5D ). At the same time, the proportion of dead plants (i.e. population mortality) is negatively correlated with the predicted variation of climatic suitability ( Fig 5E ). However, neither the number of plants per stand (i.e. population size) ( Fig 5F ) nor plant body size ( Fig 5G, 5H and 5I ) is correlated with the suitability variation.

Plots of the proportion of plants in poor (A), average (B), and good conditions (C), the proportion of plants with cones (D), the proportion of dead plants (E), the number of plants (F), the stem major (G) and minor axis (H), and the leaf length (I) as functions of the expected suitability variation in the stands. In all the plots, blue dots are values for plant stands, colored lines and areas represent the regression lines and the confidence intervals respectively, and the texts on the top of the plots indicate the results of the null-model correlation analyses (Pearson r and corrected p values for multiple comparisons). Note that suitability variation values (X-axis) are all negative; thus, the reduction of suitability increases from right to left.

The observed geographic combination of species response does not follow any simple geographic gradient. Indeed, the latitude of welwitschia stands is not correlated with any measured variable ( Table 1 ). Similarly, altitude and the measured variables are not correlated ( Table 1 ). Overall, no latitudinal or altitudinal variation is occurring as a response to climate change.

Results of the null-model correlation analyses (Pearson r and corrected p values for multiple comparisons).

The observed pattern of population conditions of welwitschia plants in northern Namibia is consistent with the expected response under climate change, and specifically with predicted variations of climatic suitability. These results strongly suggest that ongoing climate change is affecting the status of welwitschia populations in the area and producing significant changes (i.e. contraction) of the species distribution at the local scale, which translates into a threat for the long-term conservation of the species. On the other hand, the geographic combination of welwitschia response to climate change does not follow a latitudinal or altitudinal gradient. Therefore, the potential impact of climate change on this species would have passed undetected using the common testing approach based on poleward/upwards range shift.

Inter-stand variations of multiple parameters (i.e. plant health conditions and proxies for population trends) are highly correlated with estimated variation in climatic suitability, strongly suggesting a negative effect of the ongoing climate change on welwitschia trees. In this light, the high correlation between the variation of climatic suitability and plant conditions (poor, average, and good) can support a link among climate change, the distribution of plants, and the variation of plant health with observed increase of individuals in poor conditions and the reduction of plants in average or good conditions. The loss of climatic suitability could also affect future population trends, by affecting recruitment (as suggested by the observed reduction of plants with cones) and mortality (as suggested by the observed increase of dead plants). Although we measured static parameters of population condition, the geographic combination of these parameters (observed in 2019) is coherent with the dynamism of a range shift from areas that were suitable in the past (before 2000) to areas that will be suitable in the future (2050). Indeed, bad population conditions such as poor plant health, low reproduction potential, and high mortality provide a proxy of negative population dynamics, which are typically associated with the trailing edge of species range. This study indicates a clear relationship between observed and expected patterns of populations response in welwitschia. The responsivity of welwitschia to climate change as measurable by proxies of population status could make this species a ‘sentinel’ of climate change effects for Namib ecosystems. The application of the approach used in this study is thus encouraged to identify sentinel species in other desert environments around the world for an effective biomonitoring of climate-linked biotic changes.

The visual estimation of plant condition can be considered a rough estimation of chlorophyll content of the leaves and thus of the plant’s photosynthetic efficiency [ 41 , 42 ]. Alterations of photosynthesis is a well-known effect of environmental stress [ 43 , 44 ]. Heat stress in particular inhibits photosynthesis in tropical and subtropical plants [ 68 , 69 ]. This effect can be stronger in arid environments, where water shortage can hamper the leaf temperature mitigation [ 70 ]. On the other hand, studies on other populations of W . mirabilis showed that rainfall is followed by an increase in the plant’s health status [ 40 ]. As a result, we hypothesize that the observed worsening of plant condition is associated with the complex interaction between the significant increase in temperature, which is the main predicted climate alteration occurring in the area ( Fig 2 ), and the constant but limited water availability in the desert environment. However, specifically designed experiments would be needed to tease apart the different possible forces that could cause the observed responses.

Our results are consistent with the study of Bombi [ 23 ] carried out at the national level and at a much coarser spatial resolution. That study suggested a potential impact of climate change on welwitschia populations in northern Namibia and, predicted a general reduction of climatic suitability for W . mirabilis as well as potential effects on population recruitment and thus on population structure. The author postulated that adult welwitschia plants are likely to survive the expected reduction in climate suitability. In contrast, the correlation we observed between mortality and climate change would indicate a less optimistic scenario, with a progressive reduction of plant health, which translates in a potential long-term reduction of population size. This evidence could support the hypothesis that climate is changing faster and/or is becoming too hot/arid even for W . mirabilis . In this regard, it is worth noting that W . mirabilis is not a typical desert-adapted plant (it has C3 metabolism and a relatively high water demand [ 71 ]), and its ancestors probably occupied much more mesic habitats (probably even forests). It is likely that the current range fragmentation was the result of strong aridification during the Tertiary and Quaternary [ 72 ]; with current climate change predictions pointing towards further desertification (increase of 2°C in annual mean temperature is expected by 2050). Furthermore, this evidence should encourage specific management plans for northern Namibian populations and suggests that climate change should be considered among welwitschia conservation issues.

Quantitative data on plant physiological status (e.g. leaf growth rate, photosyntetic efficiency, water use efficiency), on substrate characteristics such as soil moisture profiles, and on population demographic parameters (e.g. annual recruitment, plant growth, annual mortality) are required to obtain a more detailed picture of the occurring alterations and to clarify the possible mechanistic linkage with climatic stress. Repeated physiological and demographic measurements in different sites would make it possible to follow plant responses over time. The activation of programs for the long-term monitoring of the species in the region would be particularly helpful, allowing critical situations to be detected at early stages and planning of effective recovery measures. Obviously, long-term monitoring in this remote area would be difficult and would require the involvement of local communities as well as the provision of significant resources by local and international agencies aimed at the conservation of desert ecosystems in Namibia.

Despite the great interest in W . mirabilis , which is the only living representative of an ancient lineage of gymnosperms [ 73 ], and its key-role in the Namib Desert ecosystems, several aspects of the species distribution and biology are still to be clarified for a science-based conservation strategy. First, the real level of geographic and genetic isolation of the different subranges should be verified in order to identify intra-specific evolutionary and conservation units. Second, an effort to census and make available the current knowledge on species distribution, demography, and conservation should be undertaken. Third, an analysis of climate change impacts should be extended to the other subranges and a science-based assessment of the conservation status should be made at local and global level. This set of measures could significantly contribute to conservation measures for the species that are effective in the long term.

The geographic combination of response we observed in welwitschia is more complex than the simple poleward/upward shift that was often observed for other species [ 33 , 35 , 74 ]. In the case of W . mirabilis populations of northern Namibia, the observed pattern of population conditions, which can represent a response to climate change, follows local contingencies rather than a simple latitudinal or altitudinal trends ( Table 1 ). This could be associated with the small scale of the study, which emphasized the role of local factors (e.g. land morphology, dominant winds, recurring fog), but is also in agreement with previous large-scale studies. Previous studies pointed out that specific responses to climate change can be divergent [ 36 ] and that assuming a simplified poleward/upward species movement can result in climate change impacts being underestimated [ 37 ]. In our specific case, the linkage between climate change and population conditions, which is suggested by our results, would have been completely undetected with a simplified, but frequently used approach based on the assumption of poleward/upward shifts.

The comparison of the expected pattern of response to ongoing climate change, as estimated by suitability modeling, with the observed patterns of population conditions, as measured in the field, appeared as a powerful approach for detecting impacts of climate change on wild species. This approach, proposed by Bombi et al. [ 38 ], allowed the identification of climate change as potentially a major driver of the geographical pattern of welwitschia health, reproductive status, and size we observed in the field. Clearly, such an approach is prone to a certain level of false negative when other factors, not directly related to climate change (e.g. wind, herbivory, parasites), interact, blurring the pattern generated by climate change [ 38 ]. In addition, our small sample size could further increase the probability of type II errors. On the other hand, the probability of false positives is very low and not influenced by non-climatic factors [ 38 ] or sample size, strengthening the meaning of the detection of a climate change-related pattern. This study underlines the importance of considering species responses to climate change as an emergent property of the different effects on individual populations. At a higher biodiversity level, ecosystem responses to climate change can be considered as an emergent property of the effects on individual species. Such a hierarchical relationship provides direction for the application of spatially explicit approaches, such as the one used in this study, to multiple species and across diverse ecosystems. We therefore advocate for the implementation of a large-scale program for the identification of sentinel species of climate change effects. This would allow the detection, estimation, and monitoring of climate change impacts on biodiversity, improving the long-term conservation of species at the ecosystem level.

Our study confirms that desert-adapted species can be vulnerable to the effects of climate change. Although some studies have hypothesized minor impacts on desert ecosystems by climate change [ 75 , 76 ], growing evidence, provided through different approaches, shows severe effects of climate change on species, communities, and ecosystems in arid regions worldwide [ 18 , 77 , 78 ]. Altogether these studies, in agreement with our results, indicate that desert ecosystems are likely to suffer from biodiversity loss with intensifying global warming as result of a reduction of environmental suitability for the endemic biota [ 22 ].

Supporting information

Examples of plants in the four health condition classes (A: Dead; B: Poor; C: Average; D: Good).

Color shades indicate the extent to which the predictor variables in each pixel are similar to the conditions experienced by the species in the presence sites. The black line separates zones with positive values to those with negative values. Negative MESS values indicate areas where at least one variable is outside the range of the experienced conditions and thus where model predictions can be less robust. The blue polygon represents the species extent of occurrence in the area.

S1 Appendix

Acknowledgments.

We wish to thank our Himba guides and friends Riatunga Koruhama and Mavekaumba Tjiposa; Karen Nott of IRDNC, who facilitated access to the area; Vera De Cauwer of the SCIONA project (EuropeAid/156423/DD/ACT/Multi), for the constructive collaboration. We are also grateful to the Okondjombo Communal Conservancy.

Funding Statement

This study was supported by the Mohamed bin Zayed Species Conservation Fund ( www.speciesconservation.org ) to PB (Project N 182519816).

Data Availability

Featured Topics

Featured series.

A series of random questions answered by Harvard experts.

Explore the Gazette

Read the latest.

Why AI fairness conversations must include disabled people

How did you get that frog to float?

Lifting a few with my chatbot

Exxon disputed climate findings for years. its scientists knew better..

View of ExxonMobil storage tanks of the petrochemical industry in the port of Rotterdam, Netherlands.

AP Photo/Peter Dejong

Alice McCarthy

Harvard Correspondent

Research shows that company modeled and predicted global warming with ‘shocking skill and accuracy’ starting in the 1970s

Projections created internally by ExxonMobil starting in the late 1970s on the impact of fossil fuels on climate change were very accurate, even surpassing those of some academic and governmental scientists, according to an analysis published Thursday in Science by a team of Harvard-led researchers. Despite those forecasts, team leaders say, the multinational energy giant continued to sow doubt about the gathering crisis.

In “Assessing ExxonMobil’s Global Warming Projections,” researchers from Harvard and the Potsdam Institute for Climate Impact Research show for the first time the accuracy of previously unreported forecasts created by company scientists from 1977 through 2003. The Harvard team discovered that Exxon researchers created a series of remarkably reliable models and analyses projecting global warming from carbon dioxide emissions over the coming decades. Specifically, Exxon projected that fossil fuel emissions would lead to 0.20 degrees Celsius of global warming per decade, with a margin of error of 0.04 degrees — a trend that has been proven largely accurate.

“This paper is the first ever systematic assessment of a fossil fuel company’s climate projections, the first time we’ve been able to put a number on what they knew,” said Geoffrey Supran, lead author and former research fellow in the History of Science at Harvard. “What we found is that between 1977 and 2003, excellent scientists within Exxon modeled and predicted global warming with, frankly, shocking skill and accuracy only for the company to then spend the next couple of decades denying that very climate science.”

“This paper is the first ever systematic assessment of a fossil fuel company’s climate projections, the first time we’ve been able to put a number on what they knew,” said Geoffrey Supran, lead author.

File photo by Stephanie Mitchell/Harvard Staff Photographer

“We thought this was a unique opportunity to understand what Exxon knew about this issue and what level of scientific understanding they had at the time,” added co-author Naomi Oreskes , Henry Charles Lea Professor of the History of Science whose work looks at the causes and effects of climate change denial. “We found that not only were their forecasts extremely skillful, but they were also often more skillful than forecasts made by independent academic and government scientists at the exact same time.”

Allegations that oil company executives sought to mislead the public about the industry’s role in climate change have drawn increasing scrutiny in recent years, including lawsuits by several states and cities and a recent high profile U.S. House committee investigation.

Harvard’s scientists used established Intergovernmental Panel on Climate Change (IPCC) statistical techniques to test the performance of Exxon’s models. They found that, depending on the metric used, 63-83 percent of the global warming projections reported by Exxon scientists were consistent with actual temperatures over time. Moreover, the corporation’s own projections had an average “skill score” of 72 percent, plus or minus 6 percent, with the highest scoring 99 percent. A skill score relates to how well a forecast compares to what happens in real life. For comparison, NASA scientist James Hansen’s global warming predictions presented to the U.S. Congress in 1988 had scores from 38 to 66 percent.

The researchers report that Exxon scientists correctly dismissed the possibility of a coming ice age, accurately predicted that human-caused global warming would first be detectable in the year 2000, plus or minus five years, and reasonably estimated how much CO 2 would lead to dangerous warming.

The current debate about when Exxon knew about the impact on climate change carbon emissions began in 2015 following news reports of internal company documents describing the multinational’s early knowledge of climate science.  Exxon disagreed with the reports, even providing a link to internal studies and memos from their own scientists and suggesting that interested parties should read them and make up their own minds.

“That’s exactly what we did,” said Supran, who is now at the University of Miami. Together, he and Oreskes spent a year researching those documents and in 2017 published a series of three papers analyzing Exxon’s 40-year history of climate communications . They were able to show there was a systematic discrepancy between what Exxon was saying internally and in academic circles versus what they were telling the public. “That led us to conclude that they had quantifiably misled the public, by essentially contributing quietly to climate science and yet loudly promoting doubt about that science,” said Supran.

“I think this new study is the smoking gun, the proof, because it shows the degree of understanding … this really deep, really sophisticated, really skillful understanding that was obscured by what came next,” said Harvard Professor Naomi Oreskes.

Harvard file photo

In 2021, the team published a new study in One Earth using algorithmic techniques to identify ways in which ExxonMobil used increasingly subtle but systematic language to shape the way the public talks and thinks about climate change — often in misleading ways.

These findings were hardly a surprise to Oreskes, given her long history of studying climate communications from fossil fuel companies, work that drew national attention with her 2010 bestseller, “Merchants of Doubt.” In it she and co-author, Caltech researcher Erik Conway, argued that Exxon was aware of the threat of carbon emissions on climate change yet waged a disinformation campaign about the problem.  Despite the book’s popularity and the peer-reviewed papers with Supran, however, some continued to wonder whether she could prove the effect these campaigns had, if they indeed made a difference.

“I think this new study is the smoking gun, the proof, because it shows the degree of understanding … this really deep, really sophisticated, really skillful understanding that was obscured by what came next,” Oreskes said. “It proves a point I’ve argued for years that ExxonMobil scientists knew about this problem to a shockingly fine degree as far back as the 1980s, but company spokesmen denied, challenged, and obscured this science, starting in the late 1980s/early 1990s.”

Added Supran: “Our analysis here I think seals the deal on that matter. We now have totally unimpeachable evidence that Exxon accurately predicted global warming years before it turned around and publicly attacked climate science and scientists.”

The authors of this research were supported by a Rockefeller Family Fund grant and Harvard University Faculty Development funds.

Share this article

You might like.

Tech offers promise to help yet too often perpetuates ableism, say researchers. It doesn’t have to be this way.

Ever-creative, Nobel laureate in physics Andre Geim extols fun, fanciful side of very serious science

Sociologist Sherry Turkle warns against growing trend of turning to AI for companionship, counsel

College accepts 1,937 to Class of 2028

Students represent 94 countries, all 50 states

Pushing back on DEI ‘orthodoxy’

Panelists support diversity efforts but worry that current model is too narrow, denying institutions the benefit of other voices, ideas

So what exactly makes Taylor Swift so great?

Experts weigh in on pop superstar's cultural and financial impact as her tours and albums continue to break records.

Vital Signs

Global temperature, key takeaway:.

The 10 most recent years are the warmest years on record.

This graph shows the change in global surface temperature compared to the long-term average from 1951 to 1980. Earth’s average surface temperature in 2023 was the warmest on record since recordkeeping began in 1880 (source: NASA/GISS ). NASA’s analysis generally matches independent analyses prepared by the​ National Oceanic and Atmospheric Administration (NOAA) and other research groups. Overall, Earth was about 2.45 degrees Fahrenheit (or about 1.36 degrees Celsius) warmer in 2023 than in the late 19th-century (1850-1900) preindustrial average. The 10 most recent years are the warmest on record.

The animation on the right shows the change in global surface temperatures. Dark blue shows areas cooler than average. Dark red shows areas warmer than average. Short-term variations are smoothed out using a 5-year running average to make trends more visible in this map.

The data shown are the latest available, updated annually.

GLOBAL LAND-OCEAN TEMPERATURE INDEX

Time series: 1884 to 2022.

• Share full article

The Science of Climate Change Explained: Facts, Evidence and Proof

Definitive answers to the big questions.

Credit... Photo Illustration by Andrea D'Aquino

Supported by

By Julia Rosen

Ms. Rosen is a journalist with a Ph.D. in geology. Her research involved studying ice cores from Greenland and Antarctica to understand past climate changes.

• Published April 19, 2021 Updated Nov. 6, 2021

The science of climate change is more solid and widely agreed upon than you might think. But the scope of the topic, as well as rampant disinformation, can make it hard to separate fact from fiction. Here, we’ve done our best to present you with not only the most accurate scientific information, but also an explanation of how we know it.

How do we know climate change is really happening?

How much agreement is there among scientists about climate change, do we really only have 150 years of climate data how is that enough to tell us about centuries of change, how do we know climate change is caused by humans, since greenhouse gases occur naturally, how do we know they’re causing earth’s temperature to rise, why should we be worried that the planet has warmed 2°f since the 1800s, is climate change a part of the planet’s natural warming and cooling cycles, how do we know global warming is not because of the sun or volcanoes, how can winters and certain places be getting colder if the planet is warming, wildfires and bad weather have always happened. how do we know there’s a connection to climate change, how bad are the effects of climate change going to be, what will it cost to do something about climate change, versus doing nothing.

Climate change is often cast as a prediction made by complicated computer models. But the scientific basis for climate change is much broader, and models are actually only one part of it (and, for what it’s worth, they’re surprisingly accurate ).

For more than a century , scientists have understood the basic physics behind why greenhouse gases like carbon dioxide cause warming. These gases make up just a small fraction of the atmosphere but exert outsized control on Earth’s climate by trapping some of the planet’s heat before it escapes into space. This greenhouse effect is important: It’s why a planet so far from the sun has liquid water and life!

However, during the Industrial Revolution, people started burning coal and other fossil fuels to power factories, smelters and steam engines, which added more greenhouse gases to the atmosphere. Ever since, human activities have been heating the planet.

We know this is true thanks to an overwhelming body of evidence that begins with temperature measurements taken at weather stations and on ships starting in the mid-1800s. Later, scientists began tracking surface temperatures with satellites and looking for clues about climate change in geologic records. Together, these data all tell the same story: Earth is getting hotter.

Average global temperatures have increased by 2.2 degrees Fahrenheit, or 1.2 degrees Celsius, since 1880, with the greatest changes happening in the late 20th century. Land areas have warmed more than the sea surface and the Arctic has warmed the most — by more than 4 degrees Fahrenheit just since the 1960s. Temperature extremes have also shifted. In the United States, daily record highs now outnumber record lows two-to-one.

Where it was cooler or warmer in 2020 compared with the middle of the 20th century

This warming is unprecedented in recent geologic history. A famous illustration, first published in 1998 and often called the hockey-stick graph, shows how temperatures remained fairly flat for centuries (the shaft of the stick) before turning sharply upward (the blade). It’s based on data from tree rings, ice cores and other natural indicators. And the basic picture , which has withstood decades of scrutiny from climate scientists and contrarians alike, shows that Earth is hotter today than it’s been in at least 1,000 years, and probably much longer.

In fact, surface temperatures actually mask the true scale of climate change, because the ocean has absorbed 90 percent of the heat trapped by greenhouse gases . Measurements collected over the last six decades by oceanographic expeditions and networks of floating instruments show that every layer of the ocean is warming up. According to one study , the ocean has absorbed as much heat between 1997 and 2015 as it did in the previous 130 years.

We also know that climate change is happening because we see the effects everywhere. Ice sheets and glaciers are shrinking while sea levels are rising. Arctic sea ice is disappearing. In the spring, snow melts sooner and plants flower earlier. Animals are moving to higher elevations and latitudes to find cooler conditions. And droughts, floods and wildfires have all gotten more extreme. Models predicted many of these changes, but observations show they are now coming to pass.

Back to top .

There’s no denying that scientists love a good, old-fashioned argument. But when it comes to climate change, there is virtually no debate: Numerous studies have found that more than 90 percent of scientists who study Earth’s climate agree that the planet is warming and that humans are the primary cause. Most major scientific bodies, from NASA to the World Meteorological Organization , endorse this view. That’s an astounding level of consensus given the contrarian, competitive nature of the scientific enterprise, where questions like what killed the dinosaurs remain bitterly contested .

Scientific agreement about climate change started to emerge in the late 1980s, when the influence of human-caused warming began to rise above natural climate variability. By 1991, two-thirds of earth and atmospheric scientists surveyed for an early consensus study said that they accepted the idea of anthropogenic global warming. And by 1995, the Intergovernmental Panel on Climate Change, a famously conservative body that periodically takes stock of the state of scientific knowledge, concluded that “the balance of evidence suggests that there is a discernible human influence on global climate.” Currently, more than 97 percent of publishing climate scientists agree on the existence and cause of climate change (as does nearly 60 percent of the general population of the United States).

So where did we get the idea that there’s still debate about climate change? A lot of it came from coordinated messaging campaigns by companies and politicians that opposed climate action. Many pushed the narrative that scientists still hadn’t made up their minds about climate change, even though that was misleading. Frank Luntz, a Republican consultant, explained the rationale in an infamous 2002 memo to conservative lawmakers: “Should the public come to believe that the scientific issues are settled, their views about global warming will change accordingly,” he wrote. Questioning consensus remains a common talking point today, and the 97 percent figure has become something of a lightning rod .

To bolster the falsehood of lingering scientific doubt, some people have pointed to things like the Global Warming Petition Project, which urged the United States government to reject the Kyoto Protocol of 1997, an early international climate agreement. The petition proclaimed that climate change wasn’t happening, and even if it were, it wouldn’t be bad for humanity. Since 1998, more than 30,000 people with science degrees have signed it. However, nearly 90 percent of them studied something other than Earth, atmospheric or environmental science, and the signatories included just 39 climatologists. Most were engineers, doctors, and others whose training had little to do with the physics of the climate system.

A few well-known researchers remain opposed to the scientific consensus. Some, like Willie Soon, a researcher affiliated with the Harvard-Smithsonian Center for Astrophysics, have ties to the fossil fuel industry . Others do not, but their assertions have not held up under the weight of evidence. At least one prominent skeptic, the physicist Richard Muller, changed his mind after reassessing historical temperature data as part of the Berkeley Earth project. His team’s findings essentially confirmed the results he had set out to investigate, and he came away firmly convinced that human activities were warming the planet. “Call me a converted skeptic,” he wrote in an Op-Ed for the Times in 2012.

Mr. Luntz, the Republican pollster, has also reversed his position on climate change and now advises politicians on how to motivate climate action.

A final note on uncertainty: Denialists often use it as evidence that climate science isn’t settled. However, in science, uncertainty doesn’t imply a lack of knowledge. Rather, it’s a measure of how well something is known. In the case of climate change, scientists have found a range of possible future changes in temperature, precipitation and other important variables — which will depend largely on how quickly we reduce emissions. But uncertainty does not undermine their confidence that climate change is real and that people are causing it.

Earth’s climate is inherently variable. Some years are hot and others are cold, some decades bring more hurricanes than others, some ancient droughts spanned the better part of centuries. Glacial cycles operate over many millenniums. So how can scientists look at data collected over a relatively short period of time and conclude that humans are warming the planet? The answer is that the instrumental temperature data that we have tells us a lot, but it’s not all we have to go on.

Historical records stretch back to the 1880s (and often before), when people began to regularly measure temperatures at weather stations and on ships as they traversed the world’s oceans. These data show a clear warming trend during the 20th century.

Global average temperature compared with the middle of the 20th century

+0.75°C

–0.25°

Some have questioned whether these records could be skewed, for instance, by the fact that a disproportionate number of weather stations are near cities, which tend to be hotter than surrounding areas as a result of the so-called urban heat island effect. However, researchers regularly correct for these potential biases when reconstructing global temperatures. In addition, warming is corroborated by independent data like satellite observations, which cover the whole planet, and other ways of measuring temperature changes.

Much has also been made of the small dips and pauses that punctuate the rising temperature trend of the last 150 years. But these are just the result of natural climate variability or other human activities that temporarily counteract greenhouse warming. For instance, in the mid-1900s, internal climate dynamics and light-blocking pollution from coal-fired power plants halted global warming for a few decades. (Eventually, rising greenhouse gases and pollution-control laws caused the planet to start heating up again.) Likewise, the so-called warming hiatus of the 2000s was partly a result of natural climate variability that allowed more heat to enter the ocean rather than warm the atmosphere. The years since have been the hottest on record .

Still, could the entire 20th century just be one big natural climate wiggle? To address that question, we can look at other kinds of data that give a longer perspective. Researchers have used geologic records like tree rings, ice cores, corals and sediments that preserve information about prehistoric climates to extend the climate record. The resulting picture of global temperature change is basically flat for centuries, then turns sharply upward over the last 150 years. It has been a target of climate denialists for decades. However, study after study has confirmed the results , which show that the planet hasn’t been this hot in at least 1,000 years, and probably longer.

Scientists have studied past climate changes to understand the factors that can cause the planet to warm or cool. The big ones are changes in solar energy, ocean circulation, volcanic activity and the amount of greenhouse gases in the atmosphere. And they have each played a role at times.

For example, 300 years ago, a combination of reduced solar output and increased volcanic activity cooled parts of the planet enough that Londoners regularly ice skated on the Thames . About 12,000 years ago, major changes in Atlantic circulation plunged the Northern Hemisphere into a frigid state. And 56 million years ago, a giant burst of greenhouse gases, from volcanic activity or vast deposits of methane (or both), abruptly warmed the planet by at least 9 degrees Fahrenheit, scrambling the climate, choking the oceans and triggering mass extinctions.

In trying to determine the cause of current climate changes, scientists have looked at all of these factors . The first three have varied a bit over the last few centuries and they have quite likely had modest effects on climate , particularly before 1950. But they cannot account for the planet’s rapidly rising temperature, especially in the second half of the 20th century, when solar output actually declined and volcanic eruptions exerted a cooling effect.

That warming is best explained by rising greenhouse gas concentrations . Greenhouse gases have a powerful effect on climate (see the next question for why). And since the Industrial Revolution, humans have been adding more of them to the atmosphere, primarily by extracting and burning fossil fuels like coal, oil and gas, which releases carbon dioxide.

Bubbles of ancient air trapped in ice show that, before about 1750, the concentration of carbon dioxide in the atmosphere was roughly 280 parts per million. It began to rise slowly and crossed the 300 p.p.m. threshold around 1900. CO2 levels then accelerated as cars and electricity became big parts of modern life, recently topping 420 p.p.m . The concentration of methane, the second most important greenhouse gas, has more than doubled. We’re now emitting carbon much faster than it was released 56 million years ago .

30 billion metric tons

Carbon dioxide emitted worldwide 1850-2017

Rest of world

Other developed

European Union

Developed economies

Other countries

United States

E.U. and U.K.

These rapid increases in greenhouse gases have caused the climate to warm abruptly. In fact, climate models suggest that greenhouse warming can explain virtually all of the temperature change since 1950. According to the most recent report by the Intergovernmental Panel on Climate Change, which assesses published scientific literature, natural drivers and internal climate variability can only explain a small fraction of late-20th century warming.

Another study put it this way: The odds of current warming occurring without anthropogenic greenhouse gas emissions are less than 1 in 100,000 .

But greenhouse gases aren’t the only climate-altering compounds people put into the air. Burning fossil fuels also produces particulate pollution that reflects sunlight and cools the planet. Scientists estimate that this pollution has masked up to half of the greenhouse warming we would have otherwise experienced.

Greenhouse gases like water vapor and carbon dioxide serve an important role in the climate. Without them, Earth would be far too cold to maintain liquid water and humans would not exist!

Here’s how it works: the planet’s temperature is basically a function of the energy the Earth absorbs from the sun (which heats it up) and the energy Earth emits to space as infrared radiation (which cools it down). Because of their molecular structure, greenhouse gases temporarily absorb some of that outgoing infrared radiation and then re-emit it in all directions, sending some of that energy back toward the surface and heating the planet . Scientists have understood this process since the 1850s .

Greenhouse gas concentrations have varied naturally in the past. Over millions of years, atmospheric CO2 levels have changed depending on how much of the gas volcanoes belched into the air and how much got removed through geologic processes. On time scales of hundreds to thousands of years, concentrations have changed as carbon has cycled between the ocean, soil and air.

Today, however, we are the ones causing CO2 levels to increase at an unprecedented pace by taking ancient carbon from geologic deposits of fossil fuels and putting it into the atmosphere when we burn them. Since 1750, carbon dioxide concentrations have increased by almost 50 percent. Methane and nitrous oxide, other important anthropogenic greenhouse gases that are released mainly by agricultural activities, have also spiked over the last 250 years.

We know based on the physics described above that this should cause the climate to warm. We also see certain telltale “fingerprints” of greenhouse warming. For example, nights are warming even faster than days because greenhouse gases don’t go away when the sun sets. And upper layers of the atmosphere have actually cooled, because more energy is being trapped by greenhouse gases in the lower atmosphere.

We also know that we are the cause of rising greenhouse gas concentrations — and not just because we can measure the CO2 coming out of tailpipes and smokestacks. We can see it in the chemical signature of the carbon in CO2.

Carbon comes in three different masses: 12, 13 and 14. Things made of organic matter (including fossil fuels) tend to have relatively less carbon-13. Volcanoes tend to produce CO2 with relatively more carbon-13. And over the last century, the carbon in atmospheric CO2 has gotten lighter, pointing to an organic source.

We can tell it’s old organic matter by looking for carbon-14, which is radioactive and decays over time. Fossil fuels are too ancient to have any carbon-14 left in them, so if they were behind rising CO2 levels, you would expect the amount of carbon-14 in the atmosphere to drop, which is exactly what the data show .

It’s important to note that water vapor is the most abundant greenhouse gas in the atmosphere. However, it does not cause warming; instead it responds to it . That’s because warmer air holds more moisture, which creates a snowball effect in which human-caused warming allows the atmosphere to hold more water vapor and further amplifies climate change. This so-called feedback cycle has doubled the warming caused by anthropogenic greenhouse gas emissions.

A common source of confusion when it comes to climate change is the difference between weather and climate. Weather is the constantly changing set of meteorological conditions that we experience when we step outside, whereas climate is the long-term average of those conditions, usually calculated over a 30-year period. Or, as some say: Weather is your mood and climate is your personality.

So while 2 degrees Fahrenheit doesn’t represent a big change in the weather, it’s a huge change in climate. As we’ve already seen, it’s enough to melt ice and raise sea levels, to shift rainfall patterns around the world and to reorganize ecosystems, sending animals scurrying toward cooler habitats and killing trees by the millions.

It’s also important to remember that two degrees represents the global average, and many parts of the world have already warmed by more than that. For example, land areas have warmed about twice as much as the sea surface. And the Arctic has warmed by about 5 degrees. That’s because the loss of snow and ice at high latitudes allows the ground to absorb more energy, causing additional heating on top of greenhouse warming.

Relatively small long-term changes in climate averages also shift extremes in significant ways. For instance, heat waves have always happened, but they have shattered records in recent years. In June of 2020, a town in Siberia registered temperatures of 100 degrees . And in Australia, meteorologists have added a new color to their weather maps to show areas where temperatures exceed 125 degrees. Rising sea levels have also increased the risk of flooding because of storm surges and high tides. These are the foreshocks of climate change.

And we are in for more changes in the future — up to 9 degrees Fahrenheit of average global warming by the end of the century, in the worst-case scenario . For reference, the difference in global average temperatures between now and the peak of the last ice age, when ice sheets covered large parts of North America and Europe, is about 11 degrees Fahrenheit.

Under the Paris Climate Agreement, which President Biden recently rejoined, countries have agreed to try to limit total warming to between 1.5 and 2 degrees Celsius, or 2.7 and 3.6 degrees Fahrenheit, since preindustrial times. And even this narrow range has huge implications . According to scientific studies, the difference between 2.7 and 3.6 degrees Fahrenheit will very likely mean the difference between coral reefs hanging on or going extinct, and between summer sea ice persisting in the Arctic or disappearing completely. It will also determine how many millions of people suffer from water scarcity and crop failures, and how many are driven from their homes by rising seas. In other words, one degree Fahrenheit makes a world of difference.

Earth’s climate has always changed. Hundreds of millions of years ago, the entire planet froze . Fifty million years ago, alligators lived in what we now call the Arctic . And for the last 2.6 million years, the planet has cycled between ice ages when the planet was up to 11 degrees cooler and ice sheets covered much of North America and Europe, and milder interglacial periods like the one we’re in now.

Climate denialists often point to these natural climate changes as a way to cast doubt on the idea that humans are causing climate to change today. However, that argument rests on a logical fallacy. It’s like “seeing a murdered body and concluding that people have died of natural causes in the past, so the murder victim must also have died of natural causes,” a team of social scientists wrote in The Debunking Handbook , which explains the misinformation strategies behind many climate myths.

Indeed, we know that different mechanisms caused the climate to change in the past. Glacial cycles, for example, were triggered by periodic variations in Earth’s orbit , which take place over tens of thousands of years and change how solar energy gets distributed around the globe and across the seasons.

These orbital variations don’t affect the planet’s temperature much on their own. But they set off a cascade of other changes in the climate system; for instance, growing or melting vast Northern Hemisphere ice sheets and altering ocean circulation. These changes, in turn, affect climate by altering the amount of snow and ice, which reflect sunlight, and by changing greenhouse gas concentrations. This is actually part of how we know that greenhouse gases have the ability to significantly affect Earth’s temperature.

For at least the last 800,000 years , atmospheric CO2 concentrations oscillated between about 180 parts per million during ice ages and about 280 p.p.m. during warmer periods, as carbon moved between oceans, forests, soils and the atmosphere. These changes occurred in lock step with global temperatures, and are a major reason the entire planet warmed and cooled during glacial cycles, not just the frozen poles.

Today, however, CO2 levels have soared to 420 p.p.m. — the highest they’ve been in at least three million years . The concentration of CO2 is also increasing about 100 times faster than it did at the end of the last ice age. This suggests something else is going on, and we know what it is: Since the Industrial Revolution, humans have been burning fossil fuels and releasing greenhouse gases that are heating the planet now (see Question 5 for more details on how we know this, and Questions 4 and 8 for how we know that other natural forces aren’t to blame).

Over the next century or two, societies and ecosystems will experience the consequences of this climate change. But our emissions will have even more lasting geologic impacts: According to some studies, greenhouse gas levels may have already warmed the planet enough to delay the onset of the next glacial cycle for at least an additional 50,000 years.

The sun is the ultimate source of energy in Earth’s climate system, so it’s a natural candidate for causing climate change. And solar activity has certainly changed over time. We know from satellite measurements and other astronomical observations that the sun’s output changes on 11-year cycles. Geologic records and sunspot numbers, which astronomers have tracked for centuries, also show long-term variations in the sun’s activity, including some exceptionally quiet periods in the late 1600s and early 1800s.

We know that, from 1900 until the 1950s, solar irradiance increased. And studies suggest that this had a modest effect on early 20th century climate, explaining up to 10 percent of the warming that’s occurred since the late 1800s. However, in the second half of the century, when the most warming occurred, solar activity actually declined . This disparity is one of the main reasons we know that the sun is not the driving force behind climate change.

Another reason we know that solar activity hasn’t caused recent warming is that, if it had, all the layers of the atmosphere should be heating up. Instead, data show that the upper atmosphere has actually cooled in recent decades — a hallmark of greenhouse warming .

So how about volcanoes? Eruptions cool the planet by injecting ash and aerosol particles into the atmosphere that reflect sunlight. We’ve observed this effect in the years following large eruptions. There are also some notable historical examples, like when Iceland’s Laki volcano erupted in 1783, causing widespread crop failures in Europe and beyond, and the “ year without a summer ,” which followed the 1815 eruption of Mount Tambora in Indonesia.

Since volcanoes mainly act as climate coolers, they can’t really explain recent warming. However, scientists say that they may also have contributed slightly to rising temperatures in the early 20th century. That’s because there were several large eruptions in the late 1800s that cooled the planet, followed by a few decades with no major volcanic events when warming caught up. During the second half of the 20th century, though, several big eruptions occurred as the planet was heating up fast. If anything, they temporarily masked some amount of human-caused warming.

The second way volcanoes can impact climate is by emitting carbon dioxide. This is important on time scales of millions of years — it’s what keeps the planet habitable (see Question 5 for more on the greenhouse effect). But by comparison to modern anthropogenic emissions, even big eruptions like Krakatoa and Mount St. Helens are just a drop in the bucket. After all, they last only a few hours or days, while we burn fossil fuels 24-7. Studies suggest that, today, volcanoes account for 1 to 2 percent of total CO2 emissions.

When a big snowstorm hits the United States, climate denialists can try to cite it as proof that climate change isn’t happening. In 2015, Senator James Inhofe, an Oklahoma Republican, famously lobbed a snowball in the Senate as he denounced climate science. But these events don’t actually disprove climate change.

While there have been some memorable storms in recent years, winters are actually warming across the world. In the United States, average temperatures in December, January and February have increased by about 2.5 degrees this century.

On the flip side, record cold days are becoming less common than record warm days. In the United States, record highs now outnumber record lows two-to-one . And ever-smaller areas of the country experience extremely cold winter temperatures . (The same trends are happening globally.)

So what’s with the blizzards? Weather always varies, so it’s no surprise that we still have severe winter storms even as average temperatures rise. However, some studies suggest that climate change may be to blame. One possibility is that rapid Arctic warming has affected atmospheric circulation, including the fast-flowing, high-altitude air that usually swirls over the North Pole (a.k.a. the Polar Vortex ). Some studies suggest that these changes are bringing more frigid temperatures to lower latitudes and causing weather systems to stall , allowing storms to produce more snowfall. This may explain what we’ve experienced in the U.S. over the past few decades, as well as a wintertime cooling trend in Siberia , although exactly how the Arctic affects global weather remains a topic of ongoing scientific debate .

Climate change may also explain the apparent paradox behind some of the other places on Earth that haven’t warmed much. For instance, a splotch of water in the North Atlantic has cooled in recent years, and scientists say they suspect that may be because ocean circulation is slowing as a result of freshwater streaming off a melting Greenland . If this circulation grinds almost to a halt, as it’s done in the geologic past, it would alter weather patterns around the world.

Not all cold weather stems from some counterintuitive consequence of climate change. But it’s a good reminder that Earth’s climate system is complex and chaotic, so the effects of human-caused changes will play out differently in different places. That’s why “global warming” is a bit of an oversimplification. Instead, some scientists have suggested that the phenomenon of human-caused climate change would more aptly be called “ global weirding .”

Extreme weather and natural disasters are part of life on Earth — just ask the dinosaurs. But there is good evidence that climate change has increased the frequency and severity of certain phenomena like heat waves, droughts and floods. Recent research has also allowed scientists to identify the influence of climate change on specific events.

Let’s start with heat waves . Studies show that stretches of abnormally high temperatures now happen about five times more often than they would without climate change, and they last longer, too. Climate models project that, by the 2040s, heat waves will be about 12 times more frequent. And that’s concerning since extreme heat often causes increased hospitalizations and deaths, particularly among older people and those with underlying health conditions. In the summer of 2003, for example, a heat wave caused an estimated 70,000 excess deaths across Europe. (Human-caused warming amplified the death toll .)

Climate change has also exacerbated droughts , primarily by increasing evaporation. Droughts occur naturally because of random climate variability and factors like whether El Niño or La Niña conditions prevail in the tropical Pacific. But some researchers have found evidence that greenhouse warming has been affecting droughts since even before the Dust Bowl . And it continues to do so today. According to one analysis , the drought that afflicted the American Southwest from 2000 to 2018 was almost 50 percent more severe because of climate change. It was the worst drought the region had experienced in more than 1,000 years.

Rising temperatures have also increased the intensity of heavy precipitation events and the flooding that often follows. For example, studies have found that, because warmer air holds more moisture, Hurricane Harvey, which struck Houston in 2017, dropped between 15 and 40 percent more rainfall than it would have without climate change.

It’s still unclear whether climate change is changing the overall frequency of hurricanes, but it is making them stronger . And warming appears to favor certain kinds of weather patterns, like the “ Midwest Water Hose ” events that caused devastating flooding across the Midwest in 2019 .

It’s important to remember that in most natural disasters, there are multiple factors at play. For instance, the 2019 Midwest floods occurred after a recent cold snap had frozen the ground solid, preventing the soil from absorbing rainwater and increasing runoff into the Missouri and Mississippi Rivers. These waterways have also been reshaped by levees and other forms of river engineering, some of which failed in the floods.

Wildfires are another phenomenon with multiple causes. In many places, fire risk has increased because humans have aggressively fought natural fires and prevented Indigenous peoples from carrying out traditional burning practices. This has allowed fuel to accumulate that makes current fires worse .

However, climate change still plays a major role by heating and drying forests, turning them into tinderboxes. Studies show that warming is the driving factor behind the recent increases in wildfires; one analysis found that climate change is responsible for doubling the area burned across the American West between 1984 and 2015. And researchers say that warming will only make fires bigger and more dangerous in the future.

It depends on how aggressively we act to address climate change. If we continue with business as usual, by the end of the century, it will be too hot to go outside during heat waves in the Middle East and South Asia . Droughts will grip Central America, the Mediterranean and southern Africa. And many island nations and low-lying areas, from Texas to Bangladesh, will be overtaken by rising seas. Conversely, climate change could bring welcome warming and extended growing seasons to the upper Midwest , Canada, the Nordic countries and Russia . Farther north, however, the loss of snow, ice and permafrost will upend the traditions of Indigenous peoples and threaten infrastructure.

It’s complicated, but the underlying message is simple: unchecked climate change will likely exacerbate existing inequalities . At a national level, poorer countries will be hit hardest, even though they have historically emitted only a fraction of the greenhouse gases that cause warming. That’s because many less developed countries tend to be in tropical regions where additional warming will make the climate increasingly intolerable for humans and crops. These nations also often have greater vulnerabilities, like large coastal populations and people living in improvised housing that is easily damaged in storms. And they have fewer resources to adapt, which will require expensive measures like redesigning cities, engineering coastlines and changing how people grow food.

Already, between 1961 and 2000, climate change appears to have harmed the economies of the poorest countries while boosting the fortunes of the wealthiest nations that have done the most to cause the problem, making the global wealth gap 25 percent bigger than it would otherwise have been. Similarly, the Global Climate Risk Index found that lower income countries — like Myanmar, Haiti and Nepal — rank high on the list of nations most affected by extreme weather between 1999 and 2018. Climate change has also contributed to increased human migration, which is expected to increase significantly .

Even within wealthy countries, the poor and marginalized will suffer the most. People with more resources have greater buffers, like air-conditioners to keep their houses cool during dangerous heat waves, and the means to pay the resulting energy bills. They also have an easier time evacuating their homes before disasters, and recovering afterward. Lower income people have fewer of these advantages, and they are also more likely to live in hotter neighborhoods and work outdoors, where they face the brunt of climate change.

These inequalities will play out on an individual, community, and regional level. A 2017 analysis of the U.S. found that, under business as usual, the poorest one-third of counties, which are concentrated in the South, will experience damages totaling as much as 20 percent of gross domestic product, while others, mostly in the northern part of the country, will see modest economic gains. Solomon Hsiang, an economist at University of California, Berkeley, and the lead author of the study, has said that climate change “may result in the largest transfer of wealth from the poor to the rich in the country’s history.”

Even the climate “winners” will not be immune from all climate impacts, though. Desirable locations will face an influx of migrants. And as the coronavirus pandemic has demonstrated, disasters in one place quickly ripple across our globalized economy. For instance, scientists expect climate change to increase the odds of multiple crop failures occurring at the same time in different places, throwing the world into a food crisis .

On top of that, warmer weather is aiding the spread of infectious diseases and the vectors that transmit them, like ticks and mosquitoes . Research has also identified troubling correlations between rising temperatures and increased interpersonal violence , and climate change is widely recognized as a “threat multiplier” that increases the odds of larger conflicts within and between countries. In other words, climate change will bring many changes that no amount of money can stop. What could help is taking action to limit warming.

One of the most common arguments against taking aggressive action to combat climate change is that doing so will kill jobs and cripple the economy. But this implies that there’s an alternative in which we pay nothing for climate change. And unfortunately, there isn’t. In reality, not tackling climate change will cost a lot , and cause enormous human suffering and ecological damage, while transitioning to a greener economy would benefit many people and ecosystems around the world.

Let’s start with how much it will cost to address climate change. To keep warming well below 2 degrees Celsius, the goal of the Paris Climate Agreement, society will have to reach net zero greenhouse gas emissions by the middle of this century. That will require significant investments in things like renewable energy, electric cars and charging infrastructure, not to mention efforts to adapt to hotter temperatures, rising sea-levels and other unavoidable effects of current climate changes. And we’ll have to make changes fast.

Estimates of the cost vary widely. One recent study found that keeping warming to 2 degrees Celsius would require a total investment of between $4 trillion and $60 trillion, with a median estimate of $16 trillion, while keeping warming to 1.5 degrees Celsius could cost between $10 trillion and $100 trillion, with a median estimate of $30 trillion. (For reference, the entire world economy was about $88 trillion in 2019.) Other studies have found that reaching net zero will require annual investments ranging from less than 1.5 percent of global gross domestic product to as much as 4 percent . That’s a lot, but within the range of historical energy investments in countries like the U.S.

Now, let’s consider the costs of unchecked climate change, which will fall hardest on the most vulnerable. These include damage to property and infrastructure from sea-level rise and extreme weather, death and sickness linked to natural disasters, pollution and infectious disease, reduced agricultural yields and lost labor productivity because of rising temperatures, decreased water availability and increased energy costs, and species extinction and habitat destruction. Dr. Hsiang, the U.C. Berkeley economist, describes it as “death by a thousand cuts.”

As a result, climate damages are hard to quantify. Moody’s Analytics estimates that even 2 degrees Celsius of warming will cost the world $69 trillion by 2100, and economists expect the toll to keep rising with the temperature. In a recent survey , economists estimated the cost would equal 5 percent of global G.D.P. at 3 degrees Celsius of warming (our trajectory under current policies) and 10 percent for 5 degrees Celsius. Other research indicates that, if current warming trends continue, global G.D.P. per capita will decrease between 7 percent and 23 percent by the end of the century — an economic blow equivalent to multiple coronavirus pandemics every year. And some fear these are vast underestimates .

Already, studies suggest that climate change has slashed incomes in the poorest countries by as much as 30 percent and reduced global agricultural productivity by 21 percent since 1961. Extreme weather events have also racked up a large bill. In 2020, in the United States alone, climate-related disasters like hurricanes, droughts, and wildfires caused nearly $100 billion in damages to businesses, property and infrastructure, compared to an average of $18 billion per year in the 1980s.

Given the steep price of inaction, many economists say that addressing climate change is a better deal . It’s like that old saying: an ounce of prevention is worth a pound of cure. In this case, limiting warming will greatly reduce future damage and inequality caused by climate change. It will also produce so-called co-benefits, like saving one million lives every year by reducing air pollution, and millions more from eating healthier, climate-friendly diets. Some studies even find that meeting the Paris Agreement goals could create jobs and increase global G.D.P . And, of course, reining in climate change will spare many species and ecosystems upon which humans depend — and which many people believe to have their own innate value.

The challenge is that we need to reduce emissions now to avoid damages later, which requires big investments over the next few decades. And the longer we delay, the more we will pay to meet the Paris goals. One recent analysis found that reaching net-zero by 2050 would cost the U.S. almost twice as much if we waited until 2030 instead of acting now. But even if we miss the Paris target, the economics still make a strong case for climate action, because every additional degree of warming will cost us more — in dollars, and in lives.

Veronica Penney contributed reporting.

Illustration photographs by Esther Horvath, Max Whittaker, David Maurice Smith and Talia Herman for The New York Times; Esther Horvath/Alfred-Wegener-Institut

An earlier version of this article misidentified the authors of The Debunking Handbook. It was written by social scientists who study climate communication, not a team of climate scientists.

How we handle corrections

What’s Up in Space and Astronomy

Keep track of things going on in our solar system and all around the universe..

Never miss an eclipse, a meteor shower, a rocket launch or any other 2024 event  that’s out of this world with  our space and astronomy calendar .

A new set of computer simulations, which take into account the effects of stars moving past our solar system, has effectively made it harder to predict Earth’s future and reconstruct its past.

Dante Lauretta, the planetary scientist who led the OSIRIS-REx mission to retrieve a handful of space dust , discusses his next final frontier.

A nova named T Coronae Borealis lit up the night about 80 years ago. Astronomers say it’s expected to put on another show  in the coming months.

Voyager 1, the 46-year-old first craft in interstellar space which flew by Jupiter and Saturn in its youth, may have gone dark .

Is Pluto a planet? And what is a planet, anyway? Test your knowledge here .

Advertisement

Knowledge is power

Stay in the know about climate impacts and solutions. Subscribe to our weekly newsletter.

By clicking submit, you agree to share your email address with the site owner and Mailchimp to receive emails from the site owner. Use the unsubscribe link in those emails to opt out at any time.

Yale Climate Connections

HFCs: Case Study in Interconnections Of Ozone Depletion and Climate Change

Share this:

• Click to share on Facebook (Opens in new window)
• Click to share on X (Opens in new window)

Climate change associated with atmospheric warming and depletion of the earth’s protective ozone layer are two separate but interrelated problems, intersecting in complex ways that challenge easy comprehension and also efforts to address them. Recent developments related to chemicals commonly known as HFCs illustrate the situation.

Industrial emissions of carbon dioxide, the chief greenhouse gas produced through human activity and blamed for global warming, started long before the widespread use of refrigerant chemicals later discovered to be depleting stratospheric ozone.

The threat to the ozone layer, which blocks ultraviolet radiation harmful to living organisms, was the issue that commanded world attention first, however. Mounting concerns over the “ozone hole” led to the adoption in 1987 of the Montreal Protocol, the landmark international agreement to phase out production and use of the main ozone-destroying substances – chlorofluorocarbons, or CFCs.

The frequently amended Montreal treaty is regarded as a model for successful, global-scale environmental protection. Scientists project, for example, that the most dramatic example of ozone depletion – the mammoth hole in the ozone layer over Antarctica, first documented in 1985 – now could well be closed in several decades.

Vexing problems remain, however, that illustrate how the ozone issue is connected to the science of greenhouse warming and attempts to slow it.

Besides eroding the ozone layer, CFCs are also powerful greenhouse gases. The United Nations says that banning them has helped forestall the pace of global warming by up to a dozen years. Researchers have recently warned , however, that new findings point to increased temperatures across Antarctica if the hoped-for closing of the ozone hole ends up reversing a cloud-forming phenomenon, associated with depleted ozone there, that has shielded that region from some warming.

Efforts to mitigate climate change, meanwhile, have been complicated by scientific findings that HFCs – certain “ozone-friendly” refrigerant chemicals formally known as hydrofluorocarbons, introduced to substitute for CFC-replacing, but also ozone-depleting, chemicals (hydrochlorofluorocarbons or HCFCs) – are highly potent greenhouse agents in their own right. Using more of those ozone-friendly substitutes could end up boosting global warming even as they help address the stratospheric ozone depletion problem.

Passing the HFCs Baton … But to Whom?

Because of heightened concerns about HFCs’ role in global warming, two initiatives considered at an international conference on the Montreal Protocol last November in Port Ghalib, Egypt, sought to phase down use of those chemicals. Neither was adopted.

Instead, the negotiators in Egypt handed the issue to the Copenhagen climate conference in December, which also failed to take action on it, leaving the diplomatic picture on HFCs unclear.

The House-passed American Clean Energy and Security Act (the Waxman-Markey bill) would cap HFC emissions and regulate their production and imports, but prospects for measure or any energy-climate legislation to become law this year are far from certain.

The Environmental Protection Agency’s endangerment finding on greenhouse gases in December – a necessary precursor to administratively imposed restrictions under the Clean Air Act – targeted HFCs along with carbon dioxide and other substances. The outcome of that process is also unclear, however, and some members of Congress are trying to prevent the EPA from adopting regulations, which, if fully adopted, would inevitably lead to litigation.

Even in the absence of regulatory requirements, however, a number of corporations, prodded and encouraged by environmental groups, have been moving ahead with tests of other refrigerant substances that could be used in lieu of HFCs.

Coca-Cola, for instance, announced in December that it would eliminate HFCs from its new vending machines and coolers by 2015 – an action that Reuters reported had “rais(ed) the bar for climate-friendly refrigeration in the food and beverage industry.”

“There are a number of things moving that are positive,” said Kert Davies, research director of Greenpeace USA, which issued a report in December analyzing the HFC-related actions of 18 companies, offering both praise and criticism.

“There’s a big ball of pressure and 2009 was a banner year for stigmatization of HFCs,” Davies told The Yale Forum . “There’s a lot going on that ultimately, we think, will tip the scale toward alternatives (to HFCs) coming to the market. Our work is to make that happen faster and faster.”

No Agreement at Two Diplomatic Meetings

HFCs have been dubbed “super greenhouse gases” because, molecule for molecule, their global warming effect is hundreds of times greater than carbon dioxide’s. Still, their long-recognized role in the climate issue has been relatively obscure. Trying to raise that profile, Greenpeace’s report in December called HFCs “the worst greenhouse gases you’ve never heard of.”

That may change, in part because of two new EPA rules, announced in December to comply with the Montreal Protocol. The rules will reduce the availability and use of HCFCs, whose ozone-depleting power is less than that of CFCs but still worrisome enough to prompt their restriction under the ozone treaty. The HCFC phase-down focuses more attention on HCFC-replacing HFCs, their warming potential, and on various alternatives to HFC use.

The HFC-warming issue has not gotten anything close to the media attention that the debate over reducing CO 2 emissions has received, but SolveClimate is one news outlet that has continued to follow it closely . The internet publication’s coverage last year included articles on jockeying within the Obama administration over whether the Montreal Protocol or a climate treaty is the best vehicle for reducing HFC use, then on the issue’s handling at the Egypt and Copenhagen conferences.

After the meeting on the Montreal Protocol in Egypt ended, for example, an analysis by SolveClimate founder David Sassoon assessed the issue’s diplomatic status and immediate prospects. In the article, also distributed by Reuters, he reported that the U.S. delegation was “shocked” that its proposal to phase down HFCs was “soundly rebuffed”:

Even though the stakes for the global environment are very high, the meeting ended with no amendment and no binding decision on HFCs. Instead, 41 out of 198 countries signed a weak “declaration of intent.” Advocates are doing their best to put a happy face on the outcome, but the failure to act on phasing down HFCs is a disappointment, and it provides a preview of an outcome that many fear may be repeated on a larger scale in Copenhagen.

Achim Steiner, executive director of the U.N. Environment Programme (UNEP), issued a statement after the Port Ghalib meeting, in which he described the HFC issue as increasingly linked to broader negotiations aimed at dealing with climate change: “Clearly the sooner the international community seals the deal on climate change, the sooner other related agreements can move forward.”

Covering the Copenhagen meeting, the Los Angles Times ‘ Jim Tankersley filed a report on talks there regarding HFCs and other “stealth” pollutants represented by non-CO 2 greenhouse gases:

Many scientists and environmentalists say reducing the “forgotten 50 percent” of pollutants will be faster, easier and substantially cheaper than cutting carbon dioxide, and could buy the world time in its climate clock race.

Tankersley reported that negotiators were “quietly making progress” on these lower-profile greenhouse gases. Their eventual failure to reach agreement on the issue, however, left a cloudy diplomatic road ahead. Besides another major climate conference convening in Mexico in November (where, an analysis in The Guardian concluded this month, a binding deal already seems “all but impossible,”), a separate series of U.N.-sponsored meetings this year will address possible changes to the Montreal Protocol.

A blog on the website of the scientific journal Nature reported that delegates to the Port Ghalib conference “called on a technical committee to analyze alternatives to (HFCs) in advance of a potential decision” in 2010 .

A New Chapter in an Old Debate

Scientific findings have continued to add urgency to the HFC-related decisions facing policymakers at the national and international levels.

Last June, for instance, in the journal Proceedings of the National Academy of Sciences , scientists from the Netherlands Environmental Assessment Agency, two U.S. agencies, and Du Pont published a paper projecting “substantially” growing use and emissions of HFCs – larger than in previous forecasts – if those chemicals replace HCFCs without regulation.

They wrote that the projected increases “result primarily from sustained growth in demand for refrigeration, air-conditioning and insulating foam products in developing countries,” based on anticipated economic growth there and on the prior experiences of developed nations.

Developing countries’ HFC emissions could range up to 800 percent greater than those of developed countries by mid-century, the researchers concluded.

They projected that without HFC restrictions, global emissions in 2050 would have a warming effect of up to 45 percent of projected CO 2 emissions at that time, assuming that CO 2 in the atmosphere is stabilized by then at 450 parts per million. That is the CO 2 level that other experts project is necessary to limit the earth’s average temperature increase to about 2 degrees Celsius (3.5 degrees Fahrenheit).

Last November, researchers at NASA and Purdue University published a study in the American Chemical Society’s Journal of Physical Chemistry , which extended scientific understanding of the molecular mechanism explaining HFCs’ warming capacity.

They reported that their examination of HFCs and other fluorinated compounds, including CFCs, revealed that these chemicals are much more efficient than carbon dioxide and methane at trapping warming radiation in the frequency of the infrared region known as the “atmospheric window,” blocking its return to space.

Proponents of restricting HFCs are now citing such recent studies to underscore their argument that phasing down production and use of the chemicals affords an opportunity for relatively quick and substantial action against global warming while broader and far more complex efforts to reduce CO 2 emissions proceed.

Regardless of whether the new research findings result in restrictions on HFCs through a global agreement, the question of what refrigerant substances to use in their place has been receiving increased attention.

In part, these developments represent a new chapter in a long-running debate – whether to use compounds developed by the chemical industry or “natural” refrigerants, such as hydrocarbons (without ozone-depleting and warming potential) and carbon dioxide (with far less warming capacity than HFCs).

Greenpeace has promoted “natural” substances through its GreenFreeze campaign since 1992, arguing that they are environmentally preferable to HCFCs and HFCs.

The group’s effort has included development of a GreenFreeze refrigerator, 300 million of which it says have been sold in Europe, Asia, and South America. The group also has also pressed corporations to change refrigerants. Coca-Cola, for instance, said its commitment in December to go HFC-free was the “direct result” of “increasingly cooperative” discussions with Greenpeace since 2000. The company said it has already started using hydrocarbon and CO 2 cooling in smaller and larger equipment, respectively.

An emerging point of contention involves new refrigerant chemicals called hydrofluoroolefins, or HFOs.

Manufacturers are promoting them as an environmentally superior replacement for HFCs. Honeywell’s website, for instance, describes one hydrofluoroolefin – HFO-1234yf – as “a next generation fluorinated solution” that can meet climate-related rules in the European Union. The E.U. has banned the use of the HFC gas R-134a in vehicle air conditioners, starting in 2011, because of its contribution to global warming.

“Our alternative has the lowest life-cycle climate performance rating of any R-134a replacement option – including CO 2 – due to its higher energy efficiency, which results in lower fuel consumption,” the Honeywell website text asserts.

Seeking to counter such appeals, Greenpeace issued a position paper on HFOs last October, arguing that, chemically, they are also HFCs, but are being marketed under a different name because of “the negative connotations that HFCs have acquired.”

The paper argued that HFOs are “only a short-term fix,” which presents “an unnecessary risk to the environment and human health” when compared to “natural” refrigerants.

In October, EPA proposed the listing of HFO-1234yf as a safe alternative for auto air conditioners, declaring that it does not deplete the ozone layer and “when used with proper risk mitigation technologies, will reduce the impact of (vehicle air conditioner) refrigerant emissions on the environment.”

The agency said HFO-1234yf has a “global warming potential” value of 4 – quadruple the value of 1 assigned to CO 2 , but far smaller than the value of 1,430 for R-134a and the value of 10,890 for CFC-12.

Bill Dawson

Bill Dawson is an independent journalist who edits Texas Climate News, an online magazine published by the nonprofit, nonpartisan Houston Advanced Research Center. He was previously environment writer... More by Bill Dawson

What evidence exists that Earth is warming and that humans are the main cause?

We know the world is warming because people have been recording daily high and low temperatures at thousands of weather stations worldwide, over land and ocean, for many decades and, in some locations, for more than a century. When different teams of climate scientists in different agencies (e.g., NOAA and NASA) and in other countries (e.g., the U.K.’s Hadley Centre) average these data together, they all find essentially the same result: Earth’s average surface temperature has risen by about 1.8°F (1.0°C) since 1880. 

Yearly temperature compared to the twentieth-century average (red bars mean warmer than average, blue bars mean colder than average) from 1850–2022 and atmospheric carbon dioxide amounts (gray line): 1850-1958 from IAC , 1959-2019 from NOAA ESRL . Original graph by Dr. Howard Diamond (NOAA ARL), and adapted by NOAA Climate.gov.

In addition to our surface station data, we have many different lines of evidence that Earth is warming ( learn more ). Birds are migrating earlier, and their migration patterns are changing.  Lobsters  and  other marine species  are moving north. Plants are blooming earlier in the spring. Mountain glaciers are melting worldwide, and snow cover is declining in the Northern Hemisphere (Learn more  here  and  here ). Greenland’s ice sheet—which holds about 8 percent of Earth’s fresh water—is melting at an accelerating rate ( learn more ). Mean global sea level is rising ( learn more ). Arctic sea ice is declining rapidly in both thickness and extent ( learn more ).

The Greenland Ice Sheet lost mass again in 2020, but not as much as it did 2019. Adapted from the 2020 Arctic Report Card, this graph tracks Greenland mass loss measured by NASA's GRACE satellite missions since 2002. The background photo shows a glacier calving front in western Greenland, captured from an airplane during a NASA Operation IceBridge field campaign. Full story.

We know this warming is largely caused by human activities because the key role that carbon dioxide plays in maintaining Earth’s natural greenhouse effect has been understood since the mid-1800s. Unless it is offset by some equally large cooling influence, more atmospheric carbon dioxide will lead to warmer surface temperatures. Since 1800, the amount of carbon dioxide in the atmosphere  has increased  from about 280 parts per million to 410 ppm in 2019. We know from both its rapid increase and its isotopic “fingerprint” that the source of this new carbon dioxide is fossil fuels, and not natural sources like forest fires, volcanoes, or outgassing from the ocean.

Philip James de Loutherbourg's 1801 painting, Coalbrookdale by Night , came to symbolize the start of the Industrial Revolution, when humans began to harness the power of fossil fuels—and to contribute significantly to Earth's atmospheric greenhouse gas composition. Image from Wikipedia .

Finally, no other known climate influences have changed enough to account for the observed warming trend. Taken together, these and other lines of evidence point squarely to human activities as the cause of recent global warming.

USGCRP (2017). Climate Science Special Report: Fourth National Climate Assessment, Volume 1 [Wuebbles, D.J., D.W. Fahey, K.A. Hibbard, D.J. Dokken, B.C. Stewart, and T.K. Maycock (eds.)]. U.S. Global Change Research Program, Washington, DC, USA, 470 pp, doi:  10.7930/J0J964J6 .

National Fish, Wildlife, and Plants Climate Adaptation Partnership (2012):  National Fish, Wildlife, and Plants Climate Adaptation Strategy . Association of Fish and Wildlife Agencies, Council on Environmental Quality, Great Lakes Indian Fish and Wildlife Commission, National Oceanic and Atmospheric Administration, and U.S. Fish and Wildlife Service. Washington, D.C. DOI: 10.3996/082012-FWSReport-1

IPCC (2019). Summary for Policymakers. In: IPCC Special Report on the Ocean and Cryosphere in a Changing Climate. [H.-O. Pörtner, D.C. Roberts, V. Masson-Delmotte, P. Zhai, M. Tignor, E. Poloczanska, K. Mintenbeck, A. Alegría, M. Nicolai, A. Okem, J. Petzold, B. Rama, N.M. Weyer (eds.)]. In press.

NASA JPL: "Consensus: 97% of climate scientists agree."  Global Climate Change . A website at NASA's Jet Propulsion Laboratory (climate.nasa.gov/scientific-consensus). (Accessed July 2013.)

We value your feedback

Help us improve our content

Related Content

News & features, 2017 state of the climate: mountain glaciers, warming waters shift fish communities northward in the arctic, climate & fish sticks, maps & data, past climate, land - terrestrial climate variables, future climate, teaching climate, toolbox for teaching climate & energy, student climate & conservation congress (sc3), climate youth engagement, climate resilience toolkit, arctic oceans, sea ice, and coasts, alaska and the arctic, agriculture and ecosystems.

The Effects of Climate Change

The effects of human-caused global warming are happening now, are irreversible for people alive today, and will worsen as long as humans add greenhouse gases to the atmosphere.

• We already see effects scientists predicted, such as the loss of sea ice, melting glaciers and ice sheets, sea level rise, and more intense heat waves.
• Scientists predict global temperature increases from human-made greenhouse gases will continue. Severe weather damage will also increase and intensify.

Earth Will Continue to Warm and the Effects Will Be Profound

Global climate change is not a future problem. Changes to Earth’s climate driven by increased human emissions of heat-trapping greenhouse gases are already having widespread effects on the environment: glaciers and ice sheets are shrinking, river and lake ice is breaking up earlier, plant and animal geographic ranges are shifting, and plants and trees are blooming sooner.

Effects that scientists had long predicted would result from global climate change are now occurring, such as sea ice loss, accelerated sea level rise, and longer, more intense heat waves.

The magnitude and rate of climate change and associated risks depend strongly on near-term mitigation and adaptation actions, and projected adverse impacts and related losses and damages escalate with every increment of global warming.

Intergovernmental Panel on Climate Change

Some changes (such as droughts, wildfires, and extreme rainfall) are happening faster than scientists previously assessed. In fact, according to the Intergovernmental Panel on Climate Change (IPCC) — the United Nations body established to assess the science related to climate change — modern humans have never before seen the observed changes in our global climate, and some of these changes are irreversible over the next hundreds to thousands of years.

Scientists have high confidence that global temperatures will continue to rise for many decades, mainly due to greenhouse gases produced by human activities.

The IPCC’s Sixth Assessment report, published in 2021, found that human emissions of heat-trapping gases have already warmed the climate by nearly 2 degrees Fahrenheit (1.1 degrees Celsius) since 1850-1900. 1 The global average temperature is expected to reach or exceed 1.5 degrees C (about 3 degrees F) within the next few decades. These changes will affect all regions of Earth.

The severity of effects caused by climate change will depend on the path of future human activities. More greenhouse gas emissions will lead to more climate extremes and widespread damaging effects across our planet. However, those future effects depend on the total amount of carbon dioxide we emit. So, if we can reduce emissions, we may avoid some of the worst effects.

The scientific evidence is unequivocal: climate change is a threat to human wellbeing and the health of the planet. Any further delay in concerted global action will miss the brief, rapidly closing window to secure a liveable future.

Here are some of the expected effects of global climate change on the United States, according to the Third and Fourth National Climate Assessment Reports:

Future effects of global climate change in the United States:

U.S. Sea Level Likely to Rise 1 to 6.6 Feet by 2100

Global sea level has risen about 8 inches (0.2 meters) since reliable record-keeping began in 1880. By 2100, scientists project that it will rise at least another foot (0.3 meters), but possibly as high as 6.6 feet (2 meters) in a high-emissions scenario. Sea level is rising because of added water from melting land ice and the expansion of seawater as it warms. Image credit: Creative Commons Attribution-Share Alike 4.0

Climate Changes Will Continue Through This Century and Beyond

Global climate is projected to continue warming over this century and beyond. Image credit: Khagani Hasanov, Creative Commons Attribution-Share Alike 3.0

Hurricanes Will Become Stronger and More Intense

Scientists project that hurricane-associated storm intensity and rainfall rates will increase as the climate continues to warm. Image credit: NASA

More Droughts and Heat Waves

Droughts in the Southwest and heat waves (periods of abnormally hot weather lasting days to weeks) are projected to become more intense, and cold waves less intense and less frequent. Image credit: NOAA

Longer Wildfire Season

Warming temperatures have extended and intensified wildfire season in the West, where long-term drought in the region has heightened the risk of fires. Scientists estimate that human-caused climate change has already doubled the area of forest burned in recent decades. By around 2050, the amount of land consumed by wildfires in Western states is projected to further increase by two to six times. Even in traditionally rainy regions like the Southeast, wildfires are projected to increase by about 30%.

Changes in Precipitation Patterns

Climate change is having an uneven effect on precipitation (rain and snow) in the United States, with some locations experiencing increased precipitation and flooding, while others suffer from drought. On average, more winter and spring precipitation is projected for the northern United States, and less for the Southwest, over this century. Image credit: Marvin Nauman/FEMA

Frost-Free Season (and Growing Season) will Lengthen

The length of the frost-free season, and the corresponding growing season, has been increasing since the 1980s, with the largest increases occurring in the western United States. Across the United States, the growing season is projected to continue to lengthen, which will affect ecosystems and agriculture.

Global Temperatures Will Continue to Rise

Summer of 2023 was Earth's hottest summer on record, 0.41 degrees Fahrenheit (F) (0.23 degrees Celsius (C)) warmer than any other summer in NASA’s record and 2.1 degrees F (1.2 C) warmer than the average summer between 1951 and 1980. Image credit: NASA

Arctic Is Very Likely to Become Ice-Free

Sea ice cover in the Arctic Ocean is expected to continue decreasing, and the Arctic Ocean will very likely become essentially ice-free in late summer if current projections hold. This change is expected to occur before mid-century.

U.S. Regional Effects

Climate change is bringing different types of challenges to each region of the country. Some of the current and future impacts are summarized below. These findings are from the Third 3 and Fourth 4 National Climate Assessment Reports, released by the U.S. Global Change Research Program .

• Northeast. Heat waves, heavy downpours, and sea level rise pose increasing challenges to many aspects of life in the Northeast. Infrastructure, agriculture, fisheries, and ecosystems will be increasingly compromised. Farmers can explore new crop options, but these adaptations are not cost- or risk-free. Moreover, adaptive capacity , which varies throughout the region, could be overwhelmed by a changing climate. Many states and cities are beginning to incorporate climate change into their planning.
• Northwest. Changes in the timing of peak flows in rivers and streams are reducing water supplies and worsening competing demands for water. Sea level rise, erosion, flooding, risks to infrastructure, and increasing ocean acidity pose major threats. Increasing wildfire incidence and severity, heat waves, insect outbreaks, and tree diseases are causing widespread forest die-off.
• Southeast. Sea level rise poses widespread and continuing threats to the region’s economy and environment. Extreme heat will affect health, energy, agriculture, and more. Decreased water availability will have economic and environmental impacts.
• Midwest. Extreme heat, heavy downpours, and flooding will affect infrastructure, health, agriculture, forestry, transportation, air and water quality, and more. Climate change will also worsen a range of risks to the Great Lakes.
• Southwest. Climate change has caused increased heat, drought, and insect outbreaks. In turn, these changes have made wildfires more numerous and severe. The warming climate has also caused a decline in water supplies, reduced agricultural yields, and triggered heat-related health impacts in cities. In coastal areas, flooding and erosion are additional concerns.

1. IPCC 2021, Climate Change 2021: The Physical Science Basis , the Working Group I contribution to the Sixth Assessment Report, Cambridge University Press, Cambridge, UK.

2. IPCC, 2013: Summary for Policymakers. In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.

3. USGCRP 2014, Third Climate Assessment .

4. USGCRP 2017, Fourth Climate Assessment .

Related Resources

A Degree of Difference

So, the Earth's average temperature has increased about 2 degrees Fahrenheit during the 20th century. What's the big deal?

What’s the difference between climate change and global warming?

“Global warming” refers to the long-term warming of the planet. “Climate change” encompasses global warming, but refers to the broader range of changes that are happening to our planet, including rising sea levels; shrinking mountain glaciers; accelerating ice melt in Greenland, Antarctica and the Arctic; and shifts in flower/plant blooming times.

Is it too late to prevent climate change?

Humans have caused major climate changes to happen already, and we have set in motion more changes still. However, if we stopped emitting greenhouse gases today, the rise in global temperatures would begin to flatten within a few years. Temperatures would then plateau but remain well-elevated for many, many centuries.

Discover More Topics From NASA

Explore Earth Science

Earth Science in Action

Earth Science Data

Facts About Earth

October 26, 2015

Exxon Knew about Climate Change almost 40 years ago

A new investigation shows the oil company understood the science before it became a public issue and spent millions to promote misinformation

By Shannon Hall &

Exxon was aware of climate change, as early as 1977, 11 years before it became a public issue, according to a recent investigation from InsideClimate News. This knowledge did not prevent the company (now ExxonMobil and the world’s largest oil and gas company) from spending decades refusing to publicly acknowledge climate change and even promoting climate misinformation—an approach many have likened to the lies spread by the tobacco industry regarding the health risks of smoking. Both industries were conscious that their products wouldn’t stay profitable once the world understood the risks, so much so that they used the same consultants to develop strategies on how to communicate with the public.  

Experts, however, aren’t terribly surprised. “It’s never been remotely plausible that they did not understand the science,” says Naomi Oreskes, a history of science professor at Harvard University. But as it turns out, Exxon didn’t just understand the science, the company actively engaged with it. In the 1970s and 1980s it employed top scientists to look into the issue and launched its own ambitious research program that empirically sampled carbon dioxide and built rigorous climate models. Exxon even spent more than $1 million on a tanker project that would tackle how much CO2 is absorbed by the oceans. It was one of the biggest scientific questions of the time, meaning that Exxon was truly conducting unprecedented research. 

In their eight-month-long investigation, reporters at InsideClimate News interviewed former Exxon employees, scientists and federal officials and analyzed hundreds of pages of internal documents. They found that the company’s knowledge of climate change dates back to July 1977, when its senior scientist James Black delivered a sobering message on the topic. “In the first place, there is general scientific agreement that the most likely manner in which mankind is influencing the global climate is through carbon dioxide release from the burning of fossil fuels," Black told Exxon’s management committee. A year later he warned Exxon that doubling CO2 gases in the atmosphere would increase average global temperatures by two or three degrees—a number that is consistent with the scientific consensus today. He continued to warn that “present thinking holds that man has a time window of five to 10 years before the need for hard decisions regarding changes in energy strategies might become critical." In other words, Exxon needed to act.

On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing . By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.

But ExxonMobil disagrees that any of its early statements were so stark, let alone conclusive at all. “We didn’t reach those conclusions, nor did we try to bury it like they suggest,” ExxonMobil spokesperson Allan Jeffers tells Scientific American . “The thing that shocks me the most is that we’ve been saying this for years, that we have been involved in climate research. These guys go down and pull some documents that we made available publicly in the archives and portray them as some kind of bombshell whistle-blower exposé because of the loaded language and the selective use of materials.”    One thing is certain: in June 1988, when NASA scientist James Hansen told a congressional hearing that the planet was already warming, Exxon remained publicly convinced that the science was still controversial. Furthermore, experts agree that Exxon became a leader in campaigns of confusion. By 1989 the company had helped create the Global Climate Coalition (disbanded in 2002) to question the scientific basis for concern about climate change. It also helped to prevent the U.S. from signing the international treaty on climate known as the Kyoto Protocol in 1998 to control greenhouse gases. Exxon’s tactic not only worked on the U.S. but also stopped other countries, such as China and India, from signing the treaty. At that point, “a lot of things unraveled,” Oreskes says.

But experts are still piecing together Exxon’s misconception puzzle. Last summer the Union of Concerned Scientists released a complementary investigation to the one by InsideClimate News, known as the Climate Deception Dossiers (pdf). “We included a memo of a coalition of fossil-fuel companies where they pledge basically to launch a big communications effort to sow doubt,” says union president Kenneth Kimmel. “There’s even a quote in it that says something like ‘Victory will be achieved when the average person is uncertain about climate science.’ So it’s pretty stark.”

Since then, Exxon has spent more than $30 million on think tanks that promote climate denial, according to Greenpeace . Although experts will never be able to quantify the damage Exxon’s misinformation has caused, “one thing for certain is we’ve lost a lot of ground,” Kimmell says. Half of the greenhouse gas emissions in our atmosphere were released after 1988. “I have to think if the fossil-fuel companies had been upfront about this and had been part of the solution instead of the problem, we would have made a lot of progress [today] instead of doubling our greenhouse gas emissions.”

Experts agree that the damage is huge, which is why they are likening Exxon’s deception to the lies spread by the tobacco industry. “I think there are a lot of parallels,” Kimmell says. Both sowed doubt about the science for their own means, and both worked with the same consultants to help develop a communications strategy. He notes, however, that the two diverge in the type of harm done. Tobacco companies threatened human health, but the oil companies threatened the planet’s health. “It’s a harm that is global in its reach,” Kimmel says.

To prove this, Bob Ward—who on behalf of the U.K.’s Royal Academy sent a letter to Exxon in 2006 claiming its science was “inaccurate and misleading”—thinks a thorough investigation is necessary. “Because frankly the episode with tobacco was probably the most disgraceful episode one could ever imagine,” Ward says. Kimmell agrees. These reasons “really highlight the responsibility that these companies have to come clean, acknowledge this, and work with everyone else to cut out emissions and pay for some of the cost we're going to bear as soon as possible,” Kimmell says.

It doesn’t appear, however, that Kimmell will get his retribution. Jeffers claims the investigation’s finds are “just patently untrue, misleading, and we reject them completely”—words that match Ward’s claims against them nearly a decade ago.

ENCYCLOPEDIC ENTRY

Global warming.

The causes, effects, and complexities of global warming are important to understand so that we can fight for the health of our planet.

Earth Science, Climatology

Tennessee Power Plant

Ash spews from a coal-fueled power plant in New Johnsonville, Tennessee, United States.

Photograph by Emory Kristof/ National Geographic

Global warming is the long-term warming of the planet’s overall temperature. Though this warming trend has been going on for a long time, its pace has significantly increased in the last hundred years due to the burning of fossil fuels . As the human population has increased, so has the volume of fossil fuels burned. Fossil fuels include coal, oil, and natural gas, and burning them causes what is known as the “greenhouse effect” in Earth’s atmosphere.

The greenhouse effect is when the sun’s rays penetrate the atmosphere, but when that heat is reflected off the surface cannot escape back into space. Gases produced by the burning of fossil fuels prevent the heat from leaving the atmosphere. These greenhouse gasses are carbon dioxide , chlorofluorocarbons, water vapor , methane , and nitrous oxide . The excess heat in the atmosphere has caused the average global temperature to rise overtime, otherwise known as global warming.

Global warming has presented another issue called climate change. Sometimes these phrases are used interchangeably, however, they are different. Climate change refers to changes in weather patterns and growing seasons around the world. It also refers to sea level rise caused by the expansion of warmer seas and melting ice sheets and glaciers . Global warming causes climate change, which poses a serious threat to life on Earth in the forms of widespread flooding and extreme weather. Scientists continue to study global warming and its impact on Earth.

Media Credits

The audio, illustrations, photos, and videos are credited beneath the media asset, except for promotional images, which generally link to another page that contains the media credit. The Rights Holder for media is the person or group credited.

Production Managers

Program specialists, last updated.

February 21, 2024

User Permissions

For information on user permissions, please read our Terms of Service. If you have questions about how to cite anything on our website in your project or classroom presentation, please contact your teacher. They will best know the preferred format. When you reach out to them, you will need the page title, URL, and the date you accessed the resource.

If a media asset is downloadable, a download button appears in the corner of the media viewer. If no button appears, you cannot download or save the media.

Text on this page is printable and can be used according to our Terms of Service .

Interactives

Any interactives on this page can only be played while you are visiting our website. You cannot download interactives.

Related Resources

• Election 2024
• Entertainment
• Newsletters
• Photography
• Personal Finance
• AP Buyline Personal Finance
• Press Releases
• Israel-Hamas War
• Russia-Ukraine War
• Global elections
• Asia Pacific
• Latin America
• Middle East
• Election Results
• Delegate Tracker
• AP & Elections
• March Madness
• AP Top 25 Poll
• Movie reviews
• Book reviews
• Personal finance
• Financial Markets
• Business Highlights
• Financial wellness
• Artificial Intelligence
• Social Media

UN weather agency issues ‘red alert’ on climate change after record heat, ice-melt increases in 2023

The U.N. weather agency is sounding a “red alert” about global warming, citing record-smashing increases last year in greenhouse gases, land and water temperatures and melting of glaciers and sea ice, and is warning that the world’s efforts to reverse the trend have been inadequate.

Celeste Saulo, World Meteorological Organization (WMO) Secretary-General, speaks about the state of Global Climate 2023, during a press conference at the European headquarters of the United Nations in Geneva, Switzerland, Tuesday, March 19, 2024. The U.N. weather agency is sounding a “red alert” about global warming, citing record-smashing increases last year in greenhouse gases, land and water temperatures and melting of glaciers and sea ice, and warning that the world’s efforts to reverse the trend have been inadequate. (Martial Trezzini/Keystone via AP)

• Copy Link copied

FILE - A strip of snow makes a ski slope in Saalbach, Austria, Sunday, March 17, 2024. The U.N. weather agency is sounding “a red alert” about global warming last year and beyond, citing in a new report record-smashing statistics when it comes to greenhouse gases, temperatures of land and oceans, and melting glaciers and sea-ice — even if countries, companies and citizens are getting greener. (AP Photo/Alessandro Trovati, File)

FILE - A man walks on the cracked ground of the Sau reservoir, which is only at 5 percent of its capacity, in Vilanova de Sau, about 100 km (62 miles) north of Barcelona, Spain, on Jan. 26, 2024. The U.N. weather agency is sounding “a red alert” about global warming last year and beyond, citing in a new report record-smashing statistics when it comes to greenhouse gases, temperatures of land and oceans, and melting glaciers and sea-ice — even if countries, companies and citizens are getting greener. (AP Photo/Emilio Morenatti, File)

GENEVA (AP) — The U.N. weather agency is sounding a “red alert” about global warming, citing record-smashing increases last year in greenhouse gases, land and water temperatures and melting of glaciers and sea ice, and is warning that the world’s efforts to reverse the trend have been inadequate.

The World Meteorological Organization said there is a “high probability” that 2024 will be another record-hot year.

The Geneva-based agency, in a “State of the Global Climate” report released Tuesday, ratcheted up concerns that a much-vaunted climate goal is increasingly in jeopardy: That the world can unite to limit planetary warming to no more than 1.5 degrees Celsius (2.7 degrees Fahrenheit) from pre-industrial levels.

“Never have we been so close – albeit on a temporary basis at the moment – to the 1.5° C lower limit of the Paris agreement on climate change,” said Celeste Saulo, the agency’s secretary-general. “The WMO community is sounding the red alert to the world.”

The 12-month period from March 2023 to February 2024 pushed beyond that 1.5-degree limit, averaging 1.56 C (2.81 F) higher, according to the European Union’s Copernicus Climate Service. It said the calendar year 2023 was just below 1.5 C at 1.48 C (2.66 F) , but a record hot start to this year pushed beyond that level for the 12-month average.

“Earth’s issuing a distress call,” U.N. Secretary-General Antonio Guterres said. “The latest State of the Global Climate report shows a planet on the brink. Fossil fuel pollution is sending climate chaos off the charts.”

Omar Baddour, WMO’s chief of climate monitoring, said the year after an El Niño event — the cyclical warming of the Pacific Ocean that affects global weather patterns — normally tends to be warmer.

“So we cannot say definitively about 2024 is going to be the warmest year. But what I would say: There is a high probability that 2024 will again break the record of 2023, but let’s wait and see,” he said. “January was the warmest January on record. So the records are still being broken.”

The latest WMO findings are especially stark when compiled in a single report. In 2023, over 90% of ocean waters experienced heat wave conditions at least once. Glaciers monitored since 1950 lost the most ice on record. Antarctic sea ice retreated to its lowest level ever.

“Topping all the bad news, what worries me the most is that the planet is now in a meltdown phase — literally and figuratively given the warming and mass loss from our polar ice sheets,” said Jonathan Overpeck, dean of the University of Michigan School for Environment and Sustainability, who wasn’t involved in the report.

Saulo called the climate crisis “the defining challenge that humanity faces” and said it combines with a crisis of inequality, as seen in growing food insecurity and migration.

WMO said the impact of heatwaves, floods, droughts, wildfires and tropical cyclones, exacerbated by climate change, was felt in lives and livelihoods on every continent in 2023.

“This list of record-smashing events is truly distressing, though not a surprise given the steady drumbeat of extreme events over the past year,” said University of Arizona climate scientist Kathy Jacobs, who also wasn’t involved in the WMO report. “The full cost of climate-change-accelerated events across sectors and regions has never been calculated in a meaningful way, but the cost to biodiversity and to the quality of life of future generations is incalculable.”

But the U.N. agency also acknowledged “a glimmer of hope” in trying to keep the Earth from running too high a fever. It said renewable energy generation capacity from wind, solar and waterpower rose nearly 50% from 2022 — to a total of 510 gigawatts.

“The target of 1.5C degree warming still holds, just like a speed limit on the highway still holds even if we temporarily exceed it,” said Malte Meinshausen, a professor of climate science at the University of Melbourne in Australia. “What is more urgent than ever is to grasp the economic opportunities that arise due to the low-cost renewables at our disposal, to decarbonize the electricity sector, and electrify other sectors.”

“We need to step on the brakes of ever-increasing GHG (greenhouse gas) emissions,” said Meinshausen, who also was not involved in the report. “And hopeful signs are there, that GHG emissions are about to peak.”

The report comes as climate experts and government ministers are to gather in the Danish capital, Copenhagen, on Thursday and Friday to press for greater climate action, including increased national commitments to fight global warming.

“Each year the climate story gets worse; each year WMO officials and others proclaim that the latest report is a wake-up call to decision makers,” said University of Victoria climate scientist Andrew Weaver, a former British Columbia lawmaker.

“Yet each year, once the 24-hour news cycle is over, far too many of our elected ‘leaders’ return to political grandstanding, partisan bickering and advancing policies with demonstrable short-term outcomes,” he said. “More often than not everything else ends up taking precedence over the advancement of climate policy. And so, nothing gets done.”

Borenstein reported from Washington, D.C.

The Associated Press’ climate and environmental coverage receives financial support from multiple private foundations. AP is solely responsible for all content. Find AP’s standards for working with philanthropies, a list of supporters and funded coverage areas at AP.org .

Study says since 1979 climate change has made heat waves last longer, spike hotter, hurt more people

C limate change is making giant heat waves crawl slower across the globe and they are baking more people for a longer time with higher temperatures over larger areas, a new study finds.

Since 1979, global heat waves are moving 20% more slowly — meaning more people stay hot longer — and they are happening 67% more often, according to a study in Friday's Science Advances. The study found the highest temperatures in the heat waves are warmer than 40 years ago and the area under a heat dome is larger.

Studies have shown heat waves worsening before, but this one is more comprehensive and concentrates heavily on not just temperature and area, but how long the high heat lasts and how it travels across continents, said study co-authors and climate scientists Wei Zhang of Utah State University and Gabriel Lau of Princeton University.

From 1979 to 1983, global heat waves would last eight days on average, but by 2016 to 2020 that was up to 12 days, the study said.

Eurasia was especially hit harder with longer lasting heat waves, the study said. Heat waves slowed down most in Africa, while North America and Australia saw the biggest increases in overall magnitude, which measures temperature and area, according to the study.

“This paper sends a clear warning that climate change makes heat waves yet more dangerous in more ways than one,” said Lawrence Berkeley National Lab climate scientist Michael Wehner, who wasn't part of the research.

Just like in an oven, the longer the heat lasts, the more something cooks. In this case it's people, the co-authors said.

“Those heat waves are traveling slower and so slower so that basically means that ... there's a heat wave sitting there and those heat waves could stay longer in the region," Zhang said. "And the adverse impacts on our human society would be huge and increasing over the years.”

The team conducted computer simulations showing this change was due to heat-trapping emissions that come from the burning of coal, oil and natural gas. The study found climate change's fingerprint by simulating a world without greenhouse gas emissions and concluding it could not produce the worsening heat waves observed in the last 45 years.

The study also looks at the changes in weather patterns that propagate heat waves. Atmospheric waves that move weather systems along, such as the jet stream, are weakening, so they are not moving heat waves along as quickly — west to east in most but not all continents, Zhang said.

Several outside scientists praised the big picture way Zhang and colleagues examined heat waves, showing the interaction with weather patterns and their global movement and especially how they are slowing down.

This shows “how heat waves evolve in three dimensions and move regionally and across continents rather than looking at temperatures at individual locations,” said Kathy Jacobs, a University of Arizona climate scientist who wasn't part of the study.

“One of the most direct consequences of global warming is increasing heat waves,” said Woodwell Climate Research Center scientist Jennifer Francis, who wasn't part of the study. “These results put a large exclamation point on that fact.”

Read more of AP’s climate coverage at http://www.apnews.com/climate-and-environment

Follow Seth Borenstein on X at @borenbears

The Associated Press’ climate and environmental coverage receives financial support from multiple private foundations. AP is solely responsible for all content. Find AP’s standards for working with philanthropies, a list of supporters and funded coverage areas at AP.org .