The Climate Action Monitor 2026

2. How are climate-related hazards evolving across countries?

The summer of 2026 provided a stark reminder of the growing risks posed by climate-related hazards. Western Europe experienced its hottest June on record, while successive heatwaves intensified drought conditions, fueled wildfires and disrupted economic activity. Although this chapter focuses on longer-term trends rather than individual events, recent extreme weather events underscore the mounting social and economic consequences of a changing climate.

Climate-related hazards affect livelihoods, ecosystems and economic activity through health impacts and loss of life, damage to housing and productive assets, displacement and disruptions to transport, energy, water and agricultural systems. Their impacts can be severe and unevenly distributed. While measuring their full impacts remains challenging, available evidence points to rising exposure to climate-related hazards across many countries. Hazards refer to climate events such as heatwaves, droughts and floods. Exposure captures the people, infrastructure and economic assets located in affected areas, while vulnerability reflects the predisposition to be adversely affected, together with the capacity to adapt and recover. Assessing climate risks, therefore, requires examining not only how hazards are changing, but also where people and assets are exposed and how well societies are prepared to respond.

The economic costs are already substantial and continue to rise. Across OECD members and partner countries, the annual average economic cost of climate-related disasters rose from USD 101.2 billion in 1994-1998 to USD 249.1 billion in 2021-2025, an increase of 146% in inflation-adjusted terms (Munich RE, 2025[1]). This reflects both the growing frequency and severity of extreme weather events and the rising value of exposed assets. Recent estimates suggest that the extreme weather events experienced across Europe during the summer of 2026 could reduce projected GDP growth by around 1% in several countries, including Belgium, France, Germany and Italy (Stegeman, de Wilde and Hobma, 2026[2]).

OECD indicators show that exposure to a range of climate-related hazards has increased over recent decades, although important differences remain both across and within countries. Some regions are experiencing particularly rapid changes in heat exposure, drought intensity, flood risk or wildfire danger, while differences in demographics, economic structure and adaptive capacity can significantly influence the severity of resulting impacts. National averages can mask important differences in vulnerability across regions and communities.

This chapter analyses the evolution of selected climate-related hazards, with a particular focus on extreme temperature and drought (Figure 2.1). These hazards have emerged as particularly pressing concerns due to their increasing frequency and severity, substantial human and economic costs and cascading impacts across water, food, energy, health and ecosystems. The chapter presents recent trends, subnational disparities, economic impacts, projections and policy responses to strengthen resilience.

Figure 2.1. The 2026 Climate Action Monitor focuses on extreme temperature and drought

Note: The OECD monitors a broader range of climate-related hazards, including extreme precipitation, coastal and river flooding, wind threats and wildfires. A full set of historical and forward-looking indicators is available in the OECD Data Explorer, while a detailed description of underlying methodologies can be found in the associated source.

Source: (Maes et al., 2022[3]; Maes et al., 2025[4]).

2.1. Snapshot of key trends

2.1.1. Extreme temperature

Population exposure to extreme heat has risen markedly over recent decades, with more people experiencing heat stress and increasing frequencies of hot days and tropical nights. This is increasing risks to public health, labour productivity and economic activity. In the last two decades, all regions have experienced a significant rise in the number of days with above-average temperatures (Figure 2.2). This increase is especially notable in Latin America, East Asia and the Pacific, as well as Sub-Saharan Africa, where adaptive capacity is considerably more limited (IPCC, 2022[5]).

2.1.2. Drought

Drought risk is also expanding beyond arid and semi-arid regions. Climate change is increasing the frequency, severity and geographic extent of droughts, not only intensifying in many of traditionally affected areas but also extending drought risks to regions not historically associated with chronic water scarcity, including parts of Europe, North America and the Amazon basin (OECD, 2025[6]). This widening exposure poses growing challenges for ecosystems, agriculture, water security and livelihoods.

OECD data shows that cropland soil moisture has declined in many OECD members and partner countries, increasing risks to agricultural production, food systems, water resources and rural livelihoods. These impacts are often unevenly distributed, with some regions experiencing much more severe conditions than national averages suggest. Overall, most affected regions have experienced particularly significant declines in average soil moisture during the past decades (Figure 2.3).

Declining soil moisture is one manifestation of broader pressures on land and ecosystems. Land degradation, including soil erosion, loss of soil fertility, deforestation, desertification and the decline of ecosystem functions, already affects nearly 40% of the world's land and more than 3.2 billion people. Around 1.3 billion people live directly on degraded land, underscoring the environmental and social risks (UNCCD, 2026[7]).

2.1.3. Wildfire

Drought and extreme heat increase wildfire risk by drying soils and vegetation, increasing fuel flammability and creating conditions for faster fire spread. Across OECD countries, fire-weather conditions have increased since the 1980s (Maes et al., 2022[3]). Fire-prone areas are projected to expand as warming continues. Human activity remains a major source of ignition: around 96% of wildfires in the EU are attributed to human actions, while climate change increases the conditions under which fires can spread and become more severe (European Commission, 2026[8]). Land-use and forest management practices can further modify fuel loads and landscape connectivity, affecting both the probability and potential extent of fires.

Wildfire risks are particularly high in Mexico and several other countries, including Argentina, Australia, Brazil, India, Portugal, Spain and Türkiye, where hot and dry climatic conditions, extensive forested areas and recurrent droughts contribute to elevated exposure (Figure 2.4). Exposure nonetheless varies considerably across countries. In some cases, a large share of forests is exposed to wildfire, while in others the exposed share is smaller but affects a much larger forest area in absolute terms. For example, Brazil's extensive forest area means that even a relatively low exposure rate can translate into millions of hectares exposed to wildfire. Both relative and absolute measures therefore provide important insights into wildfire risk (Figure 2.4).

However, these indicators on burned area and wildfire risk should be interpreted with caution. OECD estimates, which assess forests that have a potential risk of wildfire danger, are based on globally consistent geospatial data sources (Maes et al., 2022[3]), making them suitable for identifying broad trends and enabling international comparisons. They complement, but do not replace, national assessments, which may draw on more detailed ground-based data and locally specific methodologies.1

 

Note: Square size represents a country’s total forest area, and the colour scale captures the share of forest exposed to wildfire risk.

Source: OECD (2025), “Climate-related hazards: Historical exposure to wildfires”, OECD Environment Statistics (database).

Recent fire seasons illustrate the growing scale and geographical reach of wildfire risk. By 26 August 2026, more than 637 000 hectares (ha) had been burned across the European Union, more than twice the 2006–2025 average for the same date (European Commission, 2026[9]). France recorded more than 96 000 ha burned, Spain around 140 000 ha between 13 and 28 July 2026, while Belgium's High Fens fire burned around 3 000 ha, making it the country’s largest wildfire on record (World Weather Attribution, 2026[10]) (Fiorin et al., 2026[11]). The High Fens fire highlights that severe wildfire risk can also emerge in regions not traditionally associated with large wildfires (European Commission, 2026[12]).

2.1.4. Extreme precipitation

Increasingly intense precipitation events are placing growing pressure on agriculture and infrastructure. OECD indicators track these changes through a range of measures such as very heavy precipitation days ((>20 mm/day) and the frequency of extreme precipitation days (daily precipitation > 99th percentile of the reference period) (Maes et al., 2022[3]). These indicators show a significant increase over time in the share of land exposed to extreme precipitation events (Figure 2.5).

The increase of extreme precipitation contributes to higher risks from storms, tropical cyclones, river and coastal flooding and landslides, all of which threaten lives, infrastructure and economic assets. Exposure is particularly high in the Americas, where tropical cyclones, extensive floodplains and continuing development in coastal areas increase vulnerability (Figure 2.5). As extreme precipitation events become more frequent and widespread, resilience and adaptation measures will be necessary (Box 2.1).

Box 2.1. Heavy precipitation requires targeted investment to build climate-resilient infrastructure in West Africa

Heavy precipitation events are becoming more frequent and intense in the Sahel and West Africa, causing fatalities and significant economic losses each year. About 42% of West Africa’s roads – nearly 40 000 km – lie within extreme-precipitation zones (Figure 2.6). As rainfall intensifies, disruptions to transport infrastructure are becoming more frequent, making climate resilience an economic priority.

Unpaved roads, which account for approximately 8 700 km of the exposed road network, are especially vulnerable, as intense and prolonged rainfall causes washouts, erosion and pothole formation. In fragile and conflict-affected areas of the Sahel, transport disruptions caused by extreme rainfall can compound existing security and development challenges by limiting access to markets, public services and humanitarian assistance. Climate-resilient infrastructure investment is essential to safeguard livelihoods, economic stability and intra-regional trade across West Africa.

Figure 2.6. West Africa’s road network is highly exposed to heavy rainfall

Hotspots of road exposure to heavy rainfall across West African countries, 2019-2023

Note: Dark green represents paved roads that fall in the hotspots of heavy rain; brown represents unpaved roads that fall in the hotspots of heavy rain; and blue represents unclassified roads that fall in the hotspots of heavy rain i.e. it is unclear whether these roads are paved or unpaved. Those roads that fall outside the hot spots of heavy rain are in grey.

Source: (OECD/GGGI/The World Bank, 2026[13]).

2.2. Heatwaves are becoming more frequent, intense and costly

Prolonged periods of unusually high temperatures can affect human health, reduce labour productivity, damage infrastructure, strain energy and water systems and exacerbate existing environmental stresses (Costa, 2025[14]). Extreme heat can also interact with other hazards, increasing the likelihood and severity of droughts, wildfires and crop failures.

As average temperatures rise heatwaves are becoming more frequent, longer and more intense Their impacts are not determined by temperature alone but also by exposure and adaptive capacity. Countries and regions that have historically experienced milder summers may be particularly vulnerable where buildings, infrastructure and health systems are less adapted to sustained heat. (Lomas and Porritt, 2017[15]).

The summer of 2026 provided a stark reminder of Europe's growing vulnerability to extreme heat. June was the hottest on record in Western Europe, significant health impacts reported across the region. It is estimated that extreme heat has claimed over 200 000 lives across in Europe between 2022 and 2025 alone, with heat-related mortality up by around 30% over the past two decades (WHO, 2026[16]).

The summer of 2026 was not an isolated European event. Japan recorded five consecutive days above 40°C in late July (kokushobi (“severely hot day”),Box 2.2), its longest such streak since comparable nationwide records began in 2011, with heatstroke alerts issued across 37 of its 47 prefectures (NHK World Japan, 2026[17]). North America experienced a comparable episode in late June and early July 2026, with extreme heat producing numerous temperature records across the United States. Atlantic City, New Jersey, on the US East Coast, reached 41.1°C (106°F) on 4 July, tying its all-time temperature record (National Weather Service, 2026[18]).

2.2.1. The average temperature change is increasing overall

The increase in average temperatures over recent decades has been substantial. Across OECD members, average air surface temperatures, over the last 5 years (2021-2025), is approximately 1.06°C higher than the average temperature of the reference period 1981-2010. However, warming has been particularly pronounced in several European countries, with average temperatures surpassing 1.5°C in Austria, Romania and Switzerland. By contrast, some countries that already experience relatively warm climates, including India, Australia and Costa Rica, have recorded smaller increases over the same period. The pace and magnitude of warming vary considerably across regions, shaping countries' exposure to climate-related hazards in different ways (Figure 2.7).

2.2.2. Exposure to extreme heat is increasing, imposing growing human and economic costs

Rising temperatures affect all countries, but not all populations are affected equally (Figure 2.7). In some OECD members and partner countries, large shares of the population are now exposed to more than eight weeks of hot days each year with maximum temperature exceeds 35°C, placing growing pressure on public health, labour productivity, energy systems and economic activity (Figure 2.8). In Hungary, 90% of the population is exposed to more than 8 weeks of hot days, a value surpassed only by Saudi Arabia.

Extreme heat generates substantial human and economic costs. EM-DAT data covering 220 countries globally, indicates that heatwaves accounted for nearly 336 500 fatalities between 2000 and 2026, representing almost 20% of the 1.7 million disaster-related deaths recorded over this period (Centre for Research on the Epidemiology of Disasters (CRED), 2026[19]).2 Given reporting lags and attribution challenges for heat-related cardiovascular and respiratory deaths, the recorded heatwave share is likely underestimated. A recent industry risk assessment estimates cumulative heat-related GDP losses of 5-7% by 2030 for the largest, most exposed European economies (Krichene et al., 2026[20]).

2.2.3. Tropical nights are becoming more frequent

Tropical nights, associated with high night-time temperatures - minimum daily temperature (T_min) exceeds 20°C -, generate risks to human health due to sleep disturbances and inability to cool down at night and recover from daytime heat, thereby compounding the risk of heatwaves (Seltenrich, 2023[21]). OECD data on population-weighted tropical nights shows a clear and persistent increase in nighttime heat exposure worldwide (Maes et al., 2022[3]).

Since 1979, tropical nights globally increased by about one additional night per year - 36 additional nights. The rise is geographically consistent, with long time series recording positive trends. The increase is particularly pronounced in the Middle East and North Africa (+38 nights, from 93 to 131), and Sub-Saharan Africa (+25, from 173 to 198). The European Union (EU27) starts from a lower baseline but still records an increase of 10 nights (more than doubling, from 9 to 19). These trends indicate that the increase in nighttime heat exposure is widespread across climate zones. Although tropical countries are accustomed and often prepared for high night-time temperatures, the increase is affecting more people, surpassing 90% of people exposed in some regions.

Europe is warming roughly twice as fast as the global average, straining health systems, social care and communities already facing more frequent, intense heatwaves (United Nations, 2026[22]). Differences across subnational regions highlight important variations in how heat exposure is changing within countries (Figure 2.1). Both hot days - maximum daily temperature (T_max) exceeds 35°C - and tropical nights have increased across Europe, but their relative increases vary considerably between regions. For example, areas within central Germany are experiencing a particularly high rate of increase in hot days (more than 200% above the 1981-2010 baseline), meanwhile in the Baltic states and Scandinavia there is a more pronounced increase in tropical nights, exceeding 120% in many areas (Figure 2.9). Belgium, Hungary and Romania, as well as parts of France, Italy and Spain, stand out for increases in both indicators: in these regions, the number of hot days has more than tripled relative to the baseline period, while tropical nights have more than doubled. These patterns highlight that most changes are not necessarily those with the largest increase in hot days alone, but those where increases in tropical nights, particularly when combined with more frequent hot days, reduce opportunities for overnight recovery.

Figure 2.9. Some European regions face sharp increases in both hot days and tropical nights

Rate of increase of hot days and tropical nights across small subnational regions (OECD Territorial Level 3), 2021-2025 average compared to the climate reference period 1981-2010

Note: Hot days are days with a maximum temperature above 35°C, while tropical nights are nights with a minimum temperature above 20°C. Darker shading indicates larger increase relative to 1981-2010 base period. All estimates are population weighted. For further details on the methodology for temperature indicators, see (Maes et al., 2022[3]).

Source: OECD (2026), “Climate-related hazards: Historical exposure to extreme temperature”, OECD Environment Statistics (database).

2.2.4. Subnational disparities are critical for understanding heat exposure

National averages can conceal substantial disparities in heat exposure within countries and provide limited guidance for local adaptation planning, particularly in nations where climatic conditions, ecosystems, population densities and economic activities vary considerably across regions. Subnational analysis is therefore essential for identifying local hotspots of vulnerability and informing place-based adaptation.

Figure 2.10 presents subnational data on population exposure to extreme heat across OECD members and partner countries. The results reveal considerable disparities both across and within countries. In many cases, such as Colombia, Mexico, Spain or Türkiye, some regions are experiencing significantly higher increases in heat exposure than national averages would suggest. These indicators are based on a harmonised methodology, which uses the population-weighted change in the number of days with maximum temperatures above 35°C as a measure of exposure to extreme heat (Maes et al., 2022[3]). Consequently, the results should be interpreted as a specific indicator of population exposure and, within a national context, should not be considered a comprehensive measure of climate risk, vulnerability or the full range of climate-related impacts.3

Subnational differences have important implications for adaptation and resilience planning. Vulnerability is shaped not only by the magnitude of climate hazards but also by local characteristics, including demographic structure, urbanisation patterns, economic activities, ecosystem conditions and adaptive capacity. Identifying where heat exposure is increasing most rapidly can help target investments and adaptation measures that reflect local realities.

Figure 2.10. Heat exposure varies considerably across regions and within countries

Change in population weighted number of hot days (T_max > 35°C, days/year), 2021-2025 average compared to the climate reference period 1981-2010

Note: A population-weighted hot day is defined here as days where the maximum temperature exceeds 35°C where populations are located. National averages are shown here as a full black diamond, while the highest and lowest small subnational regions (OECD Territorial Level 3) are shown as hollow diamonds. For further details on the methodology for temperature indicators, see (Maes et al., 2022[3]).

Source: OECD (2026), "Climate-related hazards: Historical exposure to extreme temperature", OECD Environment Statistics (database).

2.2.5. Forward-looking projections point to more frequent extreme heat events

Recent heatwaves offer a glimpse of the climatic conditions that are projected to become increasingly common in the future. Under intermediate- and high-emissions scenarios (i.e., SSP2-4.5 and SSP3-7.0), climate model projections consistently show that rising global temperatures will increase the frequency, duration and intensity of extreme heat events across most OECD members and partner countries (Maes et al., 2025[4]). While the precise magnitude of future heat exposure depends on GHG emissions trajectories, all scenarios point towards a warmer world with more frequent heat extremes, more tropical nights and longer periods of heat stress. Events considered “exceptional” today are therefore expected to become increasingly commonplace over coming decades.

Projections for the European Union show a clear increase in population exposure to extreme heat by mid-century, with larger increases under higher-emissions scenarios. By 2050, population-weighted hot days increase from around 2–3 days per year under the low-emissions scenario (i.e., SSP1-2.6) to around 5-7 days under the very high-emissions scenario (i.e., SSP5-8.5) (Figure 2.11). Recent current values (such as 2022, 2025) are already higher than the modelled trajectories for the year and are above levels projected for 2050 under almost all emissions scenarios. This highlights the exceptional nature of recent extreme heat events, illustrating how current extremes may foreshadow future climate conditions.

 

Note: Hot days are defined as days with a maximum temperature above 35°C. Orange bars show observed historical values. The four Tier-1 SSPs represent the low- to very high-emissions scenarios (i.e. SSP1-2.6, SSP2-4.5, SSP3-7.0 and SSP5-8.5). Only the median value of the multi-model ensemble is visualised here. However, the 10th and 90th percentile uncertainty ranges of the multi-model ensemble are available on the OECD Data Explorer. For further details on the methodology for historical and forward-looking indicators, see (Maes et al., 2022[3]; Maes et al., 2025[4]). OECD Interactive Flourish chart, see other countries and groups.

Source: OECD (2026), “Climate-related hazards: Historical exposure to extreme temperature”, OECD Environment Statistics (database). OECD (2026), “Climate-related hazards: Projected exposure to extreme temperature”, OECD Environment Statistics (database).

2.2.6. Adapting to an already warmer present

Rising heat exposure is already reshaping how societies need to prepare for extreme temperatures. The frequency, duration and intensity of heatwaves observed over the past several years require adaptation across housing, workplaces, health systems and urban infrastructure, including measures that address increasingly disruptive night-time heat. Many OECD countries have started developing targeted responses, with Japan providing an example of a comprehensive multi-level approach to heat action planning (Box 2.2).

Box 2.2. Japan’s multi-level approach to heat action planning

Rising temperatures and more frequent extreme heat events are prompting governments to strengthen heat adaptation measures. Japan provides a notable example. In 2026, the Japan Meteorological Agency formally adopted the term kokushobi (“severely hot day”) for days exceeding 40°C, extending the existing classification that includes mōshobi ("extremely hot day") for temperatures above 35°C. Building on decades of experience and responding to the needs of one of the world’s oldest populations, Japan has developed a multi-level heat action framework that combines national heat-related health warnings with locally tailored response measures implemented by prefectures and municipalities.

Measures include:

• Nationwide Heat Stroke Alert system: The Ministry of the Environment and the Japan Meteorological Agency operate a nationwide Heat Stroke Alert system based on the Wet Bulb Globe Temperature (WBGT), which combines temperature, humidity and solar radiation to assess heat stress risk. Alerts are issued when dangerous conditions are forecasted and are accompanied by guidance on hydration, limiting outdoor activity and other support measures.

• Localised heat action plans: Under Japan’s amended 2023 Climate Change Adaptation Act, local governments convert national heatstroke alerts into climate action plans. Municipalities deploy designated cooling shelters (shirutsu sersutã), partner with volunteer networks to check on seniors and provide targeted support to vulnerable households.

• Mandatory heatstroke prevention measures in workplaces. Since June 2025, employers have been required to implement measures to prevent heat-related illness among workers, including monitoring heat conditions, establishing response procedures and ensuring rapid access to medical assistance. Workplace deaths attributed to heatstroke reportedly fell by around 50% in 2025 compared with the previous year.

• A four-month waiver of basic water charges for around 8 million households in Tokyo. Introduced in 2025, the measure aimed to reduce monthly household expenses and thereby encourage residents to use air conditioning during periods of extreme heat.

• Promotion of “heat acclimatisation” (shonetsu junka). Heat acclimatisation refers to the gradual adaptation of the body to higher temperatures before the peak summer season. Local governments and businesses, supported by media campaigns, have actively promoted practices such as light exercise, walking and warm baths to improve the body’s ability to regulate temperature and sweat efficiently.

The Japanese approach highlights the importance of combining national forecasting and warning systems with local implementation capacities. As heat impacts vary significantly across regions and population groups, locally tailored heat action plans can help ensure that vulnerable residents receive timely support and that adaptation measures reflect local needs.

Source: (Japan Meteorological Agency, 2026[23]; Government of Japan, 2023[24]; Asahi Shinbum, 2025[25]).

Vulnerability to extreme heat is not distributed equally. Older adults, outdoor and manual workers, populations living in housing without adequate cooling or insulation consistently show the highest heat-attributable mortality and morbidity across the evidence base (Ballester et al., 2023[26]) (WHO, 2026[16]). (Ballester et al., 2023[26]) (WHO, 2026[16]). Vulnerability also varies geographically and within populations according to levels of heat exposure and demographic characteristics. Combining heat exposure with the share of older residents can help identify areas where heat poses greater risks to older people and better target adaptation measures (OECD, forthcoming). An example is the case of Europe where residents of elderly care facilities are particularly exposed, as many facilities across Europe remain insufficiently equipped with adequate cooling, ventilation and heat-management systems to cope with increasingly frequent and intense heatwaves. In contrast, Japan, which has one of the world's oldest populations, has developed one of the most advanced heat action planning systems to protect vulnerable groups from extreme heat (Box 2.2).

2.3. Drought is emerging as a growing threat to water, food and energy security

Droughts develop slowly and can persist for months or years, affecting water availability, agricultural production, ecosystems, energy generation and economic activity. Their impacts often extend beyond the directly affected region, disrupting food supply chains, increasing wildfire risk, placing growing pressure on water resources and affecting livelihoods.

Climate change is altering temperatures and precipitation patterns, increasing potential evapotranspiration and creating growing imbalances between water supply and demand (IPCC, 2021[27]). As a result, droughts are becoming more frequent and severe in many regions, while also exacerbating other climate-related hazards, including wildfires, heatwaves and water shortages.

2.3.1. Declining soil moisture signals growing drought risks

The OECD tracks agricultural drought using indicators of soil water content in the topsoil layer and assesses future long-term hydrological drought using the Standardised Precipitation-Evapotranspiration Index (SPEI).4 These indicators help identify emerging risks and target adaptation efforts where they are most needed.

OECD evidence indicates that drought conditions have intensified substantially across many member and partner countries. Figure 2.12 shows the percentage change in cropland soil moisture relative to the 1981-2010 baseline for selected countries. As an indicator of agricultural drought, cropland soil moisture provides a measure of the water available for plant growth and agricultural production.

Across OECD member and partner countries, cropland soil moisture in 2021-2025 was -2.6% lower on average than during 1981-2010, indicating a broad deterioration in agricultural water availability and growing exposure to drought-related risks. Reduced soil moisture can lower crop yields, increase irrigation requirements, place additional pressure on freshwater resources and heighten vulnerability across food, energy and ecosystem systems. The magnitude of these changes varies considerably across countries, illustrating that drought risks are not evenly distributed.

2.3.2. National averages mask important subnational differences in drought risk

Longer and more frequent dry spells are projected across most OECD members and partner countries. Southern European countries are projected to be among the most exposed, with Greece, Portugal, Spain and Türkiye experiencing an average increase of 8 to 12 consecutive dry days (Figure 2.13).

However, national averages tell only part of the story, as subnational projections frequently reveal much wider ranges of outcomes. For example, Portugal's national average indicates an increase of around ten consecutive dry days, while regional projections range from approximately one to more than sixteen additional dry days. Similar patterns can be observed in countries such as Canada, where national averages appear relatively moderate but certain regions face much larger increases in the duration of dry spells.

These differences highlight the importance of place-based drought adaptation. Expanding sub-national and national drought management plans could help ensure that adaptation measures are better aligned with local conditions, thereby improving the efficiency of public investment, strengthening resilience and reducing future economic and environmental losses. Many OECD countries, particularly in Europe, have already incorporated regional drought planning; however, coverage remains uneven (Figure 2.13) (OECD, 2025[6]).

Figure 2.13. Dedicated drought plans are lacking in many regions

Change in maximum number of consecutive dry days under the intermediate-emissions scenario (SSP2-4.5) across OECD members, 2080-2099 average compared to the climate reference period 1995-2014

Note: National averages are shown as a fully coloured diamond, while the highest and lowest small subnational regions are shown as hollow diamonds. Colours indicate which countries have national and subnational drought management plans. Diamonds represent the median value of the multi-model ensemble to better capture uncertainty in climate projections, thereby addressing uncertainties due to model structure. For more information, see (Maes et al., 2025[4]). Data include OECD countries, Romania and Bulgaria, as OECD analysis on drought management plans focused on these countries (OECD, 2025[6]).

Source: OECD (2025), “Climate-related hazards: Projected exposure to drought”, OECD Environment Statistics (database). Drought management plan data from (OECD, 2025[6]).

2.3.3. Economic costs of drought are rising in various sectors

Impacts on agriculture

More frequent and intense drought conditions can be observed across OECD members and partner countries, particularly in agricultural regions. Figure 2.14. shows the change in agricultural drought between 2016–2025 and 1981–2010 for OECD members and partner countries. Most countries have experienced declines in cropland soil moisture indicating intensifying agricultural drought. Drought impacts also vary considerably within countries.

The economic implications are particularly significant in countries that depend heavily on food exports. Several countries and regions play a disproportionately important role in global food production and trade, including Argentina, Brazil, China, the United States and the Black Sea region (particularly Ukraine and the Russian Federation) (OECD, 2025[6]). Climate-related disruptions in these countries can therefore have consequences that extend far beyond national borders, increasing global food prices and undermining food security.

Argentina and Brazil stand out as the most exposed among OECD partner countries, combining a high share of food exports with a worsening agricultural drought (Figure 2.14.). In Brazil, cropland soil moisture has declined by more than 6% between 2016–2025 compared with the 1981–2010 baseline; food products accounted for roughly 38% of merchandise exports (Figure 2.14.). At the same time, Brazil has experienced an additional 22 heat stress days per year relative to the reference period 1981-2010 – a factor that can further intensify drought conditions. This combination of worsening cropland drought, high heat stress and heavy reliance on food exports points to a growing risk to future agricultural output and export revenue in Brazil and similarly exposed economies.

 

Note: Bubble size reflects cropland share of total land area (%, 2022), while colour gradient reflects change in strong heat stress days (days/year, 2016-2025). Iceland is not included in this graph because the land cover dataset does not identify cropland in Iceland. Iceland does, however, have a high share of food exports (45.5%, 2016–2025 mean), though this mostly reflects fish, crustaceans, molluscs, and aquatic invertebrates rather than crop-based agriculture. Since no cropland is identified in the country, no estimate of cropland soil moisture anomaly or cropland share could be calculated for Iceland.

Source: OECD (2026), “Climate-related hazards: Historical exposure to drought”, OECD Environment Statistics (database).

Comtrade database, United Nations Statistics Division, World Integrated Trade Solution system; Staff estimates, World Bank.

OECD (2026), “Climate-related hazards: Historical exposure to extreme temperature”, OECD Environment Statistics (database).

OECD (2025) “Land cover and land cover change”, OECD Environment Statistics (database).

Impacts on the energy sector

Drought can also disrupt energy generation, particularly in countries that rely heavily on hydropower and nuclear power. In France, for example, nuclear power generates around 70% of electricity. Low river flows and high-water temperatures can constraint reactor cooling. During the 2022 drought and heatwave, output was reduced at several reactors because environmental regulations limit the discharge of heated cooling water into already stressed rivers (Clean Air Task Force, 2023[28]). The combination of drought, extreme heat and maintenance outages contributed to a sharp decline in nuclear energy availability.

Similarly, drought conditions have also affected hydropower energy production in Brazil, where hydropower accounted for 48% of power generation in 2025 (Ember, 2026[29]). Years with exceptionally dry conditions caused by reduced rainfall have constrained Brazil’s hydropower output (Box 2.3). Diversification towards wind and solar has helped reduce reliance on hydrological conditions and enhance resilience to prolonged drought (Box 2.3).

Box 2.3. A more diversified renewable electricity mix helps Brazil manage drought risks

Brazil has experienced several years of reduced rainfall, resulting in exceptionally dry conditions during dry seasons, including in 2021, 2023, and 2025. This persistent drying has constrained hydropower output, leading to a decline in total hydro generation (Figure 2.15, Panel A). Although hydropower remained the country’s largest source of electricity, accounting for 48% of power generation in 2025, it represented less than half of Brazil’s electricity mix for several consecutive months, an uncommon occurrence in a system historically dominated by hydro resources.

Meanwhile, wind and solar power have assumed an increasingly important role in Brazil, more than doubling their combined share of generation from 9.9% in 2019 to 24% in August 2025 (Figure 2.15, Panel B). This increasing diversification of Brazil’s electricity mix strengthens the sector’s resilience by broadening the range of generation sources. Expanding wind and solar capacity has helped satisfy growing electricity demand while reducing reliance on varying hydrological conditions.

Figure 2.15. Brazil’s drought conditions affect hydropower generation

Panel A. Hydroelectric power stations and soil moisture anomaly during the dry season, % change in May – October 2025 vs. the climate reference period 1981-2010. Panel B. Brazil’s electricity generation by source, 2000-2025.

Note: Data on the geographic location of hydroelectric power plants in Brazil was last updated in August 2025. Soil moisture anomaly is an appropriate indicator for assessing short-term agricultural drought, whereas other drought indicators are better suited for evaluating longer-term hydrological drought. Nevertheless, the dry seasons of 2021, 2023 and 2025, for example, are recognised as having lasted considerably longer than the historical average.

Source: (Cavendish et al., 2026[30]), (Fulghum and Miranda, 2025[31]). Hydroelectric power station data (MapBiomas Project, 2025[32]), ERA5-Land soil moisture data in the top-soil layer (Copernicus Climate Change Service, 2022[33]) and ember Electricity yearly electricity data (Ember, 2026[29]).

Drought can also disrupt trade by affecting inland navigation. When river levels fall during prolonged droughts, navigation can become restricted, reducing transport capacity, increasing shipping costs and disrupting supply chains. The Rhine River, one of Europe's most important commercial waterways, illustrates how drought-related disruptions can have significant economic consequences that extend well beyond the directly affected regions (Box 2.4). Another example is the Panama Canal, where authorities restricted vessel transits due to declining water levels in key reservoirs caused by drought conditions, underscoring how climate-related water shortages can disrupt critical infrastructure and global supply chains (Panama Canal Authority, 2026).

Box 2.4. Low Rhine River water levels and their implications for trade and supply chains

The Rhine River is one of Europe's most important freight corridors, connecting major industrial centres in Germany, Switzerland, France and the Netherlands. Low water levels caused by drought can disrupt the transport of energy products, chemicals, raw materials and agricultural goods, generating economic impacts that extend well beyond affected regions.

Repeated drought episodes in recent years have reduced water levels at critical navigation points, forcing vessels to carry lighter loads, increasing transport costs and disrupting the movement of industrial goods, energy products and raw materials. In early August 2026, water levels at Kaub, a critical navigation point on Germany's Rhine River, dropped to a historic low with an actual channel depth of roughly 1.3 metres, below the previous record reached during the 2018 drought. At these levels, the river becomes increasingly difficult and costly to navigate, highlighting the growing vulnerability of inland waterway transport to drought conditions.

Major companies have responded by diversifying transport options, increasing the use of rail and road freight and strengthening supply-chain contingency planning. Adaptation is also driving innovation. Following earlier Rhine droughts, the chemical company BASF invested in specialised shallow-draft vessels designed to operate under low-water conditions. These vessels can continue transporting significant cargo volumes when conventional barges face severe loading restrictions, helping maintain the flow of goods during drought periods. The Rhine experience highlights a broader lesson for adaptation policy. Conditions once considered exceptional are becoming increasingly common in a changing climate.

Source: (Chemistry World, 2026[34]).

2.3.4. Projections indicate increasing long-term drought severity by 2100

Long-term hydrological drought is projected to worsen across OECD members and partner countries through the end of the century. The SPEI index captures cumulative effects of water deficits over time – with lower values indicating more severe drought conditions. Under a high-emissions scenario, average conditions across OECD countries could reach moderately dry levels as early as 2070 (Figure 2.16). Mediterranean countries such as Greece, Israel, Spain and Türkiye are projected to see the most severe declines, with extreme drought conditions expected by 2100.

 

Note: The three scenarios include the low-emissions (SSP1-2.6), intermediate-emissions (SSP2-4.5) and high-emissions (SSP3-7.0) scenarios. The very high-emissions scenario (SSP5-8.5) is not visualised here, as this climate scenario is considered increasingly implausible considering current Nationally Determined Contributions submitted by countries under the Paris Agreement. Iceland is not included in this graph because the land cover dataset does not identify cropland in Iceland. SPEI index values in this figure represent the median of the multi-model ensemble; uncertainty ranges corresponding to the 10th and 90th percentiles for the CMIP6 multi-model projections ensembles are available through the OECD Data Explorer. For further details on the methodology, see Maes et al. (2025).

Source: OECD (2025), “Climate-related hazards: Projected exposure to drought”, OECD Environment Statistics (database).

2.3.5. Strengthening resilience to drought through policy action

As drought risk intensifies, OECD members and partner countries have adopted a variety of approaches to drought governance, from dedicated drought management plans to drought-specific measures embedded in wider climate adaptation and water management policies (Figure 2.17). Common responses include water-use restriction, the development of alternative water resources and measures of managing drought episodes. Several countries, including Spain have adopted proactive approaches to drought management, such as basin-level drought plans, monitoring indicators and predefined response measures to anticipate and manage water shortages (OECD, 2025[6]). In Flanders, emergency measures are integrated with long-term water planning (Vlaamse Overheid, 2025[35]). Financial and sectoral instruments, including agricultural insurance schemes, can also complement drought preparedness and response efforts. Despite these initiatives, the growing severity, duration and frequency of droughts suggest that existing approaches may be insufficient. Adapting to a future of more persistent and widespread droughts will require increasingly sophisticated, anticipatory and systemic responses that strengthen resilience across water, agricultural and economic systems.

Figure 2.17. Many OECD countries have adopted drought national or subnational management plans

Drought management plans in OECD countries

Note: DMP: drought management plan. This map shows subnational drought management plans only for countries that do not have national plans. Countries without any dedicated drought management plans typically address drought risk through water resource management, climate adaptation or other sectoral or local strategies.

Source: (OECD, 2025[6]).

References

[26] Ballester, J. et al. (2023), “Heat-related mortality in Europe during the summer of 2022”, Nature Medicine, Vol. 29/7, pp. 1857-1866, https://doi.org/10.1038/s41591-023-02419-z.

[30] Cavendish, T. et al. (2026), “Brazil is already experiencing the brutal impacts of climate change”, Revista de Saúde Pública, Vol. 60, p. e219420, https://doi.org/10.11606/s1518-8787.2026060007009.

[19] Centre for Research on the Epidemiology of Disasters (CRED) (2026), Emergency Events Database (EM-DAT), UCLouvain / CRED, Brussels, Beligum.

[12] Centre, J. (ed.) (2026), Worsening drought and record heat grip Europe, fuelling extraordinary wildfires and extremely low river flows, https://joint-research-centre.ec.europa.eu/jrc-news-and-updates/worsening-drought-and-record-heat-grip-europe-fuelling-extraordinary-wildfires-and-extremely-low-2026-08-12_en (accessed on 20 August 2026).

[33] Copernicus Climate Change Service (2022), ERA5-Land monthly averaged data from 1950 to present, Copernicus Climate Change Service (C3S) Climate Data Store (CDS), https://doi.org/10.24381/cds.68d2bb30.

[14] Costa, H. (2025), The heat is on: Heat stress, productivity and adaptation among firms, OECD Publishing, https://doi.org/10.1787/19d94638-en.

[29] Ember (2026), Data Methodology, https://ember-energy.org/ (accessed on 6 August 2026).

[8] European Commission (2026), Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions on integrated wildfire risk management, European Commission.

[9] European Commission (2026), Current wildfire situation in Europe, Joint Research Centre, https://joint-research-centre.ec.europa.eu/scientific-activities/natural-and-man-made-hazards/forest-fires/current-wildfire-situation-europe_en.

[11] Fiorin, M. et al. (2026), “Rain aids fight against Belgium’s biggest wildfire, as blazes rip through Europe”, Reuters.

[28] Force, C. (ed.) (2023), The 2022 French nuclear outages: Lessons for nuclear energy in Europe, https://www.catf.us/2023/07/2022-french-nuclear-outages-lessons-nuclear-energy-europe/ (accessed on 20 August 2026).

[31] Fulghum, N. and R. Miranda (2025), Wind and solar generate over a third of Brazil’s electricity for the first month on record, Ember, https://ember-energy.org/ (accessed on 15 July 2026).

[24] Government of Japan (2023), Climate Change Adaptation Act, https://www.japaneselawtranslation.go.jp/en/laws/view/4910/en.

[5] IPCC (2022), IPCC Sixth Assessment Report. Climate Change 2022: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, https://doi.org/10.1017/9781009325844.011.

[27] IPCC (2021), Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change, Intergovernmental Panel on Climate Change, Geneva, https://doi.org/10.1017/9781009157896.

[23] Japan Meteorological Agency (2026), Days with a maximum temperature of 40°C or higher will be named ’extreme heat days’, https://www.jma.go.jp/jma/press/2604/17a/40degree_name.html (accessed on 21 July 2026).

[34] King, A. (ed.) (2026), Record low European river levels threaten chemical freight and process cooling, https://www.chemistryworld.com/news/record-low-european-river-levels-threaten-chemical-freight-and-process-cooling/4024021.article (accessed on 20 August 2026).

[20] Krichene, H. et al. (2026), Too hot to grow - The economic costs of extreme heat, Allianz.

[15] Lomas, K. and S. Porritt (2017), “Overheating in buildings: lessons from research”, Building Research & Information, Vol. 45/1-2, pp. 1-18, https://doi.org/10.1080/09613218.2017.1256136.

[3] Maes, M. et al. (2022), “Monitoring exposure to climate-related hazards: Indicator methodology and key results”, OECD Environment Working Papers, No. 201, OECD Publishing, Paris, https://doi.org/10.1787/da074cb6-en.

[4] Maes, M. et al. (2025), “Monitoring exposure to future climate-related hazards: Forward-looking indicator results and methods using climate scenarios”, OECD Environment Working Papers, No. 264, OECD Publishing, Paris, https://doi.org/10.1787/b9ba6ee0-en.

[32] MapBiomas Project (2025), Collection 2.1 of the Annual Land Use Land Cover Maps of Brazil, https://brasil.mapbiomas.org/en/dados-de-infraestrutura/# (accessed on 15 July 2026).

[1] Munich RE (2025), Natural disasters worldwide: Losses are on the rise as climate change strikes, https://www.munichre.com/en/risks/natural-disasters.html (accessed on 13 July 2026).

[18] National Weather Service (2026), Atlantic City, NJ Monthly Climatological Report, National Oceanic and Atmospheric Administration (NOAA), National Weather Service.

[17] NHK World Japan (2026), Japan’s high temperatures top 40 for fifth straight day, https://www3.nhk.or.jp/nhkworld/en/news/20260725_07/.

[6] OECD (2025), Global Drought Outlook: Trends, Impacts and Policies to Adapt to a Drier World, OECD Publishing, Paris, https://doi.org/10.1787/d492583a-en.

[13] OECD/GGGI/The World Bank (2026), New Data for Africa’s Changing Climate: Mapping Risks, Readiness and Policy Action, OECD Publishing, Paris, https://doi.org/10.1787/27d083ca-en.

[21] Seltenrich, N. (2023), “No Reprieve: Extreme Heat at Night Contributes to Heat Wave Mortality”, Environmental Health Perspectives, Vol. 131/7, https://doi.org/10.1289/EHP13206.

[25] Shinbum, T. (ed.) (2025), Tokyoites to get free basic water charges to cool down in summer, https://www.asahi.com/ajw/articles/15780280 (accessed on 20 August 2026).

[2] Stegeman, H., J. de Wilde and E. Hobma (2026), Hot summer economics: Economic effects of the hot European summer of 2026, Triodos Bank.

[7] UNCCD (2026), UNCCD 17th session of the Conference of the Parties, https://www.unccd.int/cop17 (accessed on 20 August 2026).

[22] United Nations (2026), ‘Heat deaths are not inevitable’: WHO pushes Europe to heat‑proof its hospitals, https://news.un.org/en/story/2026/07/1167955.

[35] Vlaamse Overheid (2025), Blue Deal 2025-2029: Kompas voor een waterwijs Vlaanderen, Vlaamse Overheid, Brussels.

[16] WHO (2026), Planning for a warmer world, https://www.who.int/europe/news/item/16-07-2026-planning-for-a-warmer-world.

[10] World Weather Attribution (2026), Climate change increases likelihood of compounding drivers of severe wildfire conditions in France and Spain, World Weather Attribution.

Notes

← 1.  For example, according to official information from CONAFOR, only 27.5% of Mexico's 138.7 million hectares of forest vegetation is classified as having a very high protection priority due to wildfire risk, while the area actually affected by forest fires in 2024 was 1.6 million hectares.

← 2.  Heat is the dominant driver in several high-mortality years, including 2003 (nearly 74 000 deaths), 2022 (~62 000) and 2024 (~67 000), when it accounted for roughly two-thirds to four-fifths of annual disaster fatalities.

← 3.  A robust assessment of climate risk requires an integrated analysis of hazards, exposure, sensitivity and adaptive capacity, as well as the interactions among these dimensions. Country-specific analyses and official assessments should therefore be consulted when interpreting these results (see the Climate Action Dashboard for country-level information)

← 4.  Drought can take several forms depending on which component of the water cycle is affected. Meteorological drought refers to a prolonged period of low precipitation; agricultural (or ecological) drought refers to a occurs when soil moisture is insufficient to meet the needs of crops and vegetation, and; hydrological drought arises when surface or groundwater levels drop below average over a prolonged period (OECD, 2025[6]).