Provisional State of the

Global Climate 2023

Key messages
• The global mean near-surface temperature in 2023 (to October) was around 1.40 ± 0.12 °C
above the 1850–1900 average. Based on the data to October, it is virtually certain that 2023
will be the warmest year in the 174-year observa�onal record, surpassing the previous joint
warmest years, 2016 at 1.29 ± 0.12 °C above the 1850–1900 average and 2020 at 1.27±0.13 °C.
• The past nine years, 2015–2023, will be the nine warmest years on record.
• Record monthly global temperatures have been observed for the ocean – from April through
to September – and, star�ng slightly later, the land – from July through to September.
• The ten-year average 2014–2023 (to October) global temperature is 1.19±0.12°C above the
1850–1900 average, the warmest 10-year period on record.
• Observed concentra�ons of the three main greenhouse gases – carbon dioxide, methane, and
nitrous oxide – reached record high levels in 2022, the latest year for which consolidated
global values are available (1984–2022). Real-�me data from specific loca�ons show that levels
of the three greenhouse gases con�nued to increase in 2023.
• Ocean heat content reached its highest level in 2022, the latest available full year of data in
the 65-year observa�onal record.
• In 2023, global mean sea level reached a record high in the satellite record (1993 to present),
reflec�ng con�nued ocean warming as well as the mel�ng of glaciers and ice sheets. The rate
of global mean sea level in rise in the past ten years (2013–2022) is more than twice the rate
of sea level rise in the first decade of the satellite record (1993–2002).
• Antarc�c sea-ice extent reached an absolute record low for the satellite era (1979 to present)
in February. Ice extent was at a record low from June onwards, and the annual maximum in
September was far below the previous record low maximum.
• Glaciers in western North America and the European Alps experienced an extreme melt
season. In Switzerland, glaciers lost around 10% of their remaining volume in the past two
years.
• Extreme weather con�nues to lead to severe socio-economic impacts. Extreme heat affected
many parts of the world. Wildfires in Hawaii, Canada and Europe led to loss of life, the
destruc�on of homes and large-scale air pollu�on. Flooding associated with extreme rainfall
from Mediterranean Cyclone Daniel affected Greece, Bulgaria, Türkiye, and Libya with
par�cularly heavy loss of life in Libya.
• Food security, popula�on displacements and impacts on vulnerable popula�ons con�nue to be
of concern in 2023, with weather and climate hazards exacerba�ng the situa�on in many parts
of the world.
• Extreme weather and climate condi�ons con�nued to trigger new, prolonged, and secondary
displacement in 2023 and increased the vulnerability of many who were already uprooted by
complex mul�-causal situa�ons of conflict and violence.
Global climate indicators
The global climate indicators provide an overview of changes in the climate system 1. The set of
interlinked physical indicators presented here connect the changing composi�on of the atmosphere
with changes in energy in the climate system and the response of land, ocean, and ice.

The global indicators are based on a wide range of data sets which comprise data from mul�ple
observing systems including satellites and in situ networks (for details on data sets used in the
report, see Data sets and methods).

Changes to the physical climate, measured here by key indicators, can have cascading impacts on
na�onal development and progress toward the Sustainable Development Goals (SDGs) 2. For
example, changes in the acidity or temperature of the ocean can affect marine life, poten�ally
impac�ng coastal communi�es that may depend on the local catch for their livelihood or food
security. On the other hand, climate science has a cri�cal role to play in facilita�ng sustainable
development. As demonstrated by the 2023 United in Science report, weather, climate, and water-
related sciences support the achievement of many of the SDGs 3. Recognizing the interconnec�ons
between climate and development can therefore lead to synergis�c ac�on—an increasing necessity
as the world gets further off-track from achieving both the SDGs and Paris Agreement 4.

Baselines
Baselines are periods of �me, usually spanning three decades or more, that are used as a fixed
benchmark against which current condi�ons can be compared. For scien�fic, policy and prac�cal
reasons, several different baselines are used in this report, and these are specified in the text and
figures. Where possible, the most recent WMO climatological standard normal, 1991–2020, is used
for consistent repor�ng.

For some indicators, however, it is not possible to use the standard normal owing to a lack of
measurements during the early part of the period. There are also two specific excep�ons. First, for
the global mean temperature �me series – and only for the global mean series – a reference period
of 1850–1900 is used. This is the baseline used in IPCC AR6 WG I as a reference period for pre-
industrial condi�ons and is relevant for understanding progress in the context of the Paris
Agreement. Second, greenhouse gas concentra�ons can be es�mated much further back in �me
using gas bubbles trapped in ice cores. Therefore, the year 1750 is used in this report to represent
pre-industrial greenhouse gas concentra�ons.

Greenhouse gases
Key message:

• Observed concentra�ons of the three main greenhouse gases – carbon dioxide, methane, and
nitrous oxide – reached record high levels in 2022, the latest year for which consolidated
global values are available (1984–2022). Real-�me data from specific loca�ons show that levels
of the three greenhouse gases con�nued to increase in 2023.

Atmospheric concentra�ons of greenhouse gases reflect a balance between emissions from human
ac�vi�es, natural sources, and sinks. Increasing levels of greenhouse gases in the atmosphere due to
human ac�vi�es have been the major driver of climate change since the industrial revolu�on. Global

1 https://journals.ametsoc.org/view/journals/bams/102/1/BAMS-D-19–0196.1.xml
2 Climate Indicators and Sustainable Development: Demonstrating the Interconnections (wmo.int)
https://library.wmo.int/records/item/56276-climate-indicators-and-sustainable-development-demonstrating-the-interconnections
3 United In Science 2023 (wmo.int) https://library.wmo.int/records/item/68235-united-in-science-2023
4 https://sdgs.un.org/sites/default/files/2023–09/UN%20Climate%20SDG%20Synergies%20Report-091223B_1.pdf
average mole frac�ons of greenhouse gases – referred to here for simplicity as the “concentra�on” in
the atmosphere – are calculated from in situ observa�ons made at mul�ple sites through the Global
Atmosphere Watch (GAW) Programme of WMO and partner networks.

In 2022 – the latest year for which consolidated global figures are available – atmospheric levels of
greenhouse gases reached new highs (Figure 1), with globally averaged concentra�ons for carbon
dioxide (CO2) at 417.9 ± 0.2 parts per million (ppm), methane (CH4) at 1923 ± 2 parts per billion (ppb)
and nitrous oxide (N2O) at 335.8 ± 0.1 ppb, respec�vely 150%, 266% and 124% of pre-industrial
(1750) levels. The rate of increase of CH4 was the second highest on record, a�er 2021 and the rate
of increase of N2O was the highest on record. The rate of increase of CO2 at 2.2 ppm was slightly
below the 10-year average of 2.46 ppm·yr-1. CO2 growth rate is typically lower in years which start
with La Niña as 2022 did, and higher in years which start with El Niño as 2016 did 5. Real-�me data
from specific loca�ons, including Mauna Loa 6 (Hawaii, United States of America) and Kennaook/Cape
Grim 7 (Tasmania, Australia) indicate that levels of CO2, CH4 and N2O con�nued to increase in 2023.

Figure 1: Top row: Monthly globally averaged mole fraction (measure of atmospheric concentration), from 1984 to 2022, of
(a) CO2 in parts per million, (b) CH4 in parts per billion and (c) N2O in parts per billion. Bottom row: the growth rates
representing increases in successive annual means of mole fractions for (d) CO2 in parts per million per year, (e) CH4 in parts
per billion per year and (f) N2O in parts per billion per year.

Temperature
Key messages:

• The global mean near-surface temperature in 2023 (to October) was around 1.40 ± 0.12 °C
above the 1850–1900 average. Based on the data to October, it is virtually certain that 2023
will be the warmest year in the 174-year observa�onal record, surpassing the previous joint
warmest years, 2016 at 1.29 ± 0.12 °C above the 1850–1900 average and 2020 at 1.27±0.13 °C.
• The past nine years, 2015–2023, will be the nine warmest years on record.

5 Betts, R., Jones, C., Knight, J. et al. El Niño and a record CO2 rise. Nature Climate Change 6, 806–810 (2016).
https://doi.org/10.1038/nclimate3063
6 https://gml.noaa.gov/ccgg/trends/mlo.html Measurements at Mauna Loa were interrupted by a volcanic eruption and the measurement

site was temporarily relocated to Maunakea observatories 21 miles to the north.

7 https://www.csiro.au/greenhouse-gases/
• Record monthly global temperatures have been observed for the ocean – from April through
to September – and, star�ng slightly later, the land – from July through to September.
• The ten-year average 2014–2023 (to October) global temperature is 1.19±0.12°C above the
1850–1900 average, the warmest 10-year period on record.

Global mean near-surface temperature in 2023 (data to October) was 1.40 ± 0.12 °C above the 1850–
1900 average 8 (Figure 2). The analysis is based on a synthesis of five global temperature datasets (see
Data sets and methods). Based on the data to October, it is virtually certain that 2023 will be the
warmest year in the 174-year instrumental record in each of the five data sets. The most recent nine
years – 2015 to 2023 – will be the nine warmest years on record. The two previous joint warmest
years were 2016 with an anomaly of 1.29 ± 0.12°C, and 2020 with an anomaly of 1.27 ± 0.13°C.

There were some noteworthy individual months, with June, July, August, and September 2023 each
surpassing the previous record for the respec�ve month by a wide margin in all datasets. The margin
increased from between 0.14 and 0.20 °C in June to between 0.46 and 0.51 °C in September. The
second-highest margin by which a September record was broken in the past 60 years (the period
covered by all datasets) was 0.02 to 0.17°C in 1983. October was also record warm. July is typically
the warmest month of the year globally, and thus July 2023 became the all-�me warmest month on
record.

The long-term increase in global temperature is due to increased concentra�ons of greenhouse gases
in the atmosphere. The shi� from La Niña, which lasted from mid-2020 to early 2023, to fully
developed El Niño condi�ons by September 2023 (see Short-term Climate Drivers) likely explains
some of the rise in temperature from 2022 to 2023. However, some areas of unusual warming such
as the Northeast Atlan�c (Figure 3) do not correspond to typical paterns of warming or cooling
associated with El Niño. Other factors, which are s�ll being inves�gated, may also have contributed
to the excep�onal warming from 2022 to 2023.

The average global temperature over the past ten years, 2014 to 2023 (data to October), was
1.19 ± 0.12 °C above the 1850–1900 average, making the past ten years the warmest among all ten-
year periods on record in all five datasets. The 2014 to 2023 average is slightly higher than the total
observed warming (1.15 [1.00 to 1.25] °C) for the period 2013 to 2022 es�mated by Forster et al.
(2023) 9, consistent with con�nued warming.

Global average sea-surface temperatures (SSTs) were at a record observed high for the �me of year,
star�ng in the late Northern Hemisphere spring. April through to September (the latest month for
which we have data) were all at a record warm high, and the records for July, August and September
were each broken by a large margin (around 0.21 to 0.27 °C). Excep�onal warmth rela�ve to the
1991–2020 baseline, was recorded in the eastern North Atlan�c, the Gulf of Mexico and the
Caribbean, and large areas of the Southern Ocean (Figure 3, see also Marine heatwaves and cold-
spells).

Global land temperature anomalies reached record observed levels in July and August, somewhat
later than for the SSTs, and the September average was a record by a large margin of 0.53 to 0.72 °C.
The second highest margin in the past 60 years was 0.21 to 0.27 °C in 2002. For the year 2023 to
date, most land areas were warmer than the 1991–2020 average (Figure 3). Unusual warmth was
reported across large areas of the eastern U.S., Mexico, and Central America, as well as western and

8 For anomalies relative to other baselines see Global mean temperature anomalies for 2023 relative to other periods.
9 Forster et al. used an update of the IPCC methodology based on four datasets, two of which are used in the current report. Forster et al.
(2023) Indicators of Global Climate Change 2022: annual update of large-scale indicators of the state of the climate system and human
influence, Earth Syst. Sci. Data, 15, 2295–2327, https://doi.org/10.5194/essd-15–2295–2023.
southern areas of South America. Western Europe and western parts of North Africa, western
Eurasia, areas of Central and southeast Asia, and Japan, were also unusually warm.

Figure 2: Annual global mean temperature anomalies (relative to 1850–1900) from 1850 to 2023. The 2023 average is
based on data to October. Data are from five data sets, see Data sets and methods for details.

Figure 3: Mean near-surface temperature anomalies (difference from the 1991–2020 average) for 2023 to October. Data are
the median of five data sets as indicated in the legend, see Data sets and methods for details.

Ocean
Increasing human emissions of CO2 and other greenhouse gases cause a posi�ve radia�ve imbalance
at the top of the atmosphere, meaning energy is being trapped within the climate system. The
imbalance leads to an accumula�on of energy in the Earth system in the form of heat that is driving
global warming 10,11. The ocean, which covers around 70% of the Earth’s surface, absorbs heat and
CO2, which can act to slow the rate of warming in the atmosphere. However, the heat absorbed by
the ocean leads to ocean warming which, together with the mel�ng of ice on land, raises sea levels.
The ocean also absorbs CO2 leading to ocean acidifica�on 12. Warming waters, sea level rise and
ocean acidifica�on all have significant effects on the ocean, as well as the plants and animals that live
in it and the people who rely upon it for their livelihoods.

Ocean heat content

Key message:

• Ocean heat content reached its highest level in 2022, the latest available full year of data in
the 65-year observa�onal record.

Around 90% of the energy that accumulated in the Earth system since 1971 was stored in the ocean.
As energy has accumulated in the ocean, it has warmed and the heat content of the ocean (Ocean
Heat Content, Figure 4) has increased.

According to a consolidated analysis based on seven individual datasets, the upper 2000 m of the
ocean con�nued to warm in 2022 (the latest full year for which we have data) 13. It is expected that
warming will con�nue – a change which is irreversible on centennial to millennial �mescales 14,15.
Ocean heat content in 2022 was the highest on record, exceeding the 2021 value by 17 ± 9 ZJ
(Figure 4). All data sets agree that ocean warming rates show a par�cularly strong increase in the
past two decades. The rate of ocean warming for the 0–2000 m layer was 0.7 ± 0.1 W·m-2 from 1971–
2022, but 1.2 ± 0.2 W·m-2 from 2006–2022 (the period covered by the Argo programme). Deep-
ocean global warming below 2000 m depth is es�mated to be 0.07 ± 0.1 W·m-2 from 1992–2022 16.

Although ocean heat content (OHC) has increased strongly through the en�re water column, the rate
of warming has not been the same everywhere 17. The strongest warming in the upper 2000 m
occurred in the Southern Ocean (60°S-35°S), North Atlan�c (20°N-50°N) and South Atlan�c (60°S-0°S)
(Figure 5). The Southern Ocean domain is the largest reservoir of heat, accoun�ng for around 36% of
the global OHC increase in the upper 2000 m since 1958. The Atlan�c Ocean accounts for
approximately 33% of the global 0-2000 m OHC increase; the Pacific Ocean around 20%.

Some rela�vely small regions are cooling, including the subpolar North Atlan�c Ocean extending
from near the surface down to a depth of over 800 m (also the only area to show centennial cooling
at the surface). The contras�ng patern of cooling (50°N-70° N) and warming (20°N-50°N) in the
North Atlan�c has been associated with a slowing of the Atlan�c Meridional Overturning Circula�on

10Hansen, J. et al. (2011). Earth’s energy imbalance and implications. Atmospheric Chemistry and Physics https://doi.org/10.5194/acp-11–
13421–2011
11 von Schuckmann, K. et al. (2016). An imperative to monitor Earth’s energy imbalance. In Nature Climate Change.

https://doi.org/10.1038/nclimate2876
12 State of the Ocean Report 2022 | UNESCO https://www.unesco.org/en/articles/state-ocean-report-2022
13 von Schuckmann et al. (2020). Heat stored in the Earth system: where does the energy go? Earth Syst. Sci. Data, 12(3), 2013–2041.

https://doi.org/10.5194/essd-12–2013–2020
14 Cheng, L.; Trenberth, K. E.; Fasullo, J. et al. Improved estimates of ocean heat content from 1960 to 2015, Science Advances 2017, 3 (3),

e1601545. https://doi.org/10.1126/sciadv.1601545.
15 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 https://doi.org/10.1175/2010JCLI3682.1.
16 Purkey, S. G., & Johnson, G. C. (2010). Warming of Global Abyssal and Deep Southern Ocean Waters between the 1990s and 2000s:

Contributions to Global Heat and Sea Level Rise Budgets. Journal of Climate, 23(23), 6336–6351. https://doi.org/10.1175/2010JCLI3682.1
17 Cheng, L., Abraham, J., Trenberth, K.E. et al. Another Year of Record Heat for the Oceans. Adv. Atmos. Sci. (2023).

https://doi.org/10.1007/s00376–023–2385–2 and Cheng, L., von Schuckmann, K., Abraham, J.P. et al. Past and future ocean warming. Nat
Rev Earth Environ 3, 776–794 (2022). https://doi.org/10.1038/s43017–022–00345–1.
and local interac�ons between the air and sea 18. Other cooling regions include the northwest Pacific,
southwest Pacific and southwest Indian Oceans.

Figure 4: 1960–2021 ensemble mean time series and ensemble standard deviation (2-standard deviations, shaded) of global
ocean heat content (OHC) anomalies relative to the 2005–2021 average for the 0–300 m (grey), 0–700 m (blue), 0–2000 m
(yellow) and 700–2000 m depth layer (green). The ensemble means OHC anomalies for the year 2022 has been added as
separate points, together with their ensemble spread, and is based on a subset of 7 datasets. Source: Mercator Ocean
international. Updated from von Schuckmann et al. (2020) 19

Figure 5: Observed upper 2000 m OHC trend from 1958 to 2022. Units: W·m–2. Data updated from Cheng et al. (2017) 20.

18 Cheng, L., von Schuckmann, K., Abraham, J. P. et al: Past and future ocean warming. Nature Reviews Earth & Environment. 2022,

https://doi.org/10.1038/s43017–022–00345–1.
19 von Schuckmann et al. (2020). Heat stored in the Earth system: where does the energy go? Earth Syst. Sci. Data, 12(3), 2013–2041.

https://doi.org/10.5194/essd-12–2013–2020
20 Cheng, L.; Trenberth, K. E.; Fasullo, J. et al. Improved estimates of ocean heat content from 1960 to 2015, Science Advances 2017, 3 (3),

e1601545. https://doi.org/10.1126/sciadv.1601545
Sea level
Key message:

• In 2023, global mean sea level reached a record high in the satellite record (1993 to present),
reflec�ng con�nued ocean warming as well as the mel�ng of glaciers and ice sheets.
• The rate of global mean sea level in rise in the past ten years (2013–2022) is more than twice
the rate of sea level rise in the first decade of the satellite record (1993–2002).

In 2023, global mean sea level (GMSL) has con�nued to rise (Figure 6). The La Niña condi�ons
between mid-2020 and early 2023 had only a small apparent effect on GMSL, unlike the 2011 La Niña
that led to a temporary decrease in the GMSL of several millimetres. The rapid rise observed in 2023
is likely due in part to the nascent El Niño and is likely to increase further as the 2023 El Niño
develops. The long-term rate of sea level rise has more than doubled since the start of the satellite
record, increasing from 2.14 mm·yr-1 between 1993 and 2002 to 4.72 mm·yr-1 between 2013 and
2022.

From January to March 2023, sea levels (Figure 7) were higher than the long-term average (1993–
2012) in the western tropical Pacific. This is characteris�c of warm seawater in the region associated
with ENSO-neutral condi�ons. Sea levels in the North Atlan�c and eastern tropical Pacific were lower
than the long-term average. Warming of the surface waters in the eastern Tropical pacific during the
early stages of the 2023 El Niño (see Short-term Climate Drivers) led to an increase in sea level
rela�ve to the long-term mean in the most eastern part of the Tropical Pacific between April and
June. By July to September, the El Niño signature was clearly visible, with sea level being above
average from the mid-tropical Pacific to the coasts of central and South America. Above average sea
levels were also observed in the tropical and north-east Atlan�c, associated with the anomalous
warming in these areas during Northern Hemisphere summer.

Figure 6: GMSL evolution between January 1993 and October 2023 based on satellite altimetry. The black line is the best
estimate, and the grey shaded area indicates the uncertainty. Near-real-time updates are indicated by a dotted line. Red and
blue annotations indicate the average rate of sea level rise during three decades of the record as indicated. (Source AVISO)
Figure 7: 3-month averages of altimetry-based sea level anomalies (relative to the 1993–2012 average, which is the product
climatology) for (top left) January to March, (top right) April to June, and (bottom left) July to September. Data downloaded
from the Copernicus Marine Service (CMEMS, https://marine.copernicus.eu).

Marine heatwaves and cold spells

As with heatwaves and cold-spells on land, marine heatwaves (MHW) and marine cold-spells (MCS)
are prolonged periods of extreme high or low temperatures in the seas and ocean that can have a
range of consequences for marine life and dependent communi�es 21. MHWs have become more
frequent, intense, and longer las�ng since the late 20th century, while MCSs have been decreasing
by those same measures. Satellite retrievals of sea-surface temperature are used to monitor MHWs
and MCSs globally, categorized here as moderate, strong, severe, extreme, or ice (for defini�ons, see
Data sets and methods).

El Niño events tend to cause wide-spread MHWs in the eastern Tropical Pacific. This region did
experience 'strong' MHWs in 2023 (Figure 8a, to late August), but yet, they have covered a smaller
area than during previous El Niño events. The area is likely to increase as the El Niño con�nues to
develop. Of par�cular concern, in 2023 were the persistent and wide-spread MHWs in the North
Atlan�c throughout Northern Hemisphere summer and early autumn. The Mediterranean Sea was
also unusually warm rela�ve to the baseline period and experienced near complete coverage of
'strong' and 'severe' MHWs for the twel�h consecu�ve year. In the southern hemisphere, the waters
surrounding New Zealand remained 1 to 2°C above the long-term average through January to
September (~270 days).

In contrast, there were almost no occurrences of MCSs within 60° North or South of the equator in
2023 to date (Figure 9a). The global ocean experienced an average daily MHW coverage of 20% (to

21 Smale, D.A., Wernberg, T., Oliver, E.C.J. et al. Marine heatwaves threaten global biodiversity and the provision of ecosystem services.

Nat. Clim. Chang. 9, 306–312 (2019). https://doi.org/10.1038/s41558–019–0412–1

date, Figure 8b), well above the previous record of 17% in 2016. In contrast, the average daily
coverage of MCS (Figure 9b) was only 2%, far below 2022 (5%).

Figure 8: (a) Global map showing the highest MHW category (for definitions, see Data sets and methods) experienced at
each pixel over 2023 (through September; reference period 1982–2011). Light grey indicates that no MHW occurred in a
pixel over the entire year; (b) Stacked bar plot showing the percentage of the surface of the ocean experiencing an MHW on
any given day of the year; (c) Stacked bar plot showing the cumulative number of MHW days averaged over the surface of
the ocean. Note: This average is calculated by dividing the cumulative sum of MHW days per pixel weighted by the surface
area of those pixels. (d) Stacked bar plot showing the total percentage of the surface of the ocean that experienced an MHW
from 1982 to present. Data are from National Oceanic and Atmospheric Administration (NOAA) Optimum Interpolation Sea-
Surface Temperature (OISST). Source: Robert Schlegel

Figure 9: As for Figure 8 but showing marine cold-spells rather than marine heatwaves. Data are from NOAA OISST. Source:
Robert Schlegel.
Ocean acidification
The ocean absorbs around one quarter of the annual emissions of anthropogenic CO2 to the
atmosphere 22,23. CO2 reacts with seawater and alters the carbonate chemistry, resul�ng in a decrease
in pH referred to as ‘ocean acidifica�on’ (Figure 10). Ocean acidifica�on affects organisms and
ecosystem services, including food security, by reducing biodiversity, degrading habitats, and
endangering fisheries and aquaculture 24. The IPCC AR6 concluded that “There is very high confidence
that open ocean surface pH is now the lowest it has been for at least 26 [thousand years] and current
rates of pH change are unprecedented since at least that time”. 25

Although global efforts, many supported by IOC-UNESCO and led by the Global Ocean Acidifica�on
Observing Network and its UN Ocean Decade Programme – Ocean Acidifica�on Research for
Sustainability, have resulted in an increase in the number of ocean acidifica�on observa�ons, many
regions remain under-sampled. Data collected for the Sustainable Development Goal 14.3.1 Indicator
(“Average marine acidity (pH) measured at agreed suite of representa�ve sampling sta�ons”) show
that the current coverage is inadequate, with �me series not long enough to determine trends and
gaps in observa�ons in all areas. The rate of change in ocean acidifica�on, its patern and scale,
shows great regional variability, highligh�ng the need for global efforts to increase high-frequency
and long-term monitoring.

Figure 10: Global annual mean ocean surface pH (purple) covering the period 1985–2022. The shaded area indicates the
estimated uncertainty in the values. Data from Copernicus Marine Environment Monitoring Service.

22 WMO. 2022. WMO Greenhouse Gas Bulletin (GHG Bulletin) – No.18: The State of Greenhouse Gases in the Atmosphere Based on Global
Observations through 2021. https://library.wmo.int/idurl/4/58743
23 Friedlingstein, P., O’Sullivan, M., Jones, M. W., Andrew, R. M., et al. 2020. Global Carbon Budget 2020. Earth System Science Data,

Vol. 12, pp. 3269–3340. https://doi.org/10.5194/essd-12–3269–2020

24 IPCC, 2023: Climate Change 2023: Synthesis Report. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the

Intergovernmental Panel on Climate Change [Core Writing Team, H. Lee and J. Romero (eds.)]. IPCC, Geneva, Switzerland, 184 pp., doi:
10.59327/IPCC/AR6–9789291691647
25 Intergovernmental Panel on Climate Change (IPCC), 2021: Climate Change 2021: The Physical Science Basis, Chapter 2,

section 2.3.3.5 Ocean pH, https://www.ipcc.ch/report/ar6/wg1/.

Cryosphere
The cryosphere comprises the frozen parts of Earth – glaciers and ice sheets, sea-ice, snow, and
permafrost. The inhospitable and o�en remote environments in which ice forms mean that it has
some�mes been challenging to undertake long-term measurements of the cryosphere.

Sea-ice
Key messages:

• Arc�c sea-ice extent remained well below normal in 2023, with the annual maximum and
annual minimum extents being respec�vely the fi�h and sixth lowest in the 45-year satellite
record.
• Antarc�c sea-ice extent reached an absolute record low for the satellite era (1979 to present)
in February. Ice extent was at a record low from June onwards, and the annual maximum in
September was far below the previous record low maximum.

Arc�c sea-ice extent reached its annual maximum of 14.62 million km2 on 6 March, the fi�h lowest in
the satellite record26. The annual minimum Arc�c sea-ice extent was reached on 19 September 27
(Figure 11), with a minimum extent of 4.23 million km2, well below the long-term average (1991–
2020) of ~5.5 million km2. This was the sixth lowest minimum Arc�c sea-ice extent in the satellite
record (1979–2023), not as extreme as 2012 or 2020 but only slightly higher than 2007, 2016 and
2019, and con�nuing the long-term trend of reduced late-summer and early autumn Arc�c ice cover.
Major nega�ve anomalies were observed in the Beaufort, Chukchi, and East Siberian Seas (Figure 11
right).

Figure 11: (left) Daily Arctic sea-ice extent from January through December, showing 2023 to date (red line)
against the climate normal (1991–2020, dark blue) and the record highest and lowest extents for each day (mid
blue). Individual years are shown in light blue. (right) Ice concentration on September 19, 2023, at the annual
minimum Arctic ice extent. The orange line indicates the median ice edge for the 1981–2010 period 28. Data and
map from the U.S. National Snow and Ice Data Center.

Antarc�c sea-ice extent declined to 1.79 million km2 on February 21, 2023 (Figure 12), an all-�me
minimum for the satellite era (1979 to present), slightly less than the previous record low that was

26 Numbers in main text are from NSIDC. OSI SAF figures are provided in Data sets and methods
https://nsidc.org/arcticseaicenews/2023/03/arctic-sea-ice-maximum-at-fifth-lowest-on-satellite-record/,
https://nsidc.org/arcticseaicenews/2023/04/polar-dawn-to-dusk/
27 https://nsidc.org/arcticseaicenews/2023/09/arctic-sea-ice-minimum-at-sixth/
28 1981–2010 is the period used by NSIDC for their monitoring.
set in 2022 29. Sea-ice extent remained below the average as the growth season commenced,
becoming record low in May. Slow growth con�nued with excep�onally low extents between July
and October, likely in associa�on with warming of the Southern Ocean 30. The maximum Antarc�c
sea-ice extent for 2023 (Figure 12) was 16.96 million km2 on 10 September, roughly 1.5 million km2
below the 1991–2020 average and 1 million km2 below the previous record low maximum, from
1986 31.

Figure 12: (left) Daily Antarctic sea-ice extent from January through December, showing 2023 (red) conditions
against the 1991–2020 climate normal (dark blue) and the record highest and lowest extents for each day (mid
blue). Individual years are shown in light blue. (right) Ice concentration on September 19, 2023, representative
of the 2023 annual maximum extent. The orange line shows the median ice edge for the 1981–2010
climatology period 32. Data and map from the U.S. National Snow and Ice Data Center.

Ice sheets
An ice sheet is 33 an expanse of ice that covers an area of more than 50 000 km2. There are two ice
sheets in the present-day climate system, in Greenland and Antarc�ca. The total mass balance (TMB)
of an ice sheet is the sum of three components: the surface mass balance (SMB), the marine mass
balance (MMB), and the basal mass balance (BMB). The SMB is the difference between snow
accumula�on and meltwater runoff from the ice sheet. The MMB is the mass loss at the edge of the
ice sheet from the calving of icebergs and the mel�ng of ice that is in contact with the ocean. BMB
consists of mel�ng at the ice sheet bed due to geothermal heat and fric�on as the ice slides over the
ground beneath. A nega�ve mass balance indicates a loss of ice mass; a posi�ve mass balance
indicates a gain.

Final numbers for the Greenland Ice Sheet mass balance for the hydrological year 2022–2023
(September 2022 to August 2023) are not yet available, but preliminary es�mates from the regional
climate model ensemble method 34 give an es�mated 2022–2023 SMB of 330 Gt, below the long-
term average but well above the extreme melt years of 2011–2012 and 2018–2019. Combined with
BMB (−27 Gt) and MMB (−490 Gt), the es�mate of the 2022–2023 ice sheet TMB is about −190 Gt.
The summer 2023 melt season was rela�vely intense, punctuated by major heat waves in July and

29 Liu, J., Z. Zhu and D. Chen, 2023. Lowest Antarctic Sea-Ice Record Broken for the Second Year in a Row. Ocean-Land-Atmos Res.

2023:2;0007. https://doi.org/10.34133/olar.0007
30 Purich, A., Doddridge, E.W. Record low Antarctic Sea-ice coverage indicates a new sea-ice state. Commun Earth Environ 4, 314 (2023).

https://doi.org/10.1038/s43247–023–00961–9.
31 https://nsidc.org/arcticseaicenews/2023/09/antarctic-sets-a-record-low-maximum-by-wide-margin/
32 The period 1981–2010 is used by NSIDC for their monitoring.
33 https://www.ipcc.ch/srocc/chapter/glossary/
34 Mankoff, K.D. et al. (2021). Greenland ice sheet mass balance from 1840 through next week, Earth Syst. Sci. Data, 13, 5001–5025,

https://doi.org/10.5194/essd-13–5001–2021.
August 35. It was the warmest summer on record at the Summit sta�on, 3.4 °C warmer than the
1991–2020 average and 1.0 °C warmer than the previous record36. Satellite melt-extent data indicate
that the ice sheet is on track for the second highest cumula�ve melt-day area 37 on record (1978–
2023), a�er the extreme melt season of 2012 (Figure 13).

The Ice Sheet Mass Balance Intercomparison Exercise (IMBIE) has documented the accelera�on in
combined mass loss from the Greenland and Antarc�c Ice Sheets over the period of the satellite
record, 1992–2020 38. The average Greenland and Antarc�c Ice Sheet TMBs over this period were
−169 and −92 Gigatonnes per year (Gt·yr-1), respec�vely, and −261 Gt·yr-1 combined. Combining the
two ice sheets, the seven highest melt years on record are all since 2010, and average rates of mass
loss increased from 105 Gt·yr-1 from 1992–1996 to 372 Gt·yr-1 from 2016–2020. This is equivalent to
~1 mm·yr-1 of global sea level rise atributed to the polar ice sheets in the later period. Preliminary
data from 2023 are consistent with these recent rates of mass loss in Greenland, but the Antarc�c ice
sheet gained mass due to higher-than-normal snow accumula�on in the last year and a half.

Figure 13: (left) Greenland Ice Sheet melt extent through the 2023 melt season, courtesy of the U.S. National Snow and Ice
Data Center. (right) Cumulative melt days for Greenland between 1 January and 29 October. White areas indicate no
melting occurred. Images and analysis courtesy of Thomas Mote, University of Georgia.

Glaciers
Key message:

• Glaciers in western North America and the European Alps experienced an extreme melt
season. In Switzerland, glaciers lost around 10% of their remaining volume in the past two
years.

Glaciers are formed from snow that has compacted to form ice, which then deforms and flows
downhill. Glaciers comprise two zones: an accumula�on zone where accumula�on of mass from
snowfall exceeds ice loss, and an abla�on zone where ice loss (abla�on) from mel�ng and other
mechanisms exceeds accumula�on. Where glaciers end in a lake or the ocean, ice loss can occur
through mel�ng where the ice meets the water, and via calving when chunks of the glacier break off.

35 Greenland Today – Late-season melt spike https://nsidc.org/greenland-today/2023/09/late-season-melt-spike/, Sudden shift to

southern heat https://nsidc.org/greenland-today/2023/07/sudden-shift-to-southern-heat/
36 https://www.dmi.dk/nyheder/2023/varmerekorder-pa-indlandsisen/
37 Cumulative melt-day area is the area of the ice sheet that experienced melting conditions each day (diagnosed by the presence of liquid

water at the surface), summed over the number of days in the melt season.
38 Otosaka, I. N., et al. (2023). Mass balance of the Greenland and Antarctic ice sheets from 1992 to 2020, Earth Syst. Sci. Data, 15, 1597–

1616, https://doi.org/10.5194/essd-15–1597–2023.
Glacier mass balance – the amount of mass gained or lost by the glacier – is commonly expressed as
the annual thickness change averaged over the glacier area, expressed in metres of water equivalent
(m w.e.) 39. Melt rates are strongly affected by the glacier albedo, the frac�on of sunlight that is
reflected by the glacier surface. Exposed glacier ice is darker and therefore has a lower albedo than
the seasonal snowpack and is sensi�ve to darkening from mineral dust, black carbon, algal ac�vity,
and fallout from forest fires. Reduced snow cover, long melt seasons, and wildfire ac�vity all serve to
concentrate darker material on the glacier surface, decreasing its albedo and thereby increasing the
melt.
Glacier mass balance data for the 2022–2023 hydrological year are not yet available, but preliminary
observa�ons indicate extremely nega�ve mass balance in both western North America and the
European Alps.

The annual mass loss for Swiss glaciers in 2022–2023 was the second largest on record (1950 to
present, Figure 14) at 4.4% of the remaining ice volume. Together with the record mass loss in 2021–
2022 of 5.9%, Swiss glaciers have lost around 10% of their remaining volume in just two years. This
has been driven by low snowpacks and warm summers each year with poten�al cumula�ve impacts
from glacier darkening associated with longer-than-normal periods of exposed glacier ice and loss of
the high-eleva�on firn 40.

Figure 14: Total annual loss of Swiss glaciers related to the current ice volume. The vertical bars indicate the percentage
change in ice volume relative to the previous year. Red and purple bars are the 8 largest relative mass losses on record. The
purple bars are the relative mass losses for 2022 and 2023. The blue shaded area in the background represents the overall
ice volume. Source: Matthias Huss based on Glacier Monitoring Switzerland, 2022: Swiss Glacier Mass Balance (Release
2023), https://doi.org/10.18750/massbalance.2022.r2022.

Western North America experienced record (1965 to present) glacier mass loss in 2023, with glacier-
averaged annual thinning of more than 3.5 m at LiDAR-monitored 41 glacier sites in the Canadian
Rockies and southern Coast Mountains. This compares with an average glacier thinning rate of

39 m w.e. is the depth of water equivalent to the change in ice thickness. Glacier ice is less dense than water, so the equivalent depth of
water is slightly less than the thickness of ice lost.
40 Firn is multi-year snow, which has a higher albedo than glacier ice.
41 LiDAR stands for Light Detection and Ranging, which uses a laser to determine the height of the glacier surface.
0.90 ± 0.13 m w.e./year in western Canada over the period 2010–2019 42. Adjus�ng the LiDAR
al�metry data for snow and ice density, preliminary es�mates of mass balance at two sites with long-
term measurements, Place and Haig Glaciers, are respec�vely, –3.1 ± 0.5 and –3.8 ± 0.6 m w.e. The
large nega�ve mass balances are due to the combina�on of a low winter snowpack, an intense spring
heatwave that led to early exposure of bare ice across most glaciers and icefields, above-average
summer temperatures, and record wildfire ac�vity in western Canada 43 (see Extreme weather and
climate events), with smoke from the wildfires darkening the glacier surface and contribu�ng to the
extreme melt. Seasonal snow had melted away by early July on many glaciers, exposing darker firn or
ice over an extended melt season (Figure 15). High rates of cumula�ve mel�ng over the past several
years may also be a factor, as impuri�es are increasingly concentrated on the glacier surfaces and the
firn zone has been lost on many mountain glaciers in western North America.

Figure 15: Aerial and ground views of Haig Glacier in the Canadian Rocky Mountains, September 15, 2023. No seasonal
snow or firn remains on the glacier and the surface albedo was anomalously low, including values of 0.07 measured in the
terminus region (albedo is between zero – very dark – and one – very light). Credit Brian Menounos and Shawn Marshall.

Snow cover
Key message:

• Seasonal snow cover in the Northern Hemisphere has been experiencing a long-term decline
in the late spring and summer. Northern Hemisphere snow cover extent for May was the
eighth lowest on record (1967–2023). North American snow cover extent for May 2023 was
the lowest on record (1967–2023).

Seasonal snow cover in the Northern Hemisphere has been experiencing a long-term decline in the
late spring and summer, which con�nued in 2023. Northern hemisphere snow-cover extent from
January through April 2023 was close to the long-term average (1991–2020), but the spring heat
wave in north-western North America drove widespread snowmelt. North American snow cover
extent for May 2023 was the lowest on record (1967–2023) at 7.47 million km2, about
1.57 million km2 (17%) below the long-term average, while overall Northern Hemisphere snow cover
extent was 16.74 million km2, the eighth lowest since 1967, and 1.47 million km2 below the long-
term average (Figure 16).

42 Hugonnet, R., McNabb, R., Berthier, E. et al. Accelerated global glacier mass loss in the early twenty-first century. Nature 592, 726–731
(2021). https://doi.org/10.1038/s41586–021–03436-z
43 Parisien, M.-A., Barber, Q. E., Flannigan, M. D., & Jain, P. (2023). Broadleaf tree phenology and springtime wildfire occurrence in boreal

Canada. Global Change Biology, 00, 1–14. https://doi.org/10.1111/gcb.16820

 Figure 16: May snow-cover extent (SCE) for the Northern Hemisphere (dark blue) and North America (light blue), 1967–
2023. The 1991–2020 average for each region is shown as a horizontal dotted line. Data from the Rutgers University Global
Snow Lab 44 (https://climate.rutgers.edu/snowcover/).

Short-term Climate Drivers

Key message:

• A prolonged period of La Niña from mid-2020 to early 2023 gave way to El Niño condi�ons
which were well established by September 2023, contribu�ng to the observed rise in global
sea-surface temperatures during 2023.

There are many different natural phenomena, o�en referred to as climate paterns or climate modes,
that affect weather and climate at �mescales ranging from days to several months or even years. In
2023, the El Niño–Southern Oscilla�on (ENSO) and the North Atlan�c Oscilla�on, highlighted here,
contributed to major weather and climate events across large areas of the world.

ENSO – El Niño Southern Oscillation

ENSO is one of the most important climate drivers of year-to-year variability in weather paterns
worldwide. It is linked to hazards such as heavy rains, floods, and drought, heatwaves, and cold
spells. El Niño, characterized by higher-than-average sea-surface temperatures in the eastern tropical
Pacific and a weakening of the trade winds, typically has a warming influence on global
temperatures. La Niña, which is characterized by below-average sea-surface temperatures in the
central and eastern tropical Pacific and a strengthening of the trade winds, has the opposite effect.

A mul�-year La Niña event began in mid-2020 and ended in early 2023. Subsequently, sea-surface
temperatures in the eastern tropical Pacific increased, crossing typical El Niño thresholds by June

44 Robinson, David A., Estilow, Thomas W., and NOAA CDR Program (2012): NOAA Climate Data Record (CDR) of Northern Hemisphere

(NH) Snow Cover Extent (SCE), Version 1. NOAA National Centers for Environmental Information. doi: 10.7289/V5N014G9 [accessed
September 19, 2023].
(Figure 17). However, the atmosphere was slower to respond, and it was not un�l early September
that El Niño condi�ons were well established in both the atmosphere and ocean.

The latest forecasts 45 from the WMO Global Producing Centres for Long-Range Forecasts indicate a
high probability (90%) of El Niño con�nuing throughout the upcoming Northern Hemisphere winter,
with condi�ons at the peak of the event corresponding to a strong El Niño. Chances of a transi�on to
ENSO-neutral are very low (10%).

Figure 17: Time series of NOAA’s Oceanic Niño index from January 1950 to September 2023 showing the presence of below-
average conditions (blue) and above average conditions (red) during 3-month average time periods. Anomalies are with
respect to the 1991–2020 average sea-surface temperature. (Data: NOAA NCEI)

El Niño has an influence on regional rainfall paterns. Precipita�on anomalies were typical of El Niño
in some regions (by September): drier than usual condi�ons in mari�me Southeast Asia and from
southern Mexico to northern South America and weter than normal condi�ons in parts of Chile
(Figure 18).

Decreased monsoon rainfall in southeast Asia is associated with El Niño. Onset of the monsoon over
Kerala, India, occurred on 8 June, 7 days later than normal. By the end of September, India received
94% of its typical monsoon rainfall 46. Higher than normal rainfall totals were observed along the
lower course of the Indus River and in central India.

In Australia, La Niña is associated with weter-than-normal condi�ons and El Niño with drier-than-
normal condi�ons. In January, rainfall for the country was 35% above normal as La Niña wound
down, but in August it was 50% below normal with a strengthening El Niño. The dryness was also
accompanied by the highest July-to-September na�onal mean temperatures in Australia’s 114-year
record. September was the driest month on record in Australia for any month.

45 https://public.wmo.int/en/our-mandate/climate/el-ni%C3%B1ola-ni%C3%B1a-update
46 https://mausam.imd.gov.in/responsive/monsooninformation.php
As average global temperature anomalies increased during the transi�on from La Niña to El Niño, the
Earth experienced its hotest June-September period and hotest month on record (July). Many
regions also experienced heatwaves, including parts of the United States, Mexico, Europe, the
Mediterranean, North Africa and the Middle East, China, Siberia, and South America. The UNEP
Fron�ers report 47 highlighted that variability associated with EI Niño has implica�ons for biomass
and fire weather 48 which increases the risk of large and intense wildfires in some places. Please refer
to the sec�on on Extreme weather and climate events for more details on extreme events.

NAO – North Atlantic Oscillation

Average sea level pressure in the North Atlan�c is characterized by an area of lower pressure close to
Iceland known as the Icelandic Low and an area of higher pressure centred over the Azores known as
the Azores High. The North Atlan�c Oscilla�on (NAO) is based on the sea-level pressure difference
between the Icelandic Low and the Azores high. It is mainly associated with driving weather
condi�ons in the North Atlan�c basin, Europe, and the Mediterranean. The posi�ve phase is
characterized by below-normal pressure over the North Atlan�c high la�tudes and above-normal
pressure over the central Atlan�c, the eastern United States, and western Europe. The nega�ve
phase has the opposite patern. Because the phases are determined by pressure changes in the
atmosphere, they can fluctuate more rapidly, on the order of days to weeks, compared with other
climate drivers such as ENSO, which are driven by slower changes in ocean temperature and
fluctuate on the order of months to years.

High pressure associated with the nega�ve NAO in late June and July 2023 contributed to a heatwave
that caused significant snow and ice melt across southern Greenland. Several widespread mel�ng
events covering 800 000 km2, up to 50% of the ice sheet, occurred during this period. Ireland
reported its warmest June on record (124 years), as did Belgium (since 1833). In eastern Canada,
many loca�ons reported record warmth for July.

Precipita�on
Accumulated precipita�on totals for the first nine months in 2023 were above the long-term average
(Figure 18) in east Asia, parts of northern Asia, the western Indian summer monsoon region, parts of
the Mari�me Con�nent, northern New Zealand, northern inland Australia, parts of West, Central and
East Africa, Southeast Europe and the eastern Mediterranean region, southern Scandinavia,
northwest, southwest and southeast North America, Greater An�lles, and parts of southwest South
America.

47 United Nations Environment Programme (2022 a). Frontiers 2022: Noise, Blazes and Mismatches-Emerging Issues of Environmental
Concern. Nairobi. https://www.unep.org/resources/frontiers-2022-noise-blazes-and-mismatches
48 Fire weather is weather conducive to wildfires, including high temperatures, low humidity, and high winds.
 Figure 18: Total precipitation in 2023 (January to September), expressed as a quantile of the 1991–2020 reference period,
for areas that would have been in the driest 20% (brown) and wettest 20% (green) of years during the reference period, with
darker shades of brown and green indicating the driest and wettest 10%, respectively (Source: Global Precipitation
Climatology Centre (GPCC), Deutscher Wetterdienst, Germany). For more details see Data sets and methods.

Regions with a marked rainfall deficit included: southeast South America, the Amazon Basin, much of
Central America, southern Canada, the western Mediterranean region and Southwest Europe, parts
of west, central, east, and southern Africa, central Asia, the eastern Indian Monsoon region, parts of
southeast Asia and the Mari�me Con�nent, southwest, southeast and coastal north Australia, and
many of the Pacific Islands.

The onset of the West African Monsoon was around normal. The start of the Gu rain season (April to
June) in the Greater Horn of Africa brought unusually large rainfalls amounts in some areas.

Extreme weather and climate events

Key messages:

• Extreme weather con�nues to lead to severe socio-economic impacts.

• Extreme heat affected many parts of the world.
• Wildfires in Hawaii, Canada and Europe led to loss of life, the destruc�on of homes and large-
scale air pollu�on.
• Flooding associated with extreme rainfall from Mediterranean Cyclone Daniel affected Greece,
Bulgaria, Türkiye, and Libya with par�cularly heavy loss of life in Libya.

Extreme weather and climate events had major impacts on all inhabited con�nents in 2023. These
included major floods (some of them associated with tropical cyclones), extreme heat and drought,
and associated wildfires which presented challenges to water and food security as well as human
welfare 49,50. Some of the most significant events are described below, with a broader range of events
described in the online supplement.

In terms of loss of life, the most significant event was the Mediterranean Cyclone (or medicane,
referred to locally as ‘Storm Daniel’) in September. In its ini�al stages the storm produced extreme

49The United Nations (2022). The Sustainable Development Goals Report 2022. New York. https://unstats.un.org/sdgs/report/2022/
50The United Nations [UN] (2023). The Sustainable Development Goals Report 2023: Special edition. Towards a Rescue Plan for People and
Planet. New York. https://unstats.un.org/sdgs/report/2023/
rainfall in Greece, southern Bulgaria, and parts of Türkiye, while at the same �me another storm
system produced significant flash flooding in Spain with adverse effects on cereal produc�on 51. The
heaviest falls were in the Thessaly region of Greece north of Athens, where Zagora Pelion received
759.6 mm on 5 September and a 5-day total of 1096.2 mm from 4 to 8 September, while in Bulgaria,
329 mm fell in 16 hours at Kos� on 4–5 September. The storm then remained slow-moving in the
eastern Mediterranean for several days before its major rainbands impacted north-eastern Libya on
10 and 11 September. Extreme rainfalls affected the coast and nearby mountains, with 414.1 mm
falling in 24 hours at Al-Bayda on 10–11 September. The intense rainfalls resulted in extreme flooding
in the region. The most extreme impacts were in the city of Derna (about 50 km east of Al-Bayda),
where much of the central city was destroyed by flooding, exacerbated by the failure of two dams. At
least 4 345 confirmed deaths in Libya 52 have been atributed to the flooding, with 8 500 people s�ll
missing (as of 11 October). There were 19 addi�onal deaths in Greece 53 and Bulgaria. Es�mates for
Storm Daniel indicate that approximately an addi�onal 43 000 individuals were displaced in
northeastern Libya 54.

Tropical Cyclone Freddy in February and March was one of the world’s longest-lived tropical cyclones.
It formed on 6 February off the western coast of Australia and, a�er earlier landfalls in Madagascar
and Mozambique, made its final landfall in Mozambique on 11 March before moving inland as a
remnant low. The major impacts of Freddy came because of flooding during the final landfall, both in
Mozambique and Malawi, as extremely heavy rain fell (up to 672 mm in Mozambique). Parts of
Mozambique and Malawi had not yet recovered from storms in 2022. Malawi was especially hard hit
with at least 679 deaths reported and over 659 000 internal displacements 55, with a further 165
deaths in Mozambique. Casual�es were also reported in Madagascar and Zimbabwe.

Tropical Cyclone Mocha, in May, was one of the most intense cyclones ever observed in the Bay of
Bengal, reaching peak 10-minute sustained winds of 115 kt. It formed on 11 May and made landfall
near the Bangladesh-Myanmar border on 14 May. Cyclone Mocha triggered 1.7 million
displacements across the sub-region from Sri Lanka to Myanmar and through India and
Bangladesh 56. In Bangladesh, displacement was reported in Cox’s Bazar, the world’s largest refugee
setlement, which is home to over 900 000 Rohingya refugees from Myanmar 57. More than 29 000
people were temporarily relocated 58. In total at least 148 lives were lost in Myanmar and over
270 000 buildings damaged or destroyed 59. The effects of Cyclone Mocha, together with an
intensifica�on of conflict and record high food prices, have severely aggravated acute food insecurity,
especially among the 3.4 million vulnerable people now in need of humanitarian assistance 60.

The North Island of New Zealand suffered repeated extreme rainfall and flooding events in January
and February. The most significant was on 13–14 February, when Cyclone Gabrielle passed just east

51 https://www.mapa.gob.es/es/prensa/ultimas-noticias/el-146--del-territorio-est%C3%A1-en-emergencia-por-escasez-de-agua-y-el-274--
en-alerta/tcm:30–659894
52 https://reliefweb.int/report/libya/who-health-supplies-arrive-libya-part-intensified-response-devastating-floods

https://reliefweb.int/report/libya/libya-flood-response-situation-report-11-october-2023
53 https://reliefweb.int/report/greece/greece-severe-weather-and-floods-update-greek-civil-protection-hellenic-national-meteorological-

service-copernicus-emsr-echo-daily-flash-11-september-2023
54 Libya – Storm Daniel Flash Update 8 (13 October 2023) | Displacement Tracking Matrix (iom.int) https://dtm.iom.int/reports/libya-

storm-daniel-flash-update-8–13-october-2023 https://www.emro.who.int/media/news/101-health-workers-killed-in-libya-in-the-
aftermath-of-storm-daniel.html
55 https://www.preventionweb.net/media/87994/download?startDownload=true
56 2023 Mid-year update (internal-displacement.org) https://story.internal-displacement.org/2023-mid-year-update/
57 htps://sen�nel-asia.org/EO/2023/ar�cle20230514MM.html
58 htps://reliefweb.int/report/bangladesh/bangladesh-cyclone-mocha-humanitarian-response-situa�on-report-14-may-2023
59 https://ahacentre.org/wp-content/uploads/2023/05/AHA-DRAFT-Situation_Report-9-TC-MOCHA-Myanmar-1.pdf
60 UNHCR. May 2023. Flash Update #2 Cyclone Mocha. Available at: https://reporting.unhcr.org/myanmar-emergency-flash-update-2-

cyclone-mocha
of the North Island as a post-tropical system. Daily rainfalls exceeded 500 mm in parts of the eastern
North Island and Auckland (971.5 hPa) had its lowest air pressure on record. 11 deaths were
reported because of Gabrielle and 4 from more localized floods in Auckland on 27–28 January. Total
economic losses from the two events were es�mated at US$5.3 to 8.6 billion 61, by far the costliest
non-earthquake natural disaster recorded in New Zealand.

Many significant heatwaves occurred in various parts of the world during 2023. Some of the most
significant were in southern Europe and North Africa, especially in the second half of July where
severe and excep�onally persistent heat occurred. Italy was par�cularly affected, with temperatures
reaching 48.2 °C on 24 July at the Sardinian sites of Lotzorai and Jerzu, only 0.6 °C below the
European record set in Sicily in 2021. Among the loca�ons experiencing record high temperatures
were Tunis (Tunisia) (49.0 °C on 24 July), Tirana (Albania) (43.0 °C on 25 July), Agadir (Morocco) (50.4
°C on 11 August), and Algiers (49.2 °C on 23 July). The extreme heat shi�ed to southeast Europe in
late July, and there were further heatwaves affec�ng west-central Europe in late August and early
September. Numerous loca�ons in southern France, northern Spain and western Switzerland set
records during these events, including Toulouse (42.4 °C on 19 August). There was also extensive
wildfire ac�vity during the summer, par�cularly in Greece (both on the mainland and islands). A fire
in north-eastern Greece in late August and early September that burned 93 000 hectares was
reported to be the largest fire ever observed in the European Union.

Canada’s wildfire season was well beyond any previously recorded. Significant fire ac�vity began in
late April, expanded during a very warm, dry May, and con�nued throughout the summer and into
early autumn. The total area burned na�onally as of 15 October was 18.5 million hectares, more
than six �mes the 10-year average (2013–2022) and far above the previous record seasonal total of
7.1 million hectares in 1995 (Figure 19). The fires also resulted in significant and widespread smoke
pollu�on, par�cularly in the heavily populated areas of eastern Canada and the north-eastern United
States in the first half of June. Six deaths were directly atributed to fires, although the broader
health impacts of the smoke are yet to be fully assessed.

Annual area burned in Canada 1983-2023

20

18

16

14
Area burned (million ha)

12

10

0
1983 1988 1993 1998 2003 2008 2013 2018 2023
Year Data from CIFFC htps://www.ciffc.ca

Figure 19: Annual Area Burned in Canada 1983–2023 (millions of hectares). Data from: Canadian Interagency Forest Fire
Centre https://www.ciffc.ca/

61 https://www.treasury.govt.nz/sites/default/files/2023–04/impacts-from-the-north-island-weather-events.pdf
The deadliest single wildfire of the year occurred in Hawaii, on the western side of the island of
Maui. Extreme fire weather condi�ons, with low humidity and strong, gusty winds driven by a
pressure gradient between strong high pressure to the north and the circula�on of Hurricane Dora
well to the south, combined with pre-exis�ng drought to favour the development and rapid spread
of intense fires. The most badly affected region was around the town of Lahaina, which was largely
destroyed with over 2 200 structures lost. Mandatory evacua�on no�ces were issued for 7 500
people across the area 62. At least 99 deaths were reported 63, the most in a wildfire in the United
States for more than 100 years. Wildfires of such intensity and speed of movement are extremely
rare in the tropics.

Long-term drought persisted in north-western Africa and parts of the Iberian Peninsula, as well as
parts of central and southwest Asia, and intensified in many parts of Central America, northern South
America and the southern United States. Among the most significant areas of drought was an area of
subtropical South America, focused on northern Argen�na and Uruguay. Rainfall from January to
August was 20 to 50% below average over much of northern and central Argen�na, with some
regions experiencing their fourth successive year of significantly below average rainfall. There were
major crop losses in Argen�na with wheat produc�on in 2022–23 more than 30% below the five-year
average. In Uruguay, water storages reached cri�cally low levels, badly affec�ng the quality of
supplies to major centres including Montevideo, although there was some improvement in the
situa�on from August.

Long term drought in the Greater Horn of Africa was followed by heavy rain in some areas at the
start of the Gu rain season (April to June). Across the Horn of Africa, weather events had displaced
1.4 million people (as of June 2023) in addi�on to the 2.7 million people displaced by the five
consecu�ve seasons of drought 64. The drought reduced the capacity of the soil to absorb water,
which increased flood risk when the Gu rains arrived in April and May. It was es�mated that over
200 000 displacements were triggered by the resul�ng floods, mostly in the Belet Weyne district in
Somalia, where the Shabelle river burst its banks. 65 Pastoralist communi�es have con�nued to be
affected by asset losses a�er two consecu�ve years of drought. These will con�nue to adversely
affect agricultural produc�on and are expected to reduce cereal produc�on 2023 compared to
2022 66.

Socio-economic impacts
Key messages:

• Food security, popula�on displacements and impacts on vulnerable popula�ons con�nue to be

of concern in 2023, with weather and climate hazards exacerba�ng the situa�on in many parts
of the world.
• Extreme weather and climate condi�ons con�nued to trigger new, prolonged, and secondary
displacement in 2023 and increased the vulnerability of many who were already uprooted by
complex mul�-causal situa�ons of conflict and violence.
• Development and implementa�on of local disaster risk reduc�on strategies have increased
since the adop�on of the Sendai Framework for Disaster Risk Reduc�on.

62 https://www.disastercenter.com/FEMA%20Daily%20Ops%20Briefing%20085–2023.pdf
63 NOAA Billion-dollar disasters https://www.ncei.noaa.gov/access/billions/events/US/2023?disasters[]=wildfire
64 East and Horn of Africa: Flood Snapshot (January - June 2023) | Displacement Tracking Matrix (iom.int)

https://dtm.iom.int/reports/east-and-horn-africa-flood-snapshot-january-june-2023
65 htps://www.internal-displacement.org/publica�ons/impacts-of-displacement-flood-displacement-in-beledweyne-somalia;

htps://reliefweb.int/report/somalia/somalia-2023-flash-and-riverine-floods-situa�on-report-no-1–14-may-2023
66 https://www.fao.org/3/cc6806en/cc6806en.pdf
• One of the essen�al components for reducing the impact of disasters is to have effec�ve mul�-
hazard early warning systems.

The events described above, and many others besides, occur in a broader context. Extreme weather
and climate events interact with and in some cases trigger or exacerbate situa�ons concerning water
and food security, popula�on mobility and environmental degrada�on 67,68.

The number of people who are acutely food insecure has more than doubled, from 135 million people
before the COVID-19 pandemic to 345 million people in 2023 (in 53 monitored countries) 69. Global
hunger levels remained unchanged from 2021 to 2022. However, these are s�ll far above pre-COVID 19
pandemic levels: in 2022, 9.2% of the global popula�on (735.1 million people) were undernourished,
compared to 7.9% of the popula�on (612.8 million people) in 2019 (Figure 20) 70. The current global
food and nutri�on crisis is the largest in modern human history 71. Protracted conflicts, economic
downturns, and high food prices, further exacerbated by high costs of agricultural inputs driven by
ongoing conflict, are at the root of high global food insecurity levels. This is aggravated by the effects
of climate and weather extremes. In southern Africa, for example, weather extremes, including the
passage of Cyclone Freddy in February 2023, have hit the region, affec�ng areas of Madagascar,
Mozambique, southern Malawi, and Zimbabwe. Flooding associated with the cyclone submerged
extensive agricultural areas and inflicted severe damage on crops. This has exacerbated a slow
economic recovery 72.

Globally, annual economic losses from climate and weather-related disasters have significantly
increased since the 2000s 73. In low and lower-middle-income countries, 82% of all damage and loss
caused by drought concerned the agriculture sector. Between 2008 and 2018, across least developed
countries and lower middle-income countries, 34% of disaster-related crop and livestock produc�on
losses were atributed to drought, followed by 19% to flooding events, 18% to severe storms and
hurricanes, 9% to crop pests and animal diseases, 6% to extreme temperatures, and 1% to wildfires 74.

67 The United Nations (2022). The Sustainable Development Goals Report 2022. New York. https://unstats.un.org/sdgs/report/2022/
68 The United Nations [UN] (2023). The Sustainable Development Goals Report 2023: Special edition. Towards a Rescue Plan for People and
Planet. New York. https://unstats.un.org/sdgs/report/2023/
69 WFP. June 2023. WFP Global Operational Response Plan 2023. Update #8. Available at: https://docs.wfp.org/api/documents/WFP-

0000150404/download/ FAO, IFAD, UNICEF, WFP and WHO. 2023. The State of Food Security and Nutrition in the World 2023.
70 Urbanization, agrifood systems transformation and healthy diets across the rural–urban continuum. Rome, FAO.

https://doi.org/10.4060/cc3017en
71 WFP. June 2023. WFP Global Operational Response Plan 2023. Update #8. Available at: docs.wfp.org/api/documents/WFP-

0000150404/download/
72 FAO. 2023. Crop Prospects and Food Situation – Quarterly Global Report No. 2, July 2023. Rome. https://doi.org/10.4060/cc6806en
73 FAO. 2021. The impact of disasters and crises on agriculture and food security: 2021. Rome. https://doi.org/10.4060/cb3673en
74 FAO. 2021. The impact of disasters and crises on agriculture and food security: 2021. Rome. https://doi.org/10.4060/cb3673en
 Figure 20: Global prevalence of undernourishment (as a%) and number of undernourished (in millions) since 2005. The
entire series was updated to reflect new information released since the publication of The State of Food Security and
Nutrition in the World 2023 75.

Across the globe, millions of people, including internally displaced persons, refugees, and migrants,
are on the move or have been forced to flee their homes and communi�es because of disasters
exacerbated by climate stresses and shocks. Weather hazards con�nued to trigger new, prolonged,
and secondary displacement in 2023 and increased the vulnerability of many who were already
uprooted by complex mul�-causal situa�ons of conflict and violence. Many migrants entering into
Somalia found themselves stranded in June and July 2023, mostly in the city of Bossaso, the main
coastal crossing loca�on to Yemen, as they were wai�ng for more favourable weather condi�ons to
cross the Gulf of Aden. In Hargeisa, a group of migrants in transi�on suffered from extreme heat, and
some died from dehydra�on 76. This is a clear indica�on of how vulnerability to climate shocks and
stresses is undermining resilience and crea�ng new protec�on risks, which threaten the achievement
of the SDGs.

In addi�on to new displacements caused by high-impact disasters in 2023, many are s�ll enduring the
prolonged effects of climate-related displacement that took place in previous years. In Pakistan, the
2022 monsoon floods, which triggered the largest disaster displacement event in a decade, con�nued
to have longstanding impacts in 2023. Displaced communi�es were s�ll recovering when heavy rains
hit some districts in June 2023, causing waterborne and vector-borne diseases such as malaria,
dengue, and cholera 77. Displacement in the context of climate change and environmental degrada�on
is o�en mul�causal. Most people move due to a combina�on of social, poli�cal, economic,
environmental, and demographic drivers, all of which are and will be affected by climate and
environmental change.

Climate related shocks and stresses in migra�on and displacement contexts affect people's
livelihoods which entrenches poverty (SDG 1) and hunger (SDG 2), pose direct threats to their lives
and well-being (SDG 3), widen inequality gaps (SDG 10), limit access to quality educa�on (SDG 4),
water and sanita�on (SDG 6) as well as clean energy (SDG 7). Women and girls are among the worst

75 FAO, IFAD, UNICEF, WFP and WHO. 2023. The State of Food Security and Nutrition in the World 2023. Urbanization, agrifood systems
transformation and healthy diets across the rural–urban continuum. Rome, FAO. https://doi.org/10.4060/cc3017en
76 https://dtm.iom.int/reports/migration-along-eastern-corridor-july-2023
77 https://reliefweb.int/report/pakistan/early-needs-identification-report-monsoon-flood-affected-areas-pakistan-august-2023
affected (SDG 5) due to pre-exis�ng gender and socio-economic inequali�es compounding their
vulnerabili�es 78.

To beter prepare for these contexts, governments, communi�es, civil society, and the UN are
engaged at all levels to strengthen climate resilience and effec�ve disaster risk reduc�on. A main
priority is the community level, where preparedness efforts including early warning systems and
emergency preparedness are picking up pace 79. With 126 countries in possession of disaster risk
management strategies and a total of 99 countries repor�ng to have local governments with disaster
risk reduc�on strategies, development and implementa�on of local disaster risk reduc�on strategies
have increased since the adop�on of the Sendai Framework for Disaster Risk Reduc�on in 2015 80.

One of the essen�al components for reducing the impact of disasters is to have effec�ve mul�-
hazard early warning systems, and to have effec�ve systems in place for genera�ng disaster risk
informa�on, crea�ng early warnings, and dissemina�ng them, as well as having plans to act on the
warnings. As of 2023, 102 countries have reported having mul�-hazard early warning systems in
place, more than half of the countries in the world. The Early Warnings for All ini�a�ve was launched
by the UN Secretary-General in March 2022 with the aim of ensuring that everyone on Earth is
protected from hazardous weather, water, or climate events through life-saving early warning
systems by the end of 2027.

78 UNHCR Gender, Displacement, and Climate Change. July 2020. https://www.unhcr.org/protection/environment/5f21565b4/gender-

displacement-and-climate-change.html
79 United Nations Environment Programme (2022b). Climate Change and Security Partnership Project Final Report: March 2017-Feburary

2022. Nairobi. https://wedocs.unep.org/handle/20.500.11822/40549

80 https://sendaiframework-mtr.undrr.org/publication/report-midterm-review-implementation-sendai-framework-disaster-risk-reduction-

2015–2030
Data sets and methods
Greenhouse gases
Es�mated concentra�ons from 1750 are used to represent pre-industrial condi�ons. Calcula�ons
assume a pre-industrial mole frac�on of 278.3 ppm for CO2, 729.2 ppb for CH4 and 270.1 ppb for
N2O.

World Data Centre for Greenhouse Gases operated by Japan Meteorological Agency
htps://gaw.kishou.go.jp/.htps://gaw.kishou.go.jp/.

[World Meteorological Organiza�on (WMO). WMO Greenhouse Gas Bulle�n – No. 19:
The State of Greenhouse Gases in the Atmosphere Based on Global Observa�ons through 2022.
Geneva, 2023]

Global temperatures
Global mean temperature series
The method for calcula�ng global mean temperature anomalies rela�ve to an 1850–1900 baseline is
based on the assessment of long-term change and its uncertainty made by Working Group I in its
contribu�on to the IPCC Sixth Assessment Report (IPCC AR6 WG I). In 2021, the IPCC AR6 WG I
assessed change from 1850–1900 to other periods based on an average of four data sets –
HadCRUT5, Berkeley Earth, NOAA Interim and Kadow et al. (2020) – which start in 1850 and are
globally or near-globally complete in the modern period.

To include shorter data sets which can help to understand recent temperature changes, in the
present report the es�mate made by the IPCC for the temperature change between 1850–1900 and
1981–2010 is combined with es�mated changes between 1981–2010 and the current year from five
data sets to calculate anomalies for 2023 rela�ve to 1850–1900. There is good agreement between
the data sets on changes from 1981–2010 to the present, as this is a period with good observa�onal
coverage.

The addi�onal uncertainty from the spread of the data sets is combined with that of the IPCC’s
es�mate of the uncertainty in the change from 1850–1900 to 1981–2010. Global mean temperature
anomalies were calculated rela�ve to an 1850–1900 baseline using the following steps star�ng from
�me series of global monthly mean temperatures for each data set:

1. For each data set, anomalies were calculated rela�ve to the 1981–2010 average by
subtrac�ng the mean for the period 1981–2010 for each month separately.
2. An annual mean anomaly was calculated from the monthly mean anomalies.
3. The amount of 0.69 °C was added to each series, based on the es�mated difference
between 1850–1900 and 1981–2010, calculated using the method from IPCC AR6 WG I
(see cap�on for Figure 1.12 in that report).
4. The mean and standard devia�on of the es�mates were calculated.
5. The uncertainty in the IPCC es�mate was combined with the standard devia�on,
assuming the two are independent and assuming the IPCC uncertainty range (0.54 °C to
0.79 °C) is representa�ve of a 90% confidence range (1.645 standard devia�ons).

The number quoted in this report for 2023 (1.40 ± 0.12 °C) was calculated in this way with 1.40 °C
being the mean of the five es�mates and 0.03 °C the standard devia�on.
Annual temperature maps
For the map of temperature anomalies for 2023, a median of five data sets was used, regridded to
the spa�al grid of the lowest resolu�on data sets (NOAAGlobalTemp, and HadCRUT5), which are
presented on a 5° la�tude by 5° longitude grid. The median is used in preference to the mean to
minimize the effect of poten�al outliers in individual grid cells. The half-range of the data sets
provides an indica�on of the uncertainty. The spread between the data sets is largest at high
la�tudes and in Central Africa, both regions with sparse data coverage.

Global mean temperature anomalies for 2023 relative to other periods

This table shows global mean temperature anomalies for individual data sets for 2023 (data to
October) rela�ve to four different baselines. The uncertainty indicated for the three modern
baselines (1981–2010, 1991–2020 and 1961–1990) are the standard devia�ons of the available
es�mates mul�plied by 1.645 to represent the 90% uncertainty range.

Period 1850–1900 1981–2010 1991–2020 1961–1990

HadCRUT5 1.40 0.71 0.52 1.05
NOAA GlobalTemp 1.37 0.68 0.50 0.99
GISTEMP 1.39 0.70 0.51 1.02
ERA5 1.43 0.74 0.55 1.07
JRA55 1.39 0.70 0.52 1.02
Mean of five 1.40±0.12 0.71±0.03 0.52±0.03 1.03±0.04
datasets

The following five data sets were used, including three tradi�onal datasets:

HadCRUT.5.0.2.0: Morice, C. P.; Kennedy, J. J.; Rayner, N. A. et al. An Updated Assessment of

NearSurface Temperature Change From 1850: The HadCRUT5 Data Set. Journal of Geophysical
Research: Atmospheres 2021, 126, e2019JD032361. htps://doi.org/10.1029/2019JD032361.
HadCRUT.5.0.2.0 data were obtained from htp://www.metoffice.gov.uk/hadobs/hadcrut5 on
24 November 2023 and are © Bri�sh Crown Copyright, Met Office 2023, provided under an Open
Government Licence, htp://www.na�onalarchives.gov.uk/doc/open-government-licence/version/3/.

NOAAGlobalTemp v5.1: NOAA Interim: Vose, R. S.; Huang, B.; Yin, X. et al. Implemen�ng Full Spa�al
Coverage in NOAA’s Global Temperature Analysis. Geophysical Research Leters 2021, 48,
e2020GL090873. htps://doi.org/10.1029/2020GL090873.

GISTEMP v4: GISTEMP Team, 2022: GISS Surface Temperature Analysis (GISTEMP), version 4. NASA
Goddard Ins�tute for Space Studies, htps://data.giss.nasa.gov/gistemp/.

Lenssen, N.; Schmidt, G.; Hansen, J. et al. Improvements in the GISTEMP Uncertainty Model. Journal
of Geophysical Research: Atmospheres 2019, 124, 6307–6326.
htps://doi.org/10.1029/2018JD029522.

And two reanalyses (note that we do not use JRA-3Q because there is not yet a peer-reviewed paper
suppor�ng this dataset; the differences between the two are small):

JRA-55: Kobayashi, S.; Ota, Y.; Harada, Y. et al. The JRA-55 Reanalysis: General Specifica�ons and Basic
Characteris�cs. Journal of the Meteorological Society of Japan. Ser. II 2015, 93, 5–48.
htps://doi.org/10.2151/jmsj.2015–001.
ERA5: Hersbach, H.; Bell, B.; Berrisford, P. et al. ERA5 Monthly Averaged Data on Single Levels from
1940 to Present; Copernicus Climate Change Service (C3S) Climate Data Store (CDS), 2023.
htps://doi.org/10.24381/cds.f17050d7.

Berkeley Earth and Kadow et al. were also used to assess monthly temperatures to September:

Berkeley Earth: Rohde, R. A.; Hausfather, Z. The Berkeley Earth Land/Ocean Temperature Record.
Earth System Science Data 2020, 12, 3469–3479. htps://doi.org/10.5194/essd-12–3469–2020.

Kadow, C., Hall, D.M. & Ulbrich, U. Ar�ficial intelligence reconstructs missing climate informa�on.
Nat. Geosci. 13, 408–413 (2020). htps://doi.org/10.1038/s41561-020-0582-5

Land temperatures and sea-surface temperatures

The land temperature assessment is based on three data sets:

CRUTEM.5.0.1.0: Osborn, T. J., P. D. Jones, D. H. Lister, C. P. Morice, I. R. Simpson and I. C. Harris

(2020), Land surface air temperature varia�ons across the globe updated to 2019: the CRUTEM5
dataset, J. Geophys. Res. htps://doi.org/10.1029/2019JD032352

CRUTEM.5.0.1.0 data were obtained from htp://www.metoffice.gov.uk/hadobs/crutem5 on

31 October 2023 and are © Bri�sh Crown Copyright, Met Office 2023, provided under an Open
Government License, htp://www.na�onalarchives.gov.uk/doc/open-government-licence/version/3/

GHCNv4: Menne, Mathew J.; Gleason, Byron E.; Lawrimore, Jay; Rennie, Jared; and Williams, Claude
N. (2017): Global Historical Climatology Network – Monthly Temperature [Global mean]. NOAA
Na�onal Centers for Environmental Informa�on. doi:10.7289/V5XW4GTH [accessed
31 October 2023].

Berkeley Earth: Rohde, R. A.; Hausfather, Z. The Berkeley Earth Land/Ocean Temperature Record.
Earth System Science Data 2020, 12, 3469–3479. htps://doi.org/10.5194/essd-12–3469–2020

The sea-surface temperature assessment is based on two data sets:

HadSST.4.0.1.0: Kennedy, J. J., Rayner, N. A., Atkinson, C. P., & Killick, R. E. (2019). An ensemble data
set of sea-surface temperature change from 1850: the Met Office Hadley Centre HadSST.4.0.0.0 data
set. Journal of Geophysical Research: Atmospheres, 124. htps://doi.org/10.1029/2018JD029867

HadSST.4.0.1.0 data were obtained from htp://www.metoffice.gov.uk/hadobs/hadsst4 on

31 October 2023 and are © Bri�sh Crown Copyright, Met Office 2023, provided under an Open
Government License, htp://www.na�onalarchives.gov.uk/doc/open-government-licence/version/3/

ERSSTv5: Boyin Huang, Peter W. Thorne, Viva F. Banzon, Tim Boyer, Gennady Chepurin, Jay H.
Lawrimore, Mathew J. Menne, Thomas M. Smith, Russell S. Vose, and Huai-Min Zhang (2017): NOAA
Extended Reconstructed Sea-Surface Temperature (ERSST), Version 5. [Global mean]. NOAA Na�onal
Centers for Environmental Informa�on. doi:10.7289/V5T72FNM [accessed 31 October 2023].

Ocean heat content

The ensemble mean is an update of the outcome of a concerted interna�onal effort 81, and all
products used are referenced here. Note that global ocean heat content values are given for the
ocean surface area between 60°S-60°N and limited to areas deeper than 300 m in each product. The

81von Schuckmann, K., Cheng, L., Palmer, M. D., Hansen, J., Tassone, C., Aich, V., Adusumilli, S., Beltrami, H., Boyer, T., Cuesta-Valero, F. J.,
Desbruyères, D., Domingues, C., García-García, A., Gentine, P., Gilson, J., Gorfer, M., Haimberger, L., Ishii, M., Johnson, G. C., … Wijffels, S.
E. (2020). Heat stored in the Earth system: where does the energy go? Earth Syst. Sci. Data, 12(3), 2013–2041.
https://doi.org/10.5194/essd-12–2013–2020
value for 2022 is based on a subset of the products for which updates are available. A baseline of
2005–2021 is used for the Ocean Heat Content �me series (Figure 4) as near-global coverage is
available in this period thanks to the network of Argo sub-surface floats.

Data used up to 2022:

Cheng, L.; Trenberth, K. E.; Fasullo, J. et al. Improved es�mates of ocean heat content from 1960 to
2015, Science Advances 2017, 3 (3), e1601545. htps://doi.org/10.1126/sciadv.1601545

Gaillard, F.; Reynaud, T.; Thierry, V. et al. In Situ–Based Reanalysis of the Global Ocean Temperature
and Salinity with ISAS: Variability of the Heat Content and Steric Height, Journal of Climate 2016, 29
(4), 1305–1323. htps://doi.org/10.1175/JCLI-D-15–0028.1

Ishii, M.; Fukuda, Y.; Hirahara, S. et al. Accuracy of Global Upper Ocean Heat Content Es�ma�on
Expected from Present Observa�onal Data Sets. SOLA 2017, 13, 163–167.
htps://doi.org/10.2151/sola.2017–030

Kuusela, M., & Giglio, D. (2023) Global Ocean Heat Content Anomalies based on Argo data (2.0.0)
[Data set]. Zenodo. htps://doi.org/10.5281/zenodo.7562281

Levitus, S.; Antonov, J. I.; Boyer, T. P. et al. World Ocean heat content and thermosteric sea level
change (0–2 000 m) 1955–2010. Geophysical Research Leters 2012, 39 (10), L10603.
htps://doi.org/10.1029/2012GL051106

Lyman, J. M.; Johnson, G. C. Es�ma�ng Global Ocean Heat Content Changes in the Upper 1800 m
since 1950 and the Influence of Climatology Choice. Journal of Climate, 2014, 27 (5), 1945–1957.
htps://doi.org/10.1175/JCLI-D-12–00752.1

von Schuckmann, K.; Le Traon, P.-Y. How well can we derive Global Ocean Indicators from Argo data?
Ocean Science 2011, 7 (6), 783–791. htps://doi.org/10.5194/os-7–783–2011. Data available at:
htps://marine.copernicus.eu/access-data/ocean-monitoring-indicators

In addi�on, data used up to 2021:

Desbruyères, D. G. ; Purkey, S. G. ; McDonagh, E. L. et al. Deep and abyssal ocean warming from 35
years of repeat hydrography, Geophysical Research Leters 2016, 43 (19), 310–356.
htps://doi.org/10.1002/2016GL070413

Desbruyères, D., McDonagh, E. L., King, B. A., & Thierry, V. (2017). Global and Full-Depth Ocean
Temperature Trends during the Early Twenty-First Century from Argo and Repeat Hydrography.
Journal of Climate, 30(6), 1985–1997. htps://doi.org/10.1175/JCLI-D-16–0396.1

Good, S. A., Mar�n, M. J., & Rayner, N. A. (2013). EN4: Quality controlled ocean temperature and
salinity profiles and monthly objec�ve analyses with uncertainty es�mates (I5197, trans.). Journal of
Geophysical Research: Oceans, 118(12), 6704–6716. htps://doi.org/10.1002/2013JC009067

Hosoda, S.; Ohira, T.; Nakamura, T. A monthly mean dataset of global oceanic temperature and
salinity derived from Argo float observa�ons. JAMSTEC Report of Research and Development, 2008,
8, 47–59. htps://www.jstage.jst.go.jp/ar�cle/jamstecr/8/0/8_0_47/_ar�cle

Kuusela M.; Stein, M. L. Locally sta�onary spa�o-temporal interpola�on of Argo profiling float data.
Proceedings of the Royal Society A 2018, 474, 20180400. htp://dx.doi.org/10.1098/rspa.2018.0400
Li, H.; Xu, F.; Zhou, W. et al. Development of a global gridded Argo data set with Barnes successive
correc�ons, Journal of Geophysical Research: Oceans 2017, 122 (2), 866–889,
htps://doi.org/10.1002/2016JC012285

Roemmich, D.; Gilson, J. The 2004–2008 mean and annual cycle of temperature, salinity, and steric
height in the global ocean from the Argo Program, Progress in Oceanography, 2009, 82 (2), 81–100.
htps://doi.org/10.1016/j.pocean.2009.03.004

Roemmich, D., Church, J., Gilson, J., Monselesan, D., Suton, P., & Wijffels, S. (2015). Unabated
planetary warming and its ocean structure since 2006 (I3631, trans.). Nature Climate Change, 5, 240.
htps://doi.org/10.1038/nclimate2513

In addi�on, data used up to 2020:

Church, J. A., White, N. J., Konikow, L. F., Domingues, C. M., Cogley, J. G., Rignot, E., Gregory, J. M.,
van den Broeke, M. R., Monaghan, A. J., & Velicogna, I. (2011). Revisi�ng the Earth’s sea-level and
energy budgets from 1961 to 2008. Geophysical Research Leters, 38(18).
htps://doi.org/10.1029/2011GL048794

Domingues, C. M., Church, J. A., White, N. J., Gleckler, P. J., Wijffels, S. E., Barker, P. M., & Dunn, J. R.
(2008). Improved es�mates of upper-ocean warming and mul�-decadal sea-level rise. Nature,
453(7198), 1090–1093. htps://doi.org/10.1038/nature07080

Li, Y., Church, J. A., McDougall, T. J., & Barker, P. M. (2022). Sensi�vity of Observa�onally Based
Es�mates of Ocean Heat Content and Thermal Expansion to Ver�cal Interpola�on Schemes.
Geophysical Research Leters, 49, e2022G. htps://doi.org/htps://doi.org/10.1029/2022GL101079

Wijffels, S., Roemmich, D., Monselesan, D., Church, J., & Gilson, J. (2016). Ocean temperatures
chronicle the ongoing warming of Earth. Nature Climate Change, 6(2), 116–118.
htps://doi.org/10.1038/nclimate2924

Sea level
GMSL from CNES/Aviso+ htps://www.aviso.al�metry.fr/en/data/products/ocean-indicators-
products/mean-sea-level/data-acces.html#c12195 Copernicus Climate Change Service (C3S), 2018:

Sea Level Daily Gridded Data from Satellite Observa�ons for the Global Ocean from 1993 to Present.
C3S CDS, htps://doi.org/10.24381/cds.4c328c78

Marine heatwaves and cold-spells

MHWs are categorized as moderate when the sea-surface temperature (SST) is above the nine�eth
percen�le of the climatological distribu�on for five days or longer; the subsequent categories are
defined with respect to the difference between the SST and the climatological distribu�on average:
strong, severe, or extreme, if that difference is, respec�vely, more than two, three or four �mes the
difference between the nine�eth percen�le and the climatological distribu�on average (Hobday et
al., 2018). MCS categories are analogous but coun�ng days below the tenth percen�le, except for the
‘ice’ category. This category is given to any MCS when the threshold for the occurrence on any given
day of the event is below -1.7°C (Schlegel et la., 2021). These are therefore considered to be
condi�ons related to sea-ice, and not extreme temperature fluctua�ons.

The baseline used for MHWs and MCSs is 1982–2011, which is shi�ed by one year from the
standard normal period of 1981–2010 because the first full year of the satellite SST series on which
it is based is 1982. This period has not been updated to the current standard normal period of 1991–
2020 because the shi�ing of the baseline has a significant effect on the results and would not allow
for comparison of MHW/MCS sta�s�cs with previous versions of this report.

All MHWs and MCSs are detected using the NOAA daily Op�mum Interpola�on Sea-Surface
Temperature (OISST) v2.1 dataset (Huang et al. 2021).

Hobday, A. J., Alexander, L. V., Perkins, S. E., Smale, D. A., Straub, S. C., Oliver, E. C.,... & Wernberg, T.
(2016). A hierarchical approach to defining marine heatwaves. Progress in Oceanography, 141, 227–
238. htps://doi.org/10.1016/j.pocean.2015.12.014

Hobday, A. J., Oliver, E. C., Gupta, A. S., Benthuysen, J. A., Burrows, M. T., Donat, M. G.,... & Smale, D.
A. (2018). Categorizing and naming marine heatwaves. Oceanography, 31(2), 162–173.
htps://www.jstor.org/stable/26542662

Huang, B., Liu, C., Banzon, V., Freeman, E., Graham, G., Hankins, B.,... & Zhang, H. M. (2021).
Improvements of the daily op�mum interpola�on sea-surface temperature (DOISST) version 2.1.
Journal of Climate, 34(8), 2923–2939. htps://doi.org/10.1175/JCLI-D-20–0166.1

Schlegel, R. W., Darmaraki, S., Benthuysen, J. A., Filbee-Dexter, K., & Oliver, E. C. (2021). Marine cold-
spells. Progress in Oceanography, 198, 102684.

Ice sheets
Mankoff, K., X. Fetweis, A. Solgaard, P. Langen, M. Stendel, B. Noël, M. van den Broeke, N. Karlsson,
J.E. Box, and K. Kjeldsen (2021) Greenland ice sheet mass balance from 1840 through next week
htps://doi.org/10.22008/FK2/OHI23Z, GEUS Dataverse, V704

Mankoff, K. D., Fetweis, X., Langen, P. L., Stendel, M., Kjeldsen, K. K., Karlsson, N. B., Noël, B., van
den Broeke, M. R., Solgaard, A., Colgan, W., Box, J. E., Simonsen, S. B., King, M. D., Ahlstrøm, A. P.,
Andersen, S. B., and Fausto, R. S.: Greenland ice sheet mass balance from 1840 through next week,
Earth Syst. Sci. Data, 13, 5001–5025, htps://doi.org/10.5194/essd-13–5001–2021, 2021.

Sea-ice
Data are from the EUMETSAT OSI SAF Sea-Ice Index v2.2 (OSI-SAF, based on Lavergne et al., 2019 –
htps://osisaf-hl.met.no/v2p2-sea-ice-index ) and the Na�onal Snow and Ice Data Center (NSIDC) v3
Sea-Ice Index (Feterer et al., 2017). Sea-ice concentra�ons are es�mated from microwave radiances
measured from satellites. Extent is the area of ocean grid cells where the sea-ice concentra�on
exceeds 15%. There are modest differences in the absolute extent between data sets, but they agree
well on the year-to-year changes and trends. In the main text of the report, NSIDC values are
reported for absolute extents, and rankings. Comparison figures for OSI SAF are given in the table.

Metric NSIDC OSI SAF

Arc�c daily minimum 4.23 million km2, 4.71 million km2,
19 September. sixth lowest on 16 September. sixth lowest on
record record
Arc�c daily maximum 14.62 million km2, 6 March, 14.64 million km2, 3 March
fi�h lowest on record
Antarc�c daily minimum 1.79 million km2, 21 February, 2.0 million km2, 16 February,
lowest on record lowest on record
Antarc�c daily maximum 16.96 million km2, 17.6 million km2,
10 September, lowest on 12 September, lowest on
record record
OSI SAF Sea-ice index 1978-onwards, version 2.2 (2023), OSI-420. EUMETSAT Ocean and Sea-Ice
Satellite Applica�on Facility. Data extracted from Sea-ice index | OSI SAF (eumetsat.int)

Feterer, F.; Knowles, K.; Meier, W. N. et al., 2017, updated daily. Sea-Ice Index, Version 3. Boulder,
Colorado, USA. Na�onal Snow and Ice Data Center (NSIDC),
htps://nsidc.org/data/G02135/versions/3.

Lavergne, T. Sørensen, A. M.; Kern, S. et al. Version 2 of the EUMETSAT OSI SAF and ESA CCI Sea-Ice
Concentra�on Climate Data Records. The Cryosphere 2019, 13 (1), 49–78.
htps://doi.org/10.5194/tc-139–2019.

Precipita�on
The following Global Precipita�on Climatology Centre (GPCC) data sets were used in the analysis:

First Guess Monthly, htps://doi.org/10.5676/DWD_GPCC/FG_M_100

Monitoring Product (Version 2022), htps://doi.org/10.5676/DWD_GPCC/MP_M_V2022_100

Full Data Monthly (Version 2022), htps://doi.org/10.5676/DWD_GPCC/FD_M_V2022_100

Precipita�on Climatology (Version 2022), htps://doi.org/10.5676/DWD_GPCC/CLIM_M_V2022_100

In Figure 18, Iceland is shown as being much drier than the long-term average and parts of China
much weter. These features are due to a change in the way that real �me data are processed and
are not reflec�ve of actual condi�ons.

Contributors
Members
Algeria, Andorra, Argen�na, Armenia, Australia, Bahrain, Barbados, Belgium, Belize, Bosnia and
Herzegovina, Brazil, Brunei, Bulgaria, Canada, Chile, China, Côte d'Ivoire, Croa�a, Cyprus, Czech
Republic, Denmark, Ecuador, Estonia, Finland, France, Georgia, Germany, Hungary, Iceland, India, Iran
(Islamic Republic of), Ireland, Israel, Italy, Japan, Jordan, Kazakhstan, Latvia, Libya, Luxembourg,
Malaysia, Mauri�us, Mexico, Morocco, Myanmar, Netherlands, New Zealand, Norway, Paraguay,
Peru, Poland, Republic of Korea, Republic of Moldova, Russian Federa�on, Saudi Arabia, Singapore,
Slovakia, Slovenia, South Africa, Sweden, Switzerland, Thailand, Türkiye, Ukraine, United Arab
Emirates, United Kingdom of Great Britain and Northern Ireland, United Republic of Tanzania, United
States of America, Uruguay, Venezuela (Bolivarian Republic of), Viet Nam.

Individual contributors
Vicente Anzellini (IDMC), Chris Atkinson (Met Office), Omar Baddour (WMO), Paul M Barker
(University of New South Wales), Jorge Alvar-Beltrán (FAO), Jana Birner (UNHCR), Nicholas Bishop
(IOM), Jessica Blunden (NOAA), Tim Boyer (NOAA NCEI), Anny Cazanave (Legos), Xuan Che (UNDRR),
Lijing Cheng (Ins�tute of Atmospheric Physics (IAP), Center for Ocean Mega-Science), John Church
(University of New South Wales), Damien Desbruyeres (Ifremer), Ca�a Domingues (NOC), Robert
Dunn (Met Office), Arianna Gialle� (FAO), Pini Giancarlo (WFP), Donata Giglio (University of
Colorado), John E Gilson (SCRIPPS), Alashiya Gordes (FAO), Atsushi Goto (WMO), Flora Gues (CELAD,
Mercator Ocean Interna�onal), Peer Hechler (WMO), Christopher Hewit (WMO), Shigeki Hosoda
(JAMSTEC), Mathias Huss (ETH Zürich), Amanda Igna�a (UNHCR), Kirsten Isensee (IOC/UNESCO),
Gregory C Johnson (NOAA PMEL), Christopher Kadow (DKRZ), Hideki Kanamaru (FAO), Maarten
Kappelle (UNEP), John Kennedy (WMO Expert), Rachel Killick (Met Office), Brian King (NOC), Nicolas
Kolodziejczyk (University of Brest), Animesh Kumar (UNDRR), Mikael Kuusela (Carnegie Mellon
University), Gernot Laganda (WFP), Lancelot Leclercq (Legos), Yuehua Li (Yunnan University), Ricardo
Locarnini (NOAA NCEI), John Lyman (NOAA PMEL), Shawn Marshall (Environment and Climate
Change Canada (ECCC) and University of Calgary), Jesse Mason (WFP), Juta May (UNDRR), Trevor
McDougall (University of New South Wales), Brian Menounos (University of Northern Bri�sh
Columbia), Audrey Minère (Mercator Ocean International), Didier Paolo Monselesan (CSIRO), Sarah
Purkey (Scripps), James Reagan (NOAA NCEI, University of Maryland), Dean Roemmich (Scripps), Lev
Nere�n (FAO), Julien Nicolas (ECMWF), Elisabeth Du Parc (IOM), Sylvain Ponserre (IDMC), Ileana
Sinziana Puscas (IOM), Claire Ransom (WMO), David Robinson (Rutgers State University of New
Jersey), Bonifacio Rogerio (WFP), Kanako Sato (JAMSTEC), Katsunari Sato (JMA), Abhishek Savita
(GEOMAR), Robert W Schlegel (Sorbonne Université, CNRS, Laboratoire d’Océanographie de
Villefranche), Katherina Schoo (IOC/UNESCO), Serhat Sensoy (Turkish State Meteorological Service),
Fumi Sezaki (JMA), Jose Álvaro Silva (WMO), Mike Sparrow (WMO/WCRP), Johan Stander (WMO),
Toshio Suga (Tohoku University, JAMSTEC), Oksana Tarasova (WMO), Caterina Tassone (WMO/GCOS),
Blair Trewin (Bureau of Meteorology), Karina von Schuckmann (Mercator Ocean Interna�onal), Ying
Wang (UNEP), Susan E Wjiffels (CSIRO, Woods Hole), Abdel-Lathif Younous (WFP), Markus Ziese
(DWD, GPCC)

Ins�tu�ons
Bureau of Meteorology, Australia; Carnegie Mellon University, USA; Center for Ocean Mega-Science,
Chinese Academy of Sciences, China; CELAD, France; Commonwealth Scientific and Industrial
Research Organization, CSIRO Oceans and Atmosphere, Tasmania, Australia; Deutsches
Klimarechenzentrum, DKRZ, Germany; Deutshce Weterdienst, DWD, Germany; European Centre for
Medium-range Weather Forecasts, ECMWF, UK; Environment and Climate Change Canada, ECCC,
Canada; ETH Zürich; GEOMAR Helmholtz Centre for Ocean Research Kiel, Germany; Global
Precipita�on Climatology Centre, GPCC, Germany; Hong Kong Observatory; Ifremer, France; IAP,
Chinese Academy of Sciences, China; International Displacement Monitoring Centre (IDMC); Japan
Marine Earth Scient and Technology, JAMSTEC, Japan; Japan Meteorological Agency, JMA, Japan;
Laboratoire d’Océanographie de Villefranche, France; Legos, France; Met Office, UK; Mercator Ocean
Interna�onal, France; Na�onal Oceanic and Atmospheric Administra�on (NOAA) Na�onal Centres for
Environmental Informa�on (NCEI), USA; Na�onal Oceanic and Atmospheric Administra�on (NOAA)
Pacific Marine Environmental Laboratory (PMEL), USA; Na�onal Oceanography Centre, Southampton,
UK ; Ocean Scope, France ; Rutgers State University of New Jersey; Scripps Institution of
Oceanography, University of California San Diego, USA; Sorbonne Université, France; Tohoku
University, Japan ; Turkish State Meteorological Service, Türkiye; University of Brest, France ;
University of Calgary, Canada; University of Colorado, USA; Cooperative Institute for Satellite Earth
Systems Studies, University of Maryland, USA ; University of New South Wales, Australia; University
of Northern Bri�sh Columbia; Woods Hole Oceanographic Institution, USA; Yunnan University, China;

UN Agencies
• Food and Agriculture Organiza�on, FAO
• Intergovernmental Oceanographic Commission of UNESCO, IOC/UNESCO
• Interna�onal Organiza�on for Migra�on, IOM
• United Na�ons Environment Programme, UNEP
• United Na�ons High Commissioner for Refugees, UNHCR
• United Na�ons Office for Disaster Risk Reduc�on, UNDRR
• World Food Programme, WFP.