The Global Climate

in 2015–2019
WEATHER CLIMATE WATER
© World Meteorological Organization, 2019

The right of publication in print, electronic and any other form and in any language is reserved by WMO. Short
extracts from WMO publications may be reproduced without authorization, provided that the complete source is
clearly indicated. Editorial correspondence and requests to publish, reproduce or translate this publication in part
or in whole should be addressed to:

Chairperson, Publications Board

World Meteorological Organization (WMO)
7 bis, avenue de la Paix Tel.: +41 (0) 22 730 84 03
P.O. Box 2300 Fax: +41 (0) 22 730 81 17
CH-1211 Geneva 2, Switzerland Email: publications@wmo.int

The Global Climate in 2015–2019 is part of the WMO Statements on Climate providing authoritative information on the state of the climate and
impacts. It builds on operational monitoring systems at global, regional and national scales. It has been authored by: Peter Siegmund, lead
author (Royal Netherlands Meteorological Institute), Jakob Abermann (University of Graz, Austria), Omar Baddour (WMO), Pep Canadell (CSIRO
Climate Science Centre, Australia), Anny Cazenave (Laboratoire d’Etudes en Géophysique et Océanographie Spatiales, Centre National d’Etudes
Spatiales and Observatoire Midi-Pyrénées, France), Chris Derksen (Environment and Climate Change Canada), Arthur Garreau (Météo-France),
Stephen Howell (Environment and Climate Change Canada), Matthias Huss (ETH Zürich), Kirsten Isensee (IOC-UNESCO), John Kennedy (UK
Met Office), Ruth Mottram (Danish Meteorological Institute), Rodica Nitu (WMO), Selvaraju Ramasamy (Food and Agriculture Organization of
the United Nations), Katherina Schoo (IOC-UNESCO), Michael Sparrow (WMO), Oksana Tarasova (WMO), Blair Trewin (Bureau of Meteorology,
Australia), Markus Ziese (Deutscher Wetterdienst)

Cover illustration: Adobe Stock, Frédérique Julliard

NOTE

The designations employed in WMO publications and the presentation of material in this publication do not imply the expression of any opinion what-
soever on the part of WMO concerning the legal status of any country, territory, city or area, or of its authorities, or concerning the delimitation of its
frontiers or boundaries.

The mention of specific companies or products does not imply that they are endorsed or recommended by WMO in preference to others of a similar
nature which are not mentioned or advertised.

The findings, interpretations and conclusions expressed in WMO publications with named authors are those of the authors alone and do not neces-
sarily reflect those of WMO or its Members.
Contents
Executive summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Key findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

Greenhouse gases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

Ocean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

Cryosphere . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

Precipitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

Extreme events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

Attribution of extreme events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

Highlights on prominent climate-related risks . . . . . . . . . . . . . . . . . . . . . . . . 18

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
THE GLOBAL CLIMATE 2015–2019
GLOBAL TEMPERATURE RISE GREENHOUSE GAS
CONCENTRATIONS INCREASE
Global mean surface concentrations 2015–2017
2015–2019
Warmest five-year period
0.2 ℃ higher than 2011–2015
CO₂ N₂O
2016 403 parts 329 parts
Is the warmest year on record, over per million per billion
1 ℃ higher than pre-industrial
period
CH₄
1852 parts
per billion
Global five-year average temperature anomalies (relative to
1981–2010) for 2015– 2019. Data are from NASA GISTEMP v4. Data for
2019 to June 2019.

OCEAN ACIDIFICATION OCEAN WARMING

Ocean acidity increasing due to rising C0₂ In 201
2
20
2018,
018,
018, global ocean heat content reached record levels

pCO₂ and pH records from three long-term ocean observation stations.

Credit: IOC-UNESCO, NOAA-PMEL, IAEA OA-ICC. Source: NOAA NCEI, UK Met Office, IAP.

SEA LEVEL CONTINUES TO RISE CRYOSPHERE

Global sea level continued to rise Ice melt is an indicator of global warming.
Ice melt major contributor Arctic Antarctic
Arctic average summer minimum Antarctic experienced its
and winter maximum sea-ice lowest and second lowest
extents were well below the summer sea-ice extent
1981–2010 average every year in 2017 and 2018,
from 2015 to 2019. respectively.

Data source: European Space Agency (ESA) Climate Change Initiative

(CCI) sea level data until December 2015,extended by data Average of observed annual specific mass-change rate of all World Glacier Monitoring
from the Copernicus Marine Service (CMEMS) as of January 2016 Service (WGMS) reference glaciers,including pentadal means.

EXTREME EVENTS
Mortality and economic losses

2017
17
2017 >US$ 125 billion
>2 000 DEATHS NORTH AMERICA, Economic losses Large-scale
attributed to CENTRAL AMERICA
and the CARIBBEAN
EUROPE
attributed to heat extremes
ASIA
Hurricane Maria, 376
13 5.5
16.6 7.4 Hurricane Harvey attributable to
4.1
Puerto Rico and human influence
2015–2019
Dominica 2016
>8 900 DEATHS AFRICA
attributed to SOUTH AMERICA >US$ 16 billion
2 2.6
heatwaves
10 0.5
SOUTH-WEST PACIFIC
Economic losses
worldwide 5 0.3 attributed to
500
100
the wildfires
10
0.5 in California
5
10
Economic Mortality
Losses (thousands of people)
(billion $)
Executive summary
Compared to the previous five-year assess- and decreased in others. Heatwaves were
ment period 2011–2015, the current five-year the deadliest meteorological hazard in the
period 2015 –2019 has seen a continued 2015–2019 period, affecting all continents
increase in carbon dioxide (CO2) emissions and resulting in new temperature records
and an accelerated increase in the atmos- in many countries accompanied by unprece-
pheric concentration of major greenhouse dented wildfires that occurred in particular in
gases (GHGs), with growth rates nearly 20% Europe, North America and other regions. The
higher. The increase in the oceanic CO2 con- 2019 northern summer saw record-breaking
centration has increased the ocean’s acidity. wildfires that expanded to the Arctic regions,
setting new records, and wide-spread fires
The five-year period 2015–20191 is likely to in the Amazon rainforest.
be the warmest of any equivalent period on
record globally, with a 1.1  °C global tem- Among all weather-related hazards, tropical
perature increase since the pre-industrial cyclones were associated with the largest
period and a 0.2  °C increase compared to economic losses, with floods, landslides and
the previous five-year period. associated loss and damage. The costliest
hazard event was Hurricane Harvey in 2017,
Continuing and accelerated trends have which led to an estimated economic loss of
also predominated among other key climate more than US$ 125 billion.
indicators, including an acceleration of rising
sea levels, a continued decline in the Arctic Climate-related risks associated with climate
sea-ice extent, an abrupt decrease in Antarctic variability and change exacerbated food
sea ice, continued ice mass loss in the glaciers insecurity in many places, in particular Africa
and the Greenland and Antarctic ice sheets, due to the impact of drought, which increased
and the clear downward trend in the northern the overall risk of climate-related illness or
hemisphere spring snow cover. death. Higher sea-surface temperatures
endangered marine life and ecosystems.
More heat is being trapped in the ocean; 2018 Higher temperatures threaten to undermine
had the largest ocean heat content values on development through adverse impacts on
record measured over the upper 700 meters. gross domestic product (GDP) in developing
Precipitation has increased in some regions countries.

For 2019 only six months of data are currently available.

1

3
 Key findings
GREENHOUSE GASES in the southern hemisphere are dominated
by the removal of CO2 by the oceans, while
CO2 EMISSIONS AND GHG the stronger sinks in the northern hemisphere
CONCENTRATIONS INCREASED have similar contributions from both land
and oceans.
CO 2 is responsible for about 66% of the
total radiative forcing from long-lived GHGs The latest analysis of observations from the
since pre-industrial time, with methane WMO Global Atmosphere Watch shows that
(CH4) responsible for about 17% and nitrous globally averaged surface concentrations
oxide (N2 O) for 6%. The global budget of calculated from this in-situ network for CO2,
anthropogenic carbon has continued to CH4 and N2O reached new highs (Figure 1).
grow since 2015 due to the increase in CO2 The growth rates of the CO2, CH4 and N2O
emissions from the combustion of fossil fuels concentrations in the atmosphere averaged
(coal, oil and gas) and cement production. CO2 over the 2015–2017 period for which data
emissions from 2015 to 2019 are estimated have been completed and processed are each
to be at least 207  Gt CO 2 , exceeding the about 20% higher than those over 2011–2015
200  Gt CO 2 2 emitted during the previous (Table  1). 3 Preliminary analysis shows that
five-year period of 2010–2014. Sinks for CO2 in 2018 the CO2 annual mean concentration
are distributed across the hemispheres, on at Mauna Loa Observatory, Hawaii, reached
land and oceans, but CO2 fluxes in the tropics 408.52  ppm and the increase from 2017 to
(30°S–30°N) are close to carbon neutral 2018 was 1.97 ppm. From January to August
due to the CO2 sink being largely offset by 2019 the increase in the concentration
emissions from deforestation. Sinks for CO2 (deseasonalized trend) was 0.85 ppm.

Table 1. Concentrations of CO 2 (ppm), CH 4 (parts per billion, ppb) and N 2 O (ppb), their growth rates
(ppm/year for CO 2 , ppb/year for CH 4 and N 2 O) averaged over 2015–2017 and 2011–2015, the relative
Figure 1. Time series change in growth rates between 2011–2015 and 2015–2017, and the percentage of 2015–2017
of globally averaged concentration to pre-industrial concentration (before 1750). Source: WMO Global Atmosphere Watch
concentrations of CO 2
in ppm (left), CH 4 in Concentration Growth rate
ppb (middle) and N 2 O
in ppm (right). Blue
2015-2017
2015-2017 2011-2015 % to pre- 2015-2017 2011-2015 % change
lines are monthly industrial
mean global averaged
concentrations, red lines CO2 403 395.5 145 2.6 2.2 +18%
are five-year running
averaged monthly mean CH4 1851.7 1826.4 256 8.7 7.2 +21%
concentrations.
Source: WMO Global N2 O 329.1 326.2 122 0.87 0.73 +19%
Atmosphere Watch

CO2 concentration CH4 concentration N2O concentration

410 1900 335

400
CH4 concentration, ppb
CO2 concentration, ppm

330
N2O concentration, ppb

1850

390 325
1800

380 320
1750
370 315
1700
360 310

1650 305
350

340 1600 300

1984 1987 1990 1993 1996 1999 2002 2005 2008 2011 2014 2017 1984 1987 1990 1993 1996 1999 2002 2005 2008 2011 2014 2017 1984 1987 1990 1993 1996 1999 2002 2005 2008 2011 2014 2017

2
1 gigaton = 1 billion tons. 3
The current growth rate of CO2 of 2.6 ppm/year corresponds
to a mass of about 50 kg CO2 per person per week worldwide.

4
 Figure 2. Five-year
TEMPERATURE running average of
global temperature
GLOBAL TEMPERATURE CONTINUES TO anomalies (relative to
RISE, 2015–2019 IS SET TO BE WARMEST pre-industrial) from 1854
FIVE-YEAR PERIOD to 2019 for five datasets:
HadCRUT.4.6.0.0,
The years 2015 to 2018 were the four warmest NOAAGlobalTemp v5,
years on record and 2019, although only six GISTEMP v4, ERA5 and
JRA-55. Data for 2019
months of data are currently available, will
to June. The anomalies
likely join them as one of the five warmest are monthly anomalies
years – most likely second or third warmest – if averaged to years.
temperature anomalies continue at the current
high levels to the end of the year. The average
Figure 3. Five-year
global temperature for 2015–2019, which is
running average of
currently estimated to be 1.1 ± 0.1 °C above continental-scale
pre-industrial (1850–1900) level, is therefore temperature anomalies
likely to be the warmest of any equivalent (relative to 1981–2010)
period on record. It is 0.20 ± 0.08 °C warmer from 1910 to 2019 for
than the average for 2011–2015 (Figure 2). North America, South
America, Europe, Africa,
Continental average temperatures typically Asia and Oceania. Data
are from NOAA. Data
show greater variability than the global mean.
for 2019 to June. The
Even so, five-year average temperatures for anomalies are monthly
2015–2019 are currently the warmest or second anomalies averaged to
warmest on record for each of the inhabited years.
continents (Figure 3). Figure 4 shows a map of above pre-industrial and 0.13 °C warmer than
temperature anomalies for 2015–2019 relative 2011–2015. Over the oceans, below-average
to the long-term average for 1981–2010. sea-surface temperatures were observed to
the south of Greenland (one of the few areas
The global mean land-surface air temperature4 globally to have seen long-term cooling), the
for 2015–2019 was approximately 1.7 °C above eastern Indian Ocean, an area off the coast
pre-industrial and 0.3 °C warmer than 2011– of West Africa, some areas of the South
2015. Nearly all land areas were warmer than Atlantic, the Drake Passage and an area of the
average, with only a few exceptions: an area Southern Ocean in the Pacific sector. Other
of Canada and an area of the Antarctic in the areas were mostly warmer than average.
Indian Ocean sector. The five-year average Record warmth was recorded over areas
temperatures were the highest on record for of the north-east Pacific, the western North
large areas of the United States including Atlantic, the western Indian Ocean, areas
Alaska, eastern parts of South America, most of the South Atlantic, and the Tasman Sea,
of Europe and the Middle East, northern which has seen a number of severe marine
Eurasia, Australia, and areas of Africa south heatwaves in the past five years.
of the Sahara.

The global mean sea-surface temperature Figure 4. Five-year

for 2015 –2019 was approximately 0.8  °C average temperature
anomalies relative to
the 1981–2010 average.
The global temperature assessment is based on five datasets:
4
Data are from National
HadCRUT.4.6.0.0 (UK Met Office Hadley Centre and Climatic Aeronautics and Space
Research Unit, University of East Anglia), GISTEMP v4
Administration (NASA)
(National Aeronautics and Space Administration Goddard
GISTEMP v4. Data
Institute for Space Studies), NOAAGlobalTemp (National
for 2019 to June. The
Oceanic and Atmospheric Administration (NOAA), National
Centers for Environmental Information (NCEI), ERA5 (Euro- anomalies are monthly
pean Centre for Medium-range Weather Forecasts), and anomalies averaged to
JRA-55 (Japan Meteorological Agency). years.

5
 OCEAN The observed rate of global mean sea-level
rise has increased from 3.04 mm/year during
SEA-LEVEL RISE IS ACCELERATING the 10-year period 1997–2006 to 4.36 mm/year
during the previous 10-year period 2007–2016
Sea level continues to rise at an accelerated (Figure 6). The sea-level budget for the period
rate as shown by altimeter satellites. The total 1993–2016 is closed to within 0.3 mm/year.
elevation of the global mean sea level over the The contribution of land ice melt from the
altimetry era (since January 1993) has reached world glaciers and ice sheets has increased
90 mm. Figure 5 shows the altimetry-based over time and now dominates the sea-level
global mean sea-level time series for the period budget (World Climate Research Programme
January 1993–May 2019. Superimposed to Sea Level Budget Group, 2018).
the long-term trend (highlighted by the thin
Figure 5. Time series black curve in Figure 5), temporary positive
of altimetry-based
global mean sea level MORE HEAT BEING TRAPPED IN THE
100
for the period January
ESA Climate Change Initiative (SL_cci) data OCEAN
1993–May 2019. The thin 90
CMEMS
Near Real Time Jason-3
black line is a quadratic 80

function showing the 70

The capacity of the ocean to absorb heat
mean sea-level rise 60
is a critical part of the climate system. It is
Sea Level (mm)

acceleration. Data 50
estimated that more than 90% of the radiative
source: European Space
40
imbalance associated with anthropogenic
Agency Climate Change climate change is taken up by the oceans. Over
30
Initiative sea-level data the last 15 years, new observation systems,
until December 2015, 20
especially the Argo series of floats, have
extended by data from
allowed systematic near-global monitoring of
10

the Copernicus Marine

ocean heat content. Prior to 2005, sampling
0

Service as of January -10

2016 and near-real-time

1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017 2019 was more infrequent and more widely spaced,
Year
Jason-3 as of April 2019 and uncertainties in ocean heat content
estimates are, therefore, much larger.
or negative deviations are related to El Niño
and La Niña events, for example in 1997/98, Ocean heat content has reached new records
2011, 2015/16. Sea level has continued to rise since 2015. Measured over the layer from the
during recent years. Over the five-year period surface to 700 meters depth (Figure 7), 2018
Figure 6. Contributions
by different components
2014–2019, the rate of global mean sea-level had the largest ocean heat content values on
to the rate of global rise has amounted to 5 mm/year. 5 This is record, with 2017 ranking second and 2015
mean sea-level rise substantially faster than the average rate since third. 2016 also had higher values than any
during the periods 1993 of 3.2 mm/year. pre-2015 year in most datasets. In the NCEI
1997–2006 (left) and
2007–2016 (right).
2007–2016
Data source: European 1997–2006
Sea-level rise Sea-level rise
Space Agency Sea 3.04 mm/yr 4.36 mm/yr
Glaciers
Glaciers Glaciers
Level Budget Closure 0.81 mm/yr
0.81 mm/yr
Antarctic
0.56 mm/yr 0.47 mm/yr
Project (v2 version). The Thermal expansion
Antarctic
1.34 mm/yr
observed rates of global 0.05 mm/yr Thermal expansion
1.47 mm/yr
mean sea-level rise are
Greenland Greenland
shown in the upper left 0.41 mm/yr Greenland
0.93 mm/yr
0.93 mm/yr
corners. The difference Land water
Land water
Land water 0.85 mm/yr
between the sum of the 0.44 mm/yr 0.85 mm/yr
various contributions
and the observed values
indicates errors in some
of the components
or contributions from
components missing The current rate of global mean sea-level rise of 5 mm/year
5

in the sea level budget corresponds to a volume of water discharged by the Amazon
computation. river in about 3 months.

6
dataset, the ocean heat content anomaly
(recomputed relative to the reference period
1981–2010) for 2018 was 13.0  x  10 22  J for
the 0–700 meter layer, and 18.2 x 1022 J for
the 0–2000 meter layer, compared with the
pre-2015 annual records of 9.5 x 1022 J and
14.3 x 1022 J, respectively.6

SEAWATER IS BECOMING MORE ACID

The ocean absorbs around 30% of the

annual emissions of anthropogenic CO2 to
the atmosphere, thereby helping to alleviate
the impacts of climate change on the planet.
Figure 7. Global ocean
The ecological costs to the ocean, however, CRYOSPHERE heat content change
are high, as the absorbed CO2 reacts with
(x 10 22 J) for the
seawater and changes the acidity of the ocean. SEA-ICE EXTENT CONTINUES TO 0–700-meter layer
This decrease in seawater pH is linked to shifts DECREASE relative to the 1981–2010
in other carbonate chemistry parameters, baseline. The lines show
such as the saturation state of aragonite, For all years from 2015 to 2018, the Arctic’s annual means from
the main form of calcium carbonate used average September minimum (summer) sea- the Levitus analysis
for the formation of shells and skeletal ice extent7 was well below the 1981–2010 produced by NOAA
material. Observations from open ocean average. Arctic sea-ice extent for July 2019 NCEI, the EN4 analysis
produced by the UK Met
sources over the last 20 to 30  years have set a new record low. Average summer sea-
Office Hadley Centre,
shown a clear trend of decreasing average ice extent during 2015–2018 was less variable and the Institute of
pH, caused by increased concentrations of compared to 2011–2015 when the record low Atmospheric Physics
CO2 in seawater (Figure 8). Trends of ocean summer sea-ice extent occurred in 2012. ocean analysis (Cheng
acidification in coastal locations are more 2015–2018 was marked by a considerable et al., 2019).
difficult to assess, due to the highly dynamic, retreat of the Arctic sea-ice extent towards
highly variable coastal environment affected the Central Arctic particularly prominent in
by temperature change, freshwater runoff, the Beaufort and Chukchi Seas. The long-
nutrient influx, biological activity and large term trend over 1979–2018 indicates that
Figure 8. pCO 2 and
ocean oscillations. There has been an overall the summer sea-ice extent in the Arctic has pH record for the
increase in acidity of 26% since the beginning declined at a rate of approximately 12% Hawaii Ocean Time-
of the industrial revolution. per decade. Series in the Pacific
Ocean, with five-year
running average pCO 2
451
8.16
and pH indicated by
8.14
black bars. Source:
395
Intergovernmental
8.12
Oceanographic
pH TOT (n situ)

375
pCO2 (µatm)

8.10
Commission of UNESCO
355
(IOC-UNESCO),
335
8.08
NOAA Pacific Marine
315 8.06
Environmental
Laboratory, International
295 8.04
Atomic Energy Agency
275 8.02 Ocean Acidification
1985 1990 1995 2000 2005 2010 2015 2020 1985 1990 1995 2000 2005 2010 2015 2020

Year Year International

Coordination Centre

The current increase rate of the world’s ocean heat content

6 7
Sea-ice extent is defined as the area covered by sea ice
in the 0–2000-meter layer is about 1 x 10 22 joules/year. This that contains an ice concentration of 15% or more.
is about 20 times as much as the world’s annual primary
energy consumption, which in 2017 was 13 511 million tons
of oil equivalent (BP, 2018), or 0.05 x 10 22 joules.

7
 2011–2015 period and the longer-term
End of growing season 1979–2018 period that exhibited increasing
10
March Arctic
10
September Antarctic
trends in both seasons. During 2015–2019,
the summer sea ice reached its lowest and
second lowest extent on record in 2017 and
5 5

% 0 % 0 2018, respectively, with 2017 also being the

second lowest winter extent. For most years
–5 –5
from 2015–2019, the sea-ice extent retreat
–10 –10
has been predominantly located in the Ross
Sea in summer and in the Ross and Weddell
1980 1990 2000 2010 2020 1980 1990 2000 2010 2020
Year Year
Seas in winter.
End of melting season
30
September Arctic
30
February Antarctic ICE SHEETS CONTINUE LOSING MASS
20

The Greenland ice sheet has undergone

20
10

significant changes over recent decades.

10
0

% –10 % 0
While the overall ice mass was rather stable
between 1981 and 2010, an accelerated loss
–20 –10

–30
–20
of ice has been observed since the turn of the
1980 1990 2000 2010 2020
–30
1980 1990 2000 2010 2020 millennium. Figure 10 gives an indication of
Year Year
the regional distribution of average surface–
mass balance (SMB) for the reference period
1986–2005 (left), the anomaly of the 2015–2018
Figure 9. Time series The Arctic’s average March maximum (winter) period (middle) and 2018–2019 (right).8 While
of sea-ice extent sea-ice extent has also declined over the most of the interior of Greenland actually
anomalies (%) for the 1979–2019 period at a rate of approximately gained a small amount of mass, strong mass
Arctic in March and 2.7% per decade. The average winter sea-ice loss occurred at the margins. In the dry north-
Antarctic in September
extent was lower for 2015–2019 compared east interior of Greenland no significant mass
(maximum ice extent),
and for the Arctic
to 2011–2015. For all years from 2015–2019, change occurred. Negative anomalies in the
in September and the average winter sea-ice extent was well years 2015–2018 particularly occurred over
Antarctic in February below the 1981–2010 mean, and the four large parts of the west coast ablation zone,
(minimum ice extent), lowest records for winter occurred in these while south-east Greenland’s margin showed
relative to the 1981–2010 five years. The largest retreat of the sea-ice a mass change that was more positive than the
mean. Black bars extent for 2015–2019 has occurred in the 1981–2010 average. This pattern also occurred
indicate five-year Barents and Bering Seas.
averages. Source: Sea
While SMB is measured directly in only a few locations
8
Ice Index, Version 3
(Fetterer et al., 2017)
In Antarctica, a remarkable feature of 2015– around the ice sheet, SMB is shown here calculated from
2019 for both the February minimum (summer) computer models. The reference periods are based on output
from a high-resolution regional climate model (HIRHAM5)
and September maximum (winter) has been
driven by climate reanalysis on the boundaries, and the most
that sea-ice extent values have become well recent year is based on output from a weather forecasting
below the 1981–2010 average since 2016. This model (HARMONIE-AROME) with full observational data
is in considerable contrast to the previous assimilated.

5°N 1.5 5°N 1 5°N 1

End of season SMB anomaly (m/yr)
End of season SMB anomaly (m/yr)

0.8 0.8
1
Annual Mean SMB (m/yr)

0.6 0.6
70°N 70°N 0.4
70°N 0.4
0.5
Figure 10. Average 0.2 0.2
Latitude

Latitude

Latitude

SMB for the reference 65°N 0 65°N 0 65°N 0

period 1986–2005 –0.5

–0.2 –0.2

60°N 60°N 60°N –0.4

(left), the anomaly for
–0.4

–0.6 –0.6
2015–2018 (middle) and 55°N
–1
55°N –0.8
55°N –0.8
DMI/polarportal.org DMI/polarportal.org
the end-of-season SMB –1.5 –1 –1
60°W 40°W 20°W 60°W 40°W 20°W 60°W 40°W 20°W
for 2018–2019 (right). Longitude Longitude Longitude
Source: DMI/polarportal.dk

8
for the season 2018/19 (right), when only the
most south-eastern part of the country ended 200 Report
period

Specific mass-change rate

with a positive SMB anomaly. The other parts 0

of Greenland, including the interior of the -200

(kg m-2 yr-1)

ice sheet, showed a notable negative SMB -400

anomaly that is strongly correlated with the -600

exceptional summer heatwave that peaked -800

Reference glaciers of the WGMS

in Greenland in early August, producing -1000

runoff from the ice sheet of up to 11 billion 1960 1970 1980 1990 2000 2010 2020
Year
tons per day.9

The amount of ice lost annually from the

Antarctic ice sheet increased at least six-fold after moderately negative values. Over the Figure 11. Time series
between 1979 and 2017. The total mass loss last decade, glaciers lost more than 300 Gt of average of observed
from the ice sheet increased from 40 Gt per per year on average, leading to a contribution annual specific mass-
year in 1979–1990 to 252 Gt per year in 2009– to sea-level rise of about 0.8 mm per year. change rate of all World
Glacier Monitoring
2017. The contribution to sea-level rise from
Service reference
Antarctica averaged 0.36 ± 0.05 mm per year SPRING SNOW COVER DECREASED glaciers, including five-
with a cumulative 14.0 ± 2.0 mm since 1979. year averages
Most of the ice loss takes place by melting of Northern hemisphere spring snow cover is
the ice shelves10 from below, due to incursions found across high-elevation, subarctic, and
of relatively warm ocean water, especially in Arctic land areas. Snow-cover extent trends
west Antarctica and to a lesser extent along over 1967–2018 are −4.2 ± 1.5% per decade
the peninsula and in east Antarctica. for May and −12.9 ± 3.6% per decade for June
(Figure  12). Approximately 800  000  km2 of
GLACIERS UNDERGO RECORD northern hemisphere spring snow cover has
MASS LOSS been lost per degree Celsius in observed
extratropical continental warming. While
Variations in glacier mass are mainly affected there are small reductions during the autumn
by summer air temperatures, solid precipita- and winter, seasonal snow loss is dominated
tion and solar radiation. Long-term cumulative by changes in May and June. Although
glacier mass changes are thus a valuable May and June snow-cover extent during
Figure 12. Time series
indicator integrating the effects of various the seasons of 2017 and 2018 were near or of northern hemisphere
components of the global climate system above historical averages, the snow loss snow-cover extent in
on snow and ice. during 2011–2015 and 2015–2018 exceeded the June, from 1967 to 2018,
reductions over the longer 2000–2018 period. including five-year
For the period 2015–2018, data from the World averages. Shading
Glacier Monitoring Service reference glaciers represents uncertainty
indicate an average specific mass change 15 based on spread among
the following five
of −908 mm water equivalent per year. This
component datasets:
is more negative than in all other five-year
12 (1) NOAA Rutgers CDR,
periods since 1950, including the previous Estilow et al., 2015;
five-year period (2011–2015) (Figure 11). (2) GlobSnow version 2.0,
9 Takala et al., 2011;
x106 km2

Global mass change estimates require extrap- (3) MERRA-2, Reichle et

olation of the scattered direct observations al., 2017;
6
to all glaciers. An assessment at the global (4) Crocus snowpack
model driven by ERA-
scale (Zemp et al., 2019) indicates an accel-
Interim, Brun et al., 2013;
eration of glacier mass loss rate since 1985 3
(5) temperature index
model driven by
This corresponds to about 0.6 times the discharge of the
9
0 ERA-Interim, Brown
Amazon river. 1970 1980 1990 2000 2010 2020 et al., 2003. Source:
10
Ice shelves are floating sheets of ice permanently attached Environment and Climate
Year Change Canada
to a land mass.

9
Figure 13. Anomaly
of the water equivalent
of snow from January
2015 to March 2019
with reference to
the long-term means
1982–2010 based on
the Global Precipitation
Climatology Centre
(GPCC) Monitoring
Product V6 (DOI:
10.5676/DWD_GPCC/
MP_M_V6_100). Source:
GPCC, Deutscher
Wetterdienst, Germany

The water equivalent of solid precipitation, PRECIPITATION

which is mainly snow in the local winter
season, computed as the product of the PRECIPITATION INCREASES IN SOME
monthly precipitation total and fraction of PARTS AND DECREASES IN OTHERS
solid precipitation is shown in Figure  13.
Above-average snowfall was observed in Precipitation totals from the 48-month period
eastern Europe and eastern United States, January 2015 to December 2018 were compared
while Canada received below-average against different reference periods. Year-to-
snow fall. Central Europe and southern year variations can counterbalance positive
South America received less than normal and negative anomalies at a certain place.
snow. Negative anomalies can be caused
by less total precipitation and by a lower Comparing the last four years 2015–2018 with
fraction of snow, which can lead to reduced the five-year period 2011–2015 shows that the
water availability by snowmelt in the warm average precipitation totals were higher in
season. the latter period than in the former in large
regions in southern South and North America,
A comparison of the most recent five-year eastern Europe and most of Asia (Figure 14).
period with previous five-year periods (not In contrast, less precipitation fell in large parts
displayed) shows that nearly all regions where of Europe, south-west and southern Africa,
snow occurs received more snow in the most northern North America and a large part of
recent period than in the previous ones, which South America, the Indian Monsoon region,
can be explained by natural variability. and northern and western Australia.

Figure 14. Difference

in monthly average
precipitation totals
between 2015–2018
and 2011–2015. Source:
GPCC, Deutscher
Wetterdienst, Germany

10
Comparing short periods can exaggerate the
effects of slowly varying circulation patterns 180 90W 0 90E 180

such as the El Niño–Southern Oscillation 60N 60N

(ENSO). Therefore, precipitation totals
from January 2015 to December 2018 were 30N 30N
compared to observed totals from 1951 to
2010. Large parts of the northern hemisphere
EQ EQ
received precipitation amounts rated in
the wettest 20% (Figure  15). Some regions
around the equator in Africa and Asia also 30S 30S

received similarly rated totals, as well as the

western interior of Australia and southern 60S 60S

South America. In contrast, large regions

180 90W 0 90E 180
with precipitation totals rated in the driest
20% were found in northern South America,
southern and south-west Africa, the Indian 0.1 0.2 0.3 0.4 0.6 0.7 0.8 0.9

Monsoon region and eastward of the Persian

Gulf, Europe, Central and northern North
America and north-east Australia. cyclones, or events that can extend over Figure 15. Total
months or years such as droughts. Some precipitation for
EXTREME EVENTS extreme events bring substantial loss of 2015–2018 expressed
life or population displacement, others may as a percentile of the
1951–2010 reference
DEADLY HEATWAVES AND COSTLY have limited casualties but major economic
period for areas that
TROPICAL CYCLONES impacts. Figure 16 provides data on mortality would have been in the
and economic losses in the six WMO Regions driest 20% (brown) and
Many of the major impacts of climate are associated with high-impact weather and wettest 20% (green)
associated with ex treme events. These climate events during the period 2015–2019 of the years during the
can be short-term events, such as tropical (to date). reference period, with
darker shades of brown
and green indicating the
driest and wettest 10%,
respectively. Source:
GPCC, Deutscher
Wetterdienst, Germany

Figure 16. Mortality

and economic losses
associated with
extreme weather in
the period 2015–2019.
The numbers apply to
88 extreme weather
events, chosen based on
mortality and economic
losses. Source: the
NCEI Billion-dollar
Weather and Climate
Disasters list, the
EM-DAT International
Disaster Database, and
the NatCatSERVICE of
Munich Re

11
 Heatwaves have been the deadliest meteor- average. The 2018 season was especially
ological hazard in the 2015–2019 period, with active, with the largest number of tropical
wildfires also featuring especially in the Arc- cyclones of any year in the twenty-first
tic, including Greenland, Alaska and Siberia, century; all northern hemisphere basins had
and in the Amazon forest. Summer 2019 saw above-average activity, with the north-east
unprecedented wildfires in the Arctic region. Pacific having its largest accumulated cyclone
In June alone, these fires emitted 50 Mt of CO2 energy (ACE) value on record. 2016 and 2017
into the atmosphere. This is more than was were slightly below-average seasons globally,
released by Arctic fires in the same month with the 2016/17 southern hemisphere season
for the totality of the period 2010–2018. being amongst the least active of the satellite
era, but both were active years in individual
The largest economic losses were associated basins.
with tropical cyclones. The 2017 Atlantic
hurricane season was one of the most
devastating on record, with more than FLOODS
US$  125  billion in losses associated with
Hurricane Harvey alone. In the Indian Ocean, While tropical cyclones are responsible for
in March and April 2019, unprecedented and many of the world’s most destructive floods,
devastating back-to-back tropical cyclones there have been many other instances of major
hit Mozambique. flooding since 2015. Some of these floods
have been relatively long-lived responses to
High impact events from a mortality and excessive rainfall in tropical regions during
economic point of view are described below, the monsoon season, but others have been
Figure 17. Europe with lists of five most prominent events per shorter-term floods, including flash floods
experienced persistent
type. The lists are chosen based on the impact associated with intense rainfall over a few
high temperatures in late
and the geographical representativeness hours. Heavy rains have also contributed to
spring and summer 2018,
as shown here for July. (Table 2). major landslides in some parts of the world.
Data source: E-OBS.
Source: Copernicus TROPICAL CYCLONES
Climate Change Service TORNADOES AND OTHER SEVERE LOCAL
(C3S)/Royal Netherlands Overall, global tropical cyclone activity STORMS
Meteorological Institute since 2015 has been close to the satellite-era
During the last five years (to date), more than
300 losses of lives and economic losses of
5.0 US$ 7.6 billion have been recorded globally as
being associated with tornadoes, extratropical
4.0
cyclones and hailstorms. In the United States,
3.0 where data are updated regularly, overall
tornado activity over the last five-year period
2.0 has been close to average, with the 2017
season having above-average and the 2018
1.0 season below-average numbers. The 2019
0.0
season to date has been very active, with
May 2019 having provisionally the second-
-1.0 largest number of tornadoes for any month
after April 2011.
-2.0

-3.0

-4.0 HEATWAVES

-5.0 Extreme heat and heatwaves were recorded

Celsius in many parts of the world during the period
2015 –2019 (to date). Heatwaves have a

12
particularly high impact on human health and
were responsible for the heaviest casualties
of any severe weather or climate during the
same period.

DROUGHT
Rainfall percentil rannking
Droughts have had major impacts, both Serious deficiency
humanitarian and economic, in numerous
Severe deficiency
parts of the world since 2015. Significant
droughts occurred on all inhabited continents Lowest on record

(Figure 18 shows rainfall deficiency for

Australia), but some of the largest impacts
were in Africa, where millions of people
required assistance after food shortages, and Rainfall percentiles (all avail. data)
significant numbers were displaced. 1 January 2017 to 30 June 2019
Distribution based on gridded data
Australian Bureau of Meteorology

WILDFIRES
unprecedented wildfires in the Arctic and Figure 18. Rainfall
While not strictly a weather phenomenon, widespread fires in the Amazon rainforest, deficiency for Australia
wildfires are strongly influenced by weather with dramatic environmental impacts. for the period January
2017−June 2019. Source:
and climate phenomena. Drought substantially
Australian Bureau of
increases the risk of wildfire in most forest
Meteorology
regions (although it can reduce the risk of COLD EVENTS
grassland fires, due to lack of fuel), with a
particularly strong influence on long-lived Despite higher temperatures overall, there
fires. The three largest economic losses were numerous significant cold events and
on record from wildfires have all occurred snowfalls over the last five years. Many of
in the last four years. In 2019 there were these occurred in North America.

Figure 19. Smoke

plume from the Camp
Fire in California,
November 2018. Source:
NASA Earth Observatory

13
 Table 2. List of five prominent events by types during 2015–2019, with date and locations
where impacts were recorded

Tropical cyclone

October 2016, North Atlantic (Haiti; USA), Hurricane Matthew

Estimated economic losses US$ 10 billion; at least 546 deaths in Haiti and 49 in the
United States

August 2017, North Atlantic (USA – Texas), Hurricane Harvey

Estimated economic losses US$ 125 billion, 89 deaths

August–September 2017, North Atlantic (Caribbean; USA), Hurricane Irma

Estimated economic losses US$ 57 billion, 134 deaths

September 2017, North Atlantic (Dominica; Puerto Rico), Hurricane Maria

Estimated economic losses over US$ 90 billion; at least 140 deaths attributed directly
to the storm but estimates of over 2 000 indirect deaths post-storm on Puerto Rico

March 2019, south-west Indian Ocean (Mozambique; Zimbabwe), Cyclones Idai and
Kenneth
At least 1 236 deaths attributed to Cyclone Idai – the most for a southern hemisphere
cyclone for at least 100 years. Kenneth was more intense than Idai but made landfall
in a sparsely populated area

Flood

June–July 2016, China – flood

At least 310 deaths and US$ 14 billion in economic losses were attributed to the floods
across the season

August 2017, India (north-east); Bangladesh; Nepal – flood

At least 1 200 deaths were reported across the three countries, and 40 million people
were affected in some way, with the spread of waterborne disease a significant factor

August 2017, Sierra Leone – landslide

Major destruction and an estimated 1 102 death

June–July 2018, Japan – flood

At least 245 deaths were reported, along with 6 767 houses destroyed.

August 2018, India (Kerala) – flood

1.4 million people were displaced and 5.4  million affected in some way. At least
223 deaths were reported, with economic losses estimated at US$ 4.3 billion

14
 Storm and tornado

April 2016, USA – Texas – hailstorm

Estimated losses US$ 3.5 billion, amongst the highest known for a hailstorm in the
United States

June 2016, China - tornado

At least 99 deaths were reported – one of the most destructive tornadoes in recorded
Chinese history

May 2018, India (north) – severe windstorm, dust

At least 112 deaths were reported, mostly in Uttar Pradesh, from wind damage and
poor air quality

June 2018, USA – Dallas, Denver – hailstorm

Losses were estimated at US$ 1.3 billion in Dallas-Fort Worth and US$ 2.2 billion in
Denver

October 2018, Mediterranean (especially Italy; Slovenia; Croatia) – extratropical cyclone

Major wind damage and flooding in several countries; 30 deaths in Italy were attributed
to the storm

Heatwave

May and June 2015, India; Pakistan – heatwave

2 248 deaths were reported due to the heat in India, and 1 229 in Pakistan

Summer 2015/16, South Africa – heatwave

There were numerous heatwaves in South Africa during the 2015/16 summer. Pretoria
broke its previous record high temperature on three separate occasions

Summer 2015 and 2018, Europe – heatwaves

In France 3 275 and 1 500 excess deaths were attributed to the heat in 2015 and 2018,
respectively

Summer 2018–19, Australia – heatwave

Hottest summer on record for Australia. There were also significant heatwaves in the
2016/17 and 2017/18 summers, especially in New South Wales

June–July 2019, Europe – heatwave

Two major long and extended heatwaves recorded in Europe in June–July 2019 with
national records broken in many countries. In southern France a national record for any
month of 46.0 °C was observed. The heat dome spread northwards through Scandinavia
and towards Greenland where it accelerated the already above-average rate of ice melt

15
 Drought

2015/16, Northwest South America; Central America; Caribbean – drought

Drought associated with the 2015/16 El Niño affected many parts of northern South
America, Central America and the Caribbean. Rainfall averaged across the Amazon
basin in Brazil in 2016 was the lowest on record

2015–2018, Africa – drought

Severely depleted water supply storages occurred in Cape Province of South Africa,
leading to Cape Town to potentially run out of water during 2018. This followed severe
drought in many parts of southern Africa in 2015 and 2016, following poor rainy seasons
in 2014–2015 and 2015–2016. In east Africa in 2016–2017, 6.7 million people in Somalia
were experiencing food insecurity at the drought’s peak, decreasing to 5.4 million by
the end of 2017 as conditions eased

2017–2019, Australia (mostly eastern) – drought

There were significant agricultural losses, as well as large-scale fish deaths after the
Darling River ceased to flow

October 2017 – March 2018, northern Argentina; Uruguay – drought

There were heavy losses to summer crops with agricultural losses estimated at
US$ 5.9 billion

2018, Europe (northern and central) – drought

There were heavy agricultural losses across numerous countries, and low flows in the
Rhine severely disrupted river transport, causing significant economic losses

Wildfire

2015, Indonesia – wildfire

Drought led to extensive wildfires in Indonesia in the second half of 2015. 2.6 million
hectares were reported to have burned. 34 deaths were directly attributed to the fires

May 2016, Canada (Alberta) – wildfire

A wildfire caused major damage in Fort McMurray, Alberta, in May. Insured losses
exceeded US$ 3 billion with indirect losses of several billion dollars more

2016, 2019, Australia (Tasmania) – wildfire

Long-lived fires, associated with severe drought in normally wet areas. The fires burned
World Heritage areas that are believed not to have experienced fire for at least several
hundred years

July 2018, Greece – wildfire

Major fast-moving wildfires affected the region around Athens, driven by strong winds
which reached 124 km/h. At least 99 deaths were reported, the heaviest loss of life in
a wildfire globally since 2009

November 2018, USA (California) – wildfire

The town of Paradise was largely destroyed by a fast-moving wildfire. At least 85
lives were lost, and economic losses were estimated at US$ 16.5 billion, the largest
on record for a wildfire globally

16
 Cold event

February 2015, eastern USA and Canada – cold

Persistent cold in the north-east United States and eastern Canada. It was the second
coldest February on record for the north-east region of the United States

January 2016, East Asia – cold

In late January, abnormally low temperatures extended south from eastern China as
far south as Thailand. Guangzhou experienced its first snow since 1967 and Nanning
its first since 1983

July 2017, Argentina – cold

Temperature at Bariloche, Argentina, fell to −25.4 °C on 16 July, 4.3 °C below the
previous lowest on record

February–March 2018, Europe – cold/snow

Abnormal cold for late winter and early spring extended across much of Europe. Eastern
Ireland had its heaviest snowfalls for more than 50 years with totals exceeding 50 cm
in places

January–February 2019, north-central USA; interior western Canada – cold

Persistent very cold conditions in late January and February in the north-central United
States and interior western areas of Canada

ATTRIBUTION OF EXTREME EVENTS Almost every study of a significant heatwave

THE LIKELIHOOD OF OCCURRENCE OF since 2015 has found that its probability has
HEATWAVES HAS BEEN SIGNIFICANTLY been significantly increased by anthropogenic
INCREASED BY ANTHROPOGENIC climate change. For example, a study found
CLIMATE CHANGE that the heatwave that affected Japan in
July  201811 would have been impossible
Determining the extent, if any, to which the without human influence. In general, the most
chance of extreme events occurring has been clear-cut results are obtained for indicators
affected by anthropogenic climate change is that cover a large area over a substantial
an active area of science, with hundreds of period of time (for example, a national mean
papers published in the last five years. Many of monthly temperature), with more uncertainty
these studies appear one to two years after the for results at single locations over periods
event, but there is also an increasing interest of a few days.
in attribution of events relatively soon after
the event, using already established methods. An increasing number of studies are also
finding a human influence on the risk of
According to recently published peer-reviewed ex treme rainfall events, sometimes in
studies in the annual supplement to the Bulletin conjunc tion with other major climate
of the American Meteorological Society, over influences such as ENSO. One example is the
the period 2015 to 2017, 62 of the 77 events extreme rainfall in eastern China in June–July
reported show a significant anthropogenic 2016, where two studies12 found that human
influence on the event's occurrence, either influence significantly increased the risk of
directly, or indirectly (through, for example, the event, with the signal less clear in a third
influencing atmospheric circulation patterns
that contributed to the event). Attribution
studies are also published in other journals 11
Imada et al., 2019.
and reports. 12
Sun and Miao, 2018; Yuan et al., 2018.

17
 increased the amount of rainfall associated
840
with tropical cyclones. In one notable
Number of undernourished people in

830
821.6
example, Hurricane Harvey in the Houston
820 area in 2017, a study concluded that human
811.7
the world (millions)

810 influence increased the amount of rainfall

800 796.5 that occurred by about 15% (8% to 19%).16
790 785.4

780
HIGHLIGHTS ON PROMINENT
770
CLIMATE-RELATED RISKS
760
2015 2016 2017 2018
RECENT RISE IN FOOD INSECURITY AND
Year GLOBAL HUNGER DUE TO DROUGHT
IMPACT

Figure 20. study.13 Some, although not all, droughts also According to the Food and Agriculture
Number of show a direct or indirect human influence, Organization (FAO) annual reports, The State
undernourished such as the 2016/17 East African drought,14 of Food Security and Nutrition in the World,
people in the world,
which was strongly influenced by warm sea- climate variability and extremes are among
2015–2018. Source:
surface temperatures in the western Indian the key drivers behind the recent rises in
FAO, International
Fund for Agricultural Ocean to which human influence contributed. global hunger and one of the leading causes
Development (IFAD), of severe food crises (Figure 20). The changing
United Nations Very few studies have yet found any human nature of climate variability and extremes is
Children’s Fund signal in small-scale severe weather events negatively affecting all dimensions of food
(UNICEF), World Food such as thunderstorms and tornadoes, and security (food availability, access, utilization
Programme (WFP) the limited studies of anthropogenic influence and stability).
and World Health on fire weather, such as the February 201715
Organization (WHO),
event in New South Wales, Australia, have The impact of the 2015 –2016 El Niño on
2019
mostly been inconclusive. While few clear agricultural vegetation is clearly visible
anthropogenic signals have been found in through the frequency of drought conditions
tropical cyclone intensity and frequency, it in 2015–2017. The map in Figure 21 shows that
has been found that human influences have large areas in Africa, parts of central America,

Figure 21. Percentage

of time (dekad is a
10-day period) with
active vegetation when
the anomaly hot spots of
agricultural production
(ASAP) was signalling
possible agricultural
production anomalies
CROP (2015-2017)
according to the
% of dekads with drought
Normalized Difference
< 4%
Vegetation Index (NDVI)
4–8%
for more than 25% of the
crop areas in 2015–2017. 8–12%
Source: FAO, IFAD, 12–16%
UNICEF, WFP and WHO, >16%
2018 No crops

13
Zhou et al., 2018. 16
Hope et al., 2017.
14
Funk et al., 2019.
15
Hope et al., 2017.

18
Brazil and the Caribbean, as well as Australia

Change in the number of the heatwave

and parts of the Near East experienced a large Change relative to 1986–2005 average

exposure events (millions per year)

200
increase in frequency of drought conditions in
2015–2017 compared to the 14-year average.
150
The risk of food insecurity and malnutrition
is greater nowadays because livelihoods
100
and livelihood assets – especially of the
poor – are more exposed and vulnerable to
changing climate variability and extremes. 50
In the Horn of Africa, rainfall deficits led
to the failure of the 2016 and 2017 rainy 0
seasons and the number of food-insecure
people rose significantly in Eastern Africa. In
Somalia, more than half of the cropland was 2000 2002 2004 2006 2008 2010 2012 2014 2016
affected by drought, and herds had reduced Year
by 40% to 60% due to increased mortality
and distress sales. In Malawi, the 2015 floods
resulted in severe losses to crops, livestock, associated with outbreaks of water-borne Figure 21. The change
fisheries and forestry assets, and production diseases or those linked to poor sanitation, in the number of people
flows. According to the United Nations High as was reported in Bangladesh during the exposed to heatwaves
in millions per year from
Commissioner for Refugees Protection and August 2017 floods.
2000 to 2017, relative to
Return Monitoring Network, some 883 000
the 1986–2005 average.
new internal displacements were recorded MARINE LIFE AND ECOSYSTEMS ARE Source: Watts et al.,
between January and December 2018, with BEING THREATENED BY HIGHER SEA- 2018
conflict the primary reason for displacement SURFACE TEMPERATURES
(36%), followed by flooding (32%) and drought
(29%). As at September 2018, up to 200 000 According to IOC-UNESCO, significantly
of the total estimated 900  000 Rohingya higher sea-surface temperatures, as much
refugees were exposed to these natural as 3  °C above average in some areas, are
hazards. implicated in dramatic changes to the
physical, chemical and biological state of
THE OVERALL RISK OF CLIMATE-RELATED the marine environment, with great impacts
ILLNESS OR DEATH HAS INCREASED on food chains and marine ecosystems, as
well as socioeconomically important fisheries.
Based on WHO data and analysis, the overall Among the areas significantly affected are
risk of heat-related illness or death has the Great Barrier Reef off the east coast of
climbed steadily since 1980, with around Australia, Pacific island countries such as
30% of the world’s population now living in Fiji and Kiribati, and the Okinawa region of
climatic conditions that deliver potentially Japan. Coral mortality of up to 50% to 70%
deadly temperatures at least 20 days a year. has been reported.

Other climate-related events such as heavy IOC-UNESCO also reported that oxygen is
rain and associated floods create favourable declining in the open and coastal oceans,
conditions for various sorts of epidemic including estuaries and semi-enclosed seas.
outbreaks. In cholera-endemic countries, an Since the middle of the last century, there
estimated 1.3 billion people are at risk, while has been an estimated 1%–2% decrease in
in Africa alone about 40 million people live the global ocean oxygen inventory. Regions
in cholera “hotspots”. WHO has recognized with historically low oxygen concentrations
that large cholera outbreaks in eastern and are expanding, and new regions are now
central, and later southern Africa were exhibiting low oxygen conditions. Global
likely enhanced by El Niño-driven weather warming is expected to contribute to this
conditions, in particular extreme rainfall decrease directly because the solubility
and floods. Flood events are also of ten of oxygen decreases in warmer waters,

19
Figure 23. Oxygen
minimum zones (blue)
and areas with coastal
hypoxia (red) in the
world’s oceans. Coastal
hypoxic sites mapped
here are systems where
oxygen concentrations
of < 2 mg/L have been
recorded and in which
anthropogenic nutrients
are a major cause of
oxygen decline. Sources:
data from Diaz and
Rosenberg (2008), figure
adapted after Breitburg
et al., 2018

200 meter depth

Coastal hypoxic areas
0.07 1.9 mg l–1 O2

and indirectly through changes in ocean Brown, R.D., B. Brasnett and D. Robinson,
dynamics that reduce ocean ventilation, 2003: Gridded North American monthly snow
which is the introduction of oxygen to the depth and snow water equivalent for GCM
ocean interior. evaluation. Atmosphere-Ocean, 41(1):1–14,
https://doi.org/10.3137/ao.410101.
GROSS DOMESTIC PRODUCT IS FALLING
IN DEVELOPING COUNTRIES DUE TO Brun, E., V. Vionnet, A. Boone, B. Decharme,
INCREASING TEMPERATURES Y. Peings, R. Valette, F. Karbou and S. Morin,
2013: Simulation of Northern Eurasian local
The International Monetary Fund found that snow depth, mass, and density using a detailed
for a median- and low-income developing snowpack model and meteorological reanalyses.
country, with an annual average temperature Journal of Hydrometeorology, 14: 203–219,
of 25  °C, the effect of a 1  °C increase in https://doi.org/10.1175/JHM-D-12-012.1.
temperature leads to a growth decrease
of 1.2%. Countries whose economies are Cheng, L., et al., 2019: 2018 Continues record
projected to be significantly adversely affected global ocean warming, Advances in Atmospheric
by an increase in temperature produced only Sciences, 36:249 –252, ht tps://doi.org/10.1007/
about 20% of global GDP in 2016; however, s00376-019-8276-x.
they are currently home to nearly 60% of
the global population and are projected to Estilow, T.W., A.H. Young and D.A. Robinson,
be home to more than 75% by the end of 2015: A long-term northern hemisphere snow
the century. cover extent data record for climate studies
and monitoring. Earth System Science Data,
REFERENCES 7:137-142, https://doi.org/10.5194/essd-7-137-2015.

Breitburg, B., L.A. Levin, A. Oschlies, M. Fetterer, F., K. Knowles, W.N. Meier, M. Savoie
Grégoire, F.P. Chavez, D.J. Conley, V. Garçon, D. and A.K. Windnagel, 2017: Sea Ice Index, Version
Gilbert, D. Gutiérrez, K. Isensee, G.S. Jacinto, 3. National Snow and Ice Data Center, https://
K.E. Limburg, I. Montes, S.W.A. Naqvi, G.C. doi.org/10.7265/N5K072F8.
Pitcher, N.N. Rabalais, M.R. Roman, K.A. Rose,
B.A. Seibel, M. Telszewski, M. Yasuhara and Funk, C., D. Pedreros, S. Nicholson, A. Hoell,
J. Zhang, 2018: Declining oxygen in the global D. Korecha, G. Galu, G. Artan, Z. Segele, A.
ocean and coastal waters. IOC Global Ocean Tadege, Z. Atheru, F. Teshome, K. Hailermariam,
Oxygen Network. Science, 359(6371), https:// L. Harrison and C. Pomposi, 2019: Examining the
doi.org/10.1126/science.aam7240. potential contributions of extreme "western V"

20
sea surface temperatures to the 2017 March- Takala, M., K. Luojus, J. Pulliainen, C. Derksen,
June east African drought. Bulletin of the J. Lemmetyinen, J.-P. Kärnä, J. Koskinen and B.
American Meteorological Society, 100:S55-S60, Bojkov, 2011: Estimating northern hemisphere
https://doi.org/10.1175/BAMS-D-18-0108.1. snow water equivalent for climate research
through assimilation of space-borne radiometer
Hope, P., M.T. Black, E. Lim, A. Dowdy, G. data and ground-based measurements. Remote
Wang, R.J. Fawcett and A.S. Pepler, 2019: Sensing of Environment, 115(12):3517–3529,
On determining the impact of increasing https://doi.org/10.1016/j.rse.2011.08.014.
atmospheric CO2 on the record fire weather
in eastern Australia in February 2017. Bulletin of Watts, N., et al., 2018: The 2018 report of The
the American Meteorological Society, 100:S111- Lancet Countdown on health and climate change:
116, https://doi.org/10.1175/BAMS-D-18-0135.1. shaping the health of nations for centuries to
come, Lancet, 392(10163):2479–2514, https://doi.
Imada, Y., M. Watanabe, H. Kawase, H. org/10.1016/S0140-6736(18)32594-7.
Shiogama and M. Arai, 2019: The July 2018
high temperature event in Japan could not Yuan, X., S. Wang and Z. Hu, 2018: Do climate
have happened without human-induced global change and El Niño increase likelihood of
warming. Scientific Online Letters on the Yangtze River rainfall? Bulletin of the American
Atmosphere, 15A:8–12, https://doi.org/10.2151/ Meteorological Society, 99:S113–S117, https://
sola.15A-002. doi.org/10.1175/BAMS-D-17-0089.1.

Oldenborgh, G.J. van, K. van der Wiel, A. Zemp, M., M. Huss, E. Thibert, N. Eckert, R.
Sebastian, R. Singh, J. Arrighi, F. O t to, McNabb, J. Huber, M. Barandun, H. Machguth,
K. Haustein, S. Li, G. Vecchi and H. Cullen, S. U. Nussbaumer, I. Gärtner-Roer, L. Thomson,
2017: Attribution of extreme rainfall from F. Paul, F. Maussion, S. Kutuzov and J.G. Cogley,
Hurricane Harvey, August 2017. Environmental 2019: Global glacier mass changes and their
Research Letters, 12(12):124009, ht tps://doi. contributions to sea-level rise from 1961 to
org/10.1088/1748-9326/aa9ef2. 2016. Nature, 568(7752), 382–386, https://doi.
org/10.1038/s41586-019-1071-0.
Reichle, R.H., C.S. Draper, Q. Liu, M. Girotto, S.P.
Mahanama, R.D. Koster and G.J. De Lannoy, 2017: Zhou, C., K. Wang and D. Qi, 2018. Attribution
Assessment of MERRA-2 land surface hydrology of the July 2016 extreme precipitation event
estimates. Journal of Climate, 30:2937–2960, over China's Wuhan. Bulletin of the American
https://doi.org/10.1175/JCLI-D-16-0720.1. Meteorological Society, 99:S107–S112, https://
doi.org/10.1175/BAMS-D-17-0090.1.
Sun, Q. and C. Miao, 2018: Extreme rainfall
(R20mm, RX5Day) in Yangtze-Huai, China,
in June-July 2016: the role of ENSO and
anthropogenic climate change. Bulletin of the
American Meteorological Society, 99:S102-S106,
https://doi.org/10.1175/BAMS-D-17-0091.1.

21
 For more information, please contact:

World Meteorological Organization

7 bis, avenue de la Paix – P.O. Box 2300 – CH 1211 Geneva 2 – Switzerland

Communication and Public Affairs Office

Tel.: +41 (0) 22 730 83 14/15 – Fax: +41 (0) 22 730 80 27
Email: cpa@wmo.int

public.wmo.int
JN 191303