Editing IPCC:AR6/WGI/Chapter-9 (section)

=== 9.2.1 Ocean Surface ===

<div id="h2-15-siblings" class="h2-siblings"></div>

<div id="9.2.1.1" class="h3-container"></div>

<span id="sea-surface-temperature"></span>
==== 9.2.1.1 Sea Surface Temperature ====

<div id="h3-1-siblings" class="h3-siblings"></div>

The IPCC Fifth Assessment Report (AR5; [[#Hartmann--2013|Hartmann et al., 2013]] ) assessed that it is ''virtually certain'' that global sea surface temperature (SST) has increased since the beginning of the 20th century ( ''very high confidence'' ). The Special Report on Ocean and Cryosphere in a Changing Climate (SROCC) did not assess past SST change. Since AR5, improvements in the understanding of recent SST biases in the observational records, especially extending ship-based observations with buoy-based observations and improved treatment of sea ice, have had important consequences for key climate change indicators such as global mean surface temperature (GMST), global surface air temperature (GSAT), and SST (Cross-Chapter Box 2.3). The AR5 assessment is confirmed, and it is now ''very likely'' that global mean SST changed by 0.88 [0.68 to 1.01] °C from 1850–1900 to 2011–2020, and 0.60 [0.44 to 0.74] °C from 1980 to 2020 (Figure 9.3 and Table 2.4).

<div id="_idContainer012" class="Basic-Text-Frame"></div>

[[File:c664066722b424728e3f3413047341a5 IPCC_AR6_WGI_Figure_9_3.png]]

'''Figure''' '''9.3 |''' '''Sea surface temperature (SST) and its changes with time. (a)''' Time series of global mean SST anomaly relative to 1950–1980 climatology. Shown are paleoclimate reconstructions and PMIP models, observational reanalyses (HadISST) and multi-model means from the Coupled Model Intercomparison Project (CMIP) historical simulations, CMIP projections, and HighResMIP experiment. '''(b)''' Map of observed SST (1995–2014 climatology HadISST). '''(c)''' Historical SST changes from observations. '''(d)''' CMIP 2005–2100 SST change rate. '''(e)''' Bias of CMIP. '''(f)''' CMIP change rate. '''(g)''' 2005–2050 change rate for SSP5-8.5 for the CMIP ensemble. '''(h)''' Bias of HighResMIP (bottom left) over 1995–2014. '''(i)''' HighResMIP change rate for 1950–2014. ( '''j)''' 2005–2050 change rate for SSP5-8.5 for the HighResMIP ensemble. No overlay indicates regions with high model agreement, where ≥80% of models agree on sign of change. Diagonal lines indicate regions with low model agreement, where &lt;80% of models agree on sign of change (see Cross-Chapter Box Atlas.1 for more information). Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).

Regions vary in the rate of SST warming, with slight cooling in some regions (Figure 9.3). The SROCC ( [[#Collins--2019|Collins et al., 2019]] ) and [[IPCC:Wg1:Chapter:Chapter-7#7.4.4|Section 7.4.4]] assess SST changes over specific regions, which are consistent with the changes reported here. The tropical ocean has been warming faster than other regions since 1950, with the fastest warming in regions of the tropical Indian and western Pacific oceans (Figure 9.3), due to a combination of local atmosphere–ocean coupling, the Indonesian Throughflow ( [[#9.2.3.4|Section 9.2.3.4]] and Figure 9.11), and trends in the Walker circulation (Sections 2.3.1.4.1 and 3.3.3.1, and Figure 3.16). The western boundary currents of the subtropical gyres have warmed faster than the global mean over the past century. There remains ''low agreement'' in the changes of the location and the dynamical changes in western boundary current extensions (Sections 2.3.3.4.2 and 9.2.3.4, and Figure 9.3). In the Arctic, the mean SST increase over the last two decades is similar to, or only slightly higher than, the global average (J.-L [[#Chen--2019|]] [[#Chen--2019|Chen et al., 2019]] ). In contrast, the eastern Pacific Ocean, subpolar North Atlantic Ocean and Southern Ocean have warmed more slowly than the global average or cooled (Figure 9.3). Surface warming in the subpolar Southern Ocean has been slower than the global average since the 1950s, and this pattern is consistent with the upwelling around Antarctica renewing surface water with pre-industrial, deeper water masses ( [[#9.2.3.2|Section 9.2.3.2]] ; [[#Frölicher--2015|Frölicher et al., 2015]] ; J. [[#Marshall--2015|]] [[#Marshall--2015|Marshall et al., 2015]] ; [[#Armour--2016|Armour et al., 2016]] ). New evidence since SROCC ( [[#Meredith--2019|Meredith et al., 2019]] ) confirms slight cooling since the 1980s around the subpolar Southern Ocean, contrasting with marked warming directly northward of it ( [[#9.2.3.2|Section 9.2.3.2]] ; [[#Haumann--2020|Haumann et al., 2020]] ; [[#Rye--2020|Rye et al., 2020]] ; [[#Auger--2021|Auger et al., 2021]] ). In eastern boundary upwelling systems, SROCC ( [[#Bindoff--2019|Bindoff et al., 2019]] ) reported ''low agreement'' between SST trends in recent decades, due to varying spatio-temporal resolution and interannual to multi-decadal variability. Satellite evidence not included in SROCC shows that 92% of these regions warmed more slowly than neighbouring offshore locations between 1982 and 2015, so upwelling may buffer the near shore from warming ( [[#9.2.3.5|Section 9.2.3.5]] ; [[#Varela--2018|Varela et al., 2018]] ). Coupled ocean-atmospheric modes of variability strongly affect regional SST (Cross-Chapter Box 3.1 and Annex IV). In summary, a positive SST trend since 1950 is evident globally, but there is ''very high confidence'' that the Indian Ocean, western equatorial Pacific Ocean, and western boundary currents have warmed faster than the global average, while the Southern Ocean, the eastern equatorial Pacific, and the North Atlantic Ocean have warmed more slowly, or have slightly cooled.

In AR5 ( [[#Flato--2013|Flato et al., 2013]] ), a marginal improvement was noted in Coupled Model Intercomparison Project Phase 5 (CMIP5) climate model SST biases compared to Phase 3 (CMIP3) models in AR4, with a reduction in the magnitude of biases. The AR5 noted that, in several regions, large SST biases are symptomatic of errors in the representation of important processes, such as dynamics in the equatorial Pacific and North Atlantic, and Southern Ocean. Common regional biases in SST or historical SST trends are not exclusively linked to the representation of the ocean ( ''high confidence'' ), but can have multiple causes, including: errors in the representation of long-term historical trends in equatorial winds ( [[#9.2.1.2|Section 9.2.1.2]] ); misrepresentation of the forced equatorial ocean response ( [[#Karnauskas--2012|Karnauskas et al., 2012]] ; [[#Kohyama--2017|Kohyama et al., 2017]] ; [[#Coats--2018|Coats and Karnauskas, 2018]] ); thermocline depth errors ( [[#Linz--2014|Linz et al., 2014]] ); errors in atmospheric model cloud-related shortwave radiation ( [[#Hyder--2018|Hyder et al., 2018]] ); biases in ocean circulation variability ( [[#Wang--2014|]] [[#Wang--2014|C. Wang et al., 2014]] ); and deficiencies in upper ocean (Q. [[#Li--2019|]] [[#Li--2019|]] [[#Li--2019|Li et al., 2019]] ) and atmospheric ( [[#Bates--2012|Bates et al., 2012]] ) boundary layer parametrizations. In CMIP6, the mid-latitude biases in the Northern Hemisphere are improved in the multi-model mean, and the inter-model standard deviation of the zonal mean SST error is significantly decreased in the northern Hemisphere south of 50°N compared to CMIP5, though biases in equatorial regions remain essentially unchanged ( [[IPCC:Wg1:Chapter:Chapter-3#3.5.1.1|Section 3.5.1.1]] and Figures 3.23, 3.24 and 9.3). Some long-standing ocean model biases have been reduced through increases in model resolution in CMIP6 ( [[#Bock--2020|Bock et al., 2020]] ) and improved parametrizations ( [[#Fox-Kemper--2011|Fox-Kemper et al., 2011]] ; Q. [[#Li--2016|]] [[#Li--2016|Li et al., 2016]] ; [[#Qiao--2016|Qiao et al., 2016]] ; [[#Reichl--2018|Reichl and Hallberg, 2018]] ). The High Resolution Model Intercomparison Project (HighResMIP) ensemble (Figure 9.3) has smaller cold biases in the North Atlantic and the tropical Pacific, and smaller warm biases in the upwelling regions off the western coasts of Africa, North and South America ( [[#Roberts--2018|Roberts et al., 2018]] , 2019; [[#Caldwell--2019|Caldwell et al., 2019]] ; [[#Docquier--2019|Docquier et al., 2019]] ). In summary, CMIP6 models show persistent regional biases in representing the climatological SST state ( ''very high confidence'' ), but higher resolution reduces some biases, particularly in the North Atlantic and eastern boundary upwelling systems (Figure 9.3; ''high confidence'' ).

The CMIP6 models represent the observed trends in SST patterns with greater fidelity than CMIP5, with the ocean area that is inconsistent with the observed trends decreasing by about three quarters from CMIP5 to CMIP6 ( [[#Olonscheck--2020|Olonscheck et al., 2020]] ). In some regions, the direction of SST changes in observations are consistent with CMIP6 only when including internal variability ( [[#Olonscheck--2020|Olonscheck et al., 2020]] ). This is notably the case in the equatorial Pacific, North Atlantic, and Southern Ocean, which are regions where SST is of known importance in controlling heat uptake ( [[#9.2.2.1|Section 9.2.2.1]] ) and the global radiative feedback parameter ( [[IPCC:Wg1:Chapter:Chapter-7#7.4.4.3|Section 7.4.4.3]] ). Overall, despite some persistent regional biases, CMIP6 coupled climate models reproduce the observed SST trends or high internal variability over the past century over a range of different multi-decadal periods (Figure 9.3; [[#Olonscheck--2020|Olonscheck et al., 2020]] ; [[#Watanabe--2021|Watanabe et al., 2021]] ), highlighting their skill to inform future large-scale SST changes at regional scale. Warming is projected at varying rates in all regions by 2050, except the North Atlantic Subpolar Region, the equatorial Pacific, and the Southern Ocean where models disagree ( ''high confidence'' ).

It is ''virtually certain'' that SST will continue to increase in the 21st century, at a rate depending on future emissions scenarios. The future global mean SST increase projected by CMIP6 models for the period 1995–2014 to 2081–2100 is 0.86 [5–95% range: 0.43–1.47] °C under SSP1-2.6, 1.51 [1.02 to 2.19] °C under SSP2-4.5, 2.19 [1.56 to 3.30] °C under SSP3-7.0, and 2.89 [2.01 to 4.07] °C under SSP5-8.5 (Figure 9.3). While under SSP1-2.6, the CMIP6 ensemble consistently projects that it is ''very likely'' at least 83% of the world ocean surface will have warmed by 2100, and under SSP5-8.5, at least 98% of the world ocean surface will have warmed. The spatial pattern of future change is consistent with observed SST change over the 20th century, though with notable regional differences (Figure 9.3). Long-term change in SST patterns is important for regional impacts but also affects radiative feedbacks, and therefore long-term change in climate sensitivity ( [[IPCC:Wg1:Chapter:Chapter-7#7.4.4.3|Section 7.4.4.3]] ). In the Southern Ocean, CMIP6 models project that SSTs will eventually consistently increase in the 21st century, at a rate dependent on future scenarios (Figure 9.3 and [[#9.2.3.2|Section 9.2.3.2]] ; [[#Bracegirdle--2020|Bracegirdle et al., 2020]] ). Yet, there is only ''low confidence'' that this Southern Ocean warming will emerge by the end of the century ( [[IPCC:Wg1:Chapter:Chapter-7#7.4.4.1|Section 7.4.4.1]] ), due to the inconsistent historical and near-term simulations and observations over the 20th century (Figure 9.3). Furthermore, the equilibrium SST pattern from proxy records or simulated by climate models under CO <sub>2</sub> forcing stand in contrast with the cooling trends in the Southern Ocean observed over the past decades ( [[IPCC:Wg1:Chapter:Chapter-7#7.4.4.1.2|Section 7.4.4.1.2]] ). Similarly, the SST change pattern observed in the tropical Pacific Ocean will transition on centennial time scales to a mean pattern resembling the El Niño pattern ( ''medium confidence'' ) (Annex IV). However, it is difficult to delineate a climate change trend ressembling an El Niño pattern and El Niño variability ( [[#Wittenberg--2009|Wittenberg, 2009]] ; [[#Collins--2010|Collins et al., 2010]] ) without large ensembles ( [[#Kay--2015|Kay et al., 2015]] ). Several Pliocene SST reconstructions indicate enhanced warming in the centre of the eastern Pacific equatorial cold tongue upwelling region, consistent with reconstruction of enhanced subsurface warming and enhanced warming in coastal upwelling regions ( [[IPCC:Wg1:Chapter:Chapter-7#7.4.4.2.2|Section 7.4.4.2.2]] ). The North Atlantic subpolar gyre is projected to continue to warm more slowly than surrounding regions ( [[#Suo--2017|Suo et al., 2017]] ), as the Gulf Stream concurrently warms rapidly (Figure 9.3; [[#Cheng--2013|Cheng et al., 2013]] ) and the Atlantic Meridional Overturning Circulation further declines under greenhouse gas forcing, although models disagree about the rate of change (Figure 9.3 and [[#9.2.3.1|Section 9.2.3.1]] ). In summary, CMIP6 models show a future pattern of SST change comparable to historical trends with intensity depending on future emissions scenario, and some of the observed cooling trends over the 20th century will eventually transition to a warming SST on centennial time scales, in particular in the Southern Ocean ( ''high confidence'' ) and in the equatorial Pacific ( ''medium confidence'' ), while the North Atlantic subpolar gyre will continue to warm more slowly than the global average ( ''high confidence'' ).

<div id="9.2.1.2" class="h3-container"></div>

<span id="airsea-fluxes"></span>
==== 9.2.1.2 Air–Sea Fluxes ====

<div id="h3-2-siblings" class="h3-siblings"></div>

Air–sea fluxes of energy, freshwater, and momentum (wind stresses) are difficult to observe directly ( [[#Cronin--2019|Cronin et al., 2019]] ), so estimates of the global mean net air–sea heat flux are inferred from observed ocean warming ( [[IPCC:Wg1:Chapter:Chapter-2#2.3.3.1|Section 2.3.3.1]] , Box 7.2, and Cross-Chapter Box 9.1). Air–sea heat fluxes resemble the warming patterns of CMIP3 ( [[#Domingues--2008|Domingues et al., 2008]] ; [[#Levitus--2012|Levitus et al., 2012]] ) and are consistent with the ensemble mean warming rate of CMIP5 ( [[#Cheng--2017|Cheng et al., 2017]] , 2019) and CMIP6 models ( [[IPCC:Wg1:Chapter:Chapter-3#3.5.1.3|Section 3.5.1.3]] ). Regional air–sea fluxes in models remain a key driver of uncertainty ( [[#Huber--2017|Huber and Zanna, 2017]] ; [[#Tsujino--2020|Tsujino et al., 2020]] ). A substantial part of the upper 700 m energy increase is ''very likely'' attributed to anthropogenic forcing via increasing radiative forcing (Sections 3.5.1.3, 7.2 and 7.3).

The SROCC ( [[#Abram--2019|Abram et al., 2019]] ) and AR5 ( [[#Rhein--2013|Rhein et al., 2013]] ) assessed that observations of air–sea fluxes had not yet reached the density or accuracy to directly detect trends beyond the noise. New evidence since SROCC confirms that direct heat and freshwater flux trends have not emerged yet as spatial (Figure 9.4), annual ( [[#Yu--2019|Yu, 2019]] ), and decadal ( [[#Zanna--2019|Zanna et al., 2019]] ) variability overwhelm detection. Since AR5, comprehensive comparisons ( [[#Bentamy--2017|Bentamy et al., 2017]] ; [[#Valdivieso--2017|Valdivieso et al., 2017]] ; [[#Yu--2017|Yu et al., 2017]] ) have used updated and new surface flux products to improve surface flux uncertainty estimates, and these comparisons note that implied global energy imbalances often exceed the observed ocean warming. Flux estimates using top of atmosphere observations and atmospheric fluxes from reanalysis have improved over past products ( [[#Trenberth--2018|Trenberth and Fasullo, 2018]] ) but require consistency adjustments ( [[#Trenberth--2019|Trenberth et al., 2019]] ) as the energy budget is not closed. Adjustments are needed for all flux products, and they remain less accurate than direct ocean heat content change measurements ( [[#Cheng--2017|Cheng et al., 2017]] ). Some regional changes are ''likely'' robust in both satellite observations and projections (Figure 9.4). Recent satellite-based surface flux products with improved retrieval algorithms and new satellites, for example, J-OFURO3 ( [[#Tomita--2019|Tomita et al., 2019]] ) and OAFlux-HR ( [[#Yu--2019|Yu, 2019]] ), provide a complete suite of turbulent fluxes including heat, moisture, and momentum. When combined with satellite-based surface radiation from Clouds and the Earth’s Radiant Energy System (CERES) Energy Balanced and Filled (EBAF; [[#Kato--2018|Kato et al., 2018]] ) and precipitation from Global Precipitation Climatology Project (GPCP; [[#Adler--2003|Adler et al., 2003]] ), full ocean-surface forcing is available since 1987 (Figure 9.4). These products agree with sparse buoy and ship observations within 30 W m <sup>–2</sup> ( [[#Bentamy--2017|Bentamy et al., 2017]] ; [[#Cronin--2019|Cronin et al., 2019]] ) ''.'' While patterns agree between models and satellites in net fluxes (Figure 9.4), the trend magnitudes are substantially weaker in models. The fluxes tending to warm the North Atlantic and Southern Ocean are consistent with the largest changes observed in the surface properties and water masses (Sections 9.2.1.1, 9.2.2.1 and 9.2.2.3). The observed trend toward a saltier Atlantic Ocean and a fresher Indian Ocean, as well as trends in evaporation minus precipitation (E-P) patterns in the equatorial Pacific (see also [[IPCC:Wg1:Chapter:Chapter-8#8.3.1|Section 8.3.1]] ) enhance the present mean pattern of wetting and drying. Elsewhere patterns are less clear, with only partial, large-scale agreement with the ‘wet gets wetter’ simplification (Sections 3.3.2.3, 4.4.1 and 4.5.1). In summary, globally integrated and large-scale fluxes are more reliably inferred from heat content and salinity change, while regional trends are rarely robust in observations; where they are robust, they tend to be underestimated or in disagreement in models ( ''very high confidence'' ).

<div id="_idContainer014" class="Basic-Text-Frame"></div>

[[File:171fa62ce903f21d239a4870f36f3f2a IPCC_AR6_WGI_Figure_9_4.png]]

'''Figure''' '''9.4 |''' '''Global maps of observed mean fluxes (a, d, g), the observed trends in these fluxes (b, e, h) and the projected rate of change in these fluxes from SSP5-8.5 (c, f, i).''' Shown are the freshwater flux '''(a–c)''' , net heat flux '''(d–f)''' , and momentum flux or wind stress magnitude '''(g–i)''' , with positive numbers indicating ocean freshening, warming, and accelerating respectively. The means and observed trends are calculated between 1995–2014 (freshwater and wind stress) or 2001–2014 (heat). The SSP5-8.5 projected rates are between 1995–2100 using 20-year averages at each end of the time period. Observations show objective interpolation from Clouds and the Earth’s Radiant Energy System (CERES) Energy Balanced and Filled (EBAF) v4 ( [[#Kato--2018|Kato et al., 2018]] ), Objectively Analyzed air–sea Fluxes-High Resolution (OAFlux-HR) ( [[#Yu--2019|Yu, 2019]] ), and Global Precipitation Climatology Project (GPCP) ( [[#Adler--2003|Adler et al., 2003]] ) of fluxes and flux trends (b, e, h). Observed trends with no overlay indicate regions where the trends are significant at p = 0.34 level. Crosses indicate regions where trends are not significant. For (c, f, i) projections, no overlay indicates regions with high model agreement, where ≥80% of models agree on the sign of change. Diagonal lines indicate regions with low model agreement, where &lt;80% of models agree on the sign of change (see Cross-Chapter Box Atlas.1 for more information). Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).

There is ''low confidence'' in long-term wind stress trends in most regions, but a few locations have ''likely'' trends over the scatterometer era and in projections, as shown in Figure 9.4 ( [[#Desbiolles--2017|Desbiolles et al., 2017]] ; [[#Young--2019|Young and Ribal, 2019]] ; [[#Yu--2019|Yu, 2019]] ). The AR5 ( [[#Rhein--2013|Rhein et al., 2013]] ) assessed with ''medium confidence'' that zonal wind stress over the Southern Ocean increased from the early 1980s to the 1990s ( ''medium confidence'' ) (Figure 9.4). Over 1995–2014, the zonal wind stress over the Southern Ocean continued to increase, westerly winds in the North Pacific and North Atlantic weakened, while the easterly equatorial Pacific winds of the Walker circulation strengthened (Figure 9.4). In historical simulations, CMIP5 models projected annular modes (Annex IV) to move poleward and strengthen in both hemispheres ( [[#Yang--2016|Yang et al., 2016]] ), while in CMIP6 models westerlies only strengthen over the Southern Ocean, with a weaker trend than recently observed (Figure 9.4 and Sections 4.5.1 and 4.5.3). In the tropical Pacific Ocean, a weakening trend in easterly winds and Walker circulation in the 20th century has been inferred based on observed sea level pressure data ( [[#Vecchi--2006|Vecchi et al., 2006]] ; [[#Vecchi--2007|Vecchi and Soden, 2007]] ) and coral proxies ( [[#Carilli--2014|Carilli et al., 2014]] ) and is projected to continue by CMIP6 models (Figure 9.4). Yet, over 1995–2014 observed winds have strengthened (Figure 9.4). The observed strengthening may have been influenced by a combination of factors ( [[IPCC:Wg1:Chapter:Chapter-7#7.4.4.2.1|Section 7.4.4.2.1]] ), but there is ''low confidence'' in the attribution of this signal to anthropogenic warming ( [[IPCC:Wg1:Chapter:Chapter-3#3.3.3.1|Section 3.3.3.1]] ) and ''medium confidence'' that it reflects internal variability ( [[IPCC:Wg1:Chapter:Chapter-8#8.3.2.3|Section 8.3.2.3]] ). Near-term projected changes over the Southern Ocean result from ozone recovery and greenhouse gases (Sections 4.3.3 and 4.4.3). Overall, there is only ''low confidence'' in observed and projected wind stress trends in most regions because trends in oceanic wind stresses during the satellite era have not emerged or are inconsistent with historical simulated changes.

Air–sea flux biases result from common causes in most models, and many are the same as during AR5 ( [[#Rhein--2013|Rhein et al., 2013]] ). Important currents (e.g., Gulf Stream, Kuroshio, Antarctic Circum-polar Current patterns) are often found in erroneous locations in models, affecting SST and flux signatures ( [[#Bates--2012|Bates et al., 2012]] ; [[#Beadling--2020|Beadling et al., 2020]] ; J.-L.F. [[#Li--2020|]] [[#Li--2020|]] [[#Li--2020|Li et al., 2020]] ), but their locations are improved in high-resolution ocean models ( [[#Chassignet--2017|Chassignet et al., 2017]] , 2020; [[#Hewitt--2020|Hewitt et al., 2020]] ), and high-resolution coupled models reduce the mean air–sea flux biases ( [[#Delworth--2012|Delworth et al., 2012]] ; [[#Sakamoto--2012|Sakamoto et al., 2012]] ; [[#Small--2014|Small et al., 2014]] ; [[#Haarsma--2016|Haarsma et al., 2016]] ; [[#Caldwell--2019|Caldwell et al., 2019]] ; L.C [[#Jackson--2020|]] [[#Jackson--2020|Jackson et al., 2020]] ). Oceanic variability stems either from internal chaotic variability or atmospheric forcing ( [[#Hasselmann--1976|Hasselmann, 1976]] ; [[#Sérazin--2016|Sérazin et al., 2016]] , 2017). Large-scale variability in the ocean tends to follow atmospheric forcing in low-resolution models, while in high-resolution coupled models ocean variability drives atmospheric variability on small scales ( [[#Bishop--2017|Bishop et al., 2017]] ; [[#Small--2019|Small et al., 2019]] ), allowing these high-resolution models to mimic the coupling with clouds, precipitation, and atmospheric and oceanic boundary layers apparent in observations ( [[#Chelton--2010|Chelton and Xie, 2010]] ; [[#Frenger--2013|Frenger et al., 2013]] ). Even coarse-resolution models, such as the ocean and sea ice components used in CMIP6, show significant sensitivity in the mean and variability of SST and sea ice to modest changes in flux forcing ( [[#Tsujino--2020|Tsujino et al., 2020]] ). Finally, there is still considerable disagreement between different parametrizations of air–sea fluxes used in models and strong scatter in direct observations ( [[#Renault--2016|Renault et al., 2016]] ; [[#Brodeau--2017|Brodeau et al., 2017]] ). In summary, there is ''very high confidence'' that air–sea heat flux and stress biases are reduced in coupled models with high ocean resolution over coarse-resolution models, although the effect on trends remain unclear.

<div id="9.2.1.3" class="h3-container"></div>

<span id="upper-ocean-stratification-and-surface-mixed-layers"></span>
==== 9.2.1.3 Upper-ocean Stratification and Surface Mixed Layers ====

<div id="h3-3-siblings" class="h3-siblings"></div>

The density difference from surface to deep ocean is the upper-ocean stratification. The AR5 ( [[#Rhein--2013|Rhein et al., 2013]] ) assessed that it is ''very likely'' that the thermal contribution to stratification over the fixed 0–200 m layer increased by about 1% per decade between 1971 and 2010 (based on linear trend consistently across reports). The SROCC ( [[#Bindoff--2019|Bindoff et al., 2019]] ) found it ''very likely'' that density stratification increased by 0.46–0.51% per decade between 60°S and 60°N from 1970 to 2017). New published estimates based on a variety of different interpolated observations show that SROCC assessed rate is too low, even using the same data and methods ( [[#Li--2020|]] [[#Li--2020|]] [[#Li--2020|Li et al., 2020]] ). The 1960–2018 stratification increase is estimated at 1.2 ± 0.1% per decade from the IAP dataset, 1.2 ± 0.4% per decade from the Ishii product, 0.7 ± 0.5% per decade from the EN4 dataset, 0.9 ± 0.5% per decade from ORAS4, and 1.2 ±0.3% per decade from the National Centers for Environmental Information (NCEI) dataset (G. [[#Li--2020|]] [[#Li--2020|]] [[#Li--2020|Li et al., 2020]] ). The improved methodology for computing stratification change on individual profiles before gridding yields a global annual mean increase of 0–200 m stratification change of 0.8 ± 0.2% per decade between 1960 and 2018 ( [[#Yamaguchi--2019|Yamaguchi and Suga, 2019]] ) and a global summer mean increase of 0–200 m stratification change of 1.3 ± 0.3% per decade between 1970 and 2018 ( [[#Sallée--2021|Sallée et al., 2021]] ) is of a similar magnitude to the long-term trend ( [[#Yamaguchi--2019|Yamaguchi and Suga, 2019]] ; G. [[#Li--2020|]] [[#Li--2020|]] [[#Li--2020|Li et al., 2020]] ). In summary, there is ''limited evidence'' that focusing on changes over a fixed depth range might hide larger increases occurring at the seasonally and regionally variable pycnocline depth. There is also ''limited evidence'' that summer stratification change within the pycnocline has occurred at a rate of 8.9 ± 2.7% per decade from 1970 to 2018, and ''limited evidence'' of a winter pycnocline stratification increase ( [[#Cummins--2020|Cummins and Ross, 2020]] ; [[#Sallée--2021|Sallée et al., 2021]] ).

While AR5 and SROCC did not assess change in mixed-layer depth, the reported changes in stratification can modulate the surface mixed-layer depth, which is set by a balance between fluxes and dynamical mixing (winds, tides, waves, convection) acting against the background stratification and restratification processes (solar and dynamical). Despite the large stratification increase observed at a global scale, new evidence shows that summer mixed-layer depth deepened consistently over the globe at a rate of 2.9 ± 0.5% per decade from 1970 to 2018, with the largest deepening observed in the Southern Ocean, corresponding to overall deepening from 3–15 m per decade depending on region ( [[#Somavilla--2017|Somavilla et al., 2017]] ; [[#Sallée--2021|Sallée et al., 2021]] ). While the shorter observational record in winter (compared to summer) does not allow global winter mixed-layer trends to be reliably assessed ( [[#Sallée--2021|Sallée et al., 2021]] ), winter mixed-layer depths deepening at rates of 10 m per decade have been reported at individual long-term mid-latitude monitoring sites ( [[#Somavilla--2017|Somavilla et al., 2017]] ). Projections agree that shoaling of mixed-layer depth is expected in the 21st century, but only for strong emissions scenarios, and only in some regions (Figure 9.5). In summary, there is ''limited'' observational ''evidence'' that the mixed layer is globally deepening, while models show no emergence of a trend until later in the 21st century under strong emissions.

<div id="_idContainer016" class="Basic-Text-Frame"></div>

[[File:ebaf291d5789321709688b78e3a6575a IPCC_AR6_WGI_Figure_9_5.png]]

'''Figure''' '''9.5 |''' '''Mixed-layer depth in (a–d) winter and (e–h) summer. (a, e)''' Observed climatological mean mixed-layer depth (based on density threshold) from the Argo Mixed Layer Depth Climatology ( [[#Holte--2017|Holte et al., 2017]] ) using observations for 2000–2019. '''(b, f)''' Bias between the observation-based estimate (2000–2019) and the 1995–2014 Coupled Model Intercomparison Project Phase 6 (CMIP6) climatological mean mixed-layer depth. '''(c, d, g, h)''' Projected mixed-layer depth (MLD) change from 1995–2014 to 2081–2100 under '''(c, g)''' SSP1-2.6 and '''(d, h)''' SSP5-8.5 scenarios. The '''(a–d)''' winter row shows December–January–February (DJF) in the Northern Hemisphere and June–July–August (JJA) in the Southern Hemisphere; The '''(e–h)''' summer row shows JJA in the Northern Hemisphere and DJF in the Southern Hemisphere. The mixed-layer depth is the depth where the potential density is 0.03 kg m <sup>–3</sup> denser than at 10 m. No overlay indicates regions with high model agreement, where ≥80% of models agree on the sign of change. Diagonal lines indicate regions with low model agreement, where &lt;80% of models agree on the sign of change (see Cross-Chapter Box Atlas.1 for more information). Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).

The SROCC assessed that upper-ocean stratification will continue to increase in the 21st century under increased radiative forcing ( ''high confidence'' ), due to increased surface temperature and high-latitude surface freshening ( [[#Bindoff--2019|Bindoff et al., 2019]] ). New climate model simulations concur with SROCC assessment of a future increase of the 0–200 m stratification under increased radiative forcing in all regions of the world ocean ( [[#Kwiatkowski--2020|Kwiatkowski et al., 2020]] ). In addition, CMIP6 climate models project a shallowing of the mixed-layer in summer and winter by the end of the century under increased radiative forcing (Figure 9.5; [[#Kwiatkowski--2020|Kwiatkowski et al., 2020]] ), with the exception of the Arctic showing deepening of the mixed layer as a result of sea ice retreat (Figure 9.5; [[#Lique--2018|Lique et al., 2018]] ). The regions of largest shallowing are associated with the deepest climatological mixed layer, in both winter and summer, particularly affecting the North Atlantic and the Southern Ocean basins (Figure 9.5). While CMIP6 models tend to project shallowing mixed layers under a warming climate, except at high latitudes (Figure 9.5; [[#Lique--2018|Lique et al., 2018]] ; [[#Kwiatkowski--2020|Kwiatkowski et al., 2020]] ), a deepening in the summer mixed-layer depth by intensification of the surface winds and storms may explain inconsistency among models in many regions (Figure 9.5; [[#Young--2019|Young and Ribal, 2019]] ), although model mixed-layer biases are large in the summer in the Southern Ocean ( [[#Belcher--2012|Belcher et al., 2012]] ; [[#Sallée--2013a|Sallée et al., 2013a]] ; [[#Li--2016|Q. Li et al., 2016]] ; [[#Tsujino--2020|Tsujino et al., 2020]] ). Lack of observed ocean turbulence and climate model limitations do not allow for direct assessment of ocean surface turbulence change and limit confidence in past and future mixed-layer change. Understanding of turbulent processes, their representation in ocean and climate models, and their effect on mixed-layer biases have been an active and rapidly evolving topic of research since AR5 ( [[#Buckingham--2019|Buckingham et al., 2019]] ; Q. [[#Li--2019|]] [[#Li--2019|]] [[#Li--2019|Li et al., 2019]] ). Small-scale mixed-layer processes are not resolved in climate models ( [[#D’Asaro--2014|D’Asaro, 2014]] ; [[#Buckingham--2019|Buckingham et al., 2019]] ; [[#McWilliams--2019|McWilliams, 2019]] ) and despite significant improvements in their parametrization over the last decade ( [[#Fox-Kemper--2011|Fox-Kemper et al., 2011]] ; [[#Jochum--2013|Jochum et al., 2013]] ; [[#Li--2016|Q. Li et al., 2016]] , 2019; [[#Qiao--2016|Qiao et al., 2016]] ) and significant improvement in some models ( [[#Li--2017|Li and Fox-Kemper, 2017]] ; [[#Dunne--2020|Dunne et al., 2020]] ), biases in mixed-layer representation generally persist ( [[#Heuzé--2017|Heuzé, 2017]] ; [[#Williams--2018|Williams et al., 2018]] ; [[#Cherchi--2019|Cherchi et al., 2019]] ; [[#Golaz--2019|Golaz et al., 2019]] ; [[#Voldoire--2019|Voldoire et al., 2019]] ; [[#Yukimoto--2019|Yukimoto et al., 2019]] ; [[#Boucher--2020|Boucher et al., 2020]] ; [[#Danabasoglu--2020|Danabasoglu et al., 2020]] ; [[#Dunne--2020|Dunne et al., 2020]] ; [[#Kelley--2020|Kelley et al., 2020]] ). In summary, the representation of upper-ocean stratification and mixed layers has improved in CMIP6 compared to CMIP5. While it is ''virtually certain'' that the global mean upper ocean will continue to stratify in the 21st century, there is only ''low confidence'' in the future evolution of mixed-layer depth, which is projected to mostly shoal under high emissions, except in high-latitude regions where sea ice retreats.

<div id="box-9.2" class="h2-container box-container"></div>

'''Box 9.2 | Marine Heatwaves'''

<div id="h2-11-siblings" class="h2-siblings"></div>

Marine heatwaves (MHW) are periods of extreme high sea temperature relative to the long-term mean seasonal cycle ( [[#Hobday--2016|Hobday et al., 2016]] ). Studies since the Special Report on the Ocean and Cryosphere in a Changing Climate (SROCC; [[#Collins--2019|Collins et al., 2019]] ) confirm the assessment that MHW can lead to severe and persistent impacts on marine ecosystems – from mass mortality of benthic communities, including coral bleaching, changes in phytoplankton blooms, shifts in species composition and geographical distribution, and toxic algal blooms, to decline in fisheries catch and mariculture ( [[#Smale--2019|Smale et al., 2019]] ; [[#Cheung--2020|Cheung and Frölicher, 2020]] ; [[#Hayashida--2020|Hayashida et al., 2020]] ; [[#Piatt--2020|Piatt et al., 2020]] ). Unlike synoptic atmospheric heatwaves [[IPCC:Wg1:Chapter:Chapter-11#11.3|Section 11.3]] ), MHWs can extend for millions of square kilometres, persist for weeks to months, and occur at subsurface ( [[#Bond--2015|Bond et al., 2015]] ; [[#Schaeffer--2017|Schaeffer and Roughan, 2017]] ; [[#Perkins-Kirkpatrick--2019|Perkins-Kirkpatrick et al., 2019]] ; [[#Laufkötter--2020|Laufkötter et al., 2020]] ).

The SROCC established that MHWs have occurred in all basins over the last decades. Additional evidence documenting widespread occurrence of marine heat waves in all basins and marginal seas continues to accumulate (Y. [[#Li--2019|]] [[#Li--2019|]] [[#Li--2019|Li et al., 2019]] ; [[#Yao--2020|Yao et al., 2020]] ). The SROCC highlighted the role of large-scale climate modes of variability in amplifying or suppressing MHW occurrences, which has since been further corroborated, increasing confidence in climate modes as important drivers of MHWs ( [[#Holbrook--2019|Holbrook et al., 2019]] ; [[#Sen%20Gupta--2020|Sen Gupta et al., 2020]] ). More generally, understanding of processes leading to MHWs has increased since SROCC, including air–sea heat flux [[#9.2.1.2|Section 9.2.1.2]] ), increased horizontal heat advection, shoaling of the mixed-layer and suppressed mixing processes [[#9.2.1.3|Section 9.2.1.3]] ), reduced coastal upwelling and Ekman pumping [[#9.2.3.5|Section 9.2.3.5]] ), changes in eddy activities and planetary waves, and the re-emergence of warm subsurface anomalies ( [[#Holbrook--2020|Holbrook et al., 2020]] ; [[#Sen%20Gupta--2020|Sen Gupta et al., 2020]] ).

The SROCC reported with ''high confidence'' that MHWs – defined as days exceeding the 99th percentile in sea surface temperature (SST) from 1982 to 2016 – have ''very likely'' doubled in frequency between 1982 and 2016. Additional observation-based evidence and acquisition of longer observation time series since SROCC have confirmed and expanded on this assessment: since the 1980s MHWs have also become more intense and longer ( [[#Frölicher--2018|Frölicher and Laufkötter, 2018]] ; [[#Smale--2019|Smale et al., 2019]] ; [[#Laufkötter--2020|Laufkötter et al., 2020]] ). Satellite observations and reanalyses of SST show an increase in intensity of 0.04°C per decade from 1982 to 2016, an increase in spatial extent of 19% per decade from 1982 to 2016, and an increase in annual MHW days of 54% between the 1987–2016 period compared to 1925–1954 ( [[#Frölicher--2018|Frölicher et al., 2018]] ; [[#Oliver--2019|Oliver, 2019]] ). The SROCC assessed that 84–90% of all MHWs that occurred between 2006 and 2015 are ''very likely'' caused by anthropogenic warming. There is new evidence since SROCC that the frequency of the most impactful marine heatwaves over the last few decades has increased more than 20-fold because of anthropogenic global warming ( [[#Laufkötter--2020|Laufkötter et al., 2020]] ). In summary, there is ''high confidence'' that MHWs have increased in frequency over the 20th century, with an approximate doubling from 1982 to 2016, and ''medium confidence'' that they have become more intense and longer since the 1980s.

Consistent with SROCC, future MHWs are defined with reference to the historical climate conditions. The SROCC assessed that MHWs will ''very likely'' further increase in frequency, duration, spatial extent and intensity under future global warming in the 21st century. The CMIP6 projections allow us to confirm this assessment and quantify future change based on global mean probability ratio change (Box 9.2, Figure 1): they project MHWs will become four times (5–95% range: 2–9 times] more frequent in 2081–2100 compared to 1995–2014 under SSP1-2.6, or eight times (3–15 times) more frequent under SSP5-8.5. The SROCC highlighted that future change of MHWs will not be globally uniform, with the largest changes in the frequency of marine heatwaves being projected to occur in the western tropical Pacific and the Arctic Ocean ( ''medium confidence'' ). New evidence from the latest generation of climate models confirms and complements SROCC assessment (Box 9.2, Figure 1). Moderate increases are projected for mid-latitudes, and only small increases are projected for the Southern Ocean ( ''medium confidence'' ) ( [[#Hayashida--2020|Hayashida et al., 2020]] ). While under the SSP5-8.5 scenario, permanent MHWs (more than 360 days per year) are projected to occur in the 21st century in parts of the tropical ocean, the Arctic Ocean and around 45°S, the occurrence of such permanent MHWs can largely be avoided under the SSP1-2.6 scenario ( [[#Frölicher--2018|Frölicher et al., 2018]] ; [[#Oliver--2019|Oliver et al., 2019]] ; [[#Plecha--2020|Plecha and Soares, 2020]] ). The resolution of current climate models (CMIP5 and CMIP6) capture the broad features of MHWs, but they may have a bias towards weaker and longer MHWs in the historical period ( ''medium confidence'' ) ( [[#Frölicher--2018|Frölicher et al., 2018]] ; [[#Pilo--2019|Pilo et al., 2019]] ; [[#Plecha--2020|Plecha and Soares, 2020]] ) and greater intensification in western boundary current regions ( [[#Hayashida--2020|Hayashida et al., 2020]] ).

<div id="_idContainer018" class="Basic-Text-Frame"></div>

Box 9.2

[[File:4de4e488565e4e63a395c016fcb1136e IPCC_AR6_WGI_Box_9_2_Figure_1.png]]

'''Box 9.2, Figur'''

'''e 1 |'''

'''Observed and simulated regional probability ratio of marine heatwaves (MHWs) for the 198''' '''5–2''' '''014 period and for the end of the 21st century under two different greenhouse gas emissions scenarios.''' The probability ratio is the proportion by which the number of MHW days per year has increased relative to pre-industrial times. An MHW is defined as a deviation beyond the daily 99th percentile (11-day window) in the deseasonalized sea surface temperature. '''(a)''' The MHW probability ratio from satellite observations (NOAA OISST V2.1; Huang et al. 2020) during 1985–2014. The mean warming pattern (difference in ERSST5 (Huang et al. 2017) sea surface temperature between the 1985–2014 and 1854–1900 periods) has been added to the satellite observations to calculate the probability ratio. '''(b–d)''' Coupled Model Intercomparison Project Phase 6 (CMIP6) simulated multi-model mean probability ratio of the '''(b)''' 1985–2014 period, and 2081–2100 period in the '''(c)''' SSP1 2.6 and '''(d)''' SSP5 8.5 scenarios. The areas with grey diagonal lines in (d) indicate permanent MHWs (&gt;360 heatwave days per year). These 14 CMIP6 models are included in the analysis: ACCESS-CM2, CESM2, CESM2-WACCM, CMCCCM2-SR5, CNRM-CM6-1, CNRM-ESM2-1, CanESM5, EC-Earth3, IPSL-CM6A-LR, MIROC6, MRI-ESM2-0, NESM3, NorESM2-LM, NorESM2-MM. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).

<div id="9.2.2" class="h2-container"></div>

<span id="changes-in-heat-and-salinity"></span>