Editing IPCC:AR6/WGI/Chapter-3 (section)

=== 3.5.1 Ocean Temperature ===

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

Ocean temperature and ocean heat content are key physical variables considered for climate model evaluation and are primary indicators of a changing ocean climate. This section assesses the performance of climate models in representing the mean state ocean temperature and heat content ( [[#3.5.1.1|Section 3.5.1.1]] ), with a particular focus on the tropical oceans given the importance of air-sea coupling in these areas ( [[#3.5.1.2|Section 3.5.1.2]] ). This is followed by an assessment of detection and attribution studies of changes in ocean temperature and heat content ( [[#3.5.1.3|Section 3.5.1.3]] ). Changes in global surface temperature are assessed in [[#3.3.1.1|Section 3.3.1.1]] .

<div id="3.5.1.1" class="h3-container"></div>

<span id="sea-surface-and-zonal-mean-ocean-temperature-evaluation"></span>
==== 3.5.1.1 Sea Surface and Zonal Mean Ocean Temperature Evaluation ====

<div id="h3-17-siblings" class="h3-siblings"></div>

In CMIP3 and CMIP5 models, large SST biases were found in the mid- and high latitudes ( [[#Flato--2013|Flato et al., 2013]] ). In CMIP6, the Northern Hemisphere mid-latitude surface temperature biases appear to be marginally improved in the multi-model mean when contrasted to CMIP5 despite large biases remaining in a few models (Figures 3.23a and 3.24). There is a decreased spread of the zonal mean SST error between 50°N and 30°S, relative to CMIP5 (Figure 3.24a). On the other hand, the Southern Ocean’s warm surface temperature bias remains (Figure 3.23a; [[#Beadling--2020|Beadling et al., 2020]] ), and is on average larger in CMIP6 than in CMIP5 models (Figures 3.23a and 3.24). This warm bias is often associated with persistent overlying atmospheric cloud biases ( [[#Hyder--2018|Hyder et al., 2018]] ). Several other large biases also appear to remain largely unresolved in CMIP6, particularly warm biases in excess of 1°C along the equatorial eastern continental boundaries of the tropical Atlantic and Pacific Oceans (Figure 3.23a).

<div id="_idContainer057" class="•-2-columns"></div>

[[File:75e0df52ad6c606b3aafe157c00f8761 IPCC_AR6_WGI_Figure_3_23.png]]

Figure 3.23 | '''Multi-model mean bias of (a) sea surface temperature and (b) near-surface salinity, defined as the difference between the CMIP6 m''' '''ulti-mo''' '''del mean and the climatology from the World Ocean Atlas 2018.''' The CMIP6 multi-model mean is constructed with one realization of 46 CMIP6 historical experiments for the period 1995–2014 and the climatology from the World Ocean Atlas 2018 is an average over all available years (1955–2017). Uncertainty is represented using the advanced approach: No overlay indicates regions with robust signal, where ≥66% of models show change greater than the variability threshold and ≥80% of all models agree on sign of change; diagonal lines indicate regions with no change or no robust signal, where &lt;66% of models show a change greater than the variability threshold; crossed lines indicate regions with conflicting signal, where ≥66% of models show change greater than the variability threshold and &lt;80% of all models agree on sign of change. For more information on the advanced approach, please refer to Cross-Chapter Box Atlas.1. Further details on data sources and processing are available in the chapter data table (Table 3.SM.1).

<div id="_idContainer058" class="_idGenObjectStyleOverride-1"></div>

[[File:37a6a2360493c05b19a9bdb94f3e2487 IPCC_AR6_WGI_Figure_3_24.png]]

'''Figure 3.24 |''' '''Biases in zonal mean and equatorial sea surface temperature (SST) in CMIP5 and CMIP6 models.''' CMIP6 (red), CMIP5 (blue) and HighResMIP (green) multi-model mean '''(a)''' zonally averaged SST bias; '''(b)''' equatorial SST bias; and '''(c)''' equatorial SST compared to observed mean SST (black line) for 1979–1999. The inter-model 5th and 95th percentiles are depicted by the respective shaded range. Model climatologies are derived from the 1979–1999 mean of the historical simulations, using one simulation per model. The Hadley Centre Sea Ice and Sea Surface Temperature version 1 (HadISST) ( [[#Rayner--2003|Rayner et al., 2003]] ) observational climatology for 1979–1999 is used as the reference for the error calculation in (a) and (b); and for observations in (c). Further details on data sources and processing are available in the chapter data table (Table 3.SM.1).

Overall, the simulated and observed trends in SST patterns are generally consistent for the historical period ( [[#Olonscheck--2020|Olonscheck et al., 2020]] ). The CMIP6 models generally represent the observed pattern of trends better than the CMIP5 models, and observed trends fall within the range of simulated trends over a larger area for CMIP6 models than for CMIP5 models ( [[#Olonscheck--2020|Olonscheck et al., 2020]] ).

The CMIP5 multi-model mean zonally averaged subsurface ocean temperature showed warm biases between 200 m and 2000 m (mid-depth) over most latitudes, with exceptions in the Southern Ocean (&gt;60°S, 100–2000 m) and upper (0–400 m) Arctic Ocean. Cold biases were simulated near the surface (0–200 m) at most latitudes ( [[#Flato--2013|Flato et al., 2013]] ). CMIP6 biases are broadly consistent with those reported in CMIP5 for the near-surface (&lt;200 m) and mid-depth (200–2000 m) ocean ( [[#Voldoire--2019b|Voldoire et al., 2019b]] ; [[#Beadling--2020|Beadling et al., 2020]] ; [[#Zhu--2020|]] [[#Zhu--2020|Y. Zhu et al., 2020]] ). The warm bias begins between 100 and 400 m depth in all three basins, however, it is most prominent in the Atlantic Ocean, with a maximum magnitude in the equatorial latitudes, as in CMIP5 (Figure 3.25). In the Pacific, the large warm biases are mostly seen in the subtropical regions (30°N–60°N and 30°S–60°S). The cool near surface tropical bias is most prominent in the Pacific Ocean and also present in the Atlantic, with a smaller magnitude (Figure 3.25). Relative to CMIP5, the most prominent difference is an increase to the mid-depth (300–2000 m) warm bias in CMIP6 and a change in sign of the bias from cold to warm for the Southern Ocean mid-depth (&gt;60°S) from CMIP5 to CMIP6 (Figure 3.25). Compared to CMIP3 and CMIP5, there is improved agreement between most CMIP6 models and observations in their representation of the zonal mean temperature of the upper 100 m of the Southern Ocean ( [[#Beadling--2020|Beadling et al., 2020]] ).

<div id="_idContainer060" class="•-2-columns"></div>

[[File:e5ffa930cefe2c9e979f700839ee286a IPCC_AR6_WGI_Figure_3_25.png]]

'''Figure 3.25 |''' '''CMIP6 potential temperature and salinity biases for the global ocean, Atlantic Ocean, Pacific Ocean and Indian Ocean.''' Shown in colour are the time-mean differences between the CMIP6 historical multi-model climatological mean and observations, zonally averaged for each basin (excluding marginal and regional seas). The observed climatological values are obtained from the World Ocean ( [[IPCC:Wg1:Chapter:Atlas|Atlas]] 2018 (WOA18, 1981–2010; Prepared by the Ocean Climate Laboratory, National Oceanographic Data Center, Silver Spring, MD, USA), and are shown as labelled black contours for each of the basins. The simulated annual mean climatologies for 1981 to 2010 are calculated from available CMIP6 historical simulations, and the WOA18 climatology utilized synthesized observed data from 1981 to 2010. Output from a total of 30 available CMIP6 models is used for the temperature panels (left column) and 28 models for the salinity panels (right column). Potential temperature units are °C and salinity units are the Practical Salinity Scale 1978 [PSS-78]. Further details on data sources and processing are available in the chapter data table (Table 3.SM.1).

Focusing on the deep ocean (&gt;2000 m), the CMIP6 ensemble mean shows a prominent and consistent warm bias (Figure 3.25), in all basins except the equatorial and northern Pacific, which contrasts to a cold bias seen in CMIP5 ( [[#Flato--2013|Flato et al., 2013]] ). We note that while an updated observational temperature dataset is used in this assessment (WOA09 was used in AR5, while WOA18, 1981–2010 is used in AR6), the deep-ocean warm bias remains and is approaching double the magnitude (about 0.5°C) of the equivalent CMIP5 multi-model mean bias, a feature which is particularly prominent in the Atlantic and southern Indian Oceans. Increased horizontal resolution as well as the choice of the vertical coordinate are reported to partly improve these biases in some models ( [[#Adcroft--2019|Adcroft et al., 2019]] ; [[#Rackow--2019|Rackow et al., 2019]] ; [[#Hewitt--2020|Hewitt et al., 2020]] ).

Since AR5, there has been growing evidence that the representation of mean surface and deeper ocean temperatures in coupled climate models can be improved by increasing the horizontal resolution both in the ocean and the atmosphere (e.g., [[#Small--2014|Small et al., 2014]] ; [[#Hewitt--2016|Hewitt et al., 2016]] ; [[#Iovino--2016|Iovino et al., 2016]] ; [[#Roberts--2019|Roberts et al., 2019]] ). At an ocean resolution of around 1°, which is typical of CMIP6 models, some processes are parameterized rather than explicitly resolved, leading to a compromise in their dynamical representation. An increase in the model resolution allows for processes to be explicitly resolved, and can for example, enhance the simulation of eddies, thus improving simulated vertical eddy transport, and reducing temperature drifts in the deeper ocean ( [[#Griffies--2015|Griffies et al., 2015]] ; [[#von%20Storch--2016|von Storch et al., 2016]] ). For some models, the mean absolute error in ocean temperature below 500 m is smaller in the high resolution version compared to the low resolution version, particularly in eddy-active regions such as the North Atlantic ( [[#Rackow--2019|Rackow et al., 2019]] ). Increasing the horizontal resolution of individual climate models often leads to an overall decrease in the surface temperature biases over regions where they persisted through earlier CMIP generations, such as the central and western equatorial Pacific, as well as the North and tropical Atlantic (Figure 3.3e; [[#Roberts--2019|Roberts et al., 2019]] ; [[#Hewitt--2020|Hewitt et al., 2020]] ). Despite this, as a group the four HighResMIP models included in Figures 3.3e and 3.24 do not on average show smaller SST biases than the CMIP6 multi-model mean, demonstrating the importance of factors other than resolution in contributing to SST biases.

In summary, there is little improvement in the multi-model mean sea surface and zonal mean ocean temperatures from CMIP5 to CMIP6 ( ''medium confidence'' ). Nevertheless, the CMIP6 models show a somewhat more realistic pattern of SST trends ( ''low confidence'' ).

<div id="3.5.1.2" class="h3-container"></div>

<span id="tropical-sea-surface-temperature-evaluation"></span>
==== 3.5.1.2 Tropical Sea Surface Temperature Evaluation ====

<div id="h3-18-siblings" class="h3-siblings"></div>

<div id="3.5.1.2.1" class="h4-container"></div>

<span id="tropical-pacific-ocean"></span>
===== 3.5.1.2.1 Tropical Pacific Ocean =====

<div id="h4-10-siblings" class="h4-siblings"></div>

In CMIP5, mean state biases in the tropical Pacific Ocean including the excessive equatorial cold tongue, erroneous mean thermocline depth and slope along the equator remained but were improved relative to CMIP3 ( [[#Flato--2013|Flato et al., 2013]] ). Misrepresentation of the interaction between the atmosphere and ocean via the Bjerknes feedback and vertical mixing parameterizations, and a bias in winds were among the suggested reasons for the persistent biases ( [[#Li--2014|Li et al., 2014]] ; [[#Zhu--2018|Zhu and Zhang, 2018]] ). Moving to CMIP6, a reduction of the cold bias in the equatorial cold tongue in the central Pacific is found on average in the CMIP6 models (Figure 3.24b; [[#Grose--2020|Grose et al., 2020]] ; [[#Planton--2021|Planton et al., 2021]] ), however, this reduced bias is not statistically significant when considered across the multi-model ensemble ( [[#Planton--2021|Planton et al., 2021]] ). It is also noteworthy that the longitude of the 28°C isotherm is closer to observed in CMIP6 than in CMIP5, with a coincident reduction in the CMIP6 inter-model standard deviation ( [[#Grose--2020|Grose et al., 2020]] ). The latter result implies that there is an improvement in the representation of the tropical Pacific mean state in CMIP6 models. Comparison of biases in individual HighResMIP models with biases in lower resolution versions of the same models indicates that there is no consistent improvement in SST biases in most of the equatorial Pacific with resolution (Figure 3.3e; [[#Bock--2020|Bock et al., 2020]] ).

<div id="3.5.1.2.2" class="h4-container"></div>

<span id="tropical-atlantic-ocean"></span>
===== 3.5.1.2.2 Tropical Atlantic Ocean =====

<div id="h4-11-siblings" class="h4-siblings"></div>

Fundamental features such as the mean zonal SST gradient in the tropical Atlantic were not reproduced in CMIP5 models. Studies have proposed that weaker than observed alongshore winds, underestimation of stratocumulus clouds, coarse model resolution, and insufficient oceanic cooling due to a deeper thermocline depth and weak vertical velocities at the base of the mixed layer in the eastern basin, underpinned these tropical Atlantic SST gradient biases ( [[#Hourdin--2015|Hourdin et al., 2015]] ; [[#Richter--2015|Richter, 2015]] ). The SST gradient biases still remain in CMIP6. On average the cold bias in the western part of the basin is reduced while the warm bias in the eastern part has slightly increased (Figure 3.24b,c; [[#Richter--2020|Richter and Tokinaga, 2020]] ). Several CMIP6 models, however, display large reductions in biases of the zonal SST gradient, such that the eastern equatorial Atlantic warm SST bias and associated westerly wind biases are mostly eliminated in these models ( [[#Richter--2020|Richter and Tokinaga, 2020]] ). The high resolution (HighResMIP) CMIP6 models show a better representation of the zonal SST gradient (Figure 3.24b,c), but some lower resolution models also perform well, suggesting that resolution is not the only factor responsible for biases in Tropical Atlantic SST ( [[#Richter--2020|Richter and Tokinaga, 2020]] ).

<div id="3.5.1.2.3" class="h4-container"></div>

<span id="tropical-indian-ocean"></span>
===== 3.5.1.2.3 Tropical Indian Ocean =====

<div id="h4-12-siblings" class="h4-siblings"></div>

The tropical Indian Ocean mean state is reasonably well simulated both in CMIP5 and CMIP6 (Figure 3.24b,c). However, CMIP5 models show a large spread in the thermocline depth, particularly in the equatorial part of the basin ( [[#Saji--2006|Saji et al., 2006]] ; [[#Fathrio--2017b|Fathrio et al., 2017b]] ), which has been linked to the parameterization of the vertical mixing and the wind structure, leading to a misrepresentation of the ventilation process in some models ( [[#Schott--2009|Schott et al., 2009]] ; [[#Richter--2015|Richter, 2015]] ; [[#Shikha--2018|Shikha and Valsala, 2018]] ). A common problem with the CMIP5 models is therefore a warm bias in the subsurface, mainly at depths around the thermocline, which is also apparent in the CMIP6 models (Figure 3.25g).

In the CMIP6 multi-model mean, the western tropical Indian Ocean shows a slightly larger warm bias compared to CMIP5 (Figure 3.24 b,c), which in part could be related to excessive supply of warm water from the Red Sea ( [[#Grose--2020|Grose et al., 2020]] ; [[#Semmler--2020|Semmler et al., 2020]] ). The HighResMIP models show decreases in SST bias across the Indian Ocean with increasing resolution (Figure 3.3e; [[#Bock--2020|Bock et al., 2020]] ), though as a group the SST biases in the HighResMIP models are no smaller than those of the full CMIP6 ensemble.

<div id="3.5.1.2.4" class="h4-container"></div>

<span id="summary-1"></span>
===== 3.5.1.2.4 Summary =====

<div id="h4-13-siblings" class="h4-siblings"></div>

In summary, the structure and magnitude of multi-model mean ocean temperature biases have not changed substantially between CMIP5 and CMIP6 ( ''medium confidence'' ). Although biases remain in the latest generation models, the broad consistency between the observed and simulated basin-scale ocean properties suggests that CMIP5 and CMIP6 models are appropriate tools for investigating ocean temperature and ocean heat content responses to forcing. This also provides ''high confidence'' in the utility of CMIP-class models for detection and attribution studies, for both ocean heat content ( [[#3.5.1.3|Section 3.5.1.3]] ) and thermosteric sea level applications ( [[#3.5.3.2|Section 3.5.3.2]] ).

<div id="3.5.1.3" class="h3-container"></div>

<span id="ocean-heat-content-change-attribution"></span>
==== 3.5.1.3 Ocean Heat Content Change Attribution ====

<div id="h3-19-siblings" class="h3-siblings"></div>

The ocean plays an important role as the Earth’s primary energy store. The AR5 and SROCC assessed that the ocean accounted for more than 90% of the Earth’s energy change since the 1970s ( [[#Rhein--2013|Rhein et al., 2013]] ; [[#Bindoff--2019|Bindoff et al., 2019]] ). These assessments are consistent with recent studies assessed in Section 7.2 and Cross-Chapter Box 9.1, which find that 91% of the observed change in Earth’s total energy from 1971 to 2018 was stored in the ocean ( [[#von%20Schuckmann--2020|von Schuckmann et al., 2020]] ). The AR5 concluded that anthropogenic forcing has ''very likely'' made a substantial contribution to ocean warming above 700 m, whereas below 700 m, limited measurements restricted the assessment of ocean heat content changes in AR5 and prevented a robust comparison between observations and models ( [[#Bindoff--2013|Bindoff et al., 2013]] ).

With the recent increase in ocean sampling by Argo to 2000 m ( [[#Roemmich--2015|Roemmich et al., 2015]] ; [[#Riser--2016|Riser et al., 2016]] ; [[#von%20Schuckmann--2016|von Schuckmann et al., 2016]] ) and the resulting improvements in estimates of ocean heat content ( [[#Abraham--2013|Abraham et al., 2013]] ; [[#Balmaseda--2013|Balmaseda et al., 2013]] ; [[#Durack--2014b|Durack et al., 2014b]] ; [[#Cheng--2017|Cheng et al., 2017]] ; [[#von%20Schuckmann--2020|von Schuckmann et al., 2020]] ), a more quantitative assessment of the global ocean heat content changes that extends into the intermediate ocean (700–2000 m) over the more recent period (from 2005 to the present; [[#Durack--2018|Durack et al., 2018]] ) can be performed. Observed ocean heat content changes are discussed in [[IPCC:Wg1:Chapter:Chapter-2#2.3.3.1|Section 2.3.3.1]] , where it is reported that it is ''virtually certain'' that the global upper ocean (0–700 m) and ''very likely'' that the global intermediate ocean (700–2000 m) warmed substantially from 1971 to the present. Further, ocean layer warming contributions are reported as 61% (0–700 m), 31% (700–2000 m) and 8% (&gt;2000 m) for the 1971 to 2018 period (Table 2.7). CMIP5 model simulations replicate this partitioning fairly well for the industrial-era (1865 to 2017) throughout the upper (0–700 m, 65%), intermediate (700–2000 m, 20%) and deep (&gt;2000 m, 15%) layers ( [[#Gleckler--2016|Gleckler et al., 2016]] ; [[#Durack--2018|Durack et al., 2018]] ). The corresponding warming percentages for the multi-model mean of a subset of CMIP6 simulations over the 1850–2014 period are 58% for the upper, 21% for the intermediate, and 22% for the deep-ocean layers (Figure 3.26). These results are consistent with SROCC which assessed that it is ''virtually certain'' that both the upper and intermediate ocean warmed from 2004 to 2016, with an increased rate of warming since 1993 ( [[#Bindoff--2019|Bindoff et al., 2019]] ). The spatial distribution of these changes for different ocean depths is assessed in Section 9.2.2.1.

<div id="_idContainer062" class="•-2-columns"></div>

[[File:dcba862d8c947f6b3c983b6ca47cb154 IPCC_AR6_WGI_Figure_3_26.png]]

Figure 3.26 | '''Global ocean heat content in CMIP6 simulations and observations.''' Time series of observed (black) and simulated (red) global ocean heat content anomalies with respect to 1995–2014 for the full ocean depth '''(left-hand panel)''' ; upper layer: 0–700 m '''(top right-hand panel)''' ; intermediate layer: 700–2000 m '''(middle right-hand panel)''' ; and the abyssal ocean: &gt;2000 m '''(bottom right-hand panel)''' . The best estimate observations (black solid line) for the period of 1971–2018, along with ''very likely'' ranges (black shading) are from [[IPCC:Wg1:Chapter:Chapter-2#2.3.3.1|Section 2.3.3.1]] . For the models (1860–2014), ensemble members from 15 CMIP6 models are used to calculate the multi-model mean values (red solid line) after averaging across simulations for each independent model. The ''very likely'' ranges in the simulations are shown in red shading. Simulation drift has been removed from all CMIP6 historical runs using a contemporaneous portion of the linear fit to each corresponding pre-industrial control run ( [[#Gleckler--2012|Gleckler et al., 2012]] ). Units are zettajoules (ZJ; 10 <sup>21</sup> joule). Further details on data sources and processing are available in the chapter data table (Table 3.SM.1).

The multi-model means of both CMIP5 and CMIP6 historical simulations forced with time varying natural and anthropogenic forcing shows robust increases in ocean heat content in the upper (0–700 m) and intermediate (700–2000 m) ocean ( ''high confidence'' ) (Figure 3.26; [[#Cheng--2016|Cheng et al., 2016]] , [[#Cheng--2019|2019]] ; [[#Gleckler--2016|Gleckler et al., 2016]] ; [[#Bilbao--2019|Bilbao et al., 2019]] ; [[#Tokarska--2019|Tokarska et al., 2019]] ). Temporary (&lt;10 years) surface and subsurface cooling during and after large volcanic eruptions is also captured in the upper-ocean, and global mean ocean heat content ( [[#Balmaseda--2013|Balmaseda et al., 2013]] ). The ocean heat content increase is also reflected in the corresponding ocean thermal expansion which is a leading contributor to global mean sea level rise (Sections 3.5.3.2 and 9.2.4, and Box 9.1).

For the period 1971–2014, the rate of ocean heat uptake for the global ocean in the CMIP6 models is about 6.43 [2.08–8.66] ZJ yr <sup>–1</sup> , with the upper, intermediate and deeper layers respectively accounting for 68%, 16% and 16% of the full depth global heat uptake (Figure 3.26). Overall, the simulated ocean heat content changes are consistent with the updated and improved observational analyses, within the ''very likely'' uncertainty range defined for each (see also ( [[IPCC:Wg1:Chapter:Chapter-2#2.3.3.1|Section 2.3.3.1]] , Table 2.7; [[#Domingues--2008|Domingues et al., 2008]] ; [[#Purkey--2010|Purkey and Johnson, 2010]] ; [[#Levitus--2012|Levitus et al., 2012]] ; [[#Good--2013|Good et al., 2013]] ; [[#Cheng--2017|Cheng et al., 2017]] ; [[#Ishii--2017|Ishii et al., 2017]] ; [[#Zanna--2019|Zanna et al., 2019]] ) as well as with the ocean components of total Earth heating assessed in Section 7.2.2.2, Table 7.1. Nevertheless, large uncertainties remain, particularly in the deeper layers due to the poor temporal and spatial sampling coverage, particularly in the Atlantic, Southern and Indian Oceans ( [[#Garry--2019|Garry et al., 2019]] ). The ''very likely'' ranges of the simulated trends for the full ocean depth and below 2000 m fall within the ''very likely'' range of observed uptake during the last two decades. In the intermediate layer, the multi-model ensemble mean mostly stays above the observed 5th–95th percentile range before the year 2000, and below that range after 2000. For the upper ocean, some individual model realizations show a reduced ocean heat content increase during the 1970s and 1980s, which is then compensated by a greater warming than the observations from the early 1990s. These discrepancies have been linked with a temporary increase in the Southern Ocean deep water formation rate, as well as with the models’ strong aerosol cooling effects and high equilibrium climate sensitivity (see also Section 7.5.6 and Box 7.2; [[#Andrews--2019|Andrews et al., 2019]] , [[#Andrews--2020|2020]] ; [[#Golaz--2019|Golaz et al., 2019]] ; [[#Dunne--2020|Dunne et al., 2020]] ; [[#Winton--2020|Winton et al., 2020]] ).

Nevertheless, simulations show that the rate of ocean heat uptake has doubled in the past few decades, when contrasted to the rate over the complete 20th century (Figure 3.26), with over a third of the accumulated heat stored below 700 m ( [[#Cheng--2016|Cheng et al., 2016]] , [[#Cheng--2019|2019]] ; [[#Gleckler--2016|Gleckler et al., 2016]] ; [[#Durack--2018|Durack et al., 2018]] ). The Southern Ocean shows the strongest ocean heat uptake that penetrates to deeper layers (Section 9.2.3.2), whereas ocean heat content increases in the Pacific and Indian Oceans largely occur in the upper layers ( [[#Bilbao--2019|Bilbao et al., 2019]] ).

Since AR5, the attribution of ocean heat content increases to anthropogenic forcing has been further supported by more detection and attribution studies. These studies have shown that contributions from natural forcing alone cannot explain the observed changes in ocean heat content in either the upper or intermediate ocean layers, and a response to anthropogenic forcing is clearly detectable in ocean heat content ( [[#Gleckler--2016|Gleckler et al., 2016]] ; [[#Bilbao--2019|Bilbao et al., 2019]] ; [[#Tokarska--2019|Tokarska et al., 2019]] ). Moreover, a response to greenhouse gas forcing is detectable independently of the response to other anthropogenic forcings ( [[#Bilbao--2019|Bilbao et al., 2019]] ; [[#Tokarska--2019|Tokarska et al., 2019]] ), which has offset part of the greenhouse gas induced warming. Further evidence is provided by the agreement between observed and simulated changes in global thermal expansion associated with the ocean heat content increase when both natural and anthropogenic forcings are included in the simulations ( [[#3.5.3.2|Section 3.5.3.2]] ), though internal variability plays a larger role in driving basin-scale thermosteric sea level trends ( [[#Bilbao--2015|Bilbao et al., 2015]] ). Over the Southern Ocean, warming is detectable over the late 20th century and is largely attributable to greenhouse gases ( [[#Swart--2018|Swart et al., 2018]] ; [[#Hobbs--2021|Hobbs et al., 2021]] ), while other anthropogenic forcings such as ozone depletion have been shown to mitigate the warming in some of the CMIP5 simulations ( [[#Swart--2018|Swart et al., 2018]] ; [[#Hobbs--2021|Hobbs et al., 2021]] ). The use of the mean temperature above a fixed isotherm rather than fixed depth further strengthens a robust detection of the anthropogenic response in the upper ocean ( [[#Weller--2016|Weller et al., 2016]] ), and better accounting for internal variability in the upper ocean ( [[#Rathore--2020|Rathore et al., 2020]] ), helps explain reported hemispheric asymmetry in ocean heat content change ( [[#Durack--2014b|Durack et al., 2014b]] ).

In summary, there is strong evidence for an improved understanding of the observed global ocean heat content increase. It is ''extremely likely'' that human influence was the main driver of the ocean heat content increase observed since the 1970s, which extends into the deeper ocean ( ''very high confidence'' ). Updated observations, like model simulations, show that warming extends throughout the entire water column ( ''high confidence'' ).

<div id="3.5.2" class="h2-container"></div>

<span id="ocean-salinity"></span>