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

== 3.4 Human Influence on the Cryosphere ==

<div id="3.4.1" class="h2-container"></div>

<span id="sea-ice"></span>
=== 3.4.1 Sea Ice ===

<div id="h2-12-siblings" class="h2-siblings"></div>

<div id="3.4.1.1" class="h3-container"></div>

<span id="arctic-sea-ice"></span>
==== 3.4.1.1 Arctic Sea Ice ====

<div id="h3-13-siblings" class="h3-siblings"></div>

The AR5 concluded that ‘anthropogenic forcings are ''very likely'' to have contributed to Arctic sea ice loss since 1979’ ( [[#Bindoff--2013|Bindoff et al., 2013]] ), based on studies showing that models can reproduce the observed decline only when including anthropogenic forcings, and formal attribution studies. Since the beginning of the modern satellite era in 1979, Northern Hemisphere sea ice extent has exhibited significant declines in all months with the largest reduction in September (see [[IPCC:Wg1:Chapter:Chapter-2#2.3.2.1.1|Section 2.3.2.1.1]] , and Figures 3.20 and 3.21 for more details on observed changes). The recent Arctic sea ice loss during summer is unprecedented since 1850 ( ''high confidence'' ), but as in AR5 and SROCC there remains only ''medium'' ''confidence'' that the recent reduction is unique during at least the past 1000 years due to sparse observations (Sections 2.3.2.1.1 and 9.3.1.1). CMIP5 models also simulate Northern Hemisphere sea ice loss over the satellite era but with large differences among models (e.g., [[#Massonnet--2012|Massonnet et al., 2012]] ; [[#Stroeve--2012|Stroeve et al., 2012]] ). The envelope of simulated ice loss across model simulations encompasses the observed change, although observations fall near the low end of the CMIP5 and CMIP6 distributions of trends (Figure 3.20). CMIP6 models on average better capture the observed Arctic sea ice decline, albeit with large inter-model spread. [[#Notz--2020|Notz et al. (2020)]] found that CMIP6 models better reproduce the sensitivity of Arctic sea ice area to CO <sub>2</sub> emissions and global warming than earlier CMIP models although the models’ underestimation of this sensitivity remains. [[#Davy--2020|Davy and Outten (2020)]] also found that CMIP6 models can simulate the seasonal cycle of Arctic sea ice extent and volume better than CMIP5 models. For the assessment of physical processes associated with changes in Arctic sea ice, see Section 9.3.1.1.

<div id="_idContainer049" class="•-2-columns"></div>

[[File:c87f4d332e1773e0790526086cdda02f IPCC_AR6_WGI_Figure_3_20.png]]

Figure 3.20 | '''Mean (x-axis) and trend (y-axis) of Arctic sea ice area (SIA) in September (left) and Antarctic SIA in February (right) for 1979–2017 from CMIP5 (upper) and CMIP6 (lower) models.''' All individual models (ensemble means) and the multi-model mean values are compared with the observations (OSISAF, NASA Team, and Bootstrap; see Figure 9.13). Solid line indicates a linear regression slope with corresponding correlation coefficient (r) and p-value provided. Note the different scales used on the y-axis for Arctic and Antarctic SIA. Results remain essentially the same when using sea ice extent (SIE; not shown). Further details on data sources and processing are available in the chapter data table (Table 3.SM.1).

Since AR5, there have been several new detection and attribution studies on Arctic sea ice. While the attribution literature has mostly used sea ice extent (SIE), it is closely proportional to sea ice area (SIA; [[#Notz--2014|Notz, 2014]] ), which is assessed in Chapters 2 and 9 and shown in Figures 3.20 and 3.21. [[#Kirchmeier-Young--2017|Kirchmeier-Young et al. (2017)]] compared the observed time series of the September SIE over the period 1979–2012 with those from different large ensemble simulations which provide a robust sampling of internal climate variability (CanESM2, CESM1, and CMIP5) using an optimal fingerprinting technique. They detected anthropogenic signals which were separable from the response to natural forcing due to solar irradiance variations and volcanic aerosol, supporting previous findings (Figure 3.21; [[#Min--2008b|Min et al., 2008b]] ; [[#Kay--2011|Kay et al., 2011]] ; [[#Notz--2012|Notz and Marotzke, 2012]] ; [[#Notz--2016|Notz and Stroeve, 2016]] ). Using selected CMIP5 models and three independently derived sets of observations, [[#Mueller--2018|Mueller et al. (2018)]] detected fingerprints from greenhouse gases, natural, and other anthropogenic forcings simultaneously in the September Arctic SIE over the period 1953–2012. They further showed that about a quarter of the greenhouse gas induced decrease in SIE has been offset by an increase due to other anthropogenic forcing (mainly aerosols). Similarly, [[#Gagné--2017b|Gagné et al. (2017b)]] suggested that the observed increase in Arctic sea ice concentration over the 1950–1975 period was primarily due to the cooling contribution of anthropogenic aerosol forcing based on single model simulations. [[#Gagné--2017a|Gagné et al. (2017a)]] identified a detectable increase in Arctic SIE in response to volcanic eruptions using CMIP5 models and four observational datasets. [[#Polvani--2020|Polvani et al. (2020)]] suggested that ozone depleting substances played a substantial role in the Arctic sea ice loss over the 1955–2005 period.

<div id="_idContainer051" class="•-2-columns"></div>

[[File:49a07fca8f0f68d6dc97709134dee999 IPCC_AR6_WGI_Figure_3_21.png]]

Figure 3.21 | '''Seasonal evolution of observed and simulated Arctic (left) and Antarctic (right) sea ice area (SIA) over 1979–2017.''' SIA anomalies relative to the 1979–2000 means from observations '''(OBS from OSISAF, NASA Team, and Bootstrap, top)''' and historical '''(ALL, middle)''' and natural only '''(NAT, bottom)''' simulations from CMIP5 and CMIP6 models. These anomalies are obtained by computing non-overlapping three-year mean SIA anomalies for March (February for Antarctic SIA), June, September, and December separately. CMIP5 historical simulations are extended by using RCP4.5 scenario simulations after 2005 while CMIP6 historical simulations are extended by using SSP2-4.5 scenario simulations after 2014. CMIP5 NAT simulations end in 2012. Numbers in brackets represent the number of models used. The multi-model mean is obtained by taking the ensemble mean for each model first and then averaging over models. Grey dots indicate multi-model mean anomalies stronger than inter-model spread (beyond ± 1 standard deviation). Results remain very similar when based on sea ice extent (SIE – not shown). Units: 10 <sup>6</sup> km <sup>2</sup> . Further details on data sources and processing are available in the chapter data table (Table 3.SM.1) and in the caption to Figure 9.13.

Differences in sea ice loss among the models (Figure 3.20) have been attributed to a number of factors (see also Section 9.3.1.1). These factors include the late 20th century simulated sea ice state ( [[#Massonnet--2012|Massonnet et al., 2012]] ), the magnitude of changing ocean heat transport ( [[#Mahlstein--2011|Mahlstein and Knutti, 2011]] ), and the rate of global warming (e.g., [[#Gregory--2002|Gregory et al., 2002]] ; [[#Mahlstein--2012|Mahlstein and Knutti, 2012]] ; [[#Rosenblum--2017|Rosenblum and Eisenman, 2017]] ). Sea ice thermodynamic considerations indicate that the magnitude of sea ice variability and loss depends on ice thickness ( [[#Bitz--2008|Bitz, 2008]] ; [[#Massonnet--2018|Massonnet et al., 2018]] ) and hence the climatology simulated by different models may influence their simulated sea ice trends ( ''medium confidence'' ), as indicated by the regression lines in Figure 3.20.

An important consideration in comparing Arctic sea ice loss in models and observations is the role of internal variability ( ''medium confidence'' ). Using ensemble simulations from a single model, [[#Kay--2011|Kay et al. (2011)]] suggested that internal variability could account for about half of the observed September ice loss. More recently, large ensemble simulations have been performed with many more ensemble members ( [[#Kay--2015|Kay et al., 2015]] ). These enable a more robust characterization of internal variability in the presence of forced anthropogenic change. Using such large ensembles, some studies discussed the influence of internal variability on Arctic sea ice trends ( [[#Swart--2015|Swart et al., 2015]] ). [[#Song--2016|Song et al. (2016)]] also compared the trends in the forced and unforced simulations using multiple climate models and found that internal variability explains about 40% of the observed September sea ice melting trend, supporting previous studies ( [[#Stroeve--2012|Stroeve et al., 2012]] ). Based on the large ensembles of CESM1 and CanESM2, the September Arctic sea ice extent variance first increases and then decreases as SIE declines from its pre-industrial value ( [[#Kirchmeier-Young--2017|Kirchmeier-Young et al., 2017]] ; [[#Mueller--2018|Mueller et al., 2018]] ) consistent with previous work ( [[#Goosse--2009|Goosse et al., 2009]] ), but neither study found a strong sensitivity of detection and attribution results to the change in variability. Further work has indicated that internally-driven summer atmospheric circulation trends with enhanced atmospheric ridges over Greenland and the Arctic Ocean, which project on the negative phase of the North Atlantic Oscillation ( [[#3.7.1|Section 3.7.1]] ), play an important role in the observed Arctic sea ice loss ( [[#Hanna--2015|Hanna et al., 2015]] ; [[#Ding--2017|Ding et al., 2017]] ). A fingerprint analysis using the CESM large ensemble suggests that this internal variability accounts for 40–50% of the observed September Arctic sea ice decline ( [[#Ding--2019|Ding et al., 2019]] ; [[#England--2019|England et al., 2019]] ). Internally-generated decadal tropical variability and associated atmospheric teleconnections were suggested to have contributed to the changing atmospheric circulation in the Arctic and the associated rapid sea ice decline from 2000 to 2014 ( [[#Meehl--2018|Meehl et al., 2018]] ).

Some recent studies evaluated the human contribution to recent record minimum SIE events in the Arctic. Analysing CMIP5 simulations, [[#Zhang--2013|Zhang and Knutson (2013)]] found that the observed 2012 record low in September Arctic SIE is inconsistent with internal climate variability alone. Based on several large ensembles, [[#Kirchmeier-Young--2017|Kirchmeier-Young et al. (2017)]] concluded that the observed 2012 SIE minimum cannot be reproduced in a simulation excluding human influence. [[#Fučkar--2016|Fučkar et al. (2016)]] showed that climate change contributed to the record low March Arctic SIE in 2015, which was accompanied by the record minimum SIE in the Sea of Okhotsk ( [[#Paik--2017|Paik et al., 2017]] ).

Based on the new attribution studies since AR5, we conclude that it is ''very likely'' that anthropogenic forcing mainly due to greenhouse gas increases was the main driver of Arctic sea ice loss since 1979. Increases in anthropogenic aerosols have offset part of the greenhouse gas induced Arctic sea ice loss since the 1950s ( ''medium confidence'' ). Despite large differences in the mean sea ice state in the Arctic, Arctic sea ice loss is captured by all CMIP5 and CMIP6 models. Nonetheless, large inter-model differences in the Arctic sea ice decline remain, limiting our ability to quantify forced changes and internal variability contributions.

<div id="3.4.1.2" class="h3-container"></div>

<span id="antarctic-sea-ice"></span>
==== 3.4.1.2 Antarctic Sea Ice ====

<div id="h3-14-siblings" class="h3-siblings"></div>

AR5 concluded that ‘there is ''low confidence'' in the attribution of the observed increase in Antarctic SIE since 1979’ ( [[#Bindoff--2013|Bindoff et al., 2013]] ) due to the limited understanding of the external forcing contribution as well as the role of internal variability. Based on a difference between the first and last decades, Antarctic sea ice cover exhibited a small increase in summer and winter over the 1979–2017 period ( [[IPCC:Wg1:Chapter:Chapter-2#2.3.2.1.2|Section 2.3.2.1.2]] , and Figures 3.20 and 3.21). However, these changes are not statistically significant and starting in late 2016, anomalously low sea ice area has been observed ( [[IPCC:Wg1:Chapter:Chapter-2#2.3.2.1.2|Section 2.3.2.1.2]] ). The mean hemispheric sea ice changes result from much larger, but partially compensating, regional changes with increases in the western Ross Sea and Weddell Sea and declines in the Bellingshausen and Amundsen Seas ( [[#Hobbs--2016|Hobbs et al., 2016]] ). Observed regional trends have been particularly large in austral autumn (see [[IPCC:Wg1:Chapter:Chapter-2#2.3.2.1.2|Section 2.3.2.1.2]] , and also Section 9.3.2.1 for more details of regional changes and related physical processes). Starting in austral spring of 2016, the ice extent decreased strongly ( [[#Turner--2017|Turner et al., 2017]] ) and has since remained anomalously low (Figure 3.21 and Figure 2.20). This decrease has been associated with anomalous atmospheric conditions associated with teleconnections from warming in the eastern Indian Ocean and a negative Southern Annular Mode ( [[#Chenoli--2017|Chenoli et al., 2017]] ; [[#Stuecker--2017|Stuecker et al., 2017]] ; [[#Schlosser--2018|Schlosser et al., 2018]] ; [[#Meehl--2019|Meehl et al., 2019]] ; [[#Purich--2019|Purich and England, 2019]] ; G. [[#Wang--2019|]] [[#Wang--2019|Wang et al., 2019]] ). A decadal-scale warming of the near-surface ocean that resulted from strengthened westerlies may also have contributed to and helped to sustain the sea ice loss ( [[#Meehl--2019|Meehl et al., 2019]] ). Before satellites and on even longer time scales, very limited observational data and proxy coverage leads to ''low confidence'' in all aspects of Antarctic sea ice (Sections 2.3.2.1.2 and 9.3.2.1).

CMIP5 climate models generally simulate Antarctic sea ice loss over the satellite era since 1979 ( [[#Mahlstein--2013|Mahlstein et al., 2013]] ; [[#Turner--2013|Turner et al., 2013]] ) in contrast to the observed change, and CMIP6 models also simulate Antarctic ice loss ( [[#Roach--2020|Roach et al., 2020]] ; Figure 3.20 and 3.21). A number of studies have suggested that this discrepancy may be in part due to the role of internal variability in the observed change ( [[#Mahlstein--2013|Mahlstein et al., 2013]] ; [[#Polvani--2013|Polvani and Smith, 2013]] ; [[#Zunz--2013|Zunz et al., 2013]] ; [[#Meehl--2016c|Meehl et al., 2016c]] ; [[#Turner--2016|Turner et al., 2016]] ), including teleconnections associated with tropical Pacific variability ( [[#Meehl--2016c|Meehl et al., 2016c]] ) and changing surface conditions resulting from multi-decadal ocean circulation variations ( [[#Singh--2019|Singh et al., 2019]] ). However, when the spatial pattern is considered, trends in the summer and autumn (from 1979–2005) appear outside the range of internal variability ( [[#Hobbs--2015|Hobbs et al., 2015]] ). This suggests that the models may exhibit an unrealistic simulation of the Antarctic sea ice forced response or the internal variability of the system. Discrepancies among the models in simulated sea ice variability ( [[#Zunz--2013|Zunz et al., 2013]] ), the sea ice climatological state ( [[#Roach--2018|Roach et al., 2018]] ), upper ocean temperature trends ( [[#Schneider--2018|Schneider and Deser, 2018]] ), Southern Hemisphere westerly wind trends ( [[#Purich--2016|Purich et al., 2016]] ), or the sea ice response to Southern Annular Mode variations ( [[#Ferreira--2014|Ferreira et al., 2014]] ; [[#Holland--2017|Holland et al., 2017]] ; [[#Kostov--2017|Kostov et al., 2017]] ; [[#Landrum--2017|Landrum et al., 2017]] ) may all play some role in explaining these differences with the observed trends. Increased fresh water fluxes caused by mass loss of the Antarctic Ice Sheet (either by melting at the front of ice shelves or via iceberg calving) have been suggested as a possible mechanism driving the multi-decadal Antarctic sea ice expansion ( [[#Bintanja--2015|Bintanja et al., 2015]] ; [[#Pauling--2016|Pauling et al., 2016]] ) but there is a lack of consensus on this mechanism’s impacts ( [[#Pauling--2017|Pauling et al., 2017]] ). A recent study based on a decadal prediction system suggests that initializing the state of the Antarctic Bottom Water cell allows the system to reproduce the observed Antarctic sea ice increase ( [[#Zhang--2017|Zhang et al., 2017]] ), consistent with the suggestion that multi-decadal variability associated with variations in deep convection has contributed to the observed increase in Antarctic sea ice since 1979 ( [[#Latif--2013|Latif et al., 2013]] ; [[#Zhang--2017|Zhang et al., 2017]] ; L. [[#Zhang--2019|]] [[#Zhang--2019|Zhang et al., 2019]] ) (see also Section 9.3.2.1).

There have been several studies that aimed to identify causes of the observed Antarctic SIE changes. [[#Gagné--2015|Gagné et al. (2015)]] assessed the consistency of observed and simulated changes in Antarctic SIE for an extended period using recovered satellite-based estimates, and found that the observed trends since the mid-1960s are not inconsistent with model simulated trends. Studies based on the satellite period also indicate that the observed trends are largely within the range of simulated internal variability ( [[#Hobbs--2016|Hobbs et al., 2016]] ). A few distinct factors that led to the weak signal-to-noise ratio in Antarctic SIE trends have been further identified, which include large multi-decadal variability ( [[#Monselesan--2015|Monselesan et al., 2015]] ), the short observational record (e.g., [[#Abram--2013|Abram et al., 2013]] ), and the limited model performance at representing the complex Antarctic climate system as discussed above ( [[#Bintanja--2013|Bintanja et al., 2013]] ; [[#Uotila--2014|Uotila et al., 2014]] ). The short period of comprehensive satellite observations, beginning in 1979, makes it challenging to set the observed increase between 1979 and 2015, or the subsequent decrease, in a long-term context, and to assess whether the difference in trend between observations and models, which mostly simulate long-term decreases, is systematic or a rare expression of internal variability on decadal to multi-decadal time scales.

In conclusion, the observed small increase in Antarctic sea ice extent during the satellite era is not generally captured by global climate models, and there is ''low confidence'' in attributing the causes of the change.

<div id="3.4.2" class="h2-container"></div>

<span id="snow-cover"></span>
=== 3.4.2 Snow Cover ===

<div id="h2-13-siblings" class="h2-siblings"></div>

Seasonal snow cover is a defining climate feature of the northern continents. It is therefore of considerable interest that climate models correctly simulate this feature. It is discussed in more detail in Section 9.5.3, and observational aspects of snow cover are assessed in [[IPCC:Wg1:Chapter:Chapter-2#2.3.2.2|Section 2.3.2.2]] .

The AR5 noted the strong linear correlation between Northern Hemisphere snow cover extent (SCE) and annual-mean surface air temperature in CMIP5 models. It was assessed as ''likely'' that there had been an anthropogenic contribution to observed reductions in Northern Hemisphere snow cover since 1970 ( [[#Bindoff--2013|Bindoff et al., 2013]] ). The AR5 assessed that CMIP5 models reproduced key features of observed snow cover well, including the seasonal cycle of snow cover over northern regions of Eurasia and North America, but had more difficulties in more southern regions with intermittent snow cover. The AR5 also found that CMIP5 models underestimated the observed reduction in spring snow cover over this period (Figure 3.22; see also [[#Brutel-Vuilmet--2013|Brutel-Vuilmet et al., 2013]] ; [[#Thackeray--2016|Thackeray et al., 2016]] ; [[#Santolaria-Otín--2020|Santolaria-Otín and Zolina, 2020]] ). This behaviour has been linked to how the snow-albedo feedback is represented in models ( [[#Thackeray--2018a|Thackeray et al., 2018a]] ). The CMIP5 multi-model ensemble has been shown to represent the snow-albedo feedback more realistically than CMIP3, although models from some individual modelling centres have not improved or have even got worse ( [[#Thackeray--2018a|Thackeray et al., 2018a]] ). There is still a systematic overestimation of the albedo of boreal forest covered by snow ( [[#Thackeray--2015|Thackeray et al., 2015]] ; Y. [[#Li--2016|]] [[#Li--2016|Li et al., 2016]] ). Consequently, the snow albedo feedback might have been overestimated by CMIP5 models (Section 9.5.3; [[#Xiao--2017|Xiao et al., 2017]] ).

<div id="_idContainer053" class="_idGenObjectStyleOverride-1"></div>

[[File:d0f9a5bbcefe63d97aa9756c30b7ef89 IPCC_AR6_WGI_Figure_3_22.png]]

Figure 3.22 | '''Time series of Northern Hemisphere March–April mean snow cover extent (SCE) from observations, CMIP5 and CMIP6 simulations.''' The observations (grey lines) are updated Brown-NOAA (Brown and Robinson, 2011), [[#Mudryk--2020|Mudryk et al. (2020)]] , and GLDAS2. CMIP5 '''(top)''' and CMIP6 '''(bottom)''' simulations of the response to natural plus anthropogenic forcing are shown in brown, natural forcing only in green, and the pre-industrial control simulation range is presented in blue. Five-year mean anomalies are shown for the 1923–2017 period with the x-axis representing the centre years of each five-year mean. CMIP5 all forcing simulations are extended by using RCP4.5 scenario simulations after 2005 while CMIP6 all forcing simulations are extended by using SSP2-4.5 scenario simulations after 2014. Shading indicates 5th–95th percentile ranges for CMIP5 and CMIP6 all and natural forcings simulations, and solid lines are ensemble means, based on all available ensemble members with equal weight given to each model ( [[#3.2|Section 3.2]] ). The blue vertical bar indicates the mean 5th–95th percentile range of pre-industrial control simulation anomalies, based on non-overlapping segments. The numbers in brackets indicate the number of models used. Anomalies are relative to the average over 1971–2000. For models, SCE is restricted to grid cells with land fraction ≥50%. Greenland is excluded from the total area summation. Figure is modified from [[#Paik--2020|Paik and Min (2020)]] , their Figure 1. Further details on data sources and processing are available in the chapter data table (Table 3.SM.1).

CMIP6 models improve on CMIP5 models in producing slightly increased SCE versus CMIP5, correcting the low bias in CMIP5 ( [[#Mudryk--2020|Mudryk et al., 2020]] ). The linear relationship noted above between GSAT and SCE also exists in CMIP6 ( [[#Mudryk--2020|Mudryk et al., 2020]] ). Like CMIP5, the CMIP6 models capture the negative trend in spring snow cover that has occurred in recent decades (Figure 3.22). However, the median CMIP6 model now produces slightly stronger post-1981 declines in the March to April mean SCE than the CMIP5 median ( [[#Mudryk--2020|Mudryk et al., 2020]] ). Until about 1980, the models produce a generally stable March to April SCE, but after that a substantial decline, reaching a loss of about 2 × 10 <sup>6</sup> km <sup>2</sup> in 2012–2017 relative to the 1971–2000 average. Compared to earlier studies which found that models underestimate observed trends for the 1979–2005 period ( [[#Brutel-Vuilmet--2013|Brutel-Vuilmet et al., 2013]] ), both CMIP5 and CMIP6 models show improved agreement with the observations over the period to 2017 (Figure 3.22). One remaining concern is a failure of most CMIP6 models to correctly represent the relationship between snow cover extent and snow mass, reflecting too slow seasonal increases and decreases of SCE in the models ( [[#Mudryk--2020|Mudryk et al., 2020]] ).

Several CMIP5 and CMIP6 based studies have consistently attributed the observed Northern Hemisphere spring SCE changes ( [[IPCC:Wg1:Chapter:Chapter-2#2.3.2.2|Section 2.3.2.2]] ) to anthropogenic influences ( [[#Rupp--2013|Rupp et al., 2013]] ; [[#Najafi--2016|Najafi et al., 2016]] ; [[#Paik--2020|Paik and Min, 2020]] ), with the observed changes being found to be inconsistent with natural variability alone. Similarly, spring snow thickness (Snow Water Equivalent) changes on the scale of the Northern Hemisphere have been attributed to greenhouse gas forcing ( [[#Jeong--2017|Jeong et al., 2017]] ). Using individual forcing simulations from multiple CMIP6 models, [[#Paik--2020|Paik and Min (2020)]] detected greenhouse gas influence in the observed decrease of early spring SCE between 1925 and 2019, which was found to be separable from the responses to other forcings.

In summary, it is ''very'' ''likely'' that anthropogenic influence contributed to the observed reductions in Northern Hemisphere springtime snow cover since 1950. CMIP6 models better represent the seasonality and geographical distribution of snow cover than CMIP5 simulations ( ''high confidence'' ). Both CMIP5 and CMIP6 models simulate strong declines in spring SCE during recent years, in general agreement with observations, causing the multi-model mean decreasing trend in spring SCE to now better agree with observations than in earlier evaluations. Evidence has yet to emerge that interactions between vegetation and snow, found problematic in CMIP5, have improved in CMIP6 models (Section 9.5.3). Such deficiencies in the representation of snow in climate models mean there is ''medium confidence'' in the simulation of snow cover over the northern continents in CMIP6 model simulations. The models consistently link snow extent to surface air temperature (Figure 9.24). With warming of near-surface air linked to anthropogenic influence, and particularly to greenhouse gas increases ( [[#3.3.1.1|Section 3.3.1.1]] ), this provides additional evidence that reductions in snow cover are also caused by human activity.

<div id="3.4.3" class="h2-container"></div>

<span id="glaciers-and-ice-sheets"></span>
=== 3.4.3 Glaciers and Ice Sheets ===

<div id="h2-14-siblings" class="h2-siblings"></div>

While ( [[IPCC:Wg1:Chapter:Chapter-9|Chapter 9]] (Sections 9.4 and 9.5) discusses process understanding for glaciers and ice sheets, as well as evaluation of global and regional-scale glacier and ice-sheet models, our focus here is on the attribution of large-scale changes in glaciers and ice sheets. Land ice in the form of glaciers has been included in CMIP climate and Earth system models as components of the land surface models for many years. However, their representation is simplified and is omitted altogether in the less complex modelling systems. In CMIP3 ( [[#Meehl--2007|Meehl et al., 2007]] ) and CMIP5 ( [[#Taylor--2012|Taylor et al., 2012]] ) land ice area fraction, a component of land surface models, was defined as a time-independent quantity, and in most model configurations was preset at the simulation initialization as a permanent land feature. In CMIP6 considerable progress has been made in improving and evaluating the representation of modelled land ice. For glaciers, an example is the expansion of the Joint UK Land Environment Simulator (JULES) land surface model to enable elevated tiles, and hence more accurately simulate the altitudinal atmospheric effects on glaciers ( [[#Shannon--2019|Shannon et al., 2019]] ). Moreover, standalone glacier models have now been systematically compared in GlacierMIP ( [[#Hock--2019a|Hock et al., 2019a]] ; [[#Marzeion--2020|Marzeion et al., 2020]] ). The Antarctic and Greenland Ice Sheets were absent in global climate models that pre-date CMIP6 ( [[#Eyring--2016a|Eyring et al., 2016a]] ), however some preliminary analyses that used results from CMIP5 to drive standalone ice-sheet models were included in AR5 ( [[#Church--2013a|Church et al., 2013a]] ). For the first time in CMIP, the latest CMIP6 phase includes a coordinated effort to simulate temporally evolving ice sheets within the Ice Sheet Model Intercomparison Project (ISMIP6; Box 9.3; [[#Nowicki--2016|Nowicki et al., 2016]] ). Our understanding of aspects of the global water storage contained in glaciers and ice sheets, and their contribution to sea-level rise, has improved since AR5 and SROCC ( [[#Hock--2019b|Hock et al., 2019b]] ; [[#Meredith--2019|Meredith et al., 2019]] ) both in models and observations (see assessment of observations and model evaluation for the Greenland Ice Sheet in Sections 2.3.2.4.1 and 9.4.1; Antarctica in Sections 2.3.2.4.2 and 9.4.2; and glaciers in Sections 2.3.2.3 and 9.5.1).

<div id="3.4.3.1" class="h3-container"></div>

<span id="glaciers"></span>
==== 3.4.3.1 Glaciers ====

<div id="h3-15-siblings" class="h3-siblings"></div>

Glaciers are defined as perennial surface land ice masses independent of the Antarctic and Greenland Ice Sheets (Sections 9.5 and 2.3.2.3). The AR5 assessed that anthropogenic influence had ''likely'' contributed to the retreat of glaciers observed since the 1960s ( [[#Bindoff--2013|Bindoff et al., 2013]] ), based on a high level of scientific understanding and robust estimates of observed mass loss, internal variability, and glacier response to climatic drivers. The SROCC ( [[#Hock--2019b|Hock et al., 2019b]] ) concluded that atmospheric warming was ''very likely'' the primary driver of glacier recession.

Simulations of glacier mass changes under climate change rely on glacier models driven by climate model output, often in collaborative research efforts such as GlacierMIP ( [[#Hock--2019a|Hock et al., 2019a]] ; [[#Marzeion--2020|Marzeion et al., 2020]] ). The GlacierMIP project is a systematic coordinated modelling effort designed to further understanding of glacier loss using global models. While the low resolution and remaining biases of climate model-derived boundary forcing data is a limitation, the release of the Randolph Glacier Inventory ( [[#Pfeffer--2014|Pfeffer et al., 2014]] ; [[#RGI%20Consortium--2017|RGI Consortium, 2017]] ) has supported more sophisticated, systematic and comprehensive modelling of glaciers worldwide ( [[#Hock--2019a|Hock et al., 2019a]] ).

A regional study considering 85 Northern Hemisphere glacier systems concluded that there is a discernible human influence on glacier mass balance, with glacier model simulations driven by CMIP5 historical and greenhouse gas-only simulations showing a glacier mass loss, whereas those driven by natural-only forced simulations showed a net glacier growth ( [[#Hirabayashi--2016|Hirabayashi et al., 2016]] ). In addition, a study of the role of climate change in glacier retreat using a simple mass-balance model for 37 glaciers worldwide, concluded that observed length changes would not have occurred without anthropogenic climate change, with observed length variations exceeding those associated with internal variability by several standard deviations in many cases ( [[#Roe--2017|Roe et al., 2017]] ). [[#Roe--2021|Roe et al. (2021)]] used the same model to estimate that at least 85% of cumulative glacier mass loss since 1850 is attributable to anthropogenic influence. While [[#Marzeion--2014|Marzeion et al. (2014)]] found that anthropogenic influence contributed only 25 ± 35% of glacier mass loss for the period 1851–2010, their naturally-forced simulations exhibited a substantial negative mass balance, which [[#Roe--2021|Roe et al. (2021)]] argued is unrealistic. Moreover, [[#Marzeion--2014|Marzeion et al. (2014)]] estimated that anthropogenic influence contributed 69 ± 24% of glacier mass loss for the period 1991 to 2010, consistent with a progressively increasing fraction of mass loss attributable to anthropogenic influence found by [[#Roe--2021|Roe et al. (2021)]] .

In summary, considering together the SROCC assessment that atmospheric warming was ''very likely'' the primary driver of glacier recession, the results of Roe et al. (2017, 2021) and our assessment of the dominant role of anthropogenic influence in driving atmospheric warming ( [[#3.3.1|Section 3.3.1]] ), we conclude that human influence is ''very likely'' the main driver of the near-universal retreat of glaciers globally since the 1990s.

<div id="3.4.3.2" class="h3-container"></div>

<span id="ice-sheets"></span>
==== 3.4.3.2 Ice Sheets ====

<div id="h3-16-siblings" class="h3-siblings"></div>

<div id="3.4.3.2.1" class="h4-container"></div>

<span id="greenland-ice-sheet"></span>
===== 3.4.3.2.1 Greenland Ice Sheet =====

<div id="h4-8-siblings" class="h4-siblings"></div>

The AR5 assessed that it is ''likely'' that anthropogenic forcing contributed to the surface melting of the Greenland Ice Sheet since 1993 ( [[#Bindoff--2013|Bindoff et al., 2013]] ). The SROCC did not directly assess the attribution of Greenland Ice Sheet change to anthropogenic forcing, but it did assess with ''medium confidence'' that summer melting of the Greenland Ice Sheet has increased to a level unprecedented over at least the last 350 years, which is two-to-fivefold the pre-industrial level (see also [[#Trusel--2018|Trusel et al., 2018]] ).

( [[IPCC:Wg1:Chapter:Chapter-2#2.3.2.4.1|Section 2.3.2.4.1]] assesses that Greenland Ice Sheet mass loss began in the latter half of the 19th century and that the rate of loss has increased substantially since the turn of the 21st century ( ''high confidence'' ), and also notes that integration of proxy evidence and modelling indicates that the last time the rate of mass loss was similar to the 20th century rate was during the early Holocene. Models of Greenland Ice Sheet evolution are evaluated in detail in Section 9.4.1.2, which assesses that there is overall ''medium confidence'' in these models. Model evaluation of surface mass balance changes over the Greenland Ice Sheet, including regional aspects, is also assessed in Atlas.11.2.3.

Detection and attribution studies of change in the Greenland Ice Sheet remain challenging ( [[#Kjeldsen--2015|Kjeldsen et al., 2015]] ; [[#Bamber--2019|Bamber et al., 2019]] ). This is in part due to the short observational record ( [[#Shepherd--2012|Shepherd et al., 2012]] , 2018, 2020; [[#Bamber--2018|Bamber et al., 2018]] ; [[#Cazenave--2018|Cazenave et al., 2018]] ; [[#Mouginot--2019|Mouginot et al., 2019]] ; [[#Rignot--2019|Rignot et al., 2019]] ) and the challenges this poses to the evaluation of modelling efforts (Section 9.4.1.2). The latter require not only dynamic ice-sheet models, but also appropriate atmospheric and oceanic conditions to use as a boundary forcing to drive the models ( [[#Nowicki--2018|Nowicki and Seroussi, 2018]] ; [[#Barthel--2020|Barthel et al., 2020]] ). Nonetheless, new literature since AR5 finds that ice-sheet mass balance calculations using reanalysis-driven regional model simulations of surface mass balance are found to agree well with the observed decrease in ice-sheet mass over the past twenty years ( [[#Fettweis--2020|Fettweis et al., 2020]] ; [[#Sasgen--2020|Sasgen et al., 2020]] ; [[#Tedesco--2020|Tedesco and Fettweis, 2020]] ), consistent with earlier studies ( [[#Flato--2013|Flato et al., 2013]] ). These studies also show that the exceptional melt events observed in 2012 and 2019 were associated with exceptional atmospheric conditions ( [[#Sasgen--2020|Sasgen et al., 2020]] ; [[#Tedesco--2020|Tedesco and Fettweis, 2020]] ). These results support the finding that increased surface melting is associated with warming, although atmospheric circulation anomalies, including the summer North Atlantic Oscillation (NAO) and variations in snowfall play an important role in driving interannual variations (Section 9.4.1.1; [[#Sasgen--2020|Sasgen et al., 2020]] ; [[#Tedesco--2020|Tedesco and Fettweis, 2020]] ). Further, a coupled ice-sheet-climate model study found emergence of decreased surface mass balance prior to the present day in coastal locations in Greenland, which dominate the integrated surface mass balance ( [[#Fyke--2014|Fyke et al., 2014]] ), suggesting that observed variations in surface mass balance in these regions might be expected to be distinguishable from internal variability. A CMIP6 simulation of the historical period showed stable Greenland surface mass balance up to the 1990s, after which it declined due to increased melt and runoff, consistent with a downscaled reanalysis ( [[#van%20Kampenhout--2020|van Kampenhout et al., 2020]] ). Further, all experts surveyed in a structured expert judgement exercise examining the causes of the increase in mass loss from the Greenland Ice Sheet over the last two decades ( [[#Bamber--2019|Bamber et al., 2019]] ) concluded that external forcing was responsible for at least 50% of the mass loss. A comparison of Greenland Ice Sheet mass loss trends from observations and AR5 model projections for the period 2007–2017 found that the magnitude of the observed surface mass balance trends was at the top of the AR5 assessed range, while mass loss due to changing ice dynamics was near the centre of the AR5 range ( [[#Slater--2020|Slater et al., 2020]] ), providing further evidence of consistent anthropogenically-forced mass loss trends in models and observations.

Drawing together the evidence from the continued and strengthened observed mass loss, the agreement between anthropogenically forced climate simulations and observations, and historical and paleo evidence for the unusualness of the observed rate of surface melting and mass loss, we assess that it is ''very likely'' that human influence has contributed to the observed surface melting of the Greenland Ice Sheet over the past two decades, and that there is ''medium confidence'' in an anthropogenic contribution to recent overall mass loss from Greenland.

<div id="3.4.3.2.2" class="h4-container"></div>

<span id="antarctic-ice-sheet"></span>
===== 3.4.3.2.2 Antarctic Ice Sheet =====

<div id="h4-9-siblings" class="h4-siblings"></div>

AR5 assessed that there was ''low confidence'' in attributing the causes of the observed mass loss from the Antarctic Ice Sheet since 1993 ( [[#Bindoff--2013|Bindoff et al., 2013]] ). The SROCC assessed that there is ''medium agreement'' but ''limited evidence'' of anthropogenic forcing of Antarctic mass balance through both surface mass balance and glacier dynamics. It further assessed that Antarctic ice loss is dominated by acceleration, retreat and rapid thinning of the major West Antarctic Ice Sheet outlet glaciers ( ''very high confidence'' ), driven by melting of ice shelves by warm ocean waters ( ''high confidence'' ). Based on updated observations, [[IPCC:Wg1:Chapter:Chapter-2|Chapter 2]] assesses that there is ''very high confidence'' that the Antarctic Ice Sheet lost mass between 1992 and 2017, and that there is ''medium confidence'' that this mass loss has accelerated. Models of Antarctic Ice Sheet evolution are evaluated in detail in Section 9.4.2.2, which assesses that there is ''medium confidence'' in many ice-sheet processes in Antarctic Ice Sheet models, but ''low confidence'' in the ocean forcing affecting basal melt rates. CMIP5 and CMIP6 models perform similarly in their simulation of Antarctic surface mass balance (Section 9.4.2.2, [[#Gorte--2020|Gorte et al., 2020]] ). Model evaluation of surface mass balance over the Antarctic Ice Sheet, including regional aspects, is also assessed in Atlas.11.1.3.

Ice discharge around the West Antarctic Ice Sheet is strongly influenced by variability in basal melt ( [[#Jenkins--2018|Jenkins et al., 2018]] ; [[#Hoffman--2019|Hoffman et al., 2019]] ), in particular at decadal and longer time scales ( [[#Snow--2017|Snow et al., 2017]] ). Basal melt rate variability can be induced by wind-driven ocean current changes, which may partly be of anthropogenic origin via greenhouse gas forcing ( [[#Holland--2019|Holland et al., 2019]] ). Moreover, ice discharge losses from the Antarctic Ice Sheet over the 2007–2017 period are close to the centre of the model-based range projected in AR5 ( [[#Slater--2020|Slater et al., 2020]] ). However, expert opinion differs as to whether recent Antarctic ice loss from the West Antarctic Ice Sheet has been driven primarily by external forcing or by internal variability, and there is no consensus ( [[#Bamber--2019|Bamber et al., 2019]] ). Anthropogenic influence on the Antarctic surface mass balance, which is expected to partially compensate for ice discharge losses through increases in snowfall, is currently masked by strong natural variability ( [[#Previdi--2016|Previdi and Polvani, 2016]] ; [[#Bodart--2019|Bodart and Bingham, 2019]] ), and observations suggest that it has been close to zero over recent years (see further discussion in Section 9.4.2.1; [[#Slater--2020|Slater et al., 2020]] ).

Overall, there is ''medium agreement'' but ''limited evidence'' of anthropogenic influence on Antarctic mass balance through changes in ice discharge.

<div id="3.5" class="h1-container"></div>

<span id="human-influence-on-the-ocean-1"></span>