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

==== 9.4.2.2 Model Evaluation ====

<div id="h3-24-siblings" class="h3-siblings"></div>

The AR5 ( [[#Church--2013b|Church et al., 2013b]] ; [[#Flato--2013|Flato et al., 2013]] ) stated that regional climate models and global models with bias-corrected SST and sea ice concentration tended to produce more accurate simulations of Antarctic SMB than coupled climate models. It also noted strong climate model temperature biases over the Antarctic, though the latter may reflect known biases in the reanalysis used ( [[#Fréville--2014|Fréville et al., 2014]] ). Section Atlas.11.1 assesses that there is ''medium confidence'' in the capacity of climate models to simulate Antarctic climatology and SMB changes.

( [[#9.2.3.2|Section 9.2.3.2]] assesses that there is ''low confidence'' in simulations of Southern Ocean temperature. Few ocean models resolve ice-shelf cavities, and biases in present-day melt rates can be substantial in some sectors, including the key region of the Amundsen Sea (e.g., an exception is the FESOM simulation in Figure 9.19 includes ice-shelf cavities and simulates ice-shelf basal melting and refreezing) ( [[#Naughten--2018|Naughten et al., 2018]] ). An increasing number of observational studies from which basal melt rates are calculated ( [[#Huhn--2018|Huhn et al., 2018]] ; [[#Adusumilli--2020|Adusumilli et al., 2020]] ; [[#Das--2020|Das et al., 2020]] ; [[#Hirano--2020|Hirano et al., 2020]] ; [[#Stevens--2020|Stevens et al., 2020]] ), combined with improved understanding of influences specific to water-masses and modes of melting or dissolving ( [[#Silvano--2018|Silvano et al., 2018]] ; [[#Adusumilli--2020|Adusumilli et al., 2020]] ; [[#Malyarenko--2020|Malyarenko et al., 2020]] ; [[#Wåhlin--2020|Wåhlin et al., 2020]] ), may help to refine these models in the future. However, given the limited number of available models and their biases, there is currently ''low confidence'' in the sub-shelf melt rates simulated by ocean models.

Improvements in the representation of grounding line evolution in ice-sheet models since AR5 (such as sub-grid schemes for basal friction and ice-shelf melt, and local grid refinement) means that most of the model simulations presented in SROCC were dominated by physical processes. Since then, these advances have been applied in several model intercomparison projects – such as ISMIP6 and LARMIP-2 (see Box 9.3); MISMIP+ (Cornford et al. 2020); and ABUMIP (Sun et al. 2020). All models participating in ISMIP6 and LARMIP-2 simulate ice-shelf and grounding-line evolution, and include sub-shelf melt parametrization, which was not the case in the Sea-level Response to Ice Sheet Evolution (SeaRISE) project intercomparison ( [[#Bindschadler--2013|Bindschadler et al., 2013]] ; [[#Nowicki--2013|Nowicki et al., 2013]] ). Simulations of grounding line evolution ( [[#Seroussi--2017|Seroussi et al., 2017]] , 2020) have benefitted from improved bedrock topography ( [[#Morlighem--2020|Morlighem et al., 2020]] ). Treatment of sub-shelf melting, however, remains one of the causes of large differences in AIS models, particularly for partially floating grid cells in models with coarse resolution ( [[#Levermann--2020|Levermann et al., 2020]] ; [[#Edwards--2021|Edwards et al., 2021]] ). Due to the limitations in resolving cavities in ocean models, as described above, basal melt rates are generally parameterized at the ice shelf base, based on ocean model simulations of temperatures and salinity instead ( [[#Nowicki--2020b|Nowicki et al., 2020b]] ; [[#Seroussi--2020|Seroussi et al., 2020]] ). While this has the advantage of connecting melt rates to emissions scenarios, a large variety of melt parametrizations exist ( [[#DeConto--2016|DeConto and Pollard, 2016]] ; [[#Lazeroms--2018|Lazeroms et al., 2018]] ; [[#Reese--2018|Reese et al., 2018]] ; [[#Hoffman--2019|Hoffman et al., 2019]] ; [[#Pelle--2019|Pelle et al., 2019]] ; [[#Jourdain--2020|Jourdain et al., 2020]] ), and there is ''low agreement'' due to limited observational constraints (ocean temperature, salinity, velocity, and ice shelf draft)( [[#Jourdain--2020|Jourdain et al., 2020]] ), uncertainty in the physics of parametrized processes, missing processes (e.g., tides), and uncertainty in the treatment of ice-sheet–climate feedbacks ( [[#Donat-Magnin--2017|Donat-Magnin et al., 2017]] ; [[#Bronselaer--2018|Bronselaer et al., 2018]] ; [[#Golledge--2019|Golledge et al., 2019]] ). Parametrizations are usually calibrated to present-day melt rates, but can respond differently to projected ocean warming ( [[#Favier--2019|Favier et al., 2019]] ; [[#Jourdain--2020|Jourdain et al., 2020]] ). Two different calibrations were used in ISMIP6 (Box 9.3; [[#Jourdain--2020|Jourdain et al., 2020]] ; [[#Nowicki--2020b|Nowicki et al., 2020b]] ): one reproducing melt rates averaged around the whole continent (MeanAnt: Figure 9.19), and the other reproducing melt rates near the grounding line of Pine Island Glacier (PIGL; see Figure 9.19), leading to large differences in melt rates. Evaluation with observations and two cavity-resolving models suggests that the MeanAnt parametrization better reproduces observed melt rates and projected increases in both the warm Amundsen Sea Embayment and cold Ronne-Filchner shelf cavity, as well as total Antarctic melting ( [[#Jourdain--2020|Jourdain et al., 2020]] ). The PIGL calibration represents the upper end for increased basal melt sensitivity that would be caused by continent-wide changes to ocean water properties and circulation under strong future forcing ( [[#Jourdain--2020|Jourdain et al., 2020]] ). The basal sliding law also has a strong influence on grounding line retreat and glacier acceleration in response to perturbations, and varies spatially ( [[#Sun--2020|Sun et al., 2020]] ). Sliding laws ( [[#Joughin--2019|Joughin et al., 2019]] ) can only be constrained with observations in regions experiencing significant change, and with sufficiently long observational records.

<div id="_idContainer048" class="Basic-Text-Frame _idGenObjectStyleOverride-1"></div>

[[File:d2fb69c7e007aee6fe14976159bbf4c8 IPCC_AR6_WGI_Figure_9_19.png]]

'''Figure 9.19''' '''|''' '''Ice-shelf basal melt rates for present-day (upper panels) and changes from present-day to the end of the 21st century under the RCP8.5 scenario (lower panels).''' Present-day melt rates were estimated through: the input-output method constrained by satellite observations and atmosphere/snow simulations ( [[#Rignot--2013|Rignot et al., 2013]] ) and representative of 2003–2008 (upper left); the Ice Sheet Model Intercomparison Project for CMIP6 (ISMIP6) non-local-PIGL parametrization constrained by observation-based ocean properties ( [[#Jourdain--2020|Jourdain et al., 2020]] ) and representative of 1995–2014 (upper centre); the Finite Element Sea ice/Ice Shelf Ocean Model (FESOM) simulation over 2006–2015, forced by atmospheric conditions from a Coupled Model Intercomparison Project Phase 5 (CMIP5) multi-model mean (MMM) under the RCP8.5 scenario ( [[#Naughten--2018|Naughten et al., 2018]] ) (upper right). Future anomalies are calculated as 2081–2100 minus present-day using the ISMIP6 non-local-MeanAnt and non-local-PIGL parametrizations ( [[#Jourdain--2020|Jourdain et al., 2020]] ) (lower left and centre, respectively) based on projections from the Norwegian Climate Center’s Earth System Model (NorESM1-M) CMIP5 model, and the FESOM-MMM projection (lower right). Note the symmetric-log colour bar (linear around zero, logarithmic for stronger negative and positive values). Inset highlights the Amundsen Sea Region. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).

The SROCC noted that AIS simulations are increasingly evaluated or formally calibrated with modern observations and/or paleodata – to obtain more realistic initial conditions (ice-sheet geometry, velocity and forcing) and to constrain uncertainty in probabilistic projections. This trend continues ( [[#Nias--2019|Nias et al., 2019]] ; [[#Gilford--2020|Gilford et al., 2020]] ; [[#Hamlington--2020b|Hamlington et al., 2020b]] ; [[#Wernecke--2020|Wernecke et al., 2020]] ). However, while the large-scale characteristics of the initial ice-sheet state have improved significantly (Box 9.3), capturing the smaller-scale rates of change, including mass trends, remains challenging for many models ( [[#Goldberg--2015|Goldberg et al., 2015]] ; [[#Reese--2020|Reese et al., 2020]] ; [[#Seroussi--2020|Seroussi et al., 2020]] ; [[#Siegert--2020|Siegert et al., 2020]] ). This increases uncertainty in projections, especially for the 21st century ( [[#9.4.2.5|Section 9.4.2.5]] ). However, uncertainties in ice-sheet model simulations have been much better quantified since AR5, through model intercomparison projects (in particular, ISMIP6 and LARMIP-2; see Box 9.3), perturbed parameter ensembles, and increasing use of statistical emulation ( [[#Gilford--2020|Gilford et al., 2020]] ; [[#Levermann--2020|Levermann et al., 2020]] ; [[#Wernecke--2020|Wernecke et al., 2020]] ; [[#DeConto--2021|DeConto et al., 2021]] ; [[#Edwards--2021|Edwards et al., 2021]] ) to better sample the parameter space. By exploring uncertainties more fully, these methods have the potential to identify better simulations of the historical period.

An important difficulty is how to evaluate simulations of processes that are: not currently observed; or rare; or indirectly deduced – in particular, the ice-shelf disintegrations and cliff failures that would drive the proposed marine ice cliff instability (MICI; [[#9.4.2.4|Section 9.4.2.4]] and Box 9.4; [[#DeConto--2016|DeConto and Pollard, 2016]] ; [[#DeConto--2021|DeConto et al., 2021]] ). Models of ice-cliff failure can only be indirectly and partially evaluated, using existing (i.e., static) cliffs and laboratory experiments ( [[#Clerc--2019|Clerc et al., 2019]] ). The SROCC stated that there was ''low agreement'' on the exact MICI mechanism and ''limited evidence'' of its occurrence in the present or the past, and that the validity of MICI remains unproven. Only one ice-sheet model represents MICI ( [[#Pollard--2015|Pollard et al., 2015]] ; [[#DeConto--2016|DeConto and Pollard, 2016]] ; [[#DeConto--2021|DeConto et al., 2021]] ). The mechanism has not been found to be essential for reproducing Mid Pliocene Warm Period and Last Interglacial reconstructions or satellite observations, though Last Interglacial data slightly favours it in this model ( [[#Edwards--2019|Edwards et al., 2019]] ; [[#Gilford--2020|Gilford et al., 2020]] ; [[#DeConto--2021|DeConto et al., 2021]] ).

In summary, there is now ''medium confidence'' in many ice-sheet processes in ice-sheet models, including grounding line evolution. However, there remains ''low confidence'' in the ocean forcing affecting the basal melt rates, and ''low confidence'' in simulating mechanisms that have the potential to cause widespread, sustained and very rapid ice loss from Antarctica through MICI.

<div id="9.4.2.3" class="h3-container"></div>

<span id="drivers-of-future-antarctic-ice-sheet-change"></span>