Reducing the uncertainty in the past, present, and future contribution of ice sheets to sea-level change requires a coordinated effort between the climate and glaciology communities. The Ice Sheet Model Intercomparison Project for CMIP6 (ISMIP6) is the primary activity within the Coupled Model Intercomparison Project – phase 6 (CMIP6) focusing on the Greenland and Antarctic ice sheets. In this paper, we describe the framework for ISMIP6 and its relationship with other activities within CMIP6. The ISMIP6 experimental design relies on CMIP6 climate models and includes, for the first time within CMIP, coupled ice-sheet–climate models as well as standalone ice-sheet models. To facilitate analysis of the multi-model ensemble and to generate a set of standard climate inputs for standalone ice-sheet models, ISMIP6 defines a protocol for all variables related to ice sheets. ISMIP6 will provide a basis for investigating the feedbacks, impacts, and sea-level changes associated with dynamic ice sheets and for quantifying the uncertainty in ice-sheet-sourced global sea-level change.
Ice sheets constitute the largest and most uncertain potential source of future sea-level rise (Church et al., 2013; Kopp et al., 2014). The Greenland and Antarctic ice sheets currently hold ice equivalents of over 7 and 57 m of sea-level rise, respectively. Observations indicate that the Greenland and Antarctic ice sheets have contributed approximately 7.5 and 4 mm of sea-level rise over the 1992–2011 period (Shepherd et al., 2012) and that their contribution to sea-level rise is accelerating (Rignot et al., 2011a). Sea-level change has been identified as a long-lasting consequence of anthropogenic climate change, as sea levels will continue to rise even if temperatures are stabilized (Meehl et al., 2012). Therefore, assessing whether the observed rate of mass loss from the ice sheets will continue at the same pace, or accelerate, is crucial for risk assessment and adaptation efforts.
In addition to their impact on sea-level change, ice sheets influence the Earth's climate through changes in freshwater fluxes, orography, surface albedo, and vegetation cover, across multiple spatial and temporal scales (Vizcaíno, 2014). Ice-sheet evolution and iceberg discharge affect ocean freshwater fluxes (e.g., Broecker, 1994), which in turn can affect oceanic circulation (e.g., Weaver et al., 2003) and marine biogeochemistry (Raiswell et al., 2006). Changes in ice-sheet orography modify near-surface temperatures by altering atmospheric circulation (Ridley et al., 2005) on both regional and global scales (e.g., Manabe and Broccoli, 1985). Surface albedo and elevation change due to the waxing and waning of ice sheets has played an important role in past interglacial–glacial transitions (e.g., Calov et al., 2009; Abe-Ouchi et al., 2013). Seasonal fluctuations in ice-sheet albedo can also exert considerable influence on local surface energy fluxes (e.g., Box et al., 2012), through both melt and snowfall. Over longer timescales, changes in ice-sheet elevation can cause a positive feedback on surface mass balance, wherein a thinning ice sheet experiences warmer temperatures at lower elevations, which causes further melting and thinning. Ice-sheet elevation changes can also alter the local climate, for instance changing the trajectory of Southern Ocean storms that penetrate onto the Antarctic Plateau (Morse et al., 1998).
Ice sheets gain mass primarily by accumulation of snowfall, and lose mass through a combination of surface meltwater runoff, surface sublimation, iceberg discharge to the ocean, and basal melting (under both grounded ice and floating ice shelves). The Antarctic Ice Sheet experiences minimal surface melt and thus loses mass primarily through basal melting and iceberg calving. Most basal mass loss in Antarctica occurs under ice shelves (e.g., Joughin and Padman, 2003; Pritchard et al., 2012), but sub-ice-sheet meltwater is also produced over large areas (Fricker et al., 2007). Together, basal melting and iceberg calving currently outweigh snowfall accumulation to the Antarctic Ice Sheet (Rignot et al., 2013; Depoorter et al., 2013). The Greenland Ice Sheet is also currently losing mass overall; this occurs primarily through iceberg calving and surface runoff. Surface mass balance changes have recently surpassed iceberg calving changes as the dominant contributor to Greenland mass loss (van den Broeke et al., 2009), with increased surface runoff now contributing 60 % of the mass loss (Enderlin et al., 2014). Due to the long response time of ice sheets, mass changes observed at present are a complex combination of the response to present climate changes as well as past climate changes as far back as several tens of thousands of years. These integrating effects of ice sheets and the vastly different timescales on which ice-sheet models and climate models operate have historically inhibited efforts to interface these two components of the Earth system.
Previously, ice sheets were not explicitly included in the CMIP process, and separate modeling studies were used to make projections of their future contributions to sea level. This has often led to mismatches between the climate data used to force these models and the contemporary version of the CMIP projections. This mismatch was perhaps acceptable when ice sheets were regarded as passive elements of the climate system on sub-millennial timescales (e.g., Church and Gregory, 2001). Observations of rapid mass loss associated with dynamic change in the ice sheets, however, have highlighted the need to couple ice sheets to the rest of the climate system. At one stage, this mismatch was such that little confidence could be placed in the projections of ice-sheet models, which were felt to omit the key processes responsible for observed changes (e.g., Meehl et al., 2007). With subsequent developments in ice-sheet modeling, many of the processes thought to affect ice-sheet dynamics on sub-centennial timescales (such as grounding-line migration, changes in basal lubrication, and, to some extent, iceberg calving) can be simulated with some confidence (e.g., Church et al., 2013). Previous ice-sheet model inter-comparison exercises have played a crucial role in this development. An excellent example is the ongoing series of inter-comparisons aimed at understanding issues associated with the numerical modeling of grounding-line motion (e.g., Pattyn et al., 2012, 2013). Two previous international efforts, the SeaRISE and ice2sea initiatives, supplied projections on which the assessments of Church et al. (2013) were based. A major criticism of both efforts, however, was that they were based on forcing from the Special Report on Emissions Scenarios (SRES, Nakićenović et al., 2000) rather than the current Representative Concentration Pathway (RCP, van Vuuren et al., 2011) framework. The Ice Sheet Model Intercomparison Project for CMIP6 (ISMIP6) is explicitly designed to ensure that ice-sheet (hence sea-level) projections are fully compatible with the CMIP6 process.
ISMIP6 brings together for the first time a consortium of international ice-sheet models and coupled ice-sheet–climate models. This effort will thoroughly explore the sea-level contribution from the Greenland and Antarctic ice sheets in a changing climate and assess the impact of large ice sheets on the climate system. In this paper, we provide an overview of the ISMIP6 effort and present the ISMIP6 framework. We begin by explaining the objectives and approach for ISMIP6 (Sect. 2), and describe the experimental design (Sect. 3). We next present an evaluation and analysis plan (Sect. 4) and finally discuss the expected outcome and impact of ISMIP6 (Sect. 5).
ISMIP6 was initiated with the help of the Climate and Cryosphere (CliC) effort of the World Climate Research Project (WCRP) and is now a targeted activity of CliC. The main goal is to better integrate ice-sheet models in climate research in general, and in the CMIP initiative in particular. ISMIP6 offers the exciting opportunity of widening the current CMIP definition of the Earth system to include ice sheets. Together with the CliC targeted activity on glacier modeling (GlacierMIP) and existing models for thermal expansion within the CMIP framework, output from ISMIP6 will add sea level to the family of variables for which CMIP can provide routine IPCC-style projections. ISMIP6 is primarily focused on the CMIP6 scientific question “How does the Earth system respond to forcing?”, but will also contribute to answering the question “How can we assess future climate change given climate variability, climate predictability, and uncertainty in climate scenarios?” for scenarios involving the mass budget of the ice sheets and its impact on global sea level.
ISMIP6 targets two Grand Science Challenges (GCs) of the WRCP: “Melting Ice and Global Consequences” and “Regional Sea-level Change and Coastal Impacts”. Specifically, the primary goal of the ISMIP6 effort is to improve our understanding of the evolution of the Greenland and Antarctic ice sheets under a changing climate. A related goal is to quantify past and future sea-level contributions from ice sheets, including the associated uncertainties. These uncertainties arise from uncertainties in both the climate input and the response of the ice sheets. A secondary goal is to investigate the role of feedbacks between ice sheets and climate in order to gain insight into how changes in the ice sheets will affect the Earth climate system.
These goals require an experimental framework that can address the following
objectives.
Develop better models of climate and ice sheets, as both coupled systems and
individual components. Improve understanding of how ice sheets respond to climate on various
timescales, both in the past and in the future. Improve understanding of how ice sheets affect local and global climate, and
explore ice-sheet–climate feedbacks. Improve simulation of sea-level change, especially projections for the
21st century and over the next 300 years.
As depicted in Fig. 1, our goals and objectives rely on three distinct
modeling efforts: (i) traditional CMIP atmosphere–ocean general circulation
models (AOGCM/AGCMs) without dynamic ice sheets, (ii) standalone dynamic
ice-sheet models (ISMs) that are driven by provided forcing fields
(“offline”), and (iii) atmosphere–ocean climate models coupled to dynamic
ice sheets (AOGCM–ISMs), which, as described in the following sections, can
be combined to form an integrated framework.
Overview of the ISMIP6 effort designed to obtain forcing from climate models, project sea-level contributions using ice-sheet models, and explore ice-sheet–climate feedbacks.
Overview of the experiments with climate models not coupled with ice-sheet models that are to be used by ISMIP6. All experiments are started on 1 January and end on 31 December of the specified years. n/a stands for not applicable.
Overview of the ISMIP6 experiments with dynamic ice sheets that are
either coupled to climate models (AOGCM-ISM,
Following the CMIP6 protocol, the ISMIP6 experiments both use and augment the
CMIP6-DECK (Diagnostic Evaluation and Characterization of Klima) and
Historical simulations (Meehl et al., 2014; Eyring et al., 2016). In
addition, ISMIP6 collaborates with the CMIP6-Endorsed Paleoclimate Model
Intercomparison effort (PMIP4, Kageyama et al., 2016) and builds on the
CMIP6-Endorsed ScenarioMIP (O'Neill et al., 2016) that focuses on future
climate experiments for CMIP6. For a selected number of AGCM/AOGCM
experiments that are already part of CMIP6 (Table 1 and described in
Sect. 3.1), three additional model configurations are proposed,
“
A first component of the ISMIP6 effort is to assess and evaluate CMIP atmosphere general circulation models (AGCMs) and coupled atmosphere–ocean general circulation models (AOGCMs) over and surrounding the polar ice sheets. This part of ISMIP6 can be viewed as diagnostic in the sense that all climate models that participate in CMIP6 will be included in this assessment without requiring extra work from the climate modeling centers. These experiments do not include dynamic ice sheets, and as explained in the CMIP6 protocol (Eyring et al., 2016), climate modeling centers that contribute to CMIP6 are required to submit simulations for the DECK and CMIP6 Historical runs. Our goals are to establish the suitability of the CMIP models for producing climate input for ice-sheet models and to assess the uncertainty in projections of sea-level change arising from such climate input. As described in Sect. 4, an additional goal is to assess past and projected changes in surface forcing (here for a fixed ice-sheet extent and topography), along with the resulting sea-level contribution from both ice sheets due to changes in surface freshwater flux alone. The largest uncertainty in century-scale sea-level projections, however, remains the dynamic ice-sheet response to changes in atmospheric and oceanic conditions, which will be addressed by the other components of ISMIP6 (Sect. 3.2 and 3.3).
The experiments with climate models not coupled to ISMs, listed in Table 1,
are central to ISMIP6 and thus briefly introduced. These AGCM/AOGCM
experiments are already part of CMIP6, such that more detailed information on
the experimental protocol is available elsewhere in this special issue.
ISMIP6 uses three of the four DECK experiments described in Eyring et
al. (2016). The Atmospheric Model Intercomparison Project (
The CMIP6 Historical simulation,
Evaluation of the climate over and surrounding the ice sheets is necessary
both to establish the suitability of current climate models to provide
forcing for ice-sheet models and to gain insight into sea-level uncertainty
arising from uncertainty in atmospheric and oceanic climate forcings. Of
particular interest is the surface climate over the ice sheets, with a focus
on temperature and surface mass balance (SMB). SMB is defined as total
precipitation minus evaporation, sublimation, and surface runoff, where runoff is meltwater less any refreezing
within the snowpack. Because the ocean condition is prescribed for the
The general approach for evaluating the atmospheric component of climate
models over the ice sheets (e.g., Yoshimori and Abe-Ouchi, 2012; Fettweis et
al., 2013; Vizcaíno et al., 2013; Cullather et al., 2014; Lenaerts et
al., 2016) is to compare the large-scale atmospheric state over the polar
regions, the local climate, and processes at the ice-sheet surface. The
latter focuses on whether the climate model can simulate snow processes,
including albedo evolution and refreezing, at a horizontal resolution that
captures the SMB gradients at ice-sheet margins. Both the atmospheric
components and factors that can affect atmospheric processes are often
evaluated. One example is determining whether sea ice conditions are
adequately captured in
The large-scale atmospheric state over the polar regions is often assessed by comparing the modeled atmospheric flow at 500 hPa to atmospheric reanalysis values. For the local climate, near-surface winds and near-surface temperatures can be compared to regional climate models (RCMs) such as RACMO2 (van Meijgaard et al., 2008; Lenaerts et al., 2012; van Angelen et al., 2014), MAR (Fettweis, 2007; Fettweis et al., 2011) or HIRHAM (Langen et al., 2015; Lucas-Picher et al., 2012), reanalysis (e.g., Agosta et al., 2015), and observations where available. RCMs are also used to evaluate the spatial pattern of surface mass balance and its components (precipitation, sublimation, and surface melt) computed by global circulation models. The surface energy budget, particularly the seasonal cycle of net shortwave and longwave radiation and the sensible and latent heat fluxes, can be evaluated against measurements taken by automatic weather stations on the ice-sheet surface. Such stations include, for example, the 15 Greenland stations known as the GC–Net (Steffen and Box, 2001), the Greenland PROMICE network with a focus on the ablation zone (Ahlstrøm et al., 2008), and, in Antarctica, the Neumayer Base (Lenaerts et al., 2010). These stations also record winds and temperatures. The surface temperature over the ice sheets may also be evaluated from satellite observations, using, for example, data derived from the Moderate Resolution Imaging Spectroradiometer (MODIS, Hall et al., 2012). These remotely sensed temperature products show the onset and/or spatial extent of surface melt (e.g., Mote et al., 1993; Hall et al., 2013), which can then be used to assess whether the climate models capture the relevant processes at the ice-sheet surface (e.g., Fettweis et al., 2011; Cullather et al., 2016). However, a full understanding of why surface melt varies from model to model may require investigations that include cloud properties (Van Tricht et al., 2016).
The current generation of climate models participating in CMIP6 is unlikely to simulate ocean circulation in ice-shelf cavities or within fjords. Thus, evaluation of the ocean state around the ice sheets involves first establishing that the climate models can reproduce certain properties of the key water masses. Ocean circulation around the Greenland Ice Sheet involves a complex interaction between polar waters of Arctic origin and Atlantic waters from the subtropical North Atlantic (Straneo et al., 2012). The mechanisms that transport warm water through fjords and toward the ice fronts remain an active area of research (Wilson and Straneo, 2015; Straneo and Cenedese, 2015). In the Southern Ocean, important water masses include Antarctic Bottom Water and Antarctic Intermediate Waters. In the coastal regions, Circumpolar Deep Water, Antarctic Surface Water, and High Salinity Shelf Water are the primary oceanic influences on ice sheets (Bracegirdle et al., 2016). Given the difficulty many CMIP5 models had in capturing high-latitude ocean properties, CMIP6 models should be evaluated using existing datasets (Bracegirdle et al., 2016). These datasets include Argo, expendable bathythermograph (XBT) and conductivity/temperature/depth (CTD) vertical temperature and salinity profiles (e.g., Dong et al., 2008), sea ice extent products sourced from passive microwave instruments (e.g., Bjørgo et al., 1997; Cavalieri and Parkinson, 2012; Parkinson and Cavalieri, 2012), sea surface temperature (SST) from WindSat and AMSR-E over the open ocean, satellite altimetry (Jason-1 and Jason-2) over the open ocean, and World Ocean Atlas 2009 climatological temperatures. For ocean models that include ice-shelf cavities and ice–ocean interactions, sub-ice-shelf basal melting can be compared with glaciological estimates of ice-shelf melting around Antarctica (Rignot et al., 2013; Depoorter et al., 2013) derived from remote-sensing observations, as well as independent tracer–oceanographic estimates (Loose et al., 2009; Rodehacke et al., 2006). Just as regional atmospheric models will be key for evaluating the atmospheric component of climate models, regionally focused ocean models (e.g., Timmermann et al., 2012) and ocean reanalysis products (e.g., Menemenlis et al., 2008) are likely to provide valuable insight for evaluating CMIP ocean models.
The second component of ISMIP6 is a suite of experiments designed to assess
the impacts of dynamic ice sheets on climate and to better understand
feedbacks between ice sheets and climate. We also aim to obtain an ensemble
of sea-level projections from fully coupled atmosphere–ocean–ice-sheet
frameworks, which can later be compared to projections from standalone
ice-sheet models (Sect. 3.3). The experiments should be identical to the
corresponding standard CMIP AOGCM experiments except for the treatment of ice
sheets, so that any observed feedbacks and impacts can be attributed to
dynamic ice sheets and not to other sources. As indicated in Table 2, four
coupled AOGCM-ISM simulations are proposed, whose experiment IDs are
In the
The
Initial conditions for both the
Ideally, the ice-sheet model should be forced with the actual SMB computed by
the climate model, rather than an SMB corrected to match observed
climatology. We accept that there may be biases in the atmospheric or land
models that can lead to an unrealistic SMB, which could result in a
steady-state ice-sheet geometry that differs substantially from present-day
observations. However, correcting for these biases can distort the feedbacks
between ice sheets and climate that we seek to investigate. We hope to learn
from and ultimately reduce these biases, in the same way that biases
elsewhere in the simulated coupled climate system are reduced by greater
understanding and improved model design. On the other hand, if the geometry
of the spun-up ice sheet is greatly different from observations, then the
initial ice sheet for the
The method used to downscale SMB (as well as oceanic forcing) from the coarse climate model grid to the finer ice-sheet model grid is left to the discretion of each group, but should be well documented. The data request for ISMIP6 in Appendix A asks modelers to report certain fields on both the atmospheric and ice-sheet grids to allow for an evaluation of the downscaling procedure. Also, ISMIP6 prefers that the surface-melt component of SMB be obtained from an energy-based method that conserves mass and energy, to facilitate interpretation of the drivers of SMB variability and change (e.g., Vizcaíno, 2014). Highly parameterized methods of computing surface melt, such as positive-degree-day (PDD) methods (e.g., Reeh, 1991; Bougamont et al., 2007), should be avoided. The choice of the ice-sheet model, its complexity in approximating ice flow, and ice-sheet-relevant boundary conditions (e.g., geothermal flux) are left to the modelers' discretion. In all experiments, however, the ice sheets should not be forced to terminate at the present-day ice margin if the simulated SMB and/or the ice-sheet dynamics cause a margin advance.
Regardless of the spinup method, the first ISMIP6 experiment to be performed
with the coupled AOGCM–ISM is the pre-industrial control,
Climate modeling centers that have expressed an interest in ISMIP6. * indicates only an interest in the diagnostic component (no AOGCM–ISM participation anticipated).
Another set of experiments repeats the CMIP6
Based on community feedback, we expect that several AOGCM–ISMs will be ready
to participate in coupled climate experiments for CMIP6. Table 3 shows
climate modeling centers that have expressed interest in participating in
ISMIP6. The primary focus is coupled ice-sheet–atmosphere simulation for the
Greenland Ice Sheet, but some groups have indicated participation only in the
diagnostic aspect of ISMIP6 (where the goal is to provide climate data for
the standalone ice-sheet work). Full coupling of ice-sheet models to climate
models remains challenging, especially for interactions with the ocean.
Accurate treatment of ice–ocean interactions requires ISMs that can simulate
grounding-line migration (which demands fine grid resolution) and iceberg
calving, and ocean models that can simulate circulation in the cavities below
ice shelves and the consequent melting or accretion of ice on the undersides
of the shelves. Accurate treatment of ice–ocean interactions will likely
also require ocean models to alter their domain (both vertically and
horizontally) as the calving front migrates and as sub-ice-shelf ocean
cavities evolve in space and time. For the Greenland Ice Sheet, ocean models
may need to capture fjord dynamics on smaller spatial scales (
Ice-sheet modeling groups that have expressed an interest in ISMIP6. x indicates a planned contribution.
The final set of ISMIP6 experiments will use standalone ice-sheet models
driven by climate model output and other datasets. Groups and models that
have expressed an interest in participating in this aspect of ISMIP6 are
listed in Table 4. The models participating in this effort will likely be
configured differently from those in the
The initMIP ice-sheet experiments are designed to explore uncertainties in sea-level projections associated with model initialization and spinup. Such uncertainties have been identified by previous model intercomparison efforts (e.g., Bindschadler et al., 2013; Nowicki et al., 2013a, b; Edwards et al., 2014a, Shannon et al., 2013; Goelzer et al., 2013; Gillet-Chaulet et al., 2012) and include the impacts of model initial conditions, sub-grid-scale processes, and poorly known parameters. The initMIP project aims to evaluate initialization procedures, to estimate trends caused by model initializations, and to investigate the impact of choices in numerical and physical parameters (e.g., stress balance approximation or model resolution). Results of the initMIP project are expected to point to specific aspects of ice-sheet initialization that have a crucial impact on sea-level projections and that may be improved.
ISM initialization methods to present-day conditions range from running paleo-climate spinup for thousands of years (e.g., Martin et al., 2011; Sato and Greve, 2012; Aschwanden et al., 2013; Fürst et al., 2015; Saito et al., 2016) to assimilating present-day observations (e.g., Morlighem et al., 2010; Gillet-Chaulet et al., 2012; Seroussi et al., 2013, Arthern et al., 2015). The choices made in this procedure affect ice-sheet extent, flow rates, volume, and volume trends, which can have substantial effects on estimates of ice-sheet contribution to sea-level rise (e.g., Aðalgeirsdóttir et al., 2014). Improving ISM initial conditions is an active area of research and a multidisciplinary effort. It requires acquisition of additional data with high spatial coverage over entire ice sheets and at increased resolution (e.g., Bamber et al., 2013; Rignot et al., 2011b; Joughin et al., 2010a; Howat et al., 2014). Ideally, all datasets used in the data assimilation are from the same period, as initializing an ice-sheet model with datasets taken at different times can cause the ice-flow model to artificially redistribute the glacier mass in unrealistic ways that serve only to reconcile these inconsistencies (Seroussi et al., 2011). This also implies that the date associated with the initial state can differ between models based on the datasets used. New algorithms that reconcile initialization datasets are being developed, most notably for bedrock elevation (e.g., Morlighem et al., 2011, 2014), which is notoriously poorly constrained.
The initMIP project consists of a Greenland component and an Antarctic
component. Following initialization, there is a set of two forward
experiments for the Greenland Ice Sheet and three forward experiments for the
Antarctic Ice Sheet, each run for at least 100 years: (i) a control run
(
Since these experiments are designed to allow comparison among the different
models, some simplifications are imposed. Neither SMB nor bedrock topography
should be adjusted in response to ice-sheet geometric changes in these
forward experiments. However, to sample the uncertainty in sea level due to
initialization, groups are encouraged to submit multiple variations of the
experiment, for example by changing the sliding law, stress balance
approximation, model resolution, or datasets (such as using different
bedrocks). While the initialization procedures used by the different
participating groups are not prescribed by ISMIP6, it is expected that
individual groups will take advantage of the initMIP results to improve their
initialization procedures. initMIP is also intended to give ice-sheet
modelers an opportunity to get involved in ISMIP6 at an early stage, before
outputs of CMIP6 AOGCM become available; hence our prescription of simplified
anomalies. We refer interested readers to the initMIP webpage
(
The
The forcing data can naturally be divided into atmospheric and oceanic forcing. Central to the former is the means to determine SMB associated with a particular CMIP6 experiment. Several methods have previously been employed to do this. Until we can assess the quality of the climate simulated by CMIP6 AOGCMs above and around the ice sheets (after the analysis of the CMIP6 DECK and Historical simulations), a definitive choice cannot be made. However, we list the options in order of preference.
Use the SMB calculated by the AOGCM directly. This has the advantage that the SMB will be entirely consistent with other parts of that AOGCM's simulation of climate. There is concern, however, that the quality of the SMB computed by the AOGCMs will make this approach unrealistic due primarily to the mismatch between the spatial resolution of AOGCMs and the characteristic length scale of variations in SMB. Several groups have, however, made recent progress in this area (e.g., Vizcaíno et al., 2013; Lipscomb et al., 2013). The use of anomalies should also be considered in this context.
In the event that AOGCM-determined SMB is shown to be inadequate, an intermediate step is required. Previously, this has been the use of Regional Climate Models (RCMs) to simulate SMB. For example, the ice2sea effort chose to generate SMB from an RCM (Edwards et al., 2014a, b; Fettweis et al., 2013). This approach, however, introduces a further link into the processing chain that may lead to delay in the production of sea-level projections. It also introduces the issue of choice of RCM and whether results from a number of RCMs should be used (further complicating the design of the ISM ensemble). Furthermore, the use of RCMs as intermediaries between AOGCMs and ISMs adds ambiguity about which biases are introduced by the AOGCMs and which biases are the result of the RCMs.
Use a parameterization or simplified process model to simulate SMB by downscaling atmospheric forcing over the ice sheet from an AOGCM. This approach was used by SeaRISE (Bindschadler et al., 2013), where the precipitation and surface temperature from 18 AOGCMs models taking part in the A1B scenario were combined to generate monthly mean values. These mean precipitation and temperature values were then passed to the SMB scheme of the ice-sheet model (generally a PDD method that accounted for the temperature aspect of the SMB–elevation feedback) to obtain SMB anomalies that were added to the ice-sheet surface conditions at initialization.
A further consideration is that the AOGCM models assume a fixed ice-sheet elevation: i.e., they neglect the effect of ice-sheet elevation change on the atmosphere and hence omit the SMB–elevation feedback. Standalone ISMs will need to include this effect by parameterizing the SMB lapse rate (Edwards et al., 2014a, b; Fettweis et al., 2013; Goelzer et al., 2013). This approach may be less of an issue for method 3 above because SMB is determined interactively within the ISM rather than being prescribed as forcing.
A second way in which the atmosphere could force dynamic change in ice sheets is through the production of large quantities of meltwater. Mechanisms have been proposed that link meltwater to both ice-shelf collapse (Banwell et al., 2013) and enhanced lubrication of ice flow (Zwally et al., 2002) (although recent modeling studies suggest a minor influence of the latter on large-scale ice flow; e.g., Shannon et al., 2013). Surface air temperature and runoff forcing will therefore also be made available.
Both Antarctica and Greenland are thought to respond to changes in proximal ocean temperatures, which affect the melt rates of floating ice shelves and the vertical faces of outlet glaciers. Obtaining suitable oceanic forcing from CMIP6 climate models will be a major challenge. Few CMIP6 models will calculate the appropriate melt rates, and even these results are likely to be inaccurate because of issues of model resolution and the unique physics of ocean circulation adjacent to melting ice. Melt rates will therefore need to be determined outside the climate model using an index for proximal ocean temperature. This index is most likely to be water temperature (and salinity) at the continental shelf break at an intermediate range of depths (equivalent to the base of ice shelves or the depth of ice grounded on bedrock). This quantity will be included in our evaluation of CMIP6 forcing (see Sect. 3.1).
A wide range of approaches has been used to calculate the required melt rate
from prescribed ocean-temperature forcing. The simplest method is to
calculate melt rate anomalies from changes in the nearest ocean temperature
using an observationally derived relation of
10 m yr
If none of the CMIP6 ocean models can accurately capture the broad-scale polar ocean circulation or produce realistic near-shelf temperatures, an alternative is to prescribe a melt rate that simply depends on the ice-shelf draft (e.g., Joughin et al., 2010b; Favier et al., 2014). This approach is less satisfactory, however, as it ignores temporal changes in ocean conditions, and typically uses coefficients calibrated to local thermal conditions. If ISMIP6 uses this approach, the provided coefficients would not be uniform but would take into account the fact that ocean waters reaching ice-shelf cavities or fronts differ regionally. In Antarctica, for example, the ice shelves of Pine Island Glacier and Thwaites Glacier lie in “warm” water, while the Filchner-Ronne or Ross ice shelves reside in “cold” water. Ocean temperatures reflect the dominant water sources, with warm waters dominated by circumpolar deep waters (Jacobs et al., 2011), while cold waters typically correspond to high-salinity shelf water (Nichols et al., 2001).
Ice–ocean interactions are an active area of research, and more sophisticated parameterizations of melt are becoming available (e.g., Jenkins, 2016; Asay-Davis et al., 2016). Simplified models of the system could be used (e.g., Payne et al., 2007), as could high-resolution ocean models that resolve ice-shelf cavities and fjords. Given this wide range of methods, ISMIP6 will leave the detailed choice of the parameterization to individual ice-sheet modelers, but will issue guidance on what constitutes an acceptable parameterization. We will organize workshops with the polar ocean community to investigate how to best derive oceanic forcing for ice-sheet models, so that by the time the CMIP6 ocean models are evaluated, a clearer protocol will be in place. The calculated melt rate will be part of the standard data request for ice-sheet models (see Appendix A), and part of our evaluation will be to determine how well the applied forcing compares to observed melt rates of Rignot et al. (2013) and Depoorter et al. (2013).
ISMIP6 will not dictate the choice of ice-sheet model complexity in terms of the ice-flow approximation, the basal sliding law, the treatment of grounding lines, the calving law, the ice-sheet-specific boundary conditions (e.g., bedrock topography), or the initialization method. An exception is that models of the Antarctic Ice Sheet should include floating ice shelves and grounding-line migration. The spatial resolution of the ISM in the vicinity of fast-flowing ice streams and the grounding line affects the dynamic response (Durand et al., 2009; Pattyn et al., 2012, 2013), and the model resolution must be fine enough to capture this response accurately. To this end, participating models are encouraged to take part in model intercomparison efforts that target specific aspects of ice-sheet modeling, such as the current MISOMIP (Marine Ice Sheet–Ocean Model Intercomparison Project; Asay-Davis et al., 2016), and are required to take part in initMIP (initialization-focused experiments that compare and evaluate the simulated present-day state; Sect. 3.3.1). The lack of a stricter protocol is a reflection of the challenges in identifying which factors are the most important when making projections, which datasets are most accurate, and how to best capture and parameterize certain ice-sheet processes. For example, although the choice of bedrock topography affects mass transport and is thus likely to influence a projection, it is currently not possible to identify a best dataset due to the difficulty in obtaining bedrock measurements. Groups are encouraged to repeat the experiments with a variety of perturbations of weakly constrained parameters, boundary conditions, etc. in order to test the sensitivity of projections to these choices.
Unlike the protocol for climate models, the
The first
We complement the experiments for the recent past and future with one paleo
experiment (
The ISMIP6 experiments listed in Table 2 are divided into three “Tiers” to indicate prioritization. Tier 1 denotes experiments that are to be completed by the ISMIP6 participants. Tier 2 experiments are highly encouraged, while Tier 3 experiments are optional.
For the coupled AOGCM–ISM experiments, the Tier 1 experiments
For the standalone
The framework described in this paper entails an evaluation of the climate system, with a particular focus on the polar regions. This framework works toward the goals of (i) assessing the effect of including dynamic ice sheets in climate models and (ii) improving confidence in projections of sea-level rise associated with mass loss from the Greenland and Antarctic ice sheets. Our evaluation and analysis will be based on key model output variables for the atmosphere, ocean, and ice sheets that form the ISMIP6 data request summarized in Appendix A.
Ice-sheet models will be evaluated using methodologies already in use by the ice-sheet modeling community. These metrics typically begin by assessing whether the volume and area of the modeled present-day ice sheet are comparable to observed values. The next step evaluates the spatial patterns of surface elevation, ice-sheet thickness, surface velocities, and positions of the ice front and grounding line. Some ice-sheet models are initialized using data assimilation methods, which precludes the use of certain observations in the evaluation. Evaluation of these models can be done by hindcasting, a method that evaluates whether recent observed trends are captured (Aschwanden et al., 2013). Examples include comparison against the gravimetry (GRACE) time series from 2003 onwards, which provides an integrated set of measurements for mass changes in Greenland and Antarctica. This approach will also enable a direct comparison between predicted sea-level rise from ISMs and the change in ocean mass observed by GRACE. The recent IMBIE effort (Ice Sheet Mass Balance Inter-comparison Exercise, Shepherd et al., 2012) facilitates this comparison by combining observations from gravimetry, altimetry, and velocity changes between 1992 and 2012 into a single dataset of annual mass budget for each ice sheet. The follow-on effort, IMBIE2 (A. Shepherd, personal communication, 2015), will extend the record in time and plans to separate the observed mass change into SMB and dynamic components.
The combination of coupled AOGCM–ISM simulations (
As ice sheets thin, the lower elevation leads to warmer surface temperatures that increase melting. This ice–elevation feedback is small on sub-century timescales (Edwards et al., 2014b), but over longer timescales, it can drive ice sheets to a point of no return, where retreat would continue unabated even if the climate returned to an unperturbed state.
Changes in ice-sheet elevation modify the regional atmospheric circulation (e.g., Ridley et al., 2005), which can either enhance or slow the rate of retreat.
Changes in land surface cover (e.g., from glaciated to vegetated) can darken and warm the surface, promoting atmospheric warming and further melting.
Increased freshwater fluxes (both solid and liquid) from retreating ice sheets can modify the density structure of the ocean, possibly suppressing convection and weakening the Atlantic meridional overturning circulation. Although some studies (e.g., Hu et al., 2009) find that this is a small effect, others suggest that increased runoff from the Greenland Ice Sheet has already reduced deep convection in the Labrador Sea (Yang et al., 2016).
The buoyancy of fresh glacial meltwater from sub-ice-shelf melting can modify the ocean circulation that drives the melting. On longer timescales, changes in the size and shape of sub-shelf cavities may also alter the circulation.
The ISMIP6 experiments will be performed on climate model runs lasting several centuries, long enough to allow a detailed analysis of at least the first four of these feedbacks. Ocean cavity feedbacks, however, may require further development of ocean models that can adjust their boundaries dynamically as marine ice sheets advance and retreat.
The SMB over the Greenland Ice Sheet is currently becoming less positive, thus resulting in an increasing contribution to sea-level rise due to increased surface runoff (van Angelen et al., 2014; Fettweis et al., 2011). This trend is expected to continue (Fettweis et al., 2013; Rae et al., 2012), although there is a large spread in AOGCMs (Yoshimori and Abe-Ouchi, 2012). The picture is less clear for the Antarctic Ice Sheet, where both accumulation and surface melt are projected to increase (Lenaerts et al., 2016). The multi-model ensemble of the surface freshwater flux from AOGCM simulation will provide insight into the resulting contribution of past and future sea level due to changes in SMB alone.
The largest uncertainty in sea level, however, remains the dynamic
contribution from the ice sheets. ISMIP6 targets the contribution of dynamic
ice sheets to global sea level, via multi-model ensemble analysis of
standalone ice-sheet models (
We also aim to quantify the uncertainty in sea level arising from uncertainties in both the ice-sheet models and the climate input; hence the need to sample across scenarios and models. For example, the ongoing initMIP project will provide insight into sea-level uncertainties resulting from ice-sheet model initialization. By repeating model runs with different datasets, sliding laws, model resolutions, etc., initMIP will allow us to constrain the sea-level contribution associated with these choices. Ice-sheet evolution will also depend on climatic drivers. For instance, given a certain number of AOGCMs that simulate present-day ice-sheet SMB reasonably well, comparing their SMB results under various climate-change simulations will allow us to quantify climate-model-driven uncertainty in SMB. If relationships between large-scale climate drivers (e.g., regional temperature and precipitation) and ice-sheet area-integral SMB can be established (e.g., Gregory and Huybrechts, 2006; Fettweis et al., 2013), this would allow estimation of SMB from AOGCM experiments for other climate scenarios. If possible, synergies with other CMIP6 efforts will allow us to further investigate the uncertainty in climate input. For example, the CMIP6-Endorsed High Resolution Model Intercomparison Project (HighResMIP, Haarsma et al., 2016) and Coordinated Regional Climate Downscaling Experiment (CORDEX, Gutowski Jr. et al., 2016) may allow us to quantify the impacts of increased resolution on SMB.
ISMIP6 has an experimental protocol and a diagnostic protocol. The experimental design uses and builds upon the core DECK and CMIP6 Historical simulations, along with selected CMIP6-Endorsed PMIP4 and ScenarioMIP simulations. The suite of ISMIP6 experiments involves three types of models: AOGCM/AGCM with no dynamic ice sheets, coupled AOGCM–ISM, and standalone ISM. The diagnostic protocol is based on ice-sheet-related model outputs, many of which are already present in the CMIP6 atmosphere and ocean diagnostics. The evaluation of the climate in the polar regions from AOGCM and AOGCM–ISM simulations will guide recommendations for existing and new ice-sheet–climate coupling efforts. ISMIP6 promotes the development of the ice-sheet component of climate models in an effort to bring both climate and ice-sheet models to greater maturity. ISMIP6 targets two of the WCRP Grand Science Challenges: “Melting Ice and Global Consequences” and “Regional Sea-level Change and Coastal Impacts”. Given the current rapid changes in the Greenland and Antarctic ice sheets, ice sheets can no longer be considered passive players in the climate system. Their contributions to future sea level will likely have considerable human and environmental impacts, and ISMIP6 will facilitate research in this critical area.
ISMIP6 will coordinate simulation and analysis of ice-sheet evolution in a changing climate. Inclusion of ice-sheet models is unique in CMIP history and is necessary to advance understanding of the sea-level contribution from the Greenland and Antarctic ice sheets, the climate system response to ice-sheet changes, and the feedbacks between ice sheets and climate. ISMIP6 is thus an important step in closing the gap between the climate and ice-sheet modeling communities. Our key output, the sea-level contribution from ice sheets, complements the projections of ocean thermal expansion that already sit within the CMIP framework. This improvement will help sea level join the family of variables for which CMIP can provide routine IPCC-style projections. Ultimately, the success of ISMIP6 relies on the broad participation of the CMIP6 modeling centers, standalone ice-sheet modeling groups, and analysts of the atmosphere, ocean, and ice sheets.
The model output from the simulations described in this paper will be
distributed through the Earth System Grid Federation (ESGF) with digital
object identifiers (DOIs) assigned. In order to document CMIP6's scientific
impact and enable ongoing support of CMIP, users are obligated to acknowledge
CMIP6, the participating modeling groups, and the ESGF centers (see details
on the CMIP Panel website at
This special issue includes a manuscript that is dedicated to the CMIP6 data request. The majority of our data request is based on CMIP5 CMOR tables Amon (Monthly Mean Atmospheric Fields), Omon (Monthly Mean Ocean Fields), LImon (Monthly Mean Land Cryosphere Fields), and OImon (Monthly Mean Ocean Cryosphere Fields), which already contained many of the outputs required to diagnose and intercompare the climate over land ice/ice sheets and to derive forcing for the ice sheets. In the CF convention, “land ice” comprises grounded ice sheets, floating ice shelves, glaciers, and ice caps, while “ice sheet” refers to grounded ice sheets and floating ice shelves. A few additional variables are needed to properly derive the forcings for ice sheets from AOGCMs, and to record outputs from the evolving ice sheets in the coupled AOGCM–ISM experiments (such as ice elevation change), or from the standalone ice-sheet simulations. In this Appendix, we briefly outline the ISMIP6 data request on the atmosphere grid (Table A1), ocean grid (Table A2), and ice-sheet grid (Table A3), and provide some context for key new variables.
The mass change in ice sheets (see Fig. A1) is a result of the surface mass balance (SMB), ice melt (or refreeze) at the base of the grounded ice sheet (BMB), and mass exchange with the ocean. The latter can be further split into frontal mass balance (FMB, defined as iceberg calving and melt (or refreeze) at the ice-shelf front) and melt (or refreeze) at the base of ice shelves (BMB). All fluxes are defined as positive when the process adds mass to the ice sheet and negative otherwise. The thermal state of the ice-sheet models is documented by the basal temperature and by the temperature at the ice-sheet–snowpack interface. Note that BMB and basal temperature are computed differently depending on whether the ice is grounded or floating, requiring the use of distinct long names but the same standard names in Table A3.
Climate models will be evaluated primarily based on how well they can simulate SMB over the ice sheets. This quantity (see Vizcaíno, 2014, and Fig. A2) can be defined as precipitation minus runoff minus evaporation (which in our context includes any sublimation, a small term over ice sheets), where precipitation is the sum of snowfall and rainfall. Runoff is the liquid water that escapes the ice sheet, while some of the water may be retained in the snowpack and possibly refreezes. The evaluation of climate models also benefits from analysis of energy fluxes, key temperatures, and area fraction of land ice, grounded ice sheet (excludes ice shelf) and snow over the land ice. Note that some variables, such as SMB, are present in both Tables A1 and A3, since in a coupled AOGCM–ISM simulation, the two will differ due to downscaling to the ice-sheet grid. The data request for the ocean serves primarily as input to construct oceanic forcing for ice-sheet models offline. It is not as extensive as the data request for the atmosphere, because marine boundary conditions for outlet glaciers and ice shelves are not routinely generated by AOGCMs. It is therefore premature to set diagnostic protocols at this stage. However, participants are asked to follow the protocols of the CMIP6-Endorsed Ocean Model Intercomparison Project (OMIP, Griffies et al., 2016) when preparing the data listed in Table A2, in particular when regridding the ocean data from a native grid to the CMIP6 standard grids. The ice-sheet data request contains key characteristics needed to evaluate the ice-sheet geometry and ice-sheet flow. It also contains key ice-sheet-specific boundary conditions that may differ between models and a record of the forcing applied to the ice-sheet model. To facilitate the analysis of the ice-sheet contribution to sea level, a number of integrated measures (for example, ice-sheet mass) are also requested.
Data in the LImon table (Monthly Mean Land
Cryosphere Fields) and/or Amon table (Monthly Mean Atmospheric Fields) needed
to capture the glaciated/ice-sheet surface realm. These fields are saved on
the atmosphere grid and contain monthly output. Tier indicates the priority
of the variable: Mandatory (1), Desirable (2), Experimental (3). These
variables are requested for climate models participating in the diagnostic
component of ISMIP6 (Table 1), and for the
Data on the Omon Tables (Monthly Mean Ocean
Fields) needed to capture the glaciated/ice-sheet surface realm or for
intercomparison of the model simulations. These fields are saved on the ocean
grid and contain monthly output. Data preparation should follow the
CMIP6-Endorsed OMIP protocol. Tier indicates the priority of the variable:
Mandatory (1), Desirable (2), Experimental (3). These variables are requested
for climate models participating in the diagnostic component of ISMIP6
(Table 1) and for the
Data on the limonAnt, limonGre, liyrAnt or
liyrGre tables needed to capture the dynamical ice-sheet model realm. These
fields are saved on the ice-sheet grid and contain monthly or yearly output.
Tier indicates the priority of the variable: Mandatory (1), Desirable (2),
Experimental (3). These variables are requested for models participating in
the
Illustration of the mass change of ice sheets and key processes that are specific to ice-sheet model evaluation or forcing. See text for details.
Illustration of key processes needed to compute atmospheric forcing for ice-sheet models and to evaluate the surface mass balance simulated by climate models.
We thank the CMIP6 panel members for their continuous leadership of the CMIP6 effort, the Working Group on Coupled Modeling (WGCM) Infrastructure Panel (WIP) for overseeing the CMIP6 and ISMIP6 infrastructure, and in particular Martin Juckes and Alison Pamment for their help with the ISMIP6 data request, and Karl Taylor for sharing his wisdom on CMIP experiment protocols. We thank the current ISMIP6 members, the modeling groups, and the wider glaciology community for their contribution in the ISMIP6 design. We acknowledge the Climate and Cryosphere (CliC) Project and the World Climate Research Programme (WCRP) for their guidance, support, and sponsorship. Heiko Goelzer has received funding from the program of the Netherlands Earth System Science Centre (NESSC), financially supported by the Dutch Ministry of Education, Culture and Science (OCW) under grant no. 024.002.001. Sophie Nowicki, Helene Seroussi, and Eric Larour were supported by grants from the NASA Cryospheric Science Program and the NASA Modeling Analysis and Prediction Program. William Lipscomb was supported by the Regional and Global Climate Modeling program of the Office of Biological and Environmental Research within the US Department of Energy's Office of Science. Anthony Payne is supported by the NERC Centre for Polar Observation and Modelling (CPOM). We thank our topical editor Philippe Huybrechts, our reviewers Christian Rodehacke and Xylar Asay-Davis, and everyone who contributed to the open discussion for constructive comments.
The article processing charges for this open-access publication were covered by the NASA Cryosphere Program and the NASA Modeling Analysis and Prediction Program. Edited by: P. Huybrechts Reviewed by: C. Rodehacke and X. Asay-Davis