logo image for top of page

a core project of


Idealized model intercomparison exercises:

MISMIP+: 3rd Marine Ice Sheet Model Intercomparison Project

Co-chairs: Stephen Cornford and Hilmar Gudmundsson

The following description has been adapted from Asay-Davis et al. (2015, under review). It should be noted that the GMD manuscript remains the definitive description of the experiments and the abbreviated description that follows is meant only to present some motivation and a brief overview of the experiments.

The previous Marine Ice Sheet Model Intercomparison Projects, MISMIP and MISMIP3d, tested the capabilities of ice sheet models to simulate advance and retreat cycles under changes in ice softness and basal sliding, respectively. MISMIP tested flowline models in one horizontal dimension (1HD), while MISMIP3d required modeling in two horizontal dimensions (2HD). Each experiment taught the community a great deal about the numerical behavior of ice-sheet models of various types as well as the similarities and differences in the results they produced. In particular, the MISMIP results showed that steady-state grounding-line positions could differ markedly depending on the resolution, type of stress approximation, and discretization methods employed. Follow-up studies found that models with fixed grids (as opposed to those that track the grounding line in time) and without sub-grid-scale parameterizations of the grounding line require grounding-line resolution on the order of hundreds of meters to accurately reproduce grounding-line dynamics.

It was clear in discussions of a follow-up intercomparison exercise that the MISMIP3d experimental design had three shortcomings as a test of 2HD marine ice sheet models. First, it started from a steady state that was invariant in the cross-flow direction—that is, 1HD—and did not involve significant lateral stresses.  Second, the initial grounding lines of the shallow-shelf approximation (SSA) models were around 80km downstream from the Stokes models, but the grounding line only moved about 20 km in the perturbation experiment. That left an obvious question entirely unanswered: in a realistic simulation with the model parameters chosen to match geometry and velocity derived from observations, and thus with prescribed initial conditions, does the SSA provide a good approximation to the Stokes model? Third, grounding line migration was driven by changes to the basal traction field, rather than the ice shelf melting that is thought to be the dominant driver of present-day grounding-line retreat in West Antarctica.

MISMIP+ has been designed to address each of the shortcomings above. Regarding the first, the chosen geometry, based on Gudmundsson et al. (2012), results in strong lateral stresses that buttress the ice stream, and, given particular parameter choices, results in a stable grounding line crossing a retrograde slope, a configuration not possible in 1HD.  Regarding the second, modelers are free to choose certain model parameters so that their initial grounding line is close to that of a reference model, and in preliminary tests two models that bracketed the high resolution MISMIP3d results have been found to have grounding lines within a few kilometers of one another in steady state. Finally, extensive grounding line retreat is driven by sub-shelf melt rates.

The MISMIP+ experiments are initialized by running the model with no melting to steady state. The first MISMIP+ experiment (Ice0) continues the run with no melting for one hundred years to demonstrate that the initial state remains steady over this time period.  The second experiment (Ice1) explores the response of the ice sheet to a strong melt perturbation by prescribes a depth-dependent parameterization of basal melting with a peak melt rate of ~75 m/a, driving 100 (optionally 1000) years of grounding-line retreat.  After 100 years, the melt rate is set back to zero, allowing the grounding line to re-advance over 100 (optionally 900) years.  The third experiment (Ice2) emulates a large calving event by prescribing a strong melt rate (100 m/a) beyond a prescribed calving front for 100 (optionally 1000) years, followed by a removal of the perturbation for an additional 100 (optionally 900) years.

iceshelfmeltEvolution of the basal traction and ice shelf melt rate fields from a BISICLES MISMIP+ run.

ISOMIP+: 2nd Ice Shelf Ocean Model Intercomparison Project

Co-Chairs: Xylar Asay-Davis, Ben Galton-Fenzi

The following description has been adapted from Asay-Davis et al. (2015, under review). It should be noted that the GMD manuscript remains the definitive description of the experiments and the abbreviated description that follows is meant only to present some motivation and a brief overview of the experiments. The following description has been adapte

The ISOMIP+ experiments have been designed to make a number of improvements on the original ISOMIP experiments.  Whereas ISOMIP used highly idealized geometry (the ocean column at the grounding line was 200 m thick, the ice draft sloped linearly with latitude and was invariant with longitude, and the bedrock was perfectly flat), ISOMIP+ makes use of relatively complex geometry from MISMIP+ BISICLES simulations, including an ocean cavity that reaches zero thickness at the grounding line.  Where ISOMIP uses a velocity-independent, two-equation formulationof the melt boundary conditions, ISOMIP+ uses the velocity-dependent three-equation formulation more commonly used in realistic model configurations. ISOMIP specified ~10 km resolution, too coarse to resolve the ~9 km Rossby radius of deformation, and large values of the horizontal viscosity and diffusivities, leading to a laminar flow that evolved toward steady state without eddies or other fluctuations.   In contrast, ISOMIP+ runs will typically use smaller horizontal viscosity and diffusivities and higher resolution (~2 km), allowing for mesoscale eddies and unsteady flow. A smaller computational domain makes the experiments computationally feasible despite the higher resolution.  ISOMIP+ should provide more appropriate test cases than the original ISOMIP for realistic experiments, particularly for those focused on the Amundsen Sea region of WAIS.

ISOMIP+ prescribes five idealized experiments (Ocean0 through Ocean4) for ocean models that support ice-shelf cavities. Ocean0-2 have fixed topography while Ocean3 and Ocean4 have prescribed, evolving ice topography.  The experiments are designed to improve community understanding of the models themselves (both their capabilities and their limitation) as well as physical processes involved ice-ocean interactions.

ISOMIP+ has three main goals:
1. Provide comparable model results as a sanity check during model development,
2. Explore the ocean-model setup on its own before coupling to an ice-sheet model in the MISOMP1 experiments (see below),
3. Provide a basic setup from which a large variety of parameter and process studies could usefully be performed.

The experiments are:

  • Ocean0: 1-2 year run with warm forcing and static ice-shelf geometry,
  • Ocean1: 20-year run with cold-to-warm forcing and static ice-shelf geometry,
  • Ocean2: 20-year run with warm-to-cold forcing and static ice-shelf geometry,
  • Ocean3: 100-year run with warm forcing and retreating ice-shelf geometry,
  • Ocean4: 100-year run with cold forcing and advancing ice-shelf geometry.

misomip temperaturetransectA  transect of temperature  through  the  center of the  ISOMIP+  domain  from  example  simulations  from  the  POP2x ocean model, showing the evolution of the Ocean1 (top) and Ocean2 (bottom) simulations

Topography data sets for each experiment comes from a MISMIP+ simulation (see above) using BISICLES with SSA.

Example output from POP2x runs in the “common” (COM) configuration includes movies as well as NetCDF files demonstrating the format of the output grid. Participants may find it useful to use these files for their own output, replacing the POP2x results with their own.

MISOMIP1: 1st Marine Ice Sheet-Ocean Model Intercomparison Project

Co-Chairs: Xylar Asay-Davis, Helene Seroussi

The following description has been adapted from Asay-Davis et al. (2015, under review).  It should be noted that the GMD manuscript remains the definitive description of the experiments and the abbreviated description that follows is meant only to present some motivation and a brief overview of the experiments.

The first Marine Ice Sheet-Ocean Model Intercomparison Project (MISOMIP1) combines MISMIP+ and ISOMIP+ (see above).  In some ways, the MISOMIP1 setup is similar to that of Goldberg et al. (2012a,b) in that it includes a narrow channel with strong ice-shelf buttressing and strong far-field restoring in the ocean.  MISOMIP1 differs from this previous work in having 1) steeper channel walls, meaning a stronger change in buttressing as the ice-shelf thickness changes, 2) a larger region of open ocean allowing for ocean dynamics both inside and outside the cavity, and 3) making use of  a bedrock topography with an upward-sloping region in the ice-flow direction, allowing us to investigate the possibility that thinning or other changes in the state of the ice sheet could trigger a marine ice-sheet instability (MISI).

MISOMIP1 prescribes two coupled ice sheet-ocean experiments (IceOcean1 and IceOcean2), each with two parts. We expect the MISOMIP1 experiment to play an analogous role in evaluating coupled ice sheet-ocean systems to that of the ISOMIP projects for standalone ocean models with ice-shelf cavities and the MISMIP projects for ice-sheet models.  

IceOcean1 begins with the ice-sheet steady state that also served as the initial conditions for the Ice0, Ice1 and Ice2 experiments. Unlike in ISOMIP+, IceOcean1 does not include a dynamic calving criterion. Ice is allowed to become as thin as the ice sheet and ocean components permit (potentially zero thickness) without calving.

For the first one hundred years, warm restoring (maximum temperature of 1 C) is applied near the ocean's northern boundary, which is expected to induce strong melting and subsequent rapid ice retreat.  After one hundred years, the restoring abruptly switches to cold conditions (constant temperature of -1.9 C) at the ocean's northern boundary, which should shut reduce melting within a decade and allow ice to re-advance for the remaining 100 years of simulation.

Specifying calving is a major problem in the design of MISOMIP1. There was general agreement in the community that ice-sheet models have not been shown to behave reliably with dynamic calving, nor is there any consensus about which calving parameterizations are appropriate or physically realistic.  Nevertheless, we felt that it was important for testing the robustness of the ice-sheet and ocean components in MISOMIP1 that there be an experiment with a dynamic, sheer cliff at the calving front. We include an optional coupled experiment, IceOcean2, that is identical to IceOcean1 except that it includes dynamic calving in the ice-sheet component.

popsiclesExample results from a POPSICLES simulation of the IceOcean1 (top) and IceOcean2 (bottom) experiment showing transects through the center of the domain of ocean temperature and ice topography.

Example output from POPSICLES runs includes movies and example NetCDF results.  Participants may find it useful to use these files for their own output, replacing the POPSICLES results with their own.

end of misomip1

Realistic modeling:

Common Year Forcing

...to be later developed for realistic coupled ice sheet-ocean modeling of the West Antarctic region.

A critical need for the modeling community in the context of glacier-ocean modeling, and particularly for the WAIS, is the availability of an easily accessible, data set with which to drive the models.  Prior to forcing with climate-change scenarios, we will establish a climatological, annually-repeating simulation, forced by a ‘common-year’ forcing data set.  This data set will consist of appropriate atmospheric, oceanic, glaciological, and bathymetric inputs that will be used as initial and boundary conditions to drive all models.  The availability of such standardized, reference forcing data sets, such as the CORE [Large & Yeager, 2009] used to drive global ocean models is common and highly successful in other modeling communities [Griffies et al., 2009].
Our first milestone will be to bring the community together to define and create such a common-year, ice-ocean reference experiment (CIORE) forcing data set.  Many of the data sets needed already exist in existing well-known data bases [e.g., NCEP/NCAR, BEDMAP-2, SODA, etc.].   The major gaps are in the oceanography, glaciology, meteorology, and bathymetry along the critical ice-ocean grounding zone of the WAIS.  

Past experience with defining ‘common-year’ forcing in other modeling communities has highlighted the issue of dealing with natural, interannual variability.  Such communities are often data rich with sufficient observational data to quantify the mean and variance of the initial and boundary conditions of the forcing data set.  For the WAIS system, we do not have such data – in fact in the grounding zone there are no observations, with only limited observations further out on the shelf, or inland on the glacier.  We plan to work with those who have or will collect relevant observations and to integrate those data sets into our COIRE.  Our efforts to define a ‘common-year’ will provide a better baseline than is currently available.
All models will run with COIRE to give us insight into intermodel differences, given identical initial and boundary conditions.  This will be instructive, and with dozen or so models in place, we will begin to establish estimates of simulation robustness, and a measure of the variance in our ensemble of models.  Later stages of our modeling work will involve the application of IPCC forcing scenarios grafted onto our CIORE forcing data set, so that we can address our key target question.


The first product will be a common-year forcing data set, hosted on an appropriate website.  First simulations will focus on steady-state simulations from the various models.  Effort will be spent on identifying the commonalities and differences in physics and numerics among all models, as a pathway to understand the similarities and differences in the sea-level change produced by each model under common-year forcing. The next simulations will focus on model runs under climate-change scenarios, and interannual forcing.  Recommendations on incorporating best-practices in physics and numerics of ice-sheet - ocean interaction will be shared with global climate modeling centers.