The Design Document page provides a description of the algorithms, implementation and planned testing including unit, verification, validation and performance testing. Please read Step 1.3 Performance Expectations that explains feature documentation requirements from the performance group point of view.
Design Document
Expand | ||
---|---|---|
| ||
The first table in Design Document gives overview of this document, from this info the Design Documents Overview page is automatically created. In the overview table below 4.Equ means Equations and Algorithms, 5.Ver means Verification, 6.Perf - Performance, 7. Val - Validation
Use the symbols below (copy and paste) to indicate if the section is in progress or done or not started. |
...
Page Properties | ||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||
|
Expand | |||
---|---|---|---|
| |||
|
Title: O26_LI Add Land Ice Meshes to ACME
Requirements and Design
ACME Ocean and Ice Group
Date: 2015-9-23
Summary
ACME will be one of the first climate models to include a fully coupled, dynamic ice sheet component. Meshes appropriate for use with the MPAS-LI model, the ice sheet model component in ACME, will need to be added to ACME for: 1) testing purposes, 2) use in configurations where the ice sheet may be dynamically inactive but coupled thermodynamically to the ocean component, and 3) use in configurations with a fully coupled and active ice sheet model component.
...
.
Requirements
Requirement: Coarse, uniform resolution meshes for testing
Date last modified: 2015-9-23
Contributors: Matt Hoffman, Stephen Price
Testing of the new ice sheet and climate model coupling in ACME will require inexpensive, coarse, uniform resolution meshes for the land ice model component. This mesh will be used for testing MPAS-LI as an active component in ACME and also for testing initial coupling to the other components of the climate model.
Requirement: Variable resolution meshes for static ice sheet configurations
Date last modified: 2015-9-23
Contributors: Matt Hoffman, Stephen Price
Some ACME simulations may choose to have an inactive ice sheet component. In this case, we can still allow for thermodynamic coupling between the ocean and ice sheet (e.g., so that the ocean can circulate under static ice shelves and exchange heat and fresh water fluxes with the ice sheet). This will require variable resolution meshes with intermediate spatial dimensions.
Requirement: Variable resolution meshes for active ice sheet configurations
Date last modified: 2015-9-23
Contributors: Matt Hoffman, Stephen Price
Accurate simulation of marine ice sheet dynamics - e.g., retreat and advance of ice sheet grounding lines - requires extremely high spatial grid resolution (>=1km) in limited portions of the computational domain covered by the ice sheet. Use of a 3d, higher-order ice sheet model on a sub-km, uniform resolution mesh of Antarctica is not a computationally efficient approach. As such, the MPAS-Land Ice model allows for the use of variable resolution meshes, to focus spatial resolution and computational power in select regions while simultaneously allowing for low spatial resolution in places of low dynamic complexity. ACME simulations targeting a better understanding of future mass loss from the Antarctic ice sheet will thus require the use of high quality, variable resolutions meshesACME simulations with an active ice sheet component will require variable resolution meshes that allow for this localized high resolution in order to insure accurate simulations of marine ice sheet evolution.
Algorithmic Formulations
Design solution:
short-description-of-proposed-solution-hereUniform and variable resolution SCVT generation
Date last modified:
// date2015-9-23
Contributors: Matt Hoffman, Stephen Price
(add your name to this list if it does not appear)
For each requirement, there is a design solution that is intended to meet that requirement. Design solutions can include detailed technical discussions of PDEs, algorithms, solvers and similar, as well as technical discussion of performance issues. In general, this section should steer away from a detailed discussion of low-level software issues such as variable declarations, interfaces and sequencing.
Design and Implementation
Implementation: short-desciption-of-implementation-here
Date last modified: // date
Contributors: (add your name to this list if it does not appear)
This section should detail the plan for implementing the design solution for requirement XXX. In general, this section is software-centric with a focus on software implementation. Pseudo code is appropriate in this section. Links to actual source code are appropriate. Project management items, such as svn branches, timelines and staffing are also appropriate. How do we typeset pseudo code?
Design and Implementation
Implementation: Coarse, uniform resolution Greenland mesh
Date last modified: 2015-9-23
Contributors: Matt Hoffman, Stephen Price
A uniform resolution, 20 km Greenland ice sheet mesh has been designed for the purposes of testing MPAS-LI as an active component in ACME and for testing the coupling between the land and land ice components.
This grid is currently available within ACME (sname="0.9x1.25_gx1v6_gis20" alias="f09_g16_g").
Implementation: Coarse, uniform resolution Antarctic mesh
Date last modified: 2015-9-23
Contributors: Matt Hoffman, Stephen Price
A uniform resolution, 20 km Antarctic ice sheet mesh has been designed for the purposes of testing MPAS-LI as an active component in ACME and for testing the "plumbing" of ice sheet - to - ocean and ice sheet - to - land coupling.
This grid is currently available on an ACME branch (sname="0.9x1.25_gx1v6_ais20" alias="f09_g16_a").
Implementation: Coarse, variable resolution Antarctic mesh
Date last modified: 2015-9-23
Contributors: Matt Hoffman, Stephen Price
A variable resolution, ~15-4 km Antarctic ice sheet mesh has been designed for the purposes of more rigorous testing of MPAS-LI as an active component in AMCE (i.e., allowing for better resolution of localized dynamics and flow features) and for use in configurations where a static, but high-resolution representation of the ice sheet geometry may be needed for thermodynamic coupling to the ocean.
This grid exists and has been tested extensively in MPAS-LI standalone simulations but has not yet been added to ACME.
Implementation: Fine, variable resolution Antarctic mesh
Date last modified: 2015-9-23
Contributors: Matt Hoffman, Stephen Price
A variable resolution, ~20-1 km Antarctic ice sheet mesh will be designed for use in any ACME simulation with an active Antarctic ice sheet component. The mesh generation process is expected to follow similarly from the process of designing and generating the ~15-4 km variable resolution mesh. The necessary density functions for generating the mesh are already in hand.
Planned Verification and Unit Testing
Verification and Unit Testing:
short-desciption-of-testing-herestand-along model runs vs. coupled model runs
Date last modified:
2015/09/25
Contributors:
(add your name to this list if it does not appear)Matt Hoffman, Stephen Price
How will XXX be tested? i.e. how will be we know when we have met requirement XXX. Will these unit tests be included in the ongoing going forward?We will verify the new model meshes by comparing results from runs of the land ice model in ACME, with idealized forcing (e.g. simplified or no climate model forcing), against output from stand-alone model runs using the same meshes. Identical output from from stand alone runs vs. runs conducted within ACME will verify that the meshes are working as expected in ACME.
Planned Validation Testing
Validation Testing:
short-desciption-of-testing-hereadequate mesh resolution in areas of dynamic complexity
Date last modified:2015/09/25
Contributors:
(add your name to this list if it does not appear)How will XXX be tested? What observational or other dataset will be used? i.e. how will be we know when we have met requirement XXX. Will these unit tests be included in the ongoing going forward?
Matt Hoffman, Stephen Price
The purpose of the fixed, coarse resolution and variable, coarse resolution meshes is largely for testing and we do not anticipate performing any "validation" of these meshes.
The purpose of the variable, fine resolution meshes is to ensure adequate spatial resolution for resolving dynamic processes of interest to ice sheet evolution (e.g., resolution of ice streams and accurate representation of grounding line retreat and advance). Because this is an entirely new model and set of processes that has never been included in a climate model before, there are no set tools or procedures for use in model validation. Under the PISCEES project, we are developing new methods and frameworks for validating ice sheet models as coupled components of climate models and we will adopt these for validation within ACME as appropriate. In the short-term, the metrics and diagnostics discussed on the /wiki/spaces/OCNICE/pages/1867925 will be used to validate ice sheet model output with respect to a particular mesh.
Planned Performance Testing
Performance Testing: short-desciption-of-testing-here
Date last modified: 2015/09/25
Contributors: Matt Hoffman, Stephen Price
How will XXX be tested? i.e. how will be we know when we have met requirement XXX. Will these unit tests be included in the ongoing going forward?
As noted above, the fixed, coarse resolution and variable, coarse resolution meshes are primarily for testing purposes and we do not anticipate evaluating their performance (aside from, e.g., their use in evaluating how other new processes, such as ice sheet - ocean - coupling and boundary layer physics might impact overall ocean model performance, which is a separate performance evaluation task).
The exact version of the variable, fine resolution ice sheet meshes could have an impact on ice sheet model performance (e.g., if one version had a significantly larger amount of small, ~km scale grid cells than another version it would likely be more expensive). Our plan for evaluating this would be to:
- obtain a performance baseline using our nominal (planned) ~20-1 km resolution mesh
- evaluate the scientific results of the model on this mesh (following the validation criteria noted above)
- if altered meshes with higher-resolution are deemed necessary, we will evaluate the performance impacts of these new meshes vs. the improvement in the simulation (following validation criteria noted above)
References
Ringler, T., L. Ju, and M. Gunzburger. 2008. A multiresolution method for climate system modeling: application of spherical centroidal Voronoi tessellations. Ocean Dynamics, 58(5), 475-498. link