Skip to main content
U.S. flag

An official website of the United States government

Publication Date
22 March 2019

LIVVkit Identifies a Bias in the Surface Mass Balance of CESM that is Due to Insufficient Melting over Greenland’s Southwest Region

Subtitle
Validation using LIVVkit 2.1 enables scientists to track the representation of surface mass balance by ice sheet models.
Print / PDF
Powerpoint Slide
Image
Surface Mass Balance from a CISM-Albany simulation as presented within LIVVkit
Science

In order to provide credible predictions of sea level rise to policymakers and stakeholders, scientists need accurate representations of land ice as simulated within Earth system models.  Confidence in the simulations is attained through a robust validation of these models that is available to the community. Presently, a collection of scientific analyses, metrics, and visualizations using the Land Ice Verification and Validation toolkit (LIVVkit), version 2.1, and the LIVVkit Extensions repository (LEX), version 0.1 is made available. LIVVkit aims to enable efficient and fully reproducible workflows for postprocessing, analysis, and visualization of observational and model-derived datasets from any model in a shareable format, whereby all data, methodologies, and output are distributed to users for evaluation. We demonstrate the capability applied to CESM1.0 with an active ice sheet model, the Community Ice Sheet Model (CISM), and an idealized stand-alone high-resolution simulation with CISM using the Albany solver (CISM-A) over Greenland. 

Impact

The paucity of observational data in the polar regions, especially data that are specifically relevant to ice sheets themselves, has prevented a comprehensive assessment of model skill, and therefore knowledge about the key climatological forcings that drive ISM evolution to date.  To address this issue, LIVVkit analyzes the degree to which models capture the surface mass balance (SMB) and identifies potential sources of bias, using recently available in situ and remotely sensed data as a comparison. Applied to the CESM1.0, LIVVkit identifies a positive SMB bias that is focused largely around Greenland’s southwest region, due to insufficient ablation, or melting. Related fields within atmosphere and land surface models, e.g., surface temperature, radiation, and cloud cover, are also diagnosed. From this initial capability, it is straightforward for ice sheet scientists to adapt the software to include new data and analyses and share their advancements with others. 

Summary

The Land Ice Verification and Validation toolkit (LIVVkit), version 2.1, has been developed to enable scientists to evaluate both stand-alone ice sheet and coupled Earth system models. It handles datasets and analyses that require high-performance computing and storage and enables efficient and fully reproducible workflows for postprocessing, analysis, and visualization of observational and model-derived datasets in a shareable format, whereby all data, methodologies, and output are distributed to users for evaluation. Extending from the initial LIVVkit software framework, we demonstrate the software by analyzing a Greenland ice sheet simulation within a coupled Community Earth System Model (CESM) as well as an idealized stand-alone high-resolution Community Ice Sheet Model (CISM-Albany). Using a suite of analyses, we are able to identify bias in the surface mass balance, locate it within the southwest region of Greenland that originates from too little melting, and further attribute the bias to the mismatch between the model and observed elevation of the ice sheet in that region.  This depth of analysis informs future model ice sheet and coupled model developments ongoing within the E3SM.

Point of Contact
Katherine Evans
Institution(s)
Oak Ridge National Laboratory (ORNL)
Publication
LIVVkit 2.1: Automated and Extensible Ice Sheet Model Validation
Evans, Katherine J, Joseph Kennedy, Dan Lu, Mary Michael Forrester, Stephen Price, Jeremy Fyke, Andrew Bennett, et al. 2019. “Livvkit 2.1: Automated And Extensible Ice Sheet Model Validation”. Geoscientific Model Development 12: 1067-1086. doi:10.5194/gmd-12-1067-2019.