Software release of the ILAMB Version 2 benchmarking system
June 1, 2016
We have released a parallel and extensible software framework, called the International Land Model Benchmarking version (ILAMBv2) package, which disseminates the biogeochemical benchmarking research conducted by the Biogeochemistry–Climate Feedbacks Scientific Focus Area (SFA). This software release disseminates the research output of the SFA as well as provides the community with the means to make contributions to the ongoing effort the make climate models better. The ease with which this may be done positions the ILAMBv2 software package to be a standard part of the workflow in modeling centers across the globe. The software framework, written in python, compares model results with benchmark observational datasets in parallel on leadership-class supercomputers. Moreover, it provides an abstraction that allows groups to develop plugin to extend its functionality. ILAMBv2 is already in use by the Accelerated Climate Modeling for Energy (ACME) model and the Community Earth System Model (CESM) land model development teams.
Research and disseminate benchmarking procedures and datasets which improve land model fidelity.
- Software framework written in python that compares model outputs to benchmark observational datasets in parallel on leadership-class supercomputers.
- Provides the community a means to comprehensively and routinely assess land model fidelity through comparison of 24 variables with 60 observational datasets.
- Framework provides an abstraction that allows groups to develop plugins to extend its functionality.
- New datasets and comparisons are easily added, encouraging routine use by modeling centers across the globe.
- Package is currently used by ACME and CESM to evaluate new model developments.
Collier, Nathaniel et al. 2016. “The ILAMB Version 2 Benchmarking System.” doi:10.18139/ILAMB.v002.00/1251621.