Using scientific workflows to calibrate an Australian land surface model (AWRA-L)
Vleeshouwer, Jamie;
Perraud, Jean-Michel
;
Collins, Daniel;
Warren, Garth
;
Gallant, Simon;
Bridgart, Robert
2013-12-31
Conference Material
International Congress on Modelling and Simulation, Adelaide Convention Centre
1-6
All models are incorrect, however model performance can be greatly improved through calibration, whereby parameters are varied so model outputs produce a best fit to observations. Model calibration is computationally expensive, requiring many thousands of simulations. It is, however, a critical process. With more data becoming available and problems becoming more complex, there is great demand for High Performance Computing (HPC). This is especially true for the Australian Water Resources Assessment (AWRA). The AWRA-L landscape hydrology model is one of three model components which form the AWRA system which aims to produce interpretable water balance estimates for Australia, and as much as possible agree with observations such as point gauging data and satellite observations. It is a 5 km grid scale model, driven by interpolated climate inputs, that produces continental surfaces representing water stores and fluxes across the landscape, as well as energy, and vegetation dynamics at daily timestep. The system is jointly developed by CSIRO and the Bureau of Meteorology with model improvements proposed regularly, requiring re-calibration. This regular execution makes scientific workflows an attractive solution for engaging in our large computation and data transformation tasks whilst also providing HPC scheduling, repeatability, traceability and monitoring. This paper describes the process we undertake to calibrate proposed changes to AWRA-L, design objectives, and current state of the workflow activities developed known as the ‘AWRA Calibration Tools’. We demonstrate this with workflows developed in Trident Workbench to automate typical processes including: parameter optimisation; model simulation; and benchmarking of results. These tools allow domain experts to execute and share workflows, enabling AWRA-L to be steered in a direction that either improves model performance or is considered to have an improved physical representation without significant degradation in performance. The paper concludes by identifying key challenges that have emerged, and suggests some improvements for the future.
Unknown
Trident Workbench, Optimisation, Workflows, AWRA, High Performance Computing
Software Engineering
EP137580
Conference Paper - Refereed
English
Vleeshouwer, Jamie; Perraud, Jean-Michel; Collins, Daniel; Warren, Garth; Gallant, Simon; Bridgart, Robert. Using scientific workflows to calibrate an Australian land surface model (AWRA-L). In: International Congress on Modelling and Simulation; Adelaide Convention Centre. Unknown; 2013. 1-6. csiro:EP137580. http://hdl.handle.net/102.100.100/95709?index=1
Loading citation data...
Citation counts
(Requires subscription to view)