call for abstract submission to the session Efficient Diagnostics, Sensitivity and Uncertainty Analysis of Complex Environmental Models at this year’s AGU fall meeting in San Francisco from 14-18 December 2015
Environmental models are key monitoring and prediction tools used in applied sciences. Modern environmental models incorporate multiple physical processes, have high-dimensional input/output spaces and tend to increase continually in computation time. Many-query applications such as sensitivity analysis, uncertainty quantification as well as model calibration usually require a large number of model evaluations leading to a high computational demand. This makes rigorous analyses often impractical. In this session we invite contributions that deal with the efficient handling of parameter identifiability, sensitivity analysis, uncertainty quantification, model calibration, or optimisation. We encourage reports of all kind of efficiency improvements that may include but are not limited to algorithmic improvements, dimensionality reductions, approximation techniques, surrogate models, or surrogate assisted methods. We further invite original works that exploit efficient methods on different types of data.
Mary Hill (University of Kansas)
Bryan Tolson (University of Waterloo)
Patrick Reed (Cornell University)
Francesca Pianosi (University of Bristol)
Abstract submission deadline is August 05, 2015