Global sensitivity analysis in an ocean general circulation model: a sparse spectral projection approach

Abstract

Polynomial chaos (PC) expansions are used to propagate parametric uncertainties in ocean global circulation model. The computations focus on short-time, high-resolution simulations of the Gulf of Mexico, using the hybrid coordinate ocean model, with wind stresses corresponding to hurricane Ivan. A sparse quadrature approach is used to determine the PC coefficients which provides a detailed representation of the stochastic model response. The quality of the PC representation is first examined through a systematic refinement of the number of resolution levels. The PC representation of the stochastic model response is then utilized to compute distributions of quantities of interest (QoIs) and to analyze the local and global sensitivity of these QoIs to uncertain parameters. Conclusions are finally drawn regarding limitations of local perturbations and variancebased assessment and concerning potential application of the present methodology to inverse problems and to uncertainty management.

Development of a reduced model of formation reactions in Zr-Al nanolaminates

“…Suitable metrics are provided by the so-called Sobol sensitivity (global) indices. [56][57][58][59][60][61] In the analysis below, we rely on the first-order indices and total sensitivity indices associated with the first and second components of the germ. The first-order sensitivity indices, S 1 and S 2 , essentially quantify the direct contribution of n 1 (D 0 ) and n 2 (E a ) to the variance of the QoI, Q, whereas the total sensitivity indices, T 1 and T 2 , combine the direct contribution with that arising from mixed terms.…”

Quantifying Prediction Fidelity in Ocean Circulation Models

“…The Bayesian Inference analysis revealed a C max of about 2.3 × 10 −3 , that occurs upward of V max = 34 m/s, and that the ITOP data was D inconclusive with regard to the slope m beyond saturation (figures 3 and 4). This work has been summarized in a manuscript that is currently under review (Sraj et al (2012)). …”

“…This calculation is the most CPU-intensive portion of the forward UQ problem, and directly impacts the accuracy of the resulting PC expansions. In our previous work (Alexanderian et al (2012)) an ensemble of 385 members were required to explore a four-dimensional parameter space where each variable was expanded in 5th-order PC expansion. Subsequent analysis, however, revealed that the quantity of interest (surface temperature) was insensitive to 3 out of the 4 parameters, that the response for 2 of the four parameters is almost linear, and that 7-th order polynomials would be preferable in one direction.…”

Single-site Lennard-Jones models via polynomial chaos surrogates of Monte Carlo molecular simulation

“…[30][31][32][33] The PC-based model provides a complete probabilistic representation of the outputs in terms of the random inputs. PC may suffer from "the curse of dimensionality" 34 which limits the application of PC to only a moderate number of stochastic parameters for surrogate model construction. Nonetheless, PC methods have become one of the standard approaches for solving stochastic problems, to propagate and quantify uncertainties in various disciplines including physical, [35][36][37] chemical, and geophysical systems.…”