Development of a behaviour-pattern based global sensitivity analysis procedure for coupled socioeconomic and environmental models

Abstract Ensemble-based uncertainty quantification and global sensitivity analysis of environmental models requires generating large ensembles of parameter-sets. This can already be difficult when analyzing moderately complex models based on partial differential equations because many parameter combinations cause an implausible model behavior even though the individual parameters are within plausible ranges. In this work, we apply Gaussian Process Emulators (GPE) as surrogate models in a sampling scheme. In an active-training phase of the surrogate model, we target the behavioral boundary of the parameter space before sampling this behavioral part of the parameter space more evenly by passive sampling. Active learning increases the subsequent sampling efficiency, but its additional costs pay off only for a sufficiently large sample size. We exemplify our idea with a catchment-scale subsurface flow model with uncertain material properties, boundary conditions, and geometric descriptors of the geological structure. We then perform a global-sensitivity analysis of the resulting behavioral dataset using the active-subspace method, which requires approximating the local sensitivities of the target quantity with respect to all parameters at all sampled locations in parameter space. The Gaussian Process Emulator implicitly provides an analytical expression for this gradient, thus improving the accuracy of the active-subspace construction. When applying the GPE-based preselection, 70–90% of the samples were confirmed to be behavioral by running the full model, whereas only 0.5% of the samples were behavioral in standard Monte-Carlo sampling without preselection. The GPE method also provided local sensitivities at minimal additional costs.

Download Full-text

Global Sensitivity Analysis of Parameter Uncertainty in Landscape Evolution Models

10.5194/gmd-2017-236 ◽

2017 ◽

Cited By ~ 1

Author(s):

Christopher J. Skinner ◽

Tom J. Coulthard ◽

Wolfgang Schwanghart ◽

Marco J. Van De Wiel ◽

Greg Hancock

Keyword(s):

Sensitivity Analysis ◽

Objective Function ◽

Global Sensitivity Analysis ◽

Landscape Evolution ◽

Sensitivity Analyses ◽

Large Set ◽

Sediment Yields ◽

Global Sensitivity ◽

Environmental Models ◽

Morris Method

Abstract. Landscape Evolution Models have a long history of use as exploratory models, providing greater understanding of the role large scale processes have on the long-term development of the Earth’s surface. As computational power has advanced so has the development and sophistication of these models. This has seen them applied at increasingly smaller scale and shorter-term simulations at greater detail. However, this has not gone hand-in-hand with more rigorous verifications that are commonplace in the applications of other types of environmental models- for example Sensitivity Analyses. This can be attributed to a paucity of data and methods available in order to calibrate, validate and verify the models, and also to the extra complexity Landscape Evolution Models represent – without these it is not possible to produce a reliable Objective Function against which model performance can be judged. To overcome this deficiency, we present a set of Model Functions – each representing an aspect of model behaviour – and use these to assess the relative sensitivity of a Landscape Evolution Model (CAESAR-Lisflood) to a large set of parameters via a global Sensitivity Analysis using the Morris Method. This novel combination of behavioural Model Functions and the Morris Method provides insight into which parameters are the greatest source of uncertainty in the model, and which have the greatest influence over different model behaviours. The method was repeated over two different catchments, showing that across both catchments and across most model behaviours the choice of Sediment Transport formula was the dominate source of uncertainty in the CAESAR-Lisflood model, although there were some differences between the two catchments. Crucially, different parameters influenced the model behaviours in different ways, with Model Functions related to internal geomorphic changes responding in different ways to those related to sediment yields from the catchment outlet. This method of behavioural sensitivity analysis provides a useful method of assessing the performance of Landscape Evolution Models in the absence of data and methods for an Objective Function approach.

Download Full-text

A comprehensive global sensitivity analysis using generic sampling designs by means of a combination of variance- and distribution-based approaches.

10.5194/egusphere-egu2020-5540 ◽

2020 ◽

Author(s):

Gabriele Baroni ◽

Till Francke

Keyword(s):

Sensitivity Analysis ◽

Global Sensitivity Analysis ◽

Present Contribution ◽

Global Sensitivity ◽

Environmental Models ◽

Model Response ◽

New Strategy ◽

Integrated Or ◽

Alternative Solutions

Global sensitivity analysis has been recognized as a fundamental tool to assess the input-output model response and evaluate the role of different sources of uncertainty. Among the different methods, variance- and distribution-based (or also called moment-independent) methods have mostly been applied. The first method relies on variance decomposition while the second method compares the entire distributions. The combination of both methods has also been recognized to provide possibly a better assessment. However, the methods rely on different assumptions and the comparison of indices is not straightforward. For these reasons, the methods are commonly not integrated or even considered as alternative solutions.&#160;In the present contribution, we show a new strategy to combine the two methods in an effective way to perform a comprehensive global sensitivity analysis based on a generic sampling design. The strategy is tested on three commonly-used analytic functions and one hydrological model. The strategy is compared to the state-of-the-art Jansen/Saltelli approach.The results show that the new strategy quantifies main effect and interactions consistently. It also outperforms current best practices by converging with a lower number of model runs. For these reasons, the new strategy can be considered as a new and simple approach to perform global sensitivity analysis that can be easily integrated in any environmental models.

Download Full-text