Moment-based metrics for global sensitivity analysis of hydrological systems

Dell’Oca, Aronne; Riva, Mònica; Guadagnini, Alberto

doi:10.5194/hess-21-6219-2017

Cited by 64 publications

(76 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Note that relying on multiple statistical moments explicitly recognizes that variance is not an exhaustive metric to quantify sensitivity (e.g., Dell'Oca et al, 2017;Pianosi and Wagner, 2015;Borgonovo, 2007;Liu et al, 2006). Figure 1 depicts a sketch of the overall concept underpinning (1) when one considers two possible model choices, M 1 and M 2 , each associated with a set of two parameters, i.e., (θ 1 1 , θ 1 2 ) for M 1 and (θ 2 1 , θ 2 2 ) for M 2 .…”

Section: Water Resources Researchmentioning

confidence: 99%

Global Sensitivity Analysis for Multiple Interpretive Models With Uncertain Parameters

Dell’Oca

Riva

Guadagnini

2020

Water Resources Research

View full text Add to dashboard Cite

We propose a set of new indices to assist global sensitivity analysis in the presence of data allowing for interpretations based on a collection of diverse models whose parameters could be affected by uncertainty. Our global sensitivity analysis metrics enable us to assess the sensitivity of various features (as rendered through statistical moments) of the probability density function of a quantity of interest with respect to imperfect knowledge of (i) the interpretive model employed to characterize the system behavior and (ii) the ensuing model parameters. We exemplify our methodology for the case of heavy metal sorption onto soil, for which we consider three broadly used (equilibrium isotherm) models. Our analyses consider (a) an unconstrained case, i.e., when no data are available to constrain parameter uncertainty and to evaluate the (relative) plausibility of each considered model, and (b) a constrained case, i.e., when the analysis is constrained against experimental observations. Our moment‐based indices are structured according to two key components: (a) a model‐choice contribution, associated with the possibility of analyzing the system of interest by taking advantage of multiple model conceptualizations (or mathematical renderings); and (b) a parameter‐choice contribution, related to the uncertainty in the parameters of a selected model. Our results indicate that a given parameter can be associated with diverse degrees of importance, depending on the considered statistical moment of the target model output. The influence on the latter of parameter and model uncertainty evolves as a function of the available level of information about the modeled system behavior.

show abstract

Section: Water Resources Researchmentioning

confidence: 99%

Global Sensitivity Analysis for Multiple Interpretive Models With Uncertain Parameters

Dell’Oca

Riva

Guadagnini

2020

Water Resources Research

View full text Add to dashboard Cite

show abstract

“…In this study we address this by limiting the parameter ranges of the multipliers where we suspect nonlinearity in the model response. In general the choice of the chosen global sensitivity method may yield different results (Dell'Oca et al, 2017). On the other hand, Janetti et al (2019) showed for a regional-scale groundwater study that different global methods showed similar results for hydraulic conductivity parameterization.…”

Section: Sensitivity Of Head and Surface Water Body Flow To Choice Inmentioning

confidence: 99%

“…Lastly, they show in which regions global groundwater models might benefit the most from efforts in improving global datasets like global hydraulic conductivity maps. (Reinecke et al, 2019) is a global groundwater model intended to be coupled with WaterGAP (Döll et al, 2003(Döll et al, , 2012Müller Schmied et al, 2014) and is based on the Open Source groundwater modeling framework G 3 M-f 1 (Reinecke, 2018). It computes lateral and vertical groundwater flows as well as surface water exchanges for all land areas of the globe except Antarctica and Greenland on a resolution of 5 arcmin with two vertical layers with a thickness of 100 m each, representing the aquifer.…”

mentioning

confidence: 99%

Spatially distributed sensitivity of simulated global groundwater heads and flows to hydraulic conductivity, groundwater recharge, and surface water body parameterization

Reinecke

Foglia

Mehl

et al. 2019

Hydrol. Earth Syst. Sci.

View full text Add to dashboard Cite

In global hydrological models, groundwater storages and flows are generally simulated by linear reservoir models. Recently, the first global gradient-based groundwater models were developed in order to improve the representation of groundwater-surface-water interactions, capillary rise, lateral flows, and human water use impacts. However, the reliability of model outputs is limited by a lack of data and by uncertain model assumptions that are necessary due to the coarse spatial resolution. The impact of data quality is presented in this study by showing the sensitivity of a groundwater model to changes in the only available global hydraulic conductivity dataset. To better understand the sensitivity of model output to uncertain spatially distributed parameters, we present the first application of a global sensitivity method for a global-scale groundwater model using nearly 2000 steady-state model runs of the global gradientbased groundwater model G 3 M. By applying the Morris method in a novel domain decomposition approach that identifies global hydrological response units, spatially distributed parameter sensitivities are determined for a computationally expensive model. Results indicate that globally simulated hydraulic heads are equally sensitive to hydraulic conductivity, groundwater recharge, and surface water body elevation, though parameter sensitivities vary regionally. For large areas of the globe, rivers are simulated to be either losing or gaining, depending on the parameter combination, indicat-ing a high uncertainty in simulating the direction of flow between the two compartments. Mountainous and dry regions show a high variance in simulated head due to numerical instabilities of the model, limiting the reliability of computed sensitivities in these regions. This is likely caused by the uncertainty in surface water body elevation. We conclude that maps of spatially distributed sensitivities can help to understand the complex behavior of models that incorporate data with varying spatial uncertainties. The findings support the selection of possible calibration parameters and help to anticipate challenges for a transient coupling of the model.

show abstract

“…Here I aim to accelerate parameter optimization and uncertainty assessment of an LSM using the technique of statistical machine learning-based surrogate modeling, which is theoretically investigated in the field of applied mathematics called uncertainty quantification (Sullivan, 2015). Although this technique has been used for the parameter sensitivity analysis of atmospheric models (e.g., Qian et al, 2018), hydrological models (e.g., Dell'Oca et al, 2017;Maina & Guadagnini, 2018;Teixeira Parente et al, 2019), and ecological models (e.g., Hawkins et al, 2019), few studies have applied it to parameter optimization and uncertainty assessment of LSMs with globally applicable satellite observations. In this study, a statistical surrogate model, which can mimic the relationship between model parameters and gaps between simulation and observation, is developed using machine learning.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning Accelerates Parameter Optimization and Uncertainty Assessment of a Land Surface Model

Sawada

2020

JGR Atmospheres

View full text Add to dashboard Cite

The performance of land surface models (LSMs) significantly affects the understanding of atmospheric and related processes. Many of the LSMs' soil and vegetation parameters were highly uncertain so that it is crucially important to efficiently optimize them. Here I present a globally applicable and computationally efficient method for parameter optimization and uncertainty assessment of the LSM by combining Markov Chain Monte Carlo (MCMC) with machine learning. First, I performed the long-term (decadal-scale) ensemble simulation of the LSM, in which each ensemble member has different parameters' values, and calculated the gap between simulation and observation, or the cost function, for each ensemble member. Second, I developed the statistical machine learning-based surrogate model, which is computationally cheap but accurately mimics the relationship between parameters and the cost function, by applying the Gaussian process regression to learn the model simulation. Third, I applied MCMC by repeatedly evaluating the surrogate model to get the posterior probabilistic distribution of parameters. Using satellite passive microwave brightness temperature observations, both synthetic and real-data experiments in the Sahel region of west Africa were performed to optimize unknown soil and vegetation parameters of the LSM. The primary findings are (1) the proposed method is 50,000 times as fast as the direct application of MCMC to the full LSM; (2) the skill of the LSM to simulate both soil moisture and vegetation dynamics can be improved; and (3) I successfully quantify the characteristics of equifinality by obtaining the full nonparametric probabilistic distribution of parameters.

show abstract

Moment-based metrics for global sensitivity analysis of hydrological systems

Cited by 64 publications

References 38 publications

Global Sensitivity Analysis for Multiple Interpretive Models With Uncertain Parameters

Global Sensitivity Analysis for Multiple Interpretive Models With Uncertain Parameters

Spatially distributed sensitivity of simulated global groundwater heads and flows to hydraulic conductivity, groundwater recharge, and surface water body parameterization

Machine Learning Accelerates Parameter Optimization and Uncertainty Assessment of a Land Surface Model

Contact Info

Product

Resources

About