High‐dimensional posterior exploration of hydrologic models using multiple‐try DREAM(ZS) and high‐performance computing

By Eric Laloy, Jasper A. Vrugt. 2012

532 citations recorded by the source corpus.

Subfields: Environmental Engineering, Global and Planetary Change, Water Science and Technology

View Paper: https://www.semanticscholar.org/paper/7c1d1cf8170b17e368e43831642a289f00686d73

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Environmental Engineering

In Global and Planetary Change

In Water Science and Technology

Citations in the corpus

Assessment of parametric uncertainty for groundwater reactive transport modeling (2014, background): “While DREAM requires at least N (the number of parameters) chains to be run in parallel, DREAM(ZS) generates candidate points from an archive of past states rather than the current locations of the N chains [Laloy and Vrugt, 2012].”
Implementation and Application of a Root Growth Module in HYDRUS (2018, background): “Based on a proposal distribution, the sampler evolves to the posterior distribution by iteratively finding solutions with stable frequencies stemming from the fixed probability distribution (Laloy and Vrugt, 2012).”
Training‐Image Based Geostatistical Inversion Using a Spatial Generative Adversarial Neural Network (2017, methodology): “…are compared the As no analytical solution of p ( θ | d ) is available for the type of non-linear inverse problems considered herein, we sample from p ( θ | d ) by MCMC simulation (see, e.g., Robert and Casella , 2004) with the DREAM (ZS) algorithm ( Vrugt et al. , 2009; Laloy and Vrugt , 2012).”
Markov chain Monte Carlo simulation using the DREAM software package: Theory, concepts, and MATLAB implementation (2016, background): “Benchmark experiments in geophysics, hydrology and hydrogeology have shown that this sampler1200 is able to sample correctly high-dimensional target distributions (Laloy and Vrugt, 2012a; Laloy et al., 2012b, 2013; Linde and Vrugt, 2013;1202 Carbajal et al., 2014; Lochbühler et al., 2014, 2015).”
Bridging the gap between GLUE and formal statistical approaches: approximate Bayesian computation (2013, background): “…of a formal hierarchical framework to formulate, build and test different watershed models (Clark et al., 2008); and algorithms for efﬁcient sampling of parameter and predictive uncertainty distributions (Kuczera and Parent, 1998; Vrugt et al., 2008a; Kuczera et al., 2010; Laloy and Vrugt, 2012).”
Efficient posterior exploration of a high‐dimensional groundwater model from two‐stage Markov chain Monte Carlo simulation and polynomial chaos expansion (2013, methodology): “In the remainder of this paper, we refer to this approach as strategy A. [41] The state-of-the-art DREAM ZSð Þ [ter Braak and Vrugt, 2008; Vrugt et al., 2009; Laloy and Vrugt, 2012] algorithm is used to generate posterior samples.”
Probabilistic inference of multi‐Gaussian fields from indirect hydrological data using circulant embedding and dimensionality reduction (2015, methodology): “…Gaussian process generation via circulant embedding [Dietrich and Newsam, 1997] to decouple the variogram from grid cell specific values, and implements dimensionality reduction by interpolation to enable MCMC simulation with the DREAM ZSð Þ algorithm [Vrugt et al., 2009; Laloy and Vrugt, 2012].”
Geological realism in hydrogeological and geophysical inverse modeling: A review (2015, methodology): “They combined adaptive MCMC, differential evolution and subspace sampling as implemented in the DREAM(ZS) algorithm [105] to invert for the porosity field and a rock physics transform using crosshole geophysical data.”
Measurement and Control of Single Nitrogen-Vacancy Center Spins above 600 K (2012, methodology): “text, we employ the DREAMZS algorithm written in MATLAB [4, 5], a Markov Chain Monte Carlo (MCMC) approach that uses adaptive proposal distribution tuning, sampling from the past,”
Modeling Soil Processes: Review, Key Challenges, and New Perspectives (2016, methodology): “Recent developments include multiple try sampling, snooker updates, and sampling from an archive of past states (Laloy and Vrugt, 2012).”
The stationarity paradigm revisited: Hypothesis testing using diagnostics, summary metrics, and DREAM(ABC) (2015, methodology): “To minimize the burn in, we use a related variant of DREAM(ABC) which uses past states in the jumping distribution of equation (1) [Laloy and Vrugt, 2012; Vrugt, 2015].”
A global scale mechanistic model of photosynthetic capacity (LUNA V1.0) (2016, methodology): “A detailed description of DREAM(ZS) appears in Vrugt et al. (2008, 2009) and Laloy and Vrugt (2012) and interested readers are referred to these publications for further details.”
Evaluating marginal likelihood with thermodynamic integration method and comparison with several other numerical methods (2016, methodology): “…MARGINAL LIKELIHOOD 741 commercial software, Mathematica; AME is evaluated using simple Monte Carlo simulation that generates random parameter samples from prior parameter distributions directly; the MCMC simulation needed by HME and TIE is conducted using MT-DREMA(ZS) [Laloy and Vrugt, 2012].”
Toward diagnostic model calibration and evaluation: Approximate Bayesian computation (2013, methodology): “The value of ̂ can be specified a priori based on knowledge of the measurements errors, or alternatively its value can be inferred simultaneously with the values of [Vrugt et al., 2008b; Laloy and Vrugt, 2012].”