https://core.ac.uk/download/pdf/374223.pdf

A manifesto for the equifinality thesis

Equifinality, data assimilation, and uncertainty estimation in mechanistic modelling of complex environmental systems using the GLUE methodology

Improvement of a parsimonious model for streamflow simulation

Author(s): Vrugt, JA; Gupta, HV; Bouten, W; Sorooshian, S | Abstract: Markov Chain Monte Carlo (MCMC) methods have become increasingly popular for estimating the posterior probability distribution of parameters in hydrologic models. However, MCMC methods require the a priori definition of a proposal or sampling distribution, which determines the explorative capabilities and efficiency of the sampler and therefore the statistical properties of the Markov Chain and its rate of convergence. In this paper we present an MCMC sampler entitled the Shuffled Complex Evolution Metropolis algorithm (SCEM-UA), which is well suited to infer the posterior distribution of hydrologic model parameters. The SCEM-UA algorithm is a modified version of the original SCE-UA global optimization algorithm developed by Duan et al. [1992]. The SCEM-UA algorithm operates by merging the strengths of the Metropolis algorithm, controlled random search, competitive evolution, and complex shuffling in order to continuously update the proposal distribution and evolve the sampler to the posterior target distribution. Three case studies demonstrate that the adaptive capability of the SCEM-UA algorithm significantly reduces the number of model simulations needed to infer the posterior distribution of the parameters when compared with the traditional Metropolis-Hastings samplers.

/pdf/a-shuffled-complex-evolution-metropolis-algorithm-for-5cb56y4idu.pdf

A Shuffled Complex Evolution Metropolis algorithm for optimization and uncertainty assessment of hydrologic model parameters

A Shuffled Complex Evolution Metropolis algorithm for optimization and uncertainty assessment of hydrological model parameters

Monte Carlo assessment of parameter uncertainty in conceptual catchment models: the Metropolis algorithm

The response of soil carbon dynamics to climate and land-use change will affect both the future climate and the quality of ecosystems. Deep soil carbon (> 20cm) is the primary component of the soil carbon pool, but the dynamics of deep soil carbon remains poorly understood. Therefore, radiocarbon activity (ΔD14 C), which is a function of the age of carbon, may help to understand rates of soil carbon biodegradation and stabilization. We analyzed the published 14 C contents in 122 profiles of mineral soil that were well distributed in most of the large world biomes, except for the boreal zone. With a multivariate extension of a linear mixed effects model whose inference was based on the parallel combination of two algorithms, the Expectation-Maximization (EM) and the Metropolis-Hasting algorithms, we expressed soil Δ14 C profiles as a four-parameter function of depth. The four-parameter model produced insightful predictions of soil Δ14 C as dependent on depth, soil type, climate, vegetation, land-use and date of sampling (R2 = 0.68). Further analysis with the model showed that the age of topsoil carbon was primarily affected by climate and cultivation. By contrast, the age of deep soil carbon was affected more by soil taxa than by climate and thus illustrated the strong dependence of soil carbon dynamics on other pedologic traits such as clay content and mineralogy.

Deep soil carbon dynamics are driven more by soil type than by climate: a worldwide meta-analysis of radiocarbon profiles.

https://rmgsc.cr.usgs.gov/outgoing/threshold_articles/Perreault2000.pdf

Bayesian change-point analysis in hydrometeorological time series. Part 1. The normal model revisited

A Bayesian state-space modelling framework for fitting a salmon stage-structured population dynamic model to multiple time series of field data

We propose a new stochastic algorithm for Bayesian-optimal design in nonlinear and high-dimensional contexts. Following Peter Muller, we solve an optimization problem by exploring the expected utility surface through Markov chain Monte Carlo simulations. The optimal design is the mode of this surface considered a probability distribution. Our algorithm relies on a “particle” method to efficiently explore high-dimensional multimodal surfaces, with simulated annealing to concentrate the samples near the modes. We first test the method on an optimal allocation problem for which the explicit solution is available, to compare its efficiency with a simpler algorithm. We then apply our method to a challenging medical case study in which an optimal protocol treatment needs to be determined. For this case, we propose a formalization of the problem in the framework of Bayesian decision theory, taking into account physicians' knowledge and motivations. We also briefly review further improvements and alternatives.

Eric Parent

Papers

Monte Carlo assessment of parameter uncertainty in conceptual catchment models: the Metropolis algorithm

Deep soil carbon dynamics are driven more by soil type than by climate: a worldwide meta-analysis of radiocarbon profiles.

Bayesian change-point analysis in hydrometeorological time series. Part 1. The normal model revisited

A Bayesian state-space modelling framework for fitting a salmon stage-structured population dynamic model to multiple time series of field data

Bayesian-Optimal Design via Interacting Particle Systems