Thomas Romary

Thomas Romary
MINES ParisTech | ParisTech

Ph.D.

About

66
Publications
12,622
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
397
Citations
Citations since 2016
18 Research Items
322 Citations
20162017201820192020202120220102030405060
20162017201820192020202120220102030405060
20162017201820192020202120220102030405060
20162017201820192020202120220102030405060

Publications

Publications (66)
Preprint
In the task of predicting spatio-temporal fields in environmental science, introducing models inspired by the physics of the underlying phenomena that are numerically efficient is of growing interest in spatial statistics. The size of space-time datasets calls for new numerical methods to efficiently process them. The SPDE (Stochastic Partial Diffe...
Article
High frequency water quality data measured by in-situ sensors allowed the development of auto-calibration methods for water quality simulation programs. However, these methods consider static parameter values, which may be unrealistic for microorganism activities. An alternative technique is to use data assimilation methods. This paper presents a f...
Preprint
Full-text available
Development of accurate water quality modeling tools is necessary for integrated water quality management of river systems. The existing water quality models can simulate dissolved oxygen (DO) concentration quite well during high flow and phytoplankton blooms in rivers; however, there are discrepancies during the summer low-flow season that are ass...
Article
We discuss the methods and results of the RESSTE team in the competition on spatial statistics for large datasets. In the first sub-competition, we implemented block approaches both for the estimation of the covariance parameters and for prediction using ordinary kriging. In the second sub-competition, a two-stage procedure was adopted. In the firs...
Article
Full-text available
Uranium In Situ Recovery (ISR) is based on the direct leaching of the uranium ore in the deposit by a mining solution. Fluid flow and geochemical reaction in the reservoir are difficult to predict due to geological, petrophysical and geochemical uncertainties. The reactive transport simulation code used to model ISR is very sensitive to the spatial...
Article
The coupling of high frequency data of water quality with physically based models of river systems is of great interest for the management of urban socio-ecosystems. One approach to exploit high frequency data is data assimilation which has received an increasing attention in the field of hydrology, but not for water quality modeling so far. We pre...
Article
The issue of identifying stratigraphic units within a sedimentary succession is of prime importance for reservoir studies, because it allows splitting the reservoir into several units with specific parameters, thus reducing the vertical nonstationarity in simulations. A new method is proposed for semi-automatic determination of the sedimentary unit...
Article
Dissolved oxygen within water column is a key variable to characterize the water quality. Water quality modeling has been extensively developed for decades. However, complex biogeochemical cycles are described using a high number of parameters. Hence, parameters' uncertainty constitutes a major problem in the application of these models. Sensitivit...
Preprint
Full-text available
Large spatial datasets are becoming ubiquitous in environmental sciences with the explosion in the amount of data produced by sensors that monitor and measure the Earth system. Consequently, the geostatistical analysis of these data requires adequate methods. Richer datasets lead to more complex modeling but may also prevent from using classical te...
Article
In this study, we rely on a Bayesian approach to estimate the seismic velocity from first arrival travel times. The advantage of the Bayesian approach compared to linearized ones is its ability to properly quantify the uncertainties associated with the solution. However, this approach remains fairly expensive, and the Markov chain-Monte Carlo algor...
Article
Full-text available
We present an overview of (geo-)statistical models, methods and techniques for the analysis and prediction of continuous spatio-temporal processes residing in continuous space. Various approaches exist for building statistical models for such processes, estimating their parameters and performing predictions. We cover the Gaussian process approach,...
Article
Markov chain Monte Carlo sampling methods are widely used for non-linear Bayesian inversion where no analytical expression for the forward relation between data and model parameters is available. Contrary to the linear(ized) approaches they naturally allow to evaluate the uncertainties on the model found. Nevertheless their use is problematic in hi...
Conference Paper
Full-text available
First arrival time tomography aims at determining the propagation velocity of seismic waves from experimental measurements of their first arrival time. This problem is usually ill-posed and is classically tackled by considering various iterative linearised approaches. However, these methods can yield wrong seismic velocity for highly nonlinear case...
Article
Stationary Random Functions have been successfully applied in geostatistical applications for decades. In some instances, the assumption of a homogeneous spatial dependence structure across the entire domain of interest is unrealistic. A practical approach for modeling and estimating non-stationary spatial dependence structure is considered. This c...
Article
With the increasing development of remote sensing platforms and the evolution of sampling facilities in mining and oil industry, spatial datasets are becoming increasingly large, inform a growing number of variables and cover wider and wider areas. Therefore, it is often necessary to split the domain of study to account for radically different beha...
Article
In the past decade, the training image (TI) has received considerable attention as a source for modeling spatial continuity in geostatistics. In this paper, the use of TIs in the context of kriging is investigated, specifically universal kriging (UK). Traditionally, kriging relies on a random function model formulation whereby the target variable i...
Article
Full-text available
Earthquake hypocentre locations are crucial in many domains of application (academic and industrial) as seismic event location maps are commonly used to delineate faults or fractures. The interpretation of these maps depends on location accuracy and on the reliability of the associated uncertainties. The largest contribution to location and uncerta...
Conference Paper
First arrival travel time tomography aims at estimating the velocity field of the subsurface. The resulting velocity field is then commonly used as a starting point for further seismological, mineralogical or tectonic analysis in a wide range of applications such as geothermal energy, volcanoes studies,... The estimated velocity field is obtained t...
Book
For its tenth edition, the Conference on Geostatistics for Environmental Applications took place at the former École des Mines de Paris, now Mines ParisTech, where Georges Matheron developed the foundations of Geostatistics and numerous renowned researchers followed in his footsteps. From July 9 until July 11, more than 170 experts on geostatistic...
Article
Stationary Random Functions have been successfully applied in geostatistical applications for decades. In some instances, the assumption of a homogeneous spatial dependence structure across the entire domain of interest is unrealistic. A practical approach for modeling and estimating non-stationary spatial dependence structure is considered. This c...
Conference Paper
Full-text available
The velocity field estimated by first arrival traveltime tomography is commonly used as a starting point for further seismological, mineralogical, tectonic or similar analysis. In order to interpret quantitatively the results, the tomography uncertainty values as well as their spatial distribution are required. The estimated velocity model is obtai...
Article
Measurement surveys using passive diffusion tubes are regularly carried out to elaborate atmospheric concentration maps over various areas. Sampling schemes must be designed to characterize both contaminant concentrations (of benzene or nitrogen dioxide for example) and their relations to environmental variables so as to obtain pollution maps as pr...
Article
Kriging of very large spatial datasets is a challenging problem. The size nn of the dataset causes problems in computing the kriging estimate: solving the kriging equations directly involves inverting an n×nn×n covariance matrix. This operation requires O(n3)O(n3) computations and a storage of O(n2)O(n2). Under these circumstances, straightforward...
Article
Geostatistics applied to radiological evaluation of nuclear premises provides methods to estimate radiological activities, together with their uncertainty. These tools enable to build a cartography given a preliminary measurement campaign. This paper provides a methodological study of geostatistical and computational approaches suitable to target a...
Conference Paper
Full-text available
Among many factors that contribute to microseismic location errors, the largest contribution is due to the lack of knowledge of the wave-propagation medium. In spite of efforts to build the “best” velocity model derived from surface seismic and/or logging data, these models are very often not adapted to the microseismic context and are characterize...
Article
Full-text available
Avec le développement des plateformes de télédétection, aéroportées ou satellites, et l'évolution des moyens d'échantillonnage des compagnies minières ou pétroli-ères, les jeux de données spatiales deviennent de plus en plus grands, renseignent un nombre croissant de variables et couvrent des étendues de plus en plus larges. De fait, il devient sou...
Article
Nuclear steam generators are subject to clogging of their internal parts which causes safety issues. Diagnosis methodologies are needed to optimize maintenance operations. Clogging alters the dynamic behaviour of steam generators and particularly the response of the wide range level (WRL – a pressure measurement) to power transients. A numerical mo...
Article
The monitoring of hydrocarbon reservoirs, geothermal reservoirs and mines commonly relies on the analysis of the induced seismicity. Even if a large amount of microseismic data have been recorded, the relationship between the exploration and the induced seismicity still needs to be better understood. This microseismicity is also interpreted to deri...
Patent
Full-text available
A method for evaluating an underground reservoir production scheme accounting for uncertainties is disclosed having applications, for example, to the development of petroleum reservoirs. Flow simulator input parameters characterizing the reservoir and the production scheme are selected. An approximate analytical model allowing the reservoir respons...
Article
Full-text available
Domaining is very often a complex and time-consuming process in mining assessment. Apart from the delineation of envelopes, a significant number of parameters (lithology, alteration, grades) are to be combined in order to characterize domains or subdomains within the envelopes. This rapidly leads to a huge combinatorial problem. Hopefully the numbe...
Article
Full-text available
Tube support plate clogging of steam generators affects their operating and requires frequent maintenance operations. A diagnosis method based on dynamic behaviour analysis is under development at EDF to provide means of optimisation of maintenance strategies. Previous work showed that the dynamic response to a power transient of the wide range lev...
Article
Measurement surveys using passive diffusion tubes are regularly carried out to elaborate atmospheric concentration maps over various areas. Sampling schemes must be designed to characterize both contaminant concentrations (of benzene or nitrogen dioxide for example) and their relations to environmental variables so as to obtain pollution maps as pr...
Article
By definition, kriging with a moving neighborhood consists in kriging each target point from a subset of data that varies with the target. When the target moves, data that were within the neighborhood are suddenly removed from the neighborhood. There is generally no screen effect, and the weight of such data goes suddenly from a non-zero value to a...
Article
Passive seismic monitoring is now a standard tool to follow the evolution of hydrocarbon, geothermal as well as CO2 geological storage reservoirs. Despite the fact that large volumes of data have already been acquired in one of the contexts just mentioned, our understanding of the relation of the microseismicity to the hydraulic fracture geometry i...
Article
Solving the kriging equations directly involves the inversion of an n×n covariance matrix C, which becomes numerically intractable when n is too large. Under these circumstances, straightforward kriging of massive datasets is not possible. A common practice is to solve the kriging problem approximately by local approaches that are based on consider...
Article
In history matching of lithofacies reservoir model, we attempt to find multiple realizations of lithofacies configuration that are conditional to dynamic data and representative of the model uncertainty space. This problem can be formalized in the Bayesian framework. Given a truncated Gaussian model as a prior and the dynamic data with its associat...
Article
Dans ce travail, nous présentons une méthode d'échantillonnage optimal dans un contexte spatial, pour une application aux campagnes de mesure de la concentration de benzène dans l'air ambiant, à l'échelle de l'agglomération. Dans un premier temps, nous nous fondons sur l'analyse des données de campagnes antérieures, réalisées sur deux agglomération...
Article
Full-text available
Markov chains Monte-Carlo (MCMC) methods are known to produce samples of virtually any distribution. They have already been widely used in the resolution of non-linear inverse problems where no analytical expression for the forward relation between data and model parameters is available, and where linearization is unsuccessful. However, in Bayesian...
Article
The history matching problem in reservoir engineering, which consists in matching the geostatistical model to production data, is an ill-posed inverse problem. Its resolution implies to infer the probability distribution of the geostatistical model conditioned to the dynamical data, considering both the geological a priori, expressed in the geostat...
Article
In oil industry and subsurface hydrology, geostatistical models are often used to represent the porosity or the permeability field. In history matching of a geostatistical reservoir model, we attempt to find multiple realizations that are conditional to dynamic data and representative of the model uncertainty space. A relevant way to simulate the c...
Article
Spatial statistics for very large spatial data sets is challenging. The size n of the data set causes problems in computing optimal spatial predictors such as kriging, since its computational cost is of order n^3 . In addition, a large data set is often defi ned on a large spatial domain, so the spatial process of interest typically exhibits non-st...
Article
Full-text available
The history matching problem in reservoir engineering, which consists in matching the geostatistical model to production data, is an ill-posed inverse problem. Its resolu- tion implies to infer the probability distribution of the geostatistical model conditioned to the dynamical data, considering both the geological a priori, expressed in the geost...
Conference Paper
In oil industry and subsurface hydrology, geostatistical models are often used to represent the spatial distribution of different lithofacies in the reservoir. Two main model families exist: multipoint and truncated Gaussian models. We focus here on the latter. In history matching of lithofacies reservoir model, we attempt to find multiple realizat...
Article
Full-text available
La problématique du calage d'historique en ingénierie de réservoir, c'est-à-dire le calage des modèles géostatistiques aux données de production, est un problème inverse mal posé. Dans un cadre bayésien, sa résolution suppose l'inférence de la distribution de probabilité du modèle géostatistique conditionné aux données dynamiques, rendant compte à...
Conference Paper
In history matching of geostatistical model, we attempt to find multiple realizations that are conditional to dynamic data and representative of the model uncertainty space. The huge dimension of the model makes the history matching problem intractable in practice. In recent years, several parameterization methods are proposed for reducing the dime...

Network

Cited By