Ondrej Tichy

Dr. Ondrej Tichy


Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
25 November 2016 in Geoscientific Model Development
Verified
Journal Article
Open access
PDF file
OTOndrej Tichy
Ondrej Tichy
VŠVáclav Šmídl
Václav Šmídl
RHRadek Hofman
Radek Hofman
ASAndreas Stohl
Andreas Stohl
Estimation of pollutant releases into the atmosphere is an important problem in the environmental sciences. It is typically formalized as an inverse problem using a linear model that can explain observable quantities (e.g., concentrations or deposition values) as a product of the source-receptor sensitivity (SRS) matrix obtained from an atmospheric transport model multiplied by the unknown source-term vector. Since this problem is typically ill-posed, current state-of-the-art methods are based on regularization of the problem and solution of a formulated optimization problem. This procedure depends on manual settings of uncertainties that are often very poorly quantified, effectively making them tuning parameters. We formulate a probabilistic model, that has the same maximum likelihood solution as the conventional method using pre-specified uncertainties. Replacement of the maximum likelihood solution by full Bayesian estimation also allows estimation of all tuning parameters from the measurements. The estimation procedure is based on the variational Bayes approximation which is evaluated by an iterative algorithm. The resulting method is thus very similar to the conventional approach, but with the possibility to also estimate all tuning parameters from the observations. The proposed algorithm is tested and compared with the standard methods on data from the European Tracer Experiment (ETEX) where advantages of the new method are demonstrated. A MATLAB implementation of the proposed algorithm is available for download.

1 December 2020 in Geoscientific Model Development
Verified
Journal Article
Open access
OTOndrej Tichy
Ondrej Tichy
LULukáš Ulrych
Lukáš Ulrych
VŠVáclav Šmídl
Václav Šmídl
NENikolaos Evangeliou
Nikolaos Evangeliou
ASAndreas Stohl
Andreas Stohl
Estimation of the temporal profile of an atmospheric release, also called the source term, is an important problem in environmental sciences. The problem can be formalized as a linear inverse problem wherein the unknown source term is optimized to minimize the difference between the measurements and the corresponding model predictions. The problem is typically ill-posed due to low sensor coverage of a release and due to uncertainties, e.g., in measurements or atmospheric transport modeling; hence, all state-of-the-art methods are based on some form of regularization of the problem using additional information. We consider two kinds of additional information: the prior source term, also known as the first guess, and regularization parameters for the shape of the source term. While the first guess is based on information independent of the measurements, such as the physics of the potential release or previous estimations, the regularization parameters are often selected by the designers of the optimization procedure. In this paper, we provide a sensitivity study of two inverse methodologies on the choice of the prior source term and regularization parameters of the methods. The sensitivity is studied in two cases: data from the European Tracer Experiment (ETEX) using FLEXPART v8.1 and the caesium-134 and caesium-137 dataset from the Chernobyl accident using FLEXPART v10.3.

25 October 2017 in Atmospheric Chemistry and Physics
Verified
Journal Article
Open access
PDF file
OTOndrej Tichy
Ondrej Tichy
VŠVáclav Šmídl
Václav Šmídl
RHRadek Hofman
Radek Hofman
KŠKateřina Šindelářová
Kateřina Šindelářová
MHMiroslav Hýža
Miroslav Hýža
ASAndreas Stohl
Andreas Stohl
In the fall of 2011, iodine-131 (131I) was detected at several radionuclide monitoring stations in central Europe. After investigation, the International Atomic Energy Agency (IAEA) was informed by Hungarian authorities that 131I was released from the Institute of Isotopes Ltd. in Budapest, Hungary. It was reported that a total activity of 342 GBq of 131I was emitted between 8 September and 16 November 2011. In this study, we use the ambient concentration measurements of 131I to determine the location of the release as well as its magnitude and temporal variation. As the location of the release and an estimate of the source strength became eventually known, this accident represents a realistic test case for inversion models. For our source reconstruction, we use no prior knowledge. Instead, we estimate the source location and emission variation using only the available 131I measurements. Subsequently, we use the partial information about the source term available from the Hungarian authorities for validation of our results. For the source determination, we first perform backward runs of atmospheric transport models and obtain source-receptor sensitivity (SRS) matrices for each grid cell of our study domain. We use two dispersion models, FLEXPART and Hysplit, driven with meteorological analysis data from the global forecast system (GFS) and from European Centre for Medium-range Weather Forecasts (ECMWF) weather forecast models. Second, we use a recently developed inverse method, least-squares with adaptive prior covariance (LS-APC), to determine the 131I emissions and their temporal variation from the measurements and computed SRS matrices. For each grid cell of our simulation domain, we evaluate the probability that the release was generated in that cell using Bayesian model selection. The model selection procedure also provides information about the most suitable dispersion model for the source term reconstruction. Third, we select the most probable location of the release with its associated source term and perform a forward model simulation to study the consequences of the iodine release. Results of these procedures are compared with the known release location and reported information about its time variation. We find that our algorithm could successfully locate the actual release site. The estimated release period is also in agreement with the values reported by IAEA and the reported total released activity of 342 GBq is within the 99 % confidence interval of the posterior distribution of our most likely model.

28 August 2014 in IEEE Transactions on Medical Imaging
Verified
Journal Article
Closed access
OTOndrej Tichy
Ondrej Tichy
VŠVáclav Šmídl
Václav Šmídl
A common problem of imaging 3-D objects into image plane is superposition of the projected structures. In dynamic imaging, projection overlaps of organs and tissues complicate extraction of signals specific to individual structures with different dynamics. The problem manifests itself also in dynamic tomography as tissue mixtures are present in voxels. Separation of signals specific to dynamic structures belongs to the category of blind source separation. It is an underdetermined problem with many possible solutions. Existing separation methods select the solution that best matches their additional assumptions on the source model. We propose a novel blind source separation method based on probabilistic model of dynamic image sequences assuming each source dynamics as convolution of an input function and a source specific kernel (modeling organ impulse response or retention function). These assumptions are formalized as a Bayesian model with hierarchical prior and solved by the Variational Bayes method. The proposed prior distribution assigns higher probability to sparse source images and sparse convolution kernels. We show that the results of separation are relevant to selected tasks of dynamic renal scintigraphy. Accuracy of tissue separation with simulated and clinical data provided by the proposed method outperformed accuracy of previously developed methods measured by the mean square and mean absolute errors of estimation of simulated sources and the sources separated by an expert physician. MATLAB implementation of the algorithm is available for download.

14 December 2023 in Atmospheric Chemistry and Physics
Verified
Journal Article
Open access
PDF file
OTOndrej Tichy
Ondrej Tichy
SESabine Eckhardt
Sabine Eckhardt
YBYves Balkanski
Yves Balkanski
DHDidier Hauglustaine
Didier Hauglustaine
NENikolaos Evangeliou
Nikolaos Evangeliou
Ammonia (NH3), a significant precursor of particulate matter, affects not only biodiversity, ecosystems, and soil acidification but also climate and human health. In addition, its concentrations are constantly rising due to increasing feeding needs and the large use of fertilization and animal farming. Despite the significance of ammonia, its emissions are associated with large uncertainties, while its atmospheric abundance is difficult to measure. Nowadays, satellite products can effectively measure ammonia with low uncertainty and a global coverage. Here, we use satellite observations of column ammonia in combination with an inversion algorithm to derive ammonia emissions with a high resolution over Europe for the period 2013–2020. Ammonia emissions peak in northern Europe due to agricultural application and livestock management, in western Europe (industrial activity), and over Spain (pig farming). Emissions have decreased by −26 % since 2013 (from 5431 Gg in 2013 to 3994 Gg in 2020), showing that the abatement strategies adopted by the European Union have been very efficient. The slight increase (+4.4 %) in 2015 is also reproduced here and is attributed to some European countries exceeding annual emission targets. Ammonia emissions are low in winter (286 Gg) and peak in summer (563 Gg) and are dominated by the temperature-dependent volatilization of ammonia from the soil. The largest emission decreases were observed in central and eastern Europe (−38 %) and in western Europe (−37 %), while smaller decreases were recorded in northern (−17 %) and southern Europe (−7.6 %). When complemented with ground observations, modelled concentrations using the posterior emissions showed improved statistics, also following the observed seasonal trends. The posterior emissions presented here also agree well with respective estimates reported in the literature and inferred from bottom-up and top-down methodologies. These results indicate that satellite measurements combined with inverse algorithms constitute a robust tool for emission estimates and can infer the evolution of ammonia emissions over large timescales.

1 October 2018 in Quarterly Journal of the Royal Meteorological Society
Verified
Journal Article
Open access
PDF file
OTOndrej Tichy
Ondrej Tichy
VŠVáclav Šmídl
Václav Šmídl
RHRadek Hofman
Radek Hofman
NENikolaos Evangeliou
Nikolaos Evangeliou
Determination of a source term of an accidental release of radioactive material into the atmosphere is very important for evaluating emergency situations and their consequences. However, knowledge of the source term and its composition is typically vague and uncertain. One possible way to obtain the source term is inverse modeling in which an atmospheric transport model is combined with field measurements. The most accessible measurements are those from gamma dose rate (GDR) detectors. However, GDR measurements represent a sum of contribution from all nuclides from both plume and deposition which makes the problem particularly difficult. The same difficulty arises when the measurements can not distinguish contribution from another species in the release, such as nuclides attached to different particle sizes. We propose a Bayesian method for recovery of the source term from GDR measurements where a priori knowledge on ratios of different species is given in the form of bounds. This knowledge is incorporated into the model of covariance matrix of the source term. The Bayesian methodology allows to handle uncertain knowledge on the nuclide ratios as well as unknown temporal correlations of the source term. We evaluate and compare the proposed method with other state‐of‐the‐art methods on a twin experiment of a non‐stationary release of 16 nuclides from the Czech nuclear power plant Temelin being registered by the Austrian GDR monitoring network. Real‐world validation of the approach is performed on the latest measurements of concentration and deposition of caesium‐137 from the Chernobyl accident, where we estimate composition of the source term from different particle sizes (species). The estimated source term is in very good agreement with previously reported results and the calculated species ratios are supported by the available observations.