Skip to main content

Sustainable Development and Technologies National Programme of the Hungarian Academy of Sciences


Multi-step modeling of well logging data combining unsupervised and deep learning algorithms for enhanced characterization of the Quaternary aquifer system in Debrecen area, Hungary

Discover the recent research within the Sustainable Development and Technologies project, where we conducted a multi-step modeling approach integrating unsupervised and deep learning algorithms to improve the interpretation of geophysical well-logging data for a more thorough characterization of the Quaternary aquifer system in the Debrecen area, Hungary. This approach facilitated the mapping of lithofacies and variations in hydraulic conductivity across the primary hydrostratigraphical units. Through this integration, we enhanced the understanding of the groundwater system, providing reliable inputs for groundwater modeling in a cost-effective and time-efficient manner.


Full article available here.

Assessing heterogeneous groundwater systems: Geostatistical interpretation of well logging data for estimating essential hydrogeological parameters

In the framework of the Sustainable Development and Technologies project, this research focuses on employing an unsupervised learning approach to interpret well-log data, aiming to characterize the primary hydrostratigraphical units within the Quaternary aquifer system in the Debrecen area of Eastern Hungary. The study introduced a novel approach that bridges the traditional and data-driven machine learning approaches, which proved to be beneficial in characterizing heterogeneous aquifer systems, thereby facilitating successful groundwater resource development and management.

Determination of the ammonia loss from croplands

Half of the applied fertilizers are not used globally and instead pollute the environment. Determining the degree of loss, which is crucial for prevention, is accomplished using models validated by measurements. For validation, knowledge of the concentration (χa) measured above the crop with the photoacoustic method is essential. In this regard, a summary article was published in which we described the models applicable to fertilized fields. We model material flux in the soil-plant-atmosphere system – specifically, the bi-directional exchange of ammonia – using an electrotechnical analogy. The ammonia flux (F), or current, is determined by the ratio of the concentration difference (Δχ) between two media and the resistance to exchange (R). The article discusses in detail the model concepts and the parameterizations of concentrations and resistances. Ammonia emitted by the soil is either directly released into the atmosphere (red line) or partly taken up by vegetation, i.e. recycled (purple line). The aim of our study is not only to measure and model ammonia loss, but also to determine rate of recycling.

Full article available here.


Hyperparameter inversion of well logs for the simultaneous estimation of volumetric and zone parameters

We present a new alternative for the joint inversion of well logs to predict the volumetric and zone parameters in reservoir rocks. Porosity, water saturation, shale content, kerogen and matrix volumes are simultaneously estimated with the tool response function constants with a hyperparameter estimation assisted inversion of open-hole well logs. We treat the zone parameters, i.e., the physical properties of rock matrix constituents, shale, kerogen, and pore-fluids, as well as some textural parameters, as hyperparameters and estimate them using genetic algorithm for the entire processing interval. The significance of the inversion method is in that zone parameters are extracted directly from wireline logs, which both improves the solution of the forward problem and reduces the cost of core sampling and laboratory measurements. In a field study, we demonstrate the feasibility of the inversion method using real well logs collected from a Miocene formation situated in the Pannonian Basin, East Hungary.

Szabó N. P., 2024: Hyperparameter inversion of well logs for the simultaneous estimation of volumetric and zone parameters. GEOPHYSICS, VOL. 89, NO. 4 (JULY-AUGUST 2024); P. D205–D219, 10.1190/GEO2023-0497.1.

Environmental, Social, and Economic Impacts of the Renewal of Urban Brownfield Areas

The study entitled “Environmental, Social, and Economic Impacts of the Renewal of Urban Brownfield Areas ” by Dr. Mariann Szabó, assistant professor and Fruzsina Bozsoki, doctoral student, was published in the 185th volume of the Magyar Tudomány journal of The Hungarian Academy of Sciences.
The study is available in full on the journal’s website.

The starting point of the study is that, even though the discourse related to the revitalization of brownfield areas has a long history in Hungarian development policy practice, its role in promoting the circular economy and environmental sustainability has appeared with little emphasis in Hungary so far, for which the article presents a few theoretical and practical arguments. In the study, the authors introduce a framework presenting the environmental, social and economic benefits of brownfield areas’ redevelopment, highlighting sustainability benefits along concrete examples.

Book chapter on the possibilities of urban brownfield sites in the service of more liveable cities

The study entitled “Possibilities of urban brownfield sites in the service of more liveable cities” was published by doctoral student Fruzsina Bozsoki and assistant professor Dr. Mariann Szabó in a book called ECONOMY, POLITICS AND LANGUAGE: ANALYSES AND RELATIONSHIPS, edited by János Sáringer, published by Apostrophe Publishing House in 2024.
The book is available in full on the publisher’s website.

The river Nile in Miskolc, Hungary

Geophysicists and hydrogeologists from Miskolc and Sopron gave climate researchers a unique tool by mathematically supplementing the gaps in the Nile water level measurements – the very first surviving and uniquely long numerical observation.

The lack of water level measurement covering thirteen centuries (measured from 622 to 1921) has hindered long-term data analysis until now. The longest continuous data series covered “only” 845 years. This study bridges the gaps with the help of a correct mathematical procedure, transforming the annual water level data of the Nile into a single, now uniformly manageable 1300-year time series.

Spectral analysis revealed various periods. The existence of an appr. 400-year period is seen, too. With this, and by sectioning the entire time series into characteristic stages, the study provides climate researchers with a unique and unbiased basis for searching for the causes of the Nile water level changes that have occurred, since there is no comparable quantitative time series in terms of duration, accuracy and awareness.

The study was published in a D1 journal. (Szűcs P, Dobróka M, Turai E, Szarka L, Ilyés Cs, Hamdy E M, Szabó N P, 2024: Journal of Hydrology, 630, 130693, .

The University of Miskolc’s method is also suitable for predicting future events in other catchment areas.

Contact: Csaba Ilyés,