János Mészáros
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
János Mészáros has a total of 196 co-authors. Here are the top co-authors by number of shared publications:
- Brigitta Szabó
HUN-REN Agrártudományi Kutatóközpont, Talajtani Intézet, Talajfizikai és Vízgazdálkodási Osztály, Budapest, Magyarország
19 shared publications - László Pásztor
Department of Soil Mapping and Environmental Informatics, Institute for Soil Sciences, HUN-REN Centre for Agricultural Research, Fehérvári út 132-144, 1116 Budapest, Hungary
19 shared publications - Gábor Szatmári
Institute for Soil Sciences, HUN-REN Centre for Agricultural Research, Budapest, Hungary
14 shared publications Annamária Laborczi
Institute for Soil Sciences, HUN-REN Centre for Agricultural Research, Budapest, Hungary
12 shared publications- Katalin Takács
Institute for Soil Sciences and Agricultural Chemistry, Centre for Agricultural Research, Hungarian Academy of Sciences, Herman Ottó út 15, 1022 Budapest, Hungary
10 shared publications Zsófia Bakacsi
Institute for Soil Sciences, HUN-REN Centre for Agricultural Research, Budapest, Hungary
9 shared publications
Latest Publications
- Open Access
- 18 September 2026
- Published by Copernicus GmbH in EGUsphere
- Vol. 2026, 1-53
A long-term multiscale Critical Zone dataset integrating environmental monitoring and an open-air laboratory: The Alento River Catchment Observatory
This paper presents an open-access dataset from the Alento River Catchment (ARC) in southern Italy, a strategic Mediterranean Critical Zone observatory integrating long-term environmental monitoring with an open-air laboratory. Long-term observations of weather variables, streamflow, reservoir water level, groundwater depth, soil water content, and soil matric potential are complemented by targeted field campaigns involving hydrogeophysical surveys, stable water isotope analyses, soil physical and hydraulic measurements, spectral measurements, and high-resolution UAV-based mapping. The principal value of the ARC dataset lies in the coordinated integration of continuous and campaign-based observations across multiple spatial and temporal scales and Critical Zone compartments, providing a basis for linking localized soil and subsurface processes with field- and catchment-scale hydrological responses. Since 2016, the ARC has been a component of the European TERrestrial Environmental Observatories (TERENO) network. The dataset includes both directly measured and derived products, with documented processing procedures, quality-control information, calibration results, and missing-data records. The principal value of the ARC dataset lies in the coordinated integration of continuous and campaign-based observations across multiple spatial and temporal scales and Critical Zone compartments. The combined observations can support the development, calibration, evaluation, and intercomparison of process-based hydrological and Critical Zone models, the evaluation of Earth Observation products, and the investigation of scale-dependent relationships among point-, field-, hillslope-, and catchment-scale observations. The complete dataset is openly accessible through Figshare under a CC BY 4.0 license. (Romano et al., 2026; https://doi.org/10.6084/m9.figshare.31229608).
- 18 September 2026
- Published by Copernicus GmbH in EGUsphere
- Vol. 2026, 1-53
Open AccessA long-term multiscale Critical Zone dataset integrating environmental monitoring and an open-air laboratory: The Alento River Catchment Observatory
This paper presents an open-access dataset from the Alento River Catchment (ARC) in southern Italy, a strategic Mediterranean Critical Zone observatory integrating long-term environmental monitoring with an open-air laboratory. Long-term observations of weather variables, streamflow, reservoir water level, groundwater depth, soil water content, and soil matric potential are complemented by targeted field campaigns involving hydrogeophysical surveys, stable water isotope analyses, soil physical and hydraulic measurements, spectral measurements, and high-resolution UAV-based mapping. The principal value of the ARC dataset lies in the coordinated integration of continuous and campaign-based observations across multiple spatial and temporal scales and Critical Zone compartments, providing a basis for linking localized soil and subsurface processes with field- and catchment-scale hydrological responses. Since 2016, the ARC has been a component of the European TERrestrial Environmental Observatories (TERENO) network. The dataset includes both directly measured and derived products, with documented processing procedures, quality-control information, calibration results, and missing-data records. The principal value of the ARC dataset lies in the coordinated integration of continuous and campaign-based observations across multiple spatial and temporal scales and Critical Zone compartments. The combined observations can support the development, calibration, evaluation, and intercomparison of process-based hydrological and Critical Zone models, the evaluation of Earth Observation products, and the investigation of scale-dependent relationships among point-, field-, hillslope-, and catchment-scale observations. The complete dataset is openly accessible through Figshare under a CC BY 4.0 license. (Romano et al., 2026; https://doi.org/10.6084/m9.figshare.31229608).
- Open Access
- 9 June 2026
- Published by Springer Nature in Scientific Data
- Vol. 13 (1)
HU-EnviroGrids: A Gridded Environmental Dataset for Spatial Analysis and Modelling at country scale in Hungary
There is a growing demand for comprehensive environmental information driven by increasingly complex environmental challenges. Addressing these issues requires multidisciplinary approaches supported by harmonized datasets derived from diverse data sources. We developed HU-EnviroGrids, a spatially exhaustive, gridded environmental dataset covering the entire territory of Hungary. The dataset was created using a multidisciplinary data integration approach that harmonized diverse data sources across multiple environmental domains, such as topography, hydrology, climate, biosphere & land surface and pedosphere. The derived quantitative and qualitative environmental variables were transformed into a common reference system, providing a consistent spatial coverage at a 100 × 100 m spatial resolution across Hungary. This dataset provides foundational inputs to a wide range of applications for applied research in ecology, hydrology, climatology, agriculture, forestry, and Earth system sciences. By making these data openly available, HU-EnviroGrids seeks to strengthen the scientific and practical basis for addressing complex environmental issues, and to foster more informed, data-driven decision-making at multiple scales.
- 9 June 2026
- Published by Springer Nature in Scientific Data
- Vol. 13 (1)
Open AccessHU-EnviroGrids: A Gridded Environmental Dataset for Spatial Analysis and Modelling at country scale in Hungary
There is a growing demand for comprehensive environmental information driven by increasingly complex environmental challenges. Addressing these issues requires multidisciplinary approaches supported by harmonized datasets derived from diverse data sources. We developed HU-EnviroGrids, a spatially exhaustive, gridded environmental dataset covering the entire territory of Hungary. The dataset was created using a multidisciplinary data integration approach that harmonized diverse data sources across multiple environmental domains, such as topography, hydrology, climate, biosphere & land surface and pedosphere. The derived quantitative and qualitative environmental variables were transformed into a common reference system, providing a consistent spatial coverage at a 100 × 100 m spatial resolution across Hungary. This dataset provides foundational inputs to a wide range of applications for applied research in ecology, hydrology, climatology, agriculture, forestry, and Earth system sciences. By making these data openly available, HU-EnviroGrids seeks to strengthen the scientific and practical basis for addressing complex environmental issues, and to foster more informed, data-driven decision-making at multiple scales.
- Open Access
Workflow of Visualisation of Mole-Rat Burrows Using 3D Datasets Derived from GPR, UAV Surveys, and Interpretative Processing
We present a concise methodology to model and visualise mole-rat burrows by integrating 3D ground-penetrating radar (GPR) volumes, high-resolution 3D surface texture, and interpretative 3D visualisation with open-code software, such as Blender and Houdini. The workflow shows the processing and conversion steps for converting surface and subsurface raw datasets into point clouds, then the amalgamation of those 3D objects into a voxelised volume. The voxelisation script creates a text file, a *.CSV file, that masks the voxels with the values of 0 and 1 depending on whether they are inside or outside a burrow. This parametrisation resulted in a total of 7,730,587 voxels generated, of which 48,952 have a value of 1 within them. This indicates the presence of one burrow system, in which there were about 60–80 burrow segments that were initially identified by GPR but remained rather interpretative than a verified geometry. The entire process enables handling and combining different, complex, 3D datasets into a simple text file and thus enables merging with covariates for further spatial modelling of burrow systems from incomplete, indirect, noisy measurements.
Open AccessWorkflow of Visualisation of Mole-Rat Burrows Using 3D Datasets Derived from GPR, UAV Surveys, and Interpretative Processing
We present a concise methodology to model and visualise mole-rat burrows by integrating 3D ground-penetrating radar (GPR) volumes, high-resolution 3D surface texture, and interpretative 3D visualisation with open-code software, such as Blender and Houdini. The workflow shows the processing and conversion steps for converting surface and subsurface raw datasets into point clouds, then the amalgamation of those 3D objects into a voxelised volume. The voxelisation script creates a text file, a *.CSV file, that masks the voxels with the values of 0 and 1 depending on whether they are inside or outside a burrow. This parametrisation resulted in a total of 7,730,587 voxels generated, of which 48,952 have a value of 1 within them. This indicates the presence of one burrow system, in which there were about 60–80 burrow segments that were initially identified by GPR but remained rather interpretative than a verified geometry. The entire process enables handling and combining different, complex, 3D datasets into a simple text file and thus enables merging with covariates for further spatial modelling of burrow systems from incomplete, indirect, noisy measurements.
- Open Access
- 31 March 2026
- Published by Springer Nature in Earth Systems and Environment
- Vol. 10 (8) , 1-23
Does Increasing Number of Field Observations Necessarily Improve the Performance of an ML-Based Soil Parent Material Prediction?
The study aimed to elaborate a large-scale soil parent material (SPM) map for a 90 km 2 pilot area in Hungary applying Digital Soil Mapping techniques. Initially, a coarse-scale geological map harmonized with Food and Agriculture Organization (FAO) SPM classes provided the reference data, later supplemented by field observations. Due to the limited ground truth data, we tested how increasing number of field observations affected prediction performance. Random Forest and Gradient Boosting Machine algorithms were applied in a 100-fold simulation-based method using a broad set of environmental covariates including satellite imagery, SRTM DEM derivatives and digital soil property maps. Virtual training and testing data were generated by sampling the FAO-based SPM map and 200 visual field observations were collected. Testing was structured by incrementally adding 50 field points to a randomly generated set. Validation of classified maps was carried out by: i) the overall mean accuracy (OMA) of the predicted maps, ii) the number of predicted classes (NPC) of each pixel and iii) the percentage of the most frequently (MFP) predicted class. Unexpectedly, the inclusion of field data led to a decrease in OMA, suggesting potential data quality issues. While some improvement was observed in reducing highly uncertain categories based on NPC, MFP and the total extent of uncertain areas did not consistently improve. These results highlight that increasing the quantity of field data only is insufficient; future efforts should focus on strategic sampling, data quality assessment, and its integration methods to improve model reliability. Graphical Abstract This graphical abstract illustrates a Digital Soil Mapping study conducted in the Dorog Basin (90 km 2 ) in Hungary. The study aimed to generate a large-scale Soil Parent Material (SPM) map of the area. The workflow integrates coarse-scale geological data harmonized with FAO SPM classes, satellite imagery, terrain derivatives, as well as digital soil maps. Machine learning algorithms –Random Forest and Gradient Boosting– were applied in a 100-fold simulation using virtual and field-based training data. The infographic highlights: (1) Reference data sources: including spectral information from satellite imagery, topographic information from digital terrain models, virtual sampling data from FAO-based SPM maps, and field observations. (2) Field observation data application strategy: Iterative testing with an increasing number of field points (from 0 to 200). (3) Machine learning algorithms: Random Forest and Gradient Boosting. (4) Validation metrics of the mappig: Overall Mean Accuracy (OMA), Number of Predicted Classes (NPC), and Most Frequently Predicted Class (MFP). (5) Key findings: Adding more field data did not consistently improve model accuracy, revealing that data quality and sampling strategy are more critical than quantity alone.
- 31 March 2026
- Published by Springer Nature in Earth Systems and Environment
- Vol. 10 (8) , 1-23
Open AccessDoes Increasing Number of Field Observations Necessarily Improve the Performance of an ML-Based Soil Parent Material Prediction?
The study aimed to elaborate a large-scale soil parent material (SPM) map for a 90 km 2 pilot area in Hungary applying Digital Soil Mapping techniques. Initially, a coarse-scale geological map harmonized with Food and Agriculture Organization (FAO) SPM classes provided the reference data, later supplemented by field observations. Due to the limited ground truth data, we tested how increasing number of field observations affected prediction performance. Random Forest and Gradient Boosting Machine algorithms were applied in a 100-fold simulation-based method using a broad set of environmental covariates including satellite imagery, SRTM DEM derivatives and digital soil property maps. Virtual training and testing data were generated by sampling the FAO-based SPM map and 200 visual field observations were collected. Testing was structured by incrementally adding 50 field points to a randomly generated set. Validation of classified maps was carried out by: i) the overall mean accuracy (OMA) of the predicted maps, ii) the number of predicted classes (NPC) of each pixel and iii) the percentage of the most frequently (MFP) predicted class. Unexpectedly, the inclusion of field data led to a decrease in OMA, suggesting potential data quality issues. While some improvement was observed in reducing highly uncertain categories based on NPC, MFP and the total extent of uncertain areas did not consistently improve. These results highlight that increasing the quantity of field data only is insufficient; future efforts should focus on strategic sampling, data quality assessment, and its integration methods to improve model reliability. Graphical Abstract This graphical abstract illustrates a Digital Soil Mapping study conducted in the Dorog Basin (90 km 2 ) in Hungary. The study aimed to generate a large-scale Soil Parent Material (SPM) map of the area. The workflow integrates coarse-scale geological data harmonized with FAO SPM classes, satellite imagery, terrain derivatives, as well as digital soil maps. Machine learning algorithms –Random Forest and Gradient Boosting– were applied in a 100-fold simulation using virtual and field-based training data. The infographic highlights: (1) Reference data sources: including spectral information from satellite imagery, topographic information from digital terrain models, virtual sampling data from FAO-based SPM maps, and field observations. (2) Field observation data application strategy: Iterative testing with an increasing number of field points (from 0 to 200). (3) Machine learning algorithms: Random Forest and Gradient Boosting. (4) Validation metrics of the mappig: Overall Mean Accuracy (OMA), Number of Predicted Classes (NPC), and Most Frequently Predicted Class (MFP). (5) Key findings: Adding more field data did not consistently improve model accuracy, revealing that data quality and sampling strategy are more critical than quantity alone.
- Open Access
- 14 March 2026
- Published by Copernicus GmbH
Some thoughts on the opportunities to expand the potential and improve the performance of digital soil mapping in Hungary
A wide range of disciplines and many national and international initiatives require spatially, and increasingly, spatiotemporally explicit information on soil properties, functions, processes and services at various scales. These emerging demands are mostly due to the recognition of the multifunctionality of soils. During the past decades a general demand for quantitative digital soil information for environmental modelling and management has compelled soil scientist to address many soil-related questions from a quantitative point of view. Digital soil mapping (DSM) together with pedometrics and proximal soil sensing will keep on playing crucial role in the forecasted near future referred as digital pedology as the upcoming era of the fundamental understanding of soil in space and time.Hungarian soil spatial data infrastructure has recently been renewed within the DOSoReMI.hu initiative. Primary soil property maps compiled according to GlobalSoilMap.net specifications provide an operational and widely accepted source of information on soil at national level. Nevertheless, more recently new challenges have arisen from temporal, thematic and spatial point of view: higher resolution, up-to-date DSM products are expected, which describe soil features (properties, functions, processes) in more details, from various aspects. Our overall aim is to address these challenges by (i) developing approaches that extend and/or rethink the three pillars of DSM, (ii) linking DSM products of different scales and (iii) bridging potential gaps between them.Based on the above background, it is both rational and socially beneficial to identify opportunities to expand the potential and improve the performance of DSM in the support of the multifaceted concept of soil security. Our objectives are twofold: (i) to continue developing and modernizing the national spatial soil data infrastructure capable of meeting the emerging needs for soil information for various purposes and at different scales, (ii) by addressing (at least some of) the ‘pedometrics ten challenges’, methodological issues that are currently at the forefront of soil research.During our forecasted activity we are going to focus on some highlighted topics keeping the possibility to extend them by identifying new challenges and solutions:Introduction of spatially dense observations together with proper pedotransfer function development into DSM to increase thematic versatility and spatial resolution.Application of ancillary data originating from high-resolution (both spatial and spectral) imagery and geophysical sensors to support e.g. pedometric zoning of agricultural plots and increase prediction accuracy on flat areas.Introducing spatially distributed, physics-based models into digital mapping of specific soil functions and processes through pedological interpretation of spatiotemporal model outputs.Testing the impact of iterative mapping and ensemble modelling on prediction performance based on detailed and refined accuracy assessment.According to our experiences, working on approaches to solve certain mapping challenges may easily conclude to newer methodological questions and/or ideas. With the present paper and the topics listed above, our aim is to stimulate DSM stakeholders (both users and producers) to identify and/or formulate new challenges and solutions, which might be worth for overthinking and/or solving within the spatial branch of the upcoming era of digital pedology.
- 14 March 2026
- Published by Copernicus GmbH
Open AccessSome thoughts on the opportunities to expand the potential and improve the performance of digital soil mapping in Hungary
A wide range of disciplines and many national and international initiatives require spatially, and increasingly, spatiotemporally explicit information on soil properties, functions, processes and services at various scales. These emerging demands are mostly due to the recognition of the multifunctionality of soils. During the past decades a general demand for quantitative digital soil information for environmental modelling and management has compelled soil scientist to address many soil-related questions from a quantitative point of view. Digital soil mapping (DSM) together with pedometrics and proximal soil sensing will keep on playing crucial role in the forecasted near future referred as digital pedology as the upcoming era of the fundamental understanding of soil in space and time.Hungarian soil spatial data infrastructure has recently been renewed within the DOSoReMI.hu initiative. Primary soil property maps compiled according to GlobalSoilMap.net specifications provide an operational and widely accepted source of information on soil at national level. Nevertheless, more recently new challenges have arisen from temporal, thematic and spatial point of view: higher resolution, up-to-date DSM products are expected, which describe soil features (properties, functions, processes) in more details, from various aspects. Our overall aim is to address these challenges by (i) developing approaches that extend and/or rethink the three pillars of DSM, (ii) linking DSM products of different scales and (iii) bridging potential gaps between them.Based on the above background, it is both rational and socially beneficial to identify opportunities to expand the potential and improve the performance of DSM in the support of the multifaceted concept of soil security. Our objectives are twofold: (i) to continue developing and modernizing the national spatial soil data infrastructure capable of meeting the emerging needs for soil information for various purposes and at different scales, (ii) by addressing (at least some of) the ‘pedometrics ten challenges’, methodological issues that are currently at the forefront of soil research.During our forecasted activity we are going to focus on some highlighted topics keeping the possibility to extend them by identifying new challenges and solutions:Introduction of spatially dense observations together with proper pedotransfer function development into DSM to increase thematic versatility and spatial resolution.Application of ancillary data originating from high-resolution (both spatial and spectral) imagery and geophysical sensors to support e.g. pedometric zoning of agricultural plots and increase prediction accuracy on flat areas.Introducing spatially distributed, physics-based models into digital mapping of specific soil functions and processes through pedological interpretation of spatiotemporal model outputs.Testing the impact of iterative mapping and ensemble modelling on prediction performance based on detailed and refined accuracy assessment.According to our experiences, working on approaches to solve certain mapping challenges may easily conclude to newer methodological questions and/or ideas. With the present paper and the topics listed above, our aim is to stimulate DSM stakeholders (both users and producers) to identify and/or formulate new challenges and solutions, which might be worth for overthinking and/or solving within the spatial branch of the upcoming era of digital pedology.
- Open Access
- 14 March 2026
- Published by Copernicus GmbH
Relationships between soil chemistry, soil moisture, and vegetation indices in grassland ecosystems
The NDVI and PRI vegetation indices (VIs) are widely used to assess grassland condition; however, our understanding of their sensitivity to soil chemical properties and moisture content needs to be expanded. Linking soil chemical properties and water availability to spectral responses increases the reliability of grassland monitoring and management. The present study aimed to analyze two distinct grassland ecosystems, where the spatial heterogeneity of soil chemical parameters and their effects on vegetation indices were investigated.The study was conducted in 2025 at two research sites, one in the Serbian Rekovac region (RS) and one in the Hungarian Balaton Uplands (HU). Both research sites covered approximately 1 ha. At each site, we collected 32 and 35 soil samples and measured soil chemical parameters (pH, soil organic carbon (SOC), total nitrogen, potassium, and phosphorus) in a laboratory. In the field, we also used non-destructive measurements, including soil elemental compositions using XRF (e.g., Fe, Ca, Cu, As, Pb), soil water content (SWC; Hydrosense II, Campbell Scientific), and vegetation NDVI and PRI values using spectral reflectance sensors (Meter Group) approximately 2 meters above ground. Soil CO2 emissions were measured using an EGM5 IR analyzer (PP Systems). Data for each sampling point were averaged prior to analysis.When we compared the two grassland sites, we found significantly higher C to N ratio, total N, K, P, SOC, CaCO3 concentration, and CO2 emissions at the HU site, while vegetation NDVI and PRI values were significantly lower (p 0.05). Given the large number of measured soil chemical parameters, cluster analysis and principal component analysis (PCA) were applied to reduce dimensionality and identify the main factors influencing vegetation indices. Cluster analysis grouped the variables into three distinct clusters, with the Serbian site in one and the Hungarian sites in two. We found that NDVI and PRI values were strongly and negatively correlated with many of the soil chemical parameters (e.g., pH: r = -0.94 and -0.89, SOC: r = -0.86 and -0.83, respectively). Soil CO2 emissions showed only moderate correlations with specific parameters, such as pH, potassium, or C to N ratio (r = 0.50-0.53). SWC, on the other hand, did not show any clear correlations with the main parameters measured.Our results indicate that variability in vegetation indices was primarily associated with soil chemical gradients rather than soil moisture conditions. Our study showed that local heterogeneity can strongly affect soil chemical data, which in turn affects VIs. Both sites were assessed twice, but further measurements are planned to provide more robust support for our findings.Acknowledgments: The research was funded by the Sustainable Development and Technologies National Programme of the Hungarian Academy of Sciences (FFT NP FTA) and the 2023-1.2.4-TÉT-2023-0009 project.
- 14 March 2026
- Published by Copernicus GmbH
Open AccessRelationships between soil chemistry, soil moisture, and vegetation indices in grassland ecosystems
The NDVI and PRI vegetation indices (VIs) are widely used to assess grassland condition; however, our understanding of their sensitivity to soil chemical properties and moisture content needs to be expanded. Linking soil chemical properties and water availability to spectral responses increases the reliability of grassland monitoring and management. The present study aimed to analyze two distinct grassland ecosystems, where the spatial heterogeneity of soil chemical parameters and their effects on vegetation indices were investigated.The study was conducted in 2025 at two research sites, one in the Serbian Rekovac region (RS) and one in the Hungarian Balaton Uplands (HU). Both research sites covered approximately 1 ha. At each site, we collected 32 and 35 soil samples and measured soil chemical parameters (pH, soil organic carbon (SOC), total nitrogen, potassium, and phosphorus) in a laboratory. In the field, we also used non-destructive measurements, including soil elemental compositions using XRF (e.g., Fe, Ca, Cu, As, Pb), soil water content (SWC; Hydrosense II, Campbell Scientific), and vegetation NDVI and PRI values using spectral reflectance sensors (Meter Group) approximately 2 meters above ground. Soil CO2 emissions were measured using an EGM5 IR analyzer (PP Systems). Data for each sampling point were averaged prior to analysis.When we compared the two grassland sites, we found significantly higher C to N ratio, total N, K, P, SOC, CaCO3 concentration, and CO2 emissions at the HU site, while vegetation NDVI and PRI values were significantly lower (p 0.05). Given the large number of measured soil chemical parameters, cluster analysis and principal component analysis (PCA) were applied to reduce dimensionality and identify the main factors influencing vegetation indices. Cluster analysis grouped the variables into three distinct clusters, with the Serbian site in one and the Hungarian sites in two. We found that NDVI and PRI values were strongly and negatively correlated with many of the soil chemical parameters (e.g., pH: r = -0.94 and -0.89, SOC: r = -0.86 and -0.83, respectively). Soil CO2 emissions showed only moderate correlations with specific parameters, such as pH, potassium, or C to N ratio (r = 0.50-0.53). SWC, on the other hand, did not show any clear correlations with the main parameters measured.Our results indicate that variability in vegetation indices was primarily associated with soil chemical gradients rather than soil moisture conditions. Our study showed that local heterogeneity can strongly affect soil chemical data, which in turn affects VIs. Both sites were assessed twice, but further measurements are planned to provide more robust support for our findings.Acknowledgments: The research was funded by the Sustainable Development and Technologies National Programme of the Hungarian Academy of Sciences (FFT NP FTA) and the 2023-1.2.4-TÉT-2023-0009 project.
- Open Access
- 13 March 2026
- Published by MDPI AG in Remote Sensing
- Vol. 18 (6) , 884
Dual-Stream 2D and 3D-SE-ResNet Architectures for Crop Mapping Using EnMAP Hyperspectral Time-Series
Deep learning-based crop mapping from hyperspectral satellite data offers immense potential for capturing subtle phenological differences, yet leveraging sparse time series remains a major methodological challenge. This study evaluates the ability of the EnMAP sensor to identify nine major crop types in the intensive agricultural landscape of Southeastern Hungary. We utilized a limited time series (November, March, August) to benchmark two modeling strategies: a single-date dual-stream spatial–spectral 2D-CNN (DSS-2D) and a multi-temporal 3D-SE-ResNet. Model performance was assessed using parcel-level spatial cross-validation to ensure realistic accuracy estimates and reduce spatial autocorrelation bias. The results demonstrate that the DSS-2D model achieved superior single-date accuracy (OA > 97%), significantly outperforming pixel-based baselines. Furthermore, the multi-temporal 3D-SE-ResNet achieved a robust seasonal accuracy of 92.9%, effectively compensating for temporal sparsity by exploiting the deep spectral information of the SWIR domain. This study confirms that treating hyperspectral data as a 3D volume enables the extraction of phenological traits even from limited observations. These findings provide a strong proof-of-concept for the operational feasibility of future missions such as Copernicus CHIME for continental-scale food security monitoring.
- 13 March 2026
- Published by MDPI AG in Remote Sensing
- Vol. 18 (6) , 884
Open AccessDual-Stream 2D and 3D-SE-ResNet Architectures for Crop Mapping Using EnMAP Hyperspectral Time-Series
Deep learning-based crop mapping from hyperspectral satellite data offers immense potential for capturing subtle phenological differences, yet leveraging sparse time series remains a major methodological challenge. This study evaluates the ability of the EnMAP sensor to identify nine major crop types in the intensive agricultural landscape of Southeastern Hungary. We utilized a limited time series (November, March, August) to benchmark two modeling strategies: a single-date dual-stream spatial–spectral 2D-CNN (DSS-2D) and a multi-temporal 3D-SE-ResNet. Model performance was assessed using parcel-level spatial cross-validation to ensure realistic accuracy estimates and reduce spatial autocorrelation bias. The results demonstrate that the DSS-2D model achieved superior single-date accuracy (OA > 97%), significantly outperforming pixel-based baselines. Furthermore, the multi-temporal 3D-SE-ResNet achieved a robust seasonal accuracy of 92.9%, effectively compensating for temporal sparsity by exploiting the deep spectral information of the SWIR domain. This study confirms that treating hyperspectral data as a 3D volume enables the extraction of phenological traits even from limited observations. These findings provide a strong proof-of-concept for the operational feasibility of future missions such as Copernicus CHIME for continental-scale food security monitoring.
- Open Access
- 3 February 2026
- Published by MDPI AG in Preprints.org
Dual-Stream 2D and 3D-SE-ResNet Architectures for Crop Mapping Using EnMAP Hyperspectral Time-Series
Deep learning-based crop mapping from hyperspectral satellite data offers immense potential for capturing subtle phenological differences, yet leveraging sparse time-series remains a major methodological challenge. This study evaluates the EnMAP sensor for identifying nine major crop types in the intensive agricultural landscape of Southeastern Hungary. We utilized a limited time series (November, March, August) to benchmark two modeling strategies: a single-date Dual-Stream Spatial-Spectral 2D-CNN (DSS-2D) and a multi-temporal 3D-SE-ResNet. Model performance was assessed using parcel-level spatial cross-validation to ensure realistic accuracy estimates and reduce spatial autocorrelation bias. Results demonstrate that the DSS-2D model achieved superior single-date accuracy (OA > 97%), significantly outperforming pixel-based baselines. Furthermore, the multi-temporal 3D-SE-ResNet achieved a robust seasonal accuracy of 92.9%, effectively compensating for temporal sparsity by exploiting the deep spectral information of the SWIR domain. The study confirms that treating hyperspectral data as a 3D volume enables the extraction of phenological traits even from limited observations. These findings provide a strong proof-of-concept for the operational feasibility of future missions like Copernicus CHIME for continental-scale food security monitoring.
- 3 February 2026
- Published by MDPI AG in Preprints.org
Open AccessDual-Stream 2D and 3D-SE-ResNet Architectures for Crop Mapping Using EnMAP Hyperspectral Time-Series
Deep learning-based crop mapping from hyperspectral satellite data offers immense potential for capturing subtle phenological differences, yet leveraging sparse time-series remains a major methodological challenge. This study evaluates the EnMAP sensor for identifying nine major crop types in the intensive agricultural landscape of Southeastern Hungary. We utilized a limited time series (November, March, August) to benchmark two modeling strategies: a single-date Dual-Stream Spatial-Spectral 2D-CNN (DSS-2D) and a multi-temporal 3D-SE-ResNet. Model performance was assessed using parcel-level spatial cross-validation to ensure realistic accuracy estimates and reduce spatial autocorrelation bias. Results demonstrate that the DSS-2D model achieved superior single-date accuracy (OA > 97%), significantly outperforming pixel-based baselines. Furthermore, the multi-temporal 3D-SE-ResNet achieved a robust seasonal accuracy of 92.9%, effectively compensating for temporal sparsity by exploiting the deep spectral information of the SWIR domain. The study confirms that treating hyperspectral data as a 3D volume enables the extraction of phenological traits even from limited observations. These findings provide a strong proof-of-concept for the operational feasibility of future missions like Copernicus CHIME for continental-scale food security monitoring.
- Open Access
- 29 September 2025
- Published by Springer Nature in Scientific Data
- Vol. 12 (1) , 1-11
National soil hydrologic groups map for environmental applications using data-driven and expert-based methods
Regional and national 3D soil hydraulic maps enhance understanding of soil hydraulic properties, essential for environmental assessments. However, data aggregation is often necessary in large-scale models to facilitate the modelling of complex soil characteristics. This study presents a soil hydrologic groups map for Hungary, derived through k-means clustering and expert-based rules. Clustering was applied to the 100 m resolution 3D HU-SoilHydroGrids database, considering eight hydraulic parameters across six depths. The accuracy of these maps is limited for rare soil types with extreme characteristics due to their small spatial extent and sparse representation in national datasets. To account for these underrepresented soil types, we refined each statistics-based cluster using expert-based rules incorporating soil profile depth, genetic type, electrical conductivity, and exchangeable sodium content. The final classification includes 68 soil hydrologic groups, defined by distinct hydraulic properties, such as van Genuchten parameters to describe water retention, and saturated hydraulic conductivity. This national map supports country-wide hydrological modelling, environmental management, and agricultural planning in Hungary by enabling consistent treatment of similar soils.
- 29 September 2025
- Published by Springer Nature in Scientific Data
- Vol. 12 (1) , 1-11
Open AccessNational soil hydrologic groups map for environmental applications using data-driven and expert-based methods
Regional and national 3D soil hydraulic maps enhance understanding of soil hydraulic properties, essential for environmental assessments. However, data aggregation is often necessary in large-scale models to facilitate the modelling of complex soil characteristics. This study presents a soil hydrologic groups map for Hungary, derived through k-means clustering and expert-based rules. Clustering was applied to the 100 m resolution 3D HU-SoilHydroGrids database, considering eight hydraulic parameters across six depths. The accuracy of these maps is limited for rare soil types with extreme characteristics due to their small spatial extent and sparse representation in national datasets. To account for these underrepresented soil types, we refined each statistics-based cluster using expert-based rules incorporating soil profile depth, genetic type, electrical conductivity, and exchangeable sodium content. The final classification includes 68 soil hydrologic groups, defined by distinct hydraulic properties, such as van Genuchten parameters to describe water retention, and saturated hydraulic conductivity. This national map supports country-wide hydrological modelling, environmental management, and agricultural planning in Hungary by enabling consistent treatment of similar soils.
- Open Access
- 14 July 2025
- Published by Copernicus GmbH in ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
- Vol. X-G-2025, 937-944
Investigation of Riverbank Peat Erosion in a small Alpine catchment by Multi-Temporal Point Cloud Change Analysis and Time Series Clustering
Peatlands play an important role as local carbon sinks and climate archives in natural environments. In addition to degradation by human activities, peatlands can also be affected by geomorphological processes such as fluvial erosion. This study used a collection of multi-temporal 3D point clouds to investigate such a situation in a valley of the Central Alps (Austria), where a dynamically migrating river is eroding peat. To detect time periods and locations where strong erosion occurred and to quantify the local peat erosion rate over the years 2006 to 2024, 3D point clouds from airborne laser scanning (ALS), uncrewed aerial vehicle laser scanning (ULS), and uncrewed aerial vehicle photogrammetry (UPH) were used (i) in a pair-wise Multiscale Model-to-Model Cloud Comparison (M3C2) change detection and quantification and (ii) in a time-series clustering approach. In the most dynamic sections of the river bank, the mean rate of peat erosion over the eighteen-year period was −0.12 ± 0.03 m/year. Moreover, the relationship between peatland erosion and main channel migration is investigated based on a set of surface elevation transects and multi-temporal mapping of the main channels. Overall, these methods provided detailed insights into the dynamics and functioning of local geomorphological processes, such as lateral undercutting with subsequent toppling and sliding of the peat bank.
- 14 July 2025
- Published by Copernicus GmbH in ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
- Vol. X-G-2025, 937-944
Open AccessInvestigation of Riverbank Peat Erosion in a small Alpine catchment by Multi-Temporal Point Cloud Change Analysis and Time Series Clustering
Peatlands play an important role as local carbon sinks and climate archives in natural environments. In addition to degradation by human activities, peatlands can also be affected by geomorphological processes such as fluvial erosion. This study used a collection of multi-temporal 3D point clouds to investigate such a situation in a valley of the Central Alps (Austria), where a dynamically migrating river is eroding peat. To detect time periods and locations where strong erosion occurred and to quantify the local peat erosion rate over the years 2006 to 2024, 3D point clouds from airborne laser scanning (ALS), uncrewed aerial vehicle laser scanning (ULS), and uncrewed aerial vehicle photogrammetry (UPH) were used (i) in a pair-wise Multiscale Model-to-Model Cloud Comparison (M3C2) change detection and quantification and (ii) in a time-series clustering approach. In the most dynamic sections of the river bank, the mean rate of peat erosion over the eighteen-year period was −0.12 ± 0.03 m/year. Moreover, the relationship between peatland erosion and main channel migration is investigated based on a set of surface elevation transects and multi-temporal mapping of the main channels. Overall, these methods provided detailed insights into the dynamics and functioning of local geomorphological processes, such as lateral undercutting with subsequent toppling and sliding of the peat bank.