A5000368717- Karst Systems and Hydrogeology
- Hydrology and Watershed Management Studies
- Groundwater flow and contamination studies
- Groundwater and Isotope Geochemistry
- Flood Risk Assessment and Management
- Groundwater and Watershed Analysis
- Hydrological Forecasting Using AI
- Privacy-Preserving Technologies in Data
- Simulation Techniques and Applications
- Image and Signal Denoising Methods
- Data-Driven Disease Surveillance
- Dam Engineering and Safety
- Mobile Crowdsensing and Crowdsourcing
- Enhanced Recovery After Surgery
- Big Data Technologies and Applications
- Urban Stormwater Management Solutions
- Scientific Computing and Data Management
- Machine Learning and Algorithms
- Soil and Water Nutrient Dynamics
- Aquatic and Environmental Studies
- Spacecraft Design and Technology
- Water resources management and optimization
- Environmental Science and Water Management
- Stellar, planetary, and galactic studies
- Nausea and vomiting management
Ruhrverband (Germany)
2021-2025
Groundwater Center
2023-2025
TU Dresden
2023
Technical University of Munich
2018-2022
Norsk Hydro (Germany)
2019-2021
Rutgers Sexual and Reproductive Health and Rights
2020
Rutgers, The State University of New Jersey
2017-2018
Ochsner Health System
2014
Karst aquifers provide drinking water for 10% of the world's population, support agriculture, groundwater-dependent activities, and ecosystems. These are characterised by complex groundwater-flow systems, hence, they extremely vulnerable protecting them requires an in-depth understanding systems. Poor data accessibility has limited advances in karst research realistic representation processes large-scale hydrological studies. In this study, we present World Spring hydrograph (WoKaS)...
Abstract Karst water resources are valuable freshwater sources for around 10% of the world's population. Nonetheless, anthropogenic impacts and global changes have seriously deteriorated karst quality dependent ecosystems. Multiscale karstic heterogeneity—referring to spatial variations aquifer's physical chemical characteristics at varying scales—is main challenge in describing flow contaminant transport dynamics. Solute models powerful tools represent predict spatiotemporal behaviors...
Abstract Continuous hourly time series of hydrochemical data can provide insights into the subsurface dynamics and main hydrological processes karst systems. This study investigates how high-resolution be used for verification robust conceptual event-based models. To match high temporal variability data, LuKARS 2.0 model was developed on an scale. The concept considers interaction between matrix conduit components to allow a flexible conceptualization binary systems characterized by...
Karst water resources are valuable freshwater sources for around 10 % of the world population. Nonetheless, anthropogenic factors and global changes have been seriously deteriorating karst quality dependent ecosystems. Solute transport models powerful tools to monitor, control, manage ecosystem functioning. By representing predicting spatiotemporal behavior solute migration in systems, enhance our understanding about processes, thus enabling us explore contamination risks potential outcomes....
Abstract Global sensitivity analysis is an important step in the process of developing and analyzing hydrological models. Measured data different variables are used to identify number sensitive model parameters better constrain output. However, scarcity a common issue hydrology. Since hydrology we dealing with multi‐scale time dependent problems, want overcome that by exploiting potential using decomposed wavelet temporal scales discharge signal for identification parameters. In proposed...
Abstract Uncertainties in hydrologic model outputs can arise for many reasons such as structural, parametric and input uncertainty. Identification of the sources uncertainties quantification their impacts on results are important to appropriately reproduce hydrodynamic processes karst aquifers support decision-making. The present study investigates time-dependent relevance uncertainties, defined conceptual affecting representation parameterization relevant groundwater recharge, i.e....
In this article, we perform a parameter study for recently developed karst hydrological model. The consists of high-dimensional Bayesian inverse problem and global sensitivity analysis. For the first time in hydrology, use active subspace method to find directions space that dominate update from prior posterior distribution order effectively reduce dimension computational efficiency. Additionally, calculated can be exploited construct metrics on each individual parameters used natural model...
Retention and detention basins are engineering constructions with multiple objectives; e.g., flood protection irrigation. Their performance is highly location-dependent, thus, optimization strategies needed. LOCASIN (Location detection of retention basins) an open-source MATLAB tool that enables automated rapid detection, characterization evaluation basin locations. The site based on a numerical raster analysis to determine the optimal dam axis orientation, geometry area volume. After...
Abstract Hydrochemical data of karst springs provide valuable insights into the internal hydrodynamical functioning systems and support model structure identification. However, collection high‐frequency time series major solute species is limited by analysis costs. In this study, we develop a method to retrieve individual concentration their uncertainty at high temporal resolution for using continuous observations electrical conductivity () low‐frequency ionic measurements. Due large ion...
Der Braunkohleausstieg im rheinischen Revier bis 2030 sowie sich wandelnde Ansprüche und Vorgaben an die Abwasserreinigung sind große Herausforderungen für den Erftverband. Technologischer Fortschritt integrales wasserwirtschaftliches Handeln gefragt.
We consider privacy-preserving learning in the context of online learning. Insettings where data instances arrive sequentially streaming fashion, incremental trainingalgorithms such as stochastic gradient descent (SGD) can be used to learn and updateprediction models. When labels are costly acquire, active methods beused select samples labeled from a stream unlabeled data. These datasamples then update machine Privacy-preserving onlinelearning predictors on streams containing sensitive...
We consider the problem of privacy-sensitive anomaly detection - screening to detect individuals, behaviors, areas, or data samples high interest. What defines an is context-specific; for example, a spoofed rather than genuine user attempting log in web site, fraudulent credit card transaction, suspicious traveler airport. The unifying assumption that number anomalous points quite small with respect population, so deep all individual would potentially be time-intensive, costly, and...