A5049193486- Hydrology and Watershed Management Studies
- Hydrological Forecasting Using AI
- Groundwater flow and contamination studies
- Flood Risk Assessment and Management
- demographic modeling and climate adaptation
- Climate variability and models
- Water-Energy-Food Nexus Studies
- Geophysics and Gravity Measurements
- Seismic Imaging and Inversion Techniques
- Hydrology and Drought Analysis
- Neuroendocrine Tumor Research Advances
- Ionosphere and magnetosphere dynamics
- Reservoir Engineering and Simulation Methods
Université de Caen Normandie
2023-2024
Université de Rouen Normandie
2023-2024
Centre National de la Recherche Scientifique
2023-2024
Bureau de Recherches Géologiques et Minières
2022-2024
Laboratoire Morphodynamique Continentale et Côtière
2022-2024
Continental (France)
2022-2024
The new scientific decade (2023-2032) of the International Association Hydrological Sciences (IAHS) aims at searching for sustainable solutions to undesired water conditions - may it be too little, much or polluted. Many current issues originate from global change, while problems must embrace local understanding and context. will explore crises by actionable knowledge within three themes: interactions, innovative cross-cutting methods. We capitalise on previous IAHS Scientific Decades...
Abstract. In this study, we use deep learning models with advanced variants of recurrent neural networks, specifically long short-term memory (LSTM), gated unit (GRU), and bidirectional LSTM (BiLSTM), to simulate large-scale groundwater level (GWL) fluctuations in northern France. We develop multi-station collective training for GWL simulations, using dynamic variables (i.e. climatic) static basin characteristics. This approach can incorporate features cover more reservoir heterogeneities...
Northern Metropolitan France. Assessing long-term changes in groundwater is crucial for understanding the impacts of climate change on aquifers and managing water resources.However, level (GWL) records are often scarce, limiting historical trends variability. In this paper, we present a deep learning approach to reconstruct GWLs up several decades back time using recurrent-based neural networks with wavelet pre-processing reanalysis data as inputs. reconstructed two different datasets...
This study aims to investigate the use of deep learning techniques, with or without data pre-processing for simulating groundwater levels. Two approaches are compared: (1) a single (local) station approach, where separate model is trained each station, and (2) multi-station using from multiple stations in area. In latter static catchment attributes dynamic meteorological (precipitation temperature) climate (sea level pressure, etc) inputs used levels Seine basin. By including variables...
Abstract. In this study, we used deep learning models with recurrent structure neural networks to simulate large-scale groundwater level (GWL) fluctuations in northern France. We developed a multi-station collective training for GWL simulations, using both “dynamic” variables (i.e. climatic) and static aquifer characteristics. This approach offers the possibility of incorporating dynamic features cover more reservoir heterogeneities study area. Further, investigated performance relevant...
Abstract. This paper presents the results of 2022 Groundwater Time Series Modelling Challenge, where 15 teams from different institutes applied various data-driven models to simulate hydraulic-head time series at four monitoring wells. Three wells were located in Europe and one was USA hydrogeological settings temperate, continental, or subarctic climates. Participants provided with approximately years measured heads (almost) regular intervals daily measurements weather data starting some 10...
Assessing long-term changes in groundwater is crucial for understanding the impacts of climate change on aquifers and managing water resources. However, level (GWL) records are often scarce, limiting historical trends variability. In this study, we present a deep learning approach to reconstruct GWLs up several decades back time using recurrent-based neural networks with wavelet pre-processing reanalysis data as inputs. reconstructed two different datasets distinct spatial resolutions (ERA5:...
Abstract. This paper presents the results of 2022 groundwater time series modeling challenge, where 15 teams from different institutes applied various data-driven models to simulate hydraulic head at four monitoring wells. Three wells were located in Europe and one USA, hydrogeological settings but all temperate or continental climates. Participants provided with approximately years measured heads (almost) regular intervals daily measurements weather data starting some 10 prior first...
Team DA CollectiveThe "DA Collective" team used an ensemble-of-transfer-function-models approach.The Pastas python module (Collenteur et al., 2019) was as the underlying transfer-function model engine and wrapped within ensemble-based data assimilation framework.The captured input parameter uncertainty well temporally-varying forcing uncertainty.The in Julian day multiplier parameters that were then transferred to forecast period so any biases learned during training could be propagated...
In a context where anticipating future trends and long-term variations in water resources is crucial, improving our knowledge about most types of aquifer responses to climate variability change necessary. Aquifers with dominated by seasonal (marked annual cycle) or low-frequency (interannual decadal driven large-scale dynamics) may encounter different sensitivities change. We investigated this hypothesis generating groundwater level projections using deep learning models for annual, inertial...
We have witnessed and experienced increasing compound extreme events resulting from simultaneous or sequential occurrence of multiple in a changing climate. In addition to growing demand for clearer explanation risks hydrological perspective, there has been lack attention paid socioeconomic factors driving impacted by these risks. Through critical review co-production approaches, we identified four types based on autocorrelated, multivariate, spatiotemporal patterns. A framework quantify...
The development of groundwater levels (GWL) simulations, based on deep learning (DL) models, is gaining traction due to their success in a wide range hydrological applications. GWL Simulations allow generating reconstructions be used for exploring past temporal variability resources or provide means generate projections under climate change decadal scales. Owing the diversity large-scale and local scale forcing factors involved explaining variability, machine even approaches reveal relevant...