Scientific Results

This catalogue is obtained by conducting a systematic literature review of scientific studies and reviews related to monitoring, forecasting, and simulating the inland water cycle. The analysis maps scientific expertise across research groups and classifies findings by the type of inland water studied, application focus, and geographical scope. A gap analysis will identify missing research areas and assess their relevance to policymaking.

ID â–Č Type Year Authors Title Venue/Journal DOI Research type Water System Technical Focus Abstract Link with Projects Link with Tools Related policies ID
publications-2241 Peer reviewed articles 2020 My V T Phan, Mariana Mendonca Melo, Els van Nood, Georgina Aron, Jolanda J C Kreeft-Voermans, Marion P G Koopmans, Chantal Reusken, Corine H GeurtsvanKessel, Matthew Cotten Shedding of Yellow Fever Virus From an Imported Case in the Netherlands After Travel to Brazil Open Forum Infectious Diseases 10.1093/ofid/ofaa020 Uncategorized Uncategorized Abstract We report yellow fever infection in a Dutch traveler returning from Brazil. Yellow fever virus (YFV) was identified in serum and urine samples over a period of 1 month. Yellow fever virus genome sequences from the patient clustered with recent Brazilian YFV and showed with limited nucleotide changes during the resolving infection. 799417
publications-2242 Peer reviewed articles 2020 Gregorius J. Sips, MariĂ«lle J. G. Dirven, Joke T. Donkervoort, Francien M. Kolfschoten, Claudia M. E. Schapendonk, My V. T. Phan, Annemieke Bloem, Anna F. Leeuwen, Mariechristine E. Trompenaars, Marion P. G. Koopmans, Annemiek A. Eijk, Miranda Graaf, Ewout B. Fanoy Norovirus outbreak in a natural playground: A One Health approach Zoonoses and Public Health 10.1111/zph.12689 Simulation & Modeling Precipitation & Ecological Systems AbstractNorovirus constitutes the most frequently identified infectious cause of disease outbreaks associated with untreated recreational water. When investigating outbreaks related to surface water, a One Health approach is insightful. Historically, there has been a focus on potential contamination of recreational water by bird droppings and a recent publication demonstrating human noroviruses in bird faeces suggested this should be investigated in future water‐related norovirus outbreaks. Here, we describe a One Health approach investigating a norovirus outbreak in a natural playground. On social media, a large amount of waterfowl were reported to defecate near these playground premises leading to speculations about their potential involvement. Surface water, as well as human and bird faecal specimens, was tested for human noroviruses. Norovirus was found to be the most likely cause of the outbreak but there was no evidence for transmission via waterfowl. Cases had become known on social media prior to notification to the public health service underscoring the potential of online media as an early warning system. In view of known risk factors, advice was given for future outbreak investigations and natural playground design. 799417
publications-2243 Peer reviewed articles 2022 Giannopoulos Michalis, Tsagkatakis Grigorios, Tsakalides Panagiotis 4D U-Nets for Multi-Temporal Remote Sensing Data Classification remote sensing 10.3390/rs14030634 Data Management & Analytics Precipitation & Ecological Systems Multispectral sensors constitute a core earth observation imaging technology generating massive high-dimensional observations acquired across multiple time instances. The collected multi-temporal remote sensed data contain rich information for Earth monitoring applications, from flood detection to crop classification. To easily classify such naturally multidimensional data, conventional low-order deep learning models unavoidably toss away valuable information residing across the available dimensions. In this work, we extend state-of-the-art convolutional network models based on the U-Net architecture to their high-dimensional analogs, which can naturally capture multi-dimensional dependencies and correlations. We introduce several model architectures, both of low as well as of high order, and we quantify the achieved classification performance vis-Ă -vis the latest state-of-the-art methods. The experimental analysis on observations from Landsat-8 reveals that approaches based on low-order U-Net models exhibit poor classification performance and are outperformed by our proposed high-dimensional U-Net scheme. 842560
publications-2244 Peer reviewed articles 2022 Drakonakis Georgios; Tsagkatakis Grigorios; Fotiadou Konstantina; Tsakalides Panagiotis OmbriaNet — Supervised Flood Mapping via Convolutional Neural Networks Using Multitemporal Sentinel-1 and Sentinel-2 Data Fusion IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 10.5281/zenodo.6674223 Data Management & Analytics River Basins No abstract available 842560
publications-2245 Peer reviewed articles 2020 Maria Aspri; Grigorios Tsagkatakis; Panagiotis Tsakalides Distributed Training and Inference of Deep Learning Models for Multi-Modal Land Cover Classification Remote Sensing 10.3390/rs12172670 Simulation & Modeling River Basins Deep Neural Networks (DNNs) have established themselves as a fundamental tool in numerous computational modeling applications, overcoming the challenge of defining use-case-specific feature extraction processing by incorporating this stage into unified end-to-end trainable models. Despite their capabilities in modeling, training large-scale DNN models is a very computation-intensive task that most single machines are often incapable of accomplishing. To address this issue, different parallelization schemes were proposed. Nevertheless, network overheads as well as optimal resource allocation pose as major challenges, since network communication is generally slower than intra-machine communication while some layers are more computationally expensive than others. In this work, we consider a novel multimodal DNN based on the Convolutional Neural Network architecture and explore several different ways to optimize its performance when training is executed on an Apache Spark Cluster. We evaluate the performance of different architectures via the metrics of network traffic and processing power, considering the case of land cover classification from remote sensing observations. Furthermore, we compare our architectures with an identical DNN architecture modeled after a data parallelization approach by using the metrics of classification accuracy and inference execution time. The experiments show that the way a model is parallelized has tremendous effect on resource allocation and hyperparameter tuning can reduce network overheads. Experimental results also demonstrate that proposed model parallelization schemes achieve more efficient resource use and more accurate predictions compared to data parallelization approaches. 842560
publications-2246 Peer reviewed articles 2022 Villia Maria Myrto; Tsagkatakis Grigorios; Moghaddam Mahta; Tsakalides Panagiotis Embedded Temporal Convolutional Networks for Essential Climate Variables Forecasting Sensors 10.3390/s22051851 Simulation & Modeling River Basins Forecasting the values of essential climate variables like land surface temperature and soil moisture can play a paramount role in understanding and predicting the impact of climate change. This work concerns the development of a deep learning model for analyzing and predicting spatial time series, considering both satellite derived and model-based data assimilation processes. To that end, we propose the Embedded Temporal Convolutional Network (E-TCN) architecture, which integrates three different networks, namely an encoder network, a temporal convolutional network, and a decoder network. The model accepts as input satellite or assimilation model derived values, such as land surface temperature and soil moisture, with monthly periodicity, going back more than fifteen years. We use our model and compare its results with the state-of-the-art model for spatiotemporal data, the ConvLSTM model. To quantify performance, we explore different cases of spatial resolution, spatial region extension, number of training examples and prediction windows, among others. The proposed approach achieves better performance in terms of prediction accuracy, while using a smaller number of parameters compared to the ConvLSTM model. Although we focus on two specific environmental variables, the method can be readily applied to other variables of interest. 842560
publications-2247 Peer reviewed articles 2020 Georgios Vernardos, Grigorios Tsagkatakis, Yannis Pantazis Quantifying the structure of strong gravitational lens potentials with uncertainty-aware deep neural networks Monthly Notices of the Royal Astronomical Society 10.1093/mnras/staa3201 Data Management & Analytics Precipitation & Ecological Systems ABSTRACTGravitational lensing is a powerful tool for constraining substructure in the mass distribution of galaxies, be it from the presence of dark matter sub-haloes or due to physical mechanisms affecting the baryons throughout galaxy evolution. Such substructure is hard to model and is either ignored by traditional, smooth modelling, approaches, or treated as well-localized massive perturbers. In this work, we propose a deep learning approach to quantify the statistical properties of such perturbations directly from images, where only the extended lensed source features within a mask are considered, without the need of any lens modelling. Our training data consist of mock lensed images assuming perturbing Gaussian Random Fields permeating the smooth overall lens potential, and, for the first time, using images of real galaxies as the lensed source. We employ a novel deep neural network that can handle arbitrary uncertainty intervals associated with the training data set labels as input, provides probability distributions as output, and adopts a composite loss function. The method succeeds not only in accurately estimating the actual parameter values, but also reduces the predicted confidence intervals by 10 per cent in an unsupervised manner, i.e. without having access to the actual ground truth values. Our results are invariant to the inherent degeneracy between mass perturbations in the lens and complex brightness profiles for the source. Hence, we can quantitatively and robustly quantify the smoothness of the mass density of thousands of lenses, including confidence intervals, and provide a consistent ranking for follow-up science. 842560
publications-2248 Peer reviewed articles 2021 Aidini Anastasia; Tsagkatakis Grigorios; Tsakalides Panagiotis Tensor Decomposition Learning for Compression of Multidimensional Signals IEEE Journal of Selected Topics in Signal Processing 10.1109/JSTSP.2021.3054314 Data Management & Analytics Precipitation & Ecological Systems No abstract available 842560
publications-2249 Peer reviewed articles 2021 José L.J. Ledesma, Guiomar Ruiz-Pérez, Anna Lupon, Sílvia Poblador, Martyn N. Futter, Francesc Sabater, Susana Bernal Future changes in the Dominant Source Layer of riparian lateral water fluxes in a subhumid Mediterranean catchment Journal of Hydrology 10.1016/j.jhydrol.2021.126014 IoT & Sensors Irrigation Systems No abstract available 834363
publications-2250 Peer reviewed articles 2021 JosĂ© L. J. Ledesma, Anna Lupon, Susana Bernal Hydrological responses to rainfall events including the extratropical cyclone Gloria in two contrasting Mediterranean headwaters in Spain; the perennial font del RegĂ s and the intermittent Fuirosos Hydrological Processes 10.1002/hyp.14451 Simulation & Modeling Precipitation & Ecological Systems AbstractCatchment hydrological responses to precipitation inputs, particularly during exceptionally large storms, are complex and variable, and our understanding of the associated runoff generation processes during those events is limited. Hydrological monitoring of climatically and hydrologically distinct catchments can help to improve this understanding by shedding light on the interplay between antecedent soil moisture conditions, hydrological connectivity, and rainfall event characteristics. This knowledge is urgently needed considering that both the frequency and magnitude of extreme precipitation events are increasing worldwide as a consequence of climate change. In autumn 2018, we installed water level sensors to monitor stream water and near‐stream groundwater levels at two Mediterranean forest headwater catchments with contrasting hydrological regimes: Font del RegĂ s (sub‐humid climate, perennial flow regime) and Fuirosos (semi‐arid climate, intermittent flow regime). Both catchments are located in northeastern Spain, where the extratropical cyclone Gloria hit in January 2020 and left in ca. 65 h outstanding accumulated rainfalls of 424 mm in Font del RegĂ s and 230 mm in Fuirosos. During rainfall events of low mean intensity, hydrological responses to precipitation inputs at the semi‐arid Fuirosos were more delayed and more variable than at the sub‐humid Font del RegĂ s. We explain these divergences by differences in antecedent soil moisture conditions and associated differences in catchment hydrological connectivity between the two catchments, which in this case are likely driven by differences in local climate rather than by differences in local topography. In contrast, during events of moderate and high mean rainfall intensities, including the storm Gloria, precipitation inputs and hydrological responses correlated similarly in the two catchments. We explain this convergence by rapid development of hydrological connectivity independently of antecedent soil moisture conditions. The data set presented here is unique and contributes to our mechanistic understanding on how streams respond to rainfall events and exceptionally large storms in catchments with contrasting flow regimes. 834363