Search

Simulation of the catchment rainfall-runoff transformation with physically based watershed models is a traditional way to predict streamflow and other hydrological variables at catchment scales. However, the calibration of such models requires large data inputs and computational power and contains many parameters which are often impossible to constrain or validate. An alternative approach is to use data-driven machine learning for streamflow prediction.In the past few years, LSTM (long short-term memory) models and its variants have been explored in rainfall-runoff modelling. Typical applications use daily climate variables as inputs and model the rainfall-runoff transformation processes with different timescales of memory. This is especially useful as delays in runoff production by snow accumulation and melt, soil water storage, evapotranspiration, etc., can be included. In contrast to feed-forward ANNs (artificial neural networks), LSTMs are capable of maintaining the sequential temporal order of inputs, and compared to RNNs (recurrent neural networks), of learning the long-term dependencies. [1]However, current work on LSTMs mostly focuses on the USA, the UK and Brazil, where CAMELS datasets are available [1, 2, 3]. Catchments at higher altitudes with snow-driven dynamics and sometimes glaciers are present in small number in these datasets (if at all). Systematic applications of LSTMs for streamflow prediction in climates where a significant part of the catchments are snow and ice dominated are missing. In this work, an FS-LSTM (fast slow-LSTM) previously applied in Brazil is adapted for Swiss catchments to fill this gap [3]. The FS-LSTM explored builds on the work of Hoedt et al. (2021) that imposed mass constraints on an LSTM, called MC-LSTM [4]. FS-LSTM adds a fast and slow part for streamflow, containing rainfall and soil moisture respectively. We will discuss benchmark results against an existing semi-distributed conceptual model widely used in Switzerland for streamflow simulation [5]. References:[1]: Kratzert et al., Rainfall-runoff modelling using Long Short-Term Memory (LSTM) networks, 2018.[2]: Lees et al., Hydrological concept formation inside long short-term memory (LSTM) networks, 2022.[3]: Quinones et al., Fast-Slow Streamflow Model Using Mass-Conserving LSTM, 2021.[4]: Hoedt et al., MC-LSTM: Mass-Conserving LSTM, 2021.[5]: Viviroli et al., An introduction to the hydrological modelling system PREVAH and its pre- and post-processing-tools, 2009.

The increased frequency of temperature anomalies and drought events in Switzerland has major ecological implications, with impacts over whole ecosystems. In Swiss forests, the 2018 drought, which was the most severe drought event recorded led to widespread leaf discoloration, premature leaf-shedding, and tree mortality. While work has been carried out to analyse droughts a posteriori, the prediction of potential drought impacts would make it possible to anticipate ecological responses, manage resources and mitigate damage. Current approaches to drought prediction include mechanistic models. However, such models are often limited by data accessibility and resolution to effectively describe local effects. Deep learning models trained on remote sensing and atmospheric data have been applied to drought fore- casting, but face the “black box” issue and often discard domain knowledge on drought mechanisms. In this work, we propose a spatio-temporal deep learning method for drought forecasting in forests based on Sentinel-2 satellite imagery and weather variables, with the inclusion of topographic and environmental information. Drought is monitored by a proxy of early leaf wilting, using the normalized difference vegetation index (NDVI) that can be derived from Sentinel-2 bands. By predicting future NDVI values of pixels, we predict the potential occurrence of droughts in the short term. Hand-crafted features based on environmental data are used as input for the model, such as high-resolution topographic features which can capture micro-climatic effects, as well as soil- vegetation-climate relationships. Environmental information is provided to the model through data on soil and forest properties. This explicit modelling with topographic and environmental features increases the model interpretability, compared to models performing feature extraction and based only on image bands. A sequence model with long short-term memory (LSTM) cells was selected for its capacity to learn long-term dependencies as required in our application. We implement a pipeline to process spatiotemporal data, including data aggregation, normalization, missing data impu- tation and sample pixel timeseries for the prediction task. The model is trained and tested on data between 2015 and 2021, using the mean squared error to evaluate performances. A month (3 timesteps at Sentinel-2 acquisition rate) is forecasted given the past 3 months (9 timesteps) at a specific location. We opt for a “guided prediction” approach where the model has also access to weather forecasts for the future timesteps. The model is trained and tested in different regions in Switzerland to assess its generalization in space. A feature importance study was performed to identify key factors for drought forecasting and further improve the model. This research combines drought predictors known to have an impact in ecology and hydrology with a guided deep learning model. We offer a method for dealing with heterogeneous spatiotemporal data and train an interpretable model for forecasting potential forest drought.

The ballooning volume and complexity of geospatial data is one of the main inhibitors for advancements in climate & sustainability research. Oftentimes, researchers need to create bespoke and time-consuming workflows to harmonize datasets, build/deploy AI and simulation models, and perform statistical analysis. It is increasingly evident that these workflows and the underlying infrastructure are failing to scale and exploit the massive amounts of data (Peta and Exa-scale) which reside across multiple data centers and continents. While there have been attempts to consolidate relevant geospatial data and tooling into single cloud infrastructures, we argue that the future of climate & sustainability research relies on networked/federated systems. Here we present recent progress towards multi-cloud technologies that can scale federated geospatial discovery and modeling services across a network of nodes. We demonstrate how the system architecture and associated tooling can simplify the discovery and modeling process in multi-cloud environments via examples of federated analytics for AI-based flood detection and efficient data dissemination inspired by AI foundation models.

Recent hot and dry summers in Europe have had a significant impact on forest functioning and structure. In 2018 and 2019, Central Europe experienced two extremely dry and hot summers. These extremes resulted in widespread canopy defoliation and tree mortality. The objective in this study is to create a predictive model for predicting the density of vegetation, as measured by the NDVI index. We predict NDVI at a horizon of a month utilising data from the previous months as input to determine where and when drought impacts are triggered. Such predictive models should take into account both spatial and temporal dependencies between environmental variables and impacts. We hereon focus on Switzerland's forests as a region of interest to leverage high-quality model input layers and applications to typical stakeholder needs. Widely used vegetation indices and mechanistic land surface models are not effectively informed by the full information contained in Earth Observation data and the observed spatial heterogeneity of land surface greenness responses at hillslope-scale resolution. Effective learning from the simultaneous evolution of climate and remotely sensed land surface properties is challenging. Modern deep learning and machine learning techniques, however, have the capacity to generate accurate predictions while also explaining the relationship between climate and its recent history, the position in the landscape, and influences on vegetation. The task is to predict the future NDVI over forest areas, given past and future weather and surface reflectance. Giving future weather predictions as an input to the model, we are going for a 'guided-prediction' approach where the aim is to exploit weather information from forecasting models in order to increase the predictive power of the model - similar to the EarthNet2021 Challenge. Models are fully data-driven, without feature engineering and trained on spatio-temporal datacubes which can be seen as stacked satellite imagery for a specific geo-location and a timestep of past Sentinel 2 surface reflectance, past (observed) and future (forecasted) climate reanalysis, time-invariant information from a digital elevation model, and land cover map. The data pre-processing step includes implementing a customized dataset for drought impact prediction task, and a customized data sampler in order to be able to sample data (scenes) both spatially and temporally. Additional data operations include aggregation of the weather data, normalization, and data imputation both on the image-level and missing-day level. For the prediction task, we used Convolutional Long-Short Term Memory models. In the temporal domain, models are trained on the period between 2015-2018, and be validated between 05-2019 and 09-2019. For the test period summer months of 2020 and 2021 will be used. However, in the spatial domain, for the sake of testing the generalizability of the model, different regions were used for train, validate and test processes. In order to asses the models performance on the temporal domain, tests with different training and testing window sizes are used. As for evaluating the performance of the model, Mean Squared Error was used. The project will lay the basis for an early warning platform to enable periodically updated near-term drought-impact forecasts.

Autonomic peripheral activity is partly governed by brain autonomic centers. However, there is still a lot of uncertainties regarding the precise link between peripheral and central autonomic biosignals. Clarifying these links could have a profound impact on the interpretability, and thus usefulness, of peripheral autonomic biosignals captured with wearable devices. In this study, we take advantage of a unique dataset consisting of intracranial stereo-electroencephalography (SEEG) and peripheral biosignals acquired simultaneously for several days from four subjects undergoing epilepsy monitoring. Compared to previous work, we apply a deep neural network to explore high-dimensional nonlinear correlations between the cerebral brainwaves and variations in heart rate and electrodermal activity (EDA). Further, neural network explainability methods were applied to identify most relevant brainwave frequencies, brain regions and temporal information to predict a specific biosignal. Strongest brain-peripheral correlations were observed from contacts located in the central autonomic network, in particular in the alpha, theta and 52 to 58 Hz frequency band. Furthermore, a temporal delay of 12 to 14 s between SEEG and EDA signal was observed. Finally, we believe that this pilot study demonstrates a promising approach to mapping brain-peripheral relationships in a data-driven manner by leveraging the expressiveness of deep neural networks.

Epilepsy is a common neurological disorder characterized by recurrent epileptic seizures. These seizures have different intensities and might lead to accidents or, in the worst case, to sudden death. Therefore, being able to predict epileptic seizures would allow patients to be prepared, reducing the risk of injury. This paper focuses on epileptic seizure prediction using EEG (Electroencephalogram) signals. In contrast to the standard approach where the preictal state is assumed to have a constant duration in all the seizures of a patient, we propose a new method that labels each seizure individually exploiting clustering. Our labeling approach, which was applicable for 38% of the selected seizures, results in substantial improvements compared to the standard one. In fact, it reduces noise in the labels and improves the performance of the binary classifier used to distinguish the interictal and preictal states. Hence, our results suggest that the preictal duration is seizure-specific, not only patient-specific. Finally, we show that our method is able to predict 17 out of 18 (94%) seizures between 15 and 85 minutes, before seizure onset.

Flip-chip interconnects made entirely from copper are promising candidates to overcome the intrinsic limits of solder-based interconnects and match the demand for increased current densities of high-performance microprocessors. To this end, dip-based all-copper interconnects have been demonstrated to be a viable approach to form electrical interconnects by sintering copper nanoparticles between the copper pillar and pad. However, the performance of this technology is limited by residual porosity of the copper joint formed between the pillar and the pad during sintering. Moreover, the applicability of this technology in the printed circuit board industry is constrained by thermomechanical stresses that arise from the sintering. Furthermore, the compatibility of the sintering approach with standard finishing layers used as diffusion-barrier layers to provide wetting and to prevent the oxidation of the pillar and pad surfaces is unknown. In this paper, we pursue the advancement of dip-based all-copper interconnect technology. We demonstrate the robustness of dip-transfer of copper paste for varying withdrawal velocities and typical nonuniformities in copper pillar heights. Moreover, we prove the applicability of this technology with fine pillar diameters and pitches down to 10 and $20~\mu \text{m}$ , respectively. In addition, we demonstrate reduced porosities of the copper joints by applying pressure during the bonding process at 200 °C. This densification leads to interconnects with four times the shear strength and half the electrical resistance compared to interconnects formed without the application of pressure. In addition, the bonding temperature is decreased to 160 °C by applying a novel copper paste formulation, reducing the thermomechanical stress in the assembly caused by the cooling from the sintering temperature. Finally, the compatibility of this technology with EPAG, ENIG, and ENEPIG finishing layers is demonstrated, with resulting shear strength of the interconnects above 70 MPa and electrical resistance comparable to the unfinished samples.

Pastes based on copper (Cu) nanoparticles (NPs) are promising electronic-packaging materials for the attachment of high-power devices. However, the rapid oxidation of nanostructured Cu requires the use of reducing agents during processing, which makes it less suitable for attaching large-area dies (> 4 mm2). Recently, the functionalization of Cu-NP surfaces with a mixture of amines prevented oxidation, allowing for sintering without the need for reducing agents. Here we investigate the sintering mechanisms involved during die attachment using pastes of passivated Cu NPs, with particular focus on the critical role of the carrier solvents. Using 1-nonanol or 1-decanol as solvents, we first demonstrate the absence of Cu-oxide phases in the pastes after fabrication and the stability of the resulting nanostructured copper for as much as 30 min in air. By measuring the evolution of the electrical characteristics of the paste during drying and sintering, we show that electrically conductive agglomerates form among the NPs between 141°C and 144°C, independent of the carrier solvent used. The carrier solvent was found to affect mainly the densification temperature of the copper agglomerates. Because they lead to uniform sintering of the material, Cu pastes based on solvents with a low boiling point and high vapor pressure are preferable for attaching dies with area greater than 25 mm2. We show that dies with an area as large as 100 mm2 can be attached using a Cu paste based on 1-nonanol. These pastes enables the formation of temperature-resistant bonding for high-power devices using a simple and cost-effective approach.

Non-isothermal liquid evaporation in micro-pore structures is studied experimentally and numerically using the lattice Boltzmann method. A hybrid thermal entropic multiple-relaxation-time multiphase lattice Boltzmann model (T-EMRT-MP LBM) is implemented and validated with experiments of droplet evaporation on a heated hydrophobic substrate. Then liquid evaporation is investigated in two specific pore structures, i.e. spiral-shaped and gradient-shaped micro-pillar cavities, referred to as SMS and GMS, respectively. In SMS, the liquid receding front follows the spiral pattern; while in GMS, the receding front moves layer by layer from the pillar rows with large pitch to the rows with small one. Both simulations agree well with experiments. Moreover, evaporative cooling effects in liquid and vapour are observed and explained with simulation results. Quantitatively, in both SMS and GMS, the change of liquid mass with time coincides with experimental measurements. The evaporation rate generally decreases slightly with time mainly because of the reduction of liquid–vapour interface. Isolated liquid films in SMS increase the evaporation rate temporarily resulting in local peaks in evaporation rate. Reynolds and capillary numbers show that the liquid internal flow is laminar and that the capillary forces are dominant resulting in menisci pinned to the pillars. Similar Péclet number is found in simulations and experiments, indicating a diffusive type of heat, liquid and vapour transport. Our numerical and experimental studies indicate a method for controlling liquid evaporation paths in micro-pore structures and maintaining high evaporation rate by specific geometry designs.