PEOPLEAlexander Jacob

de en it

Alexander Jacob

Vice Head of Institute
Institute for Earth Observation

T +39 0471 055 371

alexander.jacob@eurac.edu

https://orcid.org/0000-0003-4434-7244

download or scan my contact information (vCard Format)

About

Forschungsthemen

Skalierbares Rechnen (Fokus auf effiziente Datenorganisation und interoperable Verarbeitung von EO-Daten, einschließlich Themen wie Metadaten und FAIR-Prinzipien, Datenwürfel, Cloud- und HPC-Computing sowie wissenschaftliche Workflow-Beschreibung)
Datengetriebene Modellierung (Fokus auf Aspekte der erklärbaren KI und Unsicherheit bei Verwendung von Erdbeobachtungsdatenströmen, einschließlich der Integration zwischen physikalischem und datengetriebenem Modellieren hin zu digitalen Zwillingen)
Multisensor-Datenfusion (Fokus auf die Integration von SAR- und optischen Daten, einschließlich SAR-Interferometrie, für verschiedene Anwendungen wie Landbedeckungskartierung, Schneekartierung, Hydrologie und Überwachung des Klimawandels)

Kurzbiografie

Alexander Jacob ist seit 2023 stellvertretender Leiter des Instituts für Erdbeobachtung.

Er erhielt 2012 und 2014 einen Master of Science und ein Lizenziat in Geoinformatik vom Königlichen Institut für Technologie in Stockholm (Schweden). Zuvor studierte er Vermessungswesen an der Technischen Universität Darmstadt (Deutschland). Beruflich sammelte er wertvolle Erfahrungen als wissenschaftlicher Mitarbeiter (2003-2007) am Institut für Geodäsie und am Institut für Physikalische Geodäsie der TU Darmstadt, wo er an Themen wie ortsbezogenen Diensten und Softwareentwicklung für das Management von GNSS-Daten arbeitete. In diesem Zusammenhang verbrachte er auch mehrere Monate (2006) bei Trimble Terrasat in München. Später (2009) setzte er seine Erfahrungen in der Softwareentwicklung bei der Visual Meta GmbH in Berlin fort, wo er Server-seitige Entwicklung für Suchmaschinen durchführte. Von hier aus konzentrierte er sich auf die Forschung und spezialisierte sich auf Fernerkundung am Königlichen Institut für Technologie. Während seines Studiums unterrichtete er dort Kurse in vielen Fächern, darunter Geoinformatik, Photogrammmetrie, Fernerkundung, sowohl in Vorlesungen als auch in Laborübungen. 2014 erhielt er auch den Preis für den besten Lehrer an der Fakultät für Architektur und die gebaute Umwelt. Zusätzlich wurde er mehrfach als Gastdozent an die Universität Karlstad (Schweden) eingeladen, um Kurse in Photogrammmetrie zu unterrichten. 2015 trat er dem Eurac Research am Institut für Angewandte Fernerkundung in der Forschungsgruppe Technologie für Umwelt bei. 2018 gründete er die Forschungsgruppe Advanced Computing am inzwischen neubenannten Institut für Erdbeobachtung. Er ist italienischer Vertreter in der Group on Earth Observation (GEO) für die Arbeitsgruppe Daten & Wissen sowie einer der Co-Vorsitzenden der Open Geospatial Consortium (OGC) GeoDataCubes Standards Working Group. Er hat als Berater für die Europäische Weltraumorganisation (ESA) und das Gemeinsame Forschungszentrum (JRC) der Europäischen Kommission (EC) in Fragen des Erdbeobachtungsdatenmanagements und -computings gedient.

Links

https://www.linkedin.com/in/alexander-jacob-07b96670/

https://scholar.google.com/citations?user=VL6ZmeAAAAAJ&hl=en

Scientific Publications

Journal article2026

A Transformer-based Convolutional Regressor to include SAR Backscatter Signals in Monitoring Alpine Grasslands

Singh A, Göhner CS, Stendardi L, Claus M, Cuozzo G, Jacob A, Castelli M

Giovanni Cuozzo, Alexander Jacob, Mariapina Castelli, Michele Claus, Abhishek Singh

Projekt: Innovative Methoden zur Verbesserung der Überwachung von Agrarumweltbedingungen unter Verwendung von Sensordaten auf verschiedenen Skalen

Monitoring grasslands over the phenological season in the alpine environment using Synthetic Aperture Radar (SAR) backscatter data requires innovative approaches that can combine domain-specific ancillary information with advanced deep learning techniques. Alpine grasslands are characterized by complex topography, seasonal variations, and heterogeneous agricultural management, making traditional monitoring methods less effective. The Leaf Area Index (LAI) can generally be used to study the above ground biomass of grasslands, which is estimated rapidly and accurately from the Sentinel-2 multispectral data. However, the frequent unavailability of cloud-free imagery poses a significant challenge, creating temporal gaps. To address this limitation, there is growing interest in incorporating Sentinel-1 SAR backscatter data to perform temporal gap-filling, enhancing the temporal resolution of grassland LAI derived from Sentinel-2 and enabling continuous monitoring of grasslands throughout the growing season. In this paper, we propose a domain-specific transformer-based convolutional network that extracts features from SAR data and its derivatives to capture long-range dependencies, and we preliminarily investigate its application to temporal gap-filling of Sentinel-2 LAI. The effectiveness of the proposed method for predicting LAI has been validated by testing it against Sentinel-2-derived LAI. In the results, we achieved an R2 around 0.59 on the test sets, between the related predicted LAI and Sentinel-2 LAI.

DOI Quelle

DOI: 10.1109/LGRS.2025.3645675

Journal article2025

An Innovative Approach for Calibrating Hydrological Surrogate Deep Learning Models

Aieb A, Liotta A, Jacob A, Ferrario IF, Yaqub MA

Alexander Jacob, Iacopo Ferrario

Projekt: Eine interdisziplinäre Digital Twin Engine für die Wissenschaft

Developing data-driven models for spatiotemporal hydrological prediction presents challenges in managing complexity, capturing fine spatial and temporal resolution, and ensuring model resilience across diverse regions. This study introduces an innovative surrogate deep learning (SDL) architecture designed to predict daily soil moisture (DSM) and daily actual evapotranspiration (DAE) by integrating climate data and geophysical insights, with a focus on mountainous areas such as the Adige catchment. The proposed framework aims to enhance the parameter-calibration quality. The process begins by mapping the statistical characteristics of DAE and DSM across the whole region using an unsupervised fusion technique. Model accuracy is assessed by comparing the similarity of Fuzzy C-Means (FCM) clusters before and after fusion, providing a metric for feature reduction. A data transformation technique using Gradient Boosting Regression (GBR) is then applied to each homogeneous subregion identified by the Random Forest classifier (RFC), based on elevation parameters (Wflow_dem). Furthermore, Kernel density estimation is used to ensure the reproducibility of the RFC-GBR process across large-scale applications. A comparative analysis is conducted across multiple SDL architectures, including LSTM, GRU, TCN, and ConvLSTM, over 50 epochs to better evaluate the beneficial effect of the transformed parameters on model performance and accuracy. Results indicate that adjusted parameter calibration improves model performance in all cases, with better alignment to Wflow ground truth during both wet and dry periods. The proposed model increases the accuracy by 20% to 42% when using simpler SDL models like LSTM and GRU, even with fewer epochs.

DOI Quelle

DOI: 10.3390/rs17111916

Conference proceedings article2025

Feature-Level Uncertainty Response Analysis in Multi-Sensor Deep Learning Models

Liu X, Aieb A, Jacob A, Song W, Liotta A

Alexander Jacob

As environmental monitoring systems increasingly rely on multi-sensor data, quantifying predictive uncertainty at the feature level is crucial for building robust and trustworthy deep learning models. In this study, we propose a perturbation-based framework for feature-wise uncertainty estimation in sensor networks, using the Intel Lab dataset as a case study. By applying controlled perturbations to individual input features and conducting Monte Carlo Dropout inference, the framework systematically quantifies how predictive uncertainty—computed as the mean variance over stochastic forward passes—varies as a function of perturbed feature values at each sensor location. This perturbation based mean uncertainty identifies critical value ranges where the sensor-specific model becomes less confident. Based on the continuously distributed spatial feature values obtained via inverse distance weighting interpolation, we further map the perturbation-based mean uncertainty onto the spatial domain to visualize the spatial distribution patterns of predictive uncertainty across sensor locations. Experimental results demonstrate that our method effectively reveals how a certain feature contribute to the mean model uncertainty offering actionable insights for data acquisition, sensor placement, and model refinement.

DOI Quelle

DOI: 10.1145/3762329.3762361

Journal article2025

Federated and reusable processing of Earth observation data

Mohr M, Pebesman E, Dries J, Lippens S, Janssen B, Thiex D, Milcinski G, Schumacher B,

Alexander Jacob, Michele Claus

Projekt: Eine einheitliche, open source Schnittstelle zwischen Dateninfrastruktur und Frontendanwendung für Erdbeobachtung, R4openEO – Integration von R Analytics mit openEO Platformen, openEO Platform: Eine föderierte Erdbeobachtungsplattform

The unprecedented and continuously growing volume of Earth Observation (EO) and geospatial data has necessitated a paradigm change where compute is collocated with the data archives in public clouds. However, as no single cloud platform can host all of this data, federated processing solutions that work across multiple cloud platforms are becoming increasingly relevant. A community-based approach to federated processing has started using openEO, a common Application Programming Interface (API) and set of well-defined processes that simplifies reuse and provides a valuable level of abstraction when handling large EO data volumes. We present key concepts for federated processing and related interoperability aspects based on openEO Platform, a federated public cloud platform.

DOI Quelle

DOI: 10.1038/s41597-025-04513-y

Database2025

OEMC project use case: High resolution Snow Water Equivalent in selected mountain regions

Eurac Research, Premier V, Claus M, Jacob A, Notarnicola C, Marin C, Singh A, Zvolenský J

Alexander Jacob, Claudia Notarnicola, Carlo Marin, Juraj Zvolenský, Michele Claus, Valentina Premier, Abhishek Singh

We propose to produce a snow water equivalent (SWE) product at a spatial resolution of 20 meters. The product will be available for selected mountain basins in South Tyrol on a daily temporal resolution, updated on a yearly basis. We rely on a multi-sensor EO SWE reconstruction method using Sentinel-2, MODIS, potentially Sentinel-1 and in-situ data. The information will be of value for different applications such as earth system models, irrigation planning and hydro power planning (e.g. Eumetsat/ZAMG). The Province of South Tyrol is a local stakeholder planning to evaluate the data for their hydrological planning activities.

Stakeholder needs:

High-resolution and daily data over some selected hydrological catchments that are relevant for the Hydrological Office of the Province of Bolzano;
Open access data that respect the FAIR principle, easy to access and download.

Planned implementation:

Implementation steps include:

High-resolution snow cover classification (Sentinel-2 or Landsat data);
Data fusion algorithm to obtain daily high-resolution snow cover data (merging high-resolution snow cover maps with low-resolution snow cover maps, such as MODIS);
SWE reconstruction using a degree-day model forced with available in-situ temperature or re-analysis data;
Implementation in OpenEO to upscale the algorithm.

DOI Quelle

DOI: 10.60566/z84j1-yjc55

Conference proceedings article2025

Time Series Gaps Filling based on Heterogeneous Sensor Data Fusion

Aieb A, Liu X, Jacob A, Yaqub MA, Liotta A

Alexander Jacob

Missing data (MD) in heterogeneous sensor networks presents a major challenge, particularly with non-stationary climate data that exhibit inconsistent correlations with other variables. This issue stems from variations in spatial and temporal resolutions across sensors, along with complex interdependencies among climate factors. To overcome these challenges, a novel framework is introduced that combines Seasonal-Trend decomposition using LOESS (STL) for data harmonization, a Convolutional Autoencoder (CAE) for data imputation, and a Gradient boosting based Neuron Network (G_NN) for final data reconstruction. The STL component enhances the alignment between multisource datasets by isolating trend and seasonal components, enabling more reliable synthesis across heterogeneous sensors. The CAE enables the framework to capture multiscale dependencies used to impute MD, while G_NN refines the final reconstruction by adjusting residual errors. Precipitation is used as a case study given its variability and the difficulty of aligning it with other climate variables across different scales. The model was evaluated under different percentages of MD, exhibiting remarkable performance. At around 30% MD, it achieved a high R2 , ranging from 0.98 to 0.85. Even with 50% MD, the model retained strong accuracy and scalability, maintaining an R2 of approximately 0.7. Overall, the proposed method demonstrated notable improvements of 26% and 32% compared to advanced approaches such as GAN and MICE, respectively

DOI Quelle

DOI: 10.1145/3762329.3762362

Conference proceedings article2025

Towards Provenance-Aware Earth Observation Workflows: the openEO Case Study

Omidi H, Sacco L, Hutter V, Irsiegler G, Claus M, Schobben M, Jacob A, Schramm M,

Alexander Jacob, Michele Claus

Projekt: Eine interdisziplinäre Digital Twin Engine für die Wissenschaft

Capturing the history of operations and activities during a computational workflow is significantly important for Earth Observation (EO). The data provenance helps to collect the metadata that records the lineage of data products, providing information about how data are generated, transferred, manipulated, by whom all these operations are performed and through which processes, parameters, and datasets. This paper presents an approach to improve those aspects, by integrating the data provenance library yProv4WFs within openEO, a platform to let users connect to Earth Observation cloud back-ends in a simple and unified way. In addition, it is demonstrated how the integration of data provenance concepts across EO processing chains enables researchers and stakeholders to better understand the flow, the dependencies, and the transformations involved in analytical workflows.

DOI Quelle

DOI: 10.1109/eScience65000.2025.00016

Presentation2024

Cubes & Clouds – A Massive Open Online Course for Cloud Native Open Data Sciences in Earth Observation

Zellner PJ, Balogun RO, Eckardt R, Hodam H, Dolezalova T, Meißl S, Eberle J, Claus M,

Mattia Callegari, Alexander Jacob, Michele Claus

Projekt: MOOC EO Open Data Science

Earth Observation scientists are confronted with unprecedent data volumes that are constantly growing as the number of satellite missions increases, as well as spatial and temporal resolution do. Traditional working modes, such as downloading satellite data and processing it on a local computer, do not apply anymore and EO science is moving quickly towards cloud technologies and open science practices. Even though these new technologies and practices are evolving quickly and are becoming community standard, there is not much educational material available to educate the next generation of EO scientists.

The Massive Open Online Course Cubes & Clouds - Cloud Native Open Data Sciences for Earth Observation teaches the concepts of data cubes, cloud platforms and open science in the context of earth observation. It targets Earth Science students and researchers who want to increase their technical capabilities onto the newest standards in EO computing, as well as Data Scientists who want to dive into the world of EO and apply their technical background to a new field. Before starting, prerequisites are a general knowledge of EO and python programming. The course explains the concepts of data cubes, EO cloud platforms and open science by applying them to a typical EO workflow from data discovery, data processing up to sharing the results in an open and FAIR (Findable, Accessible, Interoperable, Reusable) way. An engaging mixture of videos, animated content, lectures, hands-on exercises and quizzes transmits the content. After finishing the participant will understand the theoretical concepts of cloud native EO processing and have gained practical experience by conducting an end-to-end EO workflow. The participant will be capable of independently using cloud platforms to approach EO related research questions and be confident in how to share research by adhering to the concepts of open science.
The hands on exercises are carried out on the EO cloud platform Copernicus Data Space Ecosystem and are leveraging the Copernicus Data available through the STAC Catalogue and the cloud processing API openEO. In the final exercise the participants collaborate on a community mapping project adhering to open science and FAIR standards. A snow cover map is jointly created where every participants maps a small area of the alps and submits it to a STAC catalogue and web viewer. Ultimately, the map grows in space and time and every participant has contributed proving they are capable of EO cloud computing and open science practices.
The talk will guide through the topics covered in the MOOC and show how they are presented in the EOCollege e-learning platform, the links to the exercises carried out on CDSE will be explored and the open science aspect will be shown in the community mapping project and the invitation to collaborate on the courses completely open GitHub repository.

DOI Quelle

DOI: 10.5194/egusphere-egu24-21733

Conference proceedings article2024

Enhancing Seasonal Climate Forecasting for the Alpine Region Through Machine Learning Statistical Downscaling

Dhinakaran S, Crespi A, Jacob A, Pebesma E

Alexander Jacob, Suriyah Dhinakaran, Alice Crespi

Projekt: Eine interdisziplinäre Digital Twin Engine für die Wissenschaft

This study aims to address the critical need for refined, highresolution seasonal climate forecasts in the Alpine region in support to risk management and decision-making processes, especially in the challenging context of climate change. By leveraging regression-based Machine Learning (ML) algorithms, a Perfect Prognosis approach is applied to statistically downscale the daily fields of 2-metre temperature and total precipitation of ECMWF SEAS5 seasonal forecasts over the Alps. In particular, four different ML methods are considered: Random Forest, Light Gradient Boosting Machine (LGBM), Adaptive Boosting (AdaBoost) and Extreme Gradient Boosting (XGBoost). The daily fields of the European CERRA reanalysis (5.5 km) are used as reference target, while a set of meteorological predictors from the coarser grid are considered. In a preparatory phase, all ML methods and configurations are implemented and validated starting from the predictor fields of the ERA5 reanalysis. LGBM displayed the best results during training and validation for both temperature and precipitation, with superior computational speed and efficiency with respect to the other methods. It demonstrates prowess in capturing daily variations, with R2 scores of 0.95 mean temperature and 0.67 for precipitation, with generally low biases (-0.05 °C and 5.34% for daily mean temperature and precipitation, respectively, as yearly averages). Further optimization to increase the prediction accuracy of extreme values and annual precipitation averages are discussed. The best performing LGBM method is finally applied to downscale the SEAS5 seasonal forecast data and will represent a crucial component of a drought predicting model for the Alps in the framework of the EU-funded interTwin project (101058386).

DOI

DOI: 10.1109/IGARSS53475.2024.10642272

Conference proceedings article2024

Interpreting Image Super-Resolution in Artificial Neural Networks from Global and Local Views

Liu X, Jacob A, Song W, Liotta A

Alexander Jacob

Work on image super-resolution (SR), to construct higher-resolution images starting from low-quality ones, has focused primarily on reconstruction algorithms and specific application domains. In this work, we aim at methods to aid interpreting SR inner-working, with a view to improve explain- ability. We propose a novel gradient-based attribution approach, to provide interpretations from global and local perspectives, dubbed glocal attribution map (GL-AM). After verification with five different SR models, we show that GL-AM: (1) is a powerful tool to understand the principles of SR networks from both global and local views; (2) provides the consensus and variation sensitivity of different models to the input; (3) is more effective to emphasize the features captured by the attention mechanism (for the SR model) through feature re-calibration; (4) is more computationally efficient and more effective as the region of interest increases.

DOI

DOI: 10.1109/IDSTA62194.2024.10746988

Conference proceedings article2024

MOOC Cubes and Clouds - Cloud Native Open Data Sciences for Earth Observation

Zellner PJ, Claus M, Dolezalova T, Balogun RO, Eberle J, Hodam H, Eckhardt R, Meißl S,

Alexander Jacob, Michele Claus

Projekt: MOOC EO Open Data Science

The "Cubes and Clouds" Massive Open Online Course (MOOC) addresses the growing challenges and opportunities in Earth Observation (EO) data analysis posed by the exponential growth of satellite missions and data volumes. This course introduces cloud-native EO concepts and open science principles, facilitating collaboration and data sharing within the EO community. Key topics covered in the MOOC include data discovery, processing workflows, and data sharing using real-world examples like snow cover mapping in the Alps. Participants engage with interactive lessons, videos, and hands-on coding exercises, leveraging freely available geospatial data and emphasizing open data principles and interoperability. The course infrastructure seamlessly integrates with cloud platforms like the Copernicus Data Space Ecosystem, enabling learners to apply concepts in a practical, cloud-based environment. Initial user statistics indicate strong interest, particularly among early-career professionals and researchers, with participant surveys suggesting increased confidence in using EO cloud platforms and embracing open science practices upon course completion. Overall, "Cubes and Clouds" serves as a valuable resource for the EO community, fostering transparency, collaboration, and innovation in geospatial science. Future developments aim to expand course content and attract diverse user groups, ensuring broader accessibility and long-term impact in advancing cloud-native EO and open science initiatives.

DOI Quelle

DOI: 10.5194/isprs-archives-XLVIII-4-W12-2024-157-2024

Conference proceedings article2024

Short-Term Forecasting of Non-Stationary Time Series

Aieb A, Liotta A, Jacob A, Yaqub MA

Alexander Jacob

Forecasting climate events is crucial for mitigating and managing risks related to climate change; however, the problem of non-stationarity in time series (NTS) arises, making it difficult to capture and model the underlying trends. This task requires a complex procedure to address the challenge of creating a strong model that can effectively handle the non-uniform variability in various climate datasets. In this work, we use a daily standardized precipitation index dataset as an example of NTS, whereby the heterogeneous variability of daily precipitation poses complexities for traditional machine-learning models in predicting future events. To address these challenges, we introduce a novel approach, aiming to adjust the non-uniform distribution and simplify the detection time lags using autocorrelation. Our study employs a range of statistical techniques, including sampling-based seasonality, mathematical transformation, and normalization, to preprocess the data to increase the time lag window. Through the exploration of linear and sinusoidal transformation, we aim to assess their impact on increasing the accuracy of forecasting models. A strong performance is effectively observed by using the proposed approach to capture more than one year of time delay across all the seasonal subsets. Furthermore, improved model accuracy is observed, notably with K-Nearest Neighbors (KNN) and Random Forest (RF). This study underscores RF’s consistently strong performance across all the transformations, while KNN only demonstrates optimal results when the data have been linearized.

DOI Quelle

DOI: 10.3390/engproc2024068034

Technische
Diese Cookies sind für das Funktionieren der Website erforderlich und können nicht deaktiviert werden.

Service	Abgelehnt/Akzeptiert	Terms Link
Youtube		https://support.google.com/youtube/answer/7671399?...
Vimeo		https://vimeo.com/terms
Flourish		https://flourish.studio/terms/
Datawrapper		https://www.datawrapper.de/terms
Power BI		https://www.microsoft.com/en/servicesagreement/
Facebook		https://www.facebook.com/policies/cookies/
Instagram		https://help.instagram.com/1896641480634370?ref=ig
X		https://help.twitter.com/en/rules-and-policies/twi...
Google Maps		https://maps.google.com/help/terms_maps/
LinkedIn		https://www.linkedin.com/legal/cookie-policy

Presentation2024

Cubes & Clouds – A Massive Open Online Course for Cloud Native Open Data Sciences in Earth Observation

Zellner PJ, Balogun RO, Eckardt R, Hodam H, Dolezalova T, Meißl S, Eberle J, Claus M,

Mattia Callegari, Alexander Jacob, Michele Claus

Projekt: MOOC EO Open Data Science

DOI Quelle

DOI: 10.5194/egusphere-egu24-21733

Alexander Jacob

About

Scientific Publications

A Transformer-based Convolutional Regressor to include SAR Backscatter Signals in Monitoring Alpine Grasslands

An Innovative Approach for Calibrating Hydrological Surrogate Deep Learning Models

Feature-Level Uncertainty Response Analysis in Multi-Sensor Deep Learning Models

Federated and reusable processing of Earth observation data

OEMC project use case: High resolution Snow Water Equivalent in selected mountain regions

Time Series Gaps Filling based on Heterogeneous Sensor Data Fusion

Towards Provenance-Aware Earth Observation Workflows: the openEO Case Study

Cubes & Clouds – A Massive Open Online Course for Cloud Native Open Data Sciences in Earth Observation

Enhancing Seasonal Climate Forecasting for the Alpine Region Through Machine Learning Statistical Downscaling

Interpreting Image Super-Resolution in Artificial Neural Networks from Global and Local Views

MOOC Cubes and Clouds - Cloud Native Open Data Sciences for Earth Observation

Short-Term Forecasting of Non-Stationary Time Series

Science Shots Eurac Research Newsletter

Was wir tun

Wo wir sind

Institute & Center

Folge uns

Meine Cookie Einstellungen

Externe Inhalte

Scientific Publications

A Transformer-based Convolutional Regressor to include SAR Backscatter Signals in Monitoring Alpine Grasslands

An Innovative Approach for Calibrating Hydrological Surrogate Deep Learning Models

Feature-Level Uncertainty Response Analysis in Multi-Sensor Deep Learning Models

Federated and reusable processing of Earth observation data

OEMC project use case: High resolution Snow Water Equivalent in selected mountain regions

Time Series Gaps Filling based on Heterogeneous Sensor Data Fusion

Towards Provenance-Aware Earth Observation Workflows: the openEO Case Study

Cubes & Clouds – A Massive Open Online Course for Cloud Native Open Data Sciences in Earth Observation

Enhancing Seasonal Climate Forecasting for the Alpine Region Through Machine Learning Statistical Downscaling

Interpreting Image Super-Resolution in Artificial Neural Networks from Global and Local Views

MOOC Cubes and Clouds - Cloud Native Open Data Sciences for Earth Observation

Short-Term Forecasting of Non-Stationary Time Series