Data will be made available on request.

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.

Research in image super-resolution (SR), which seeks to enhance image quality by producing higher-resolution versions from low-quality inputs, has primarily focused on developing reconstruction algorithms rather than enhancing interpretability. SR networks continue to exhibit the opaque, black-box characteristics typical of deep learning, with limited research dedicated to investigating their internal mechanisms. This study aims to deepen the understanding of SR and conduct attribution analysis of SR networks from a holistic reconstruction perspective. We introduce a novel attribution method based on gradient and optical flow, termed GOFlow. Following verification with five SR models, we demonstrate that (1) GOFlow proves to be an effective tool for analyzing attributed pixels in SR neural networks from a comprehensive perspective; (2) Compared to the Layer Attribution Method (LAM), GOFlow produces a more detailed attribution map from a global perspective; (3) GOFlow is capable of exploring how texture and colorfulness influence the outputs of SR, proving that GOFlow can interpret SR models at the feature level.

This study aims to address the critical need for refined, high-resolution 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).

The usage of Sentinel-1 Single Look Complex (SLC) and its derived interferometric products was limited due to the complexity of the workflows and computational resources required to access, download, process, store, and exploit them. We propose SAR2Cube as a framework based on open source software capable to efficiently store Sentinel-1 Single Look Complex (SLC) images into data cubes and produce on-the-fly Interferometric Synthetic Aperture Radar (InSAR) products, also providing a benchmark with the official European Space Agency (ESA) SNAP processing toolbox.

Due to the success of the Copernicus program and the general awareness towards satellite Earth Observation (EO) data, a growing number of cloud-based EO services are now available to users for processing and analyzing the available EO data. From a user perspective this is currently creating confusion due to the large number of available services and the lack of comparability between the offers. In addition, offers based on proprietary APIs without open source components have an inherent risk of vendor lock-in. The openEO standard has been developed to address this, and enabled creating openEO platform (see https://openeo.cloud), a federation of openEO backends, where the combined offering supersedes the capabilities of the individual parts.

Over the past decade, a multitude of independent initiatives have developed solutions to answer the need for Analysis-Ready Data, in order to reduce the time and effort required in order to generate added value out of raw, heterogeneous data. These initiatives resulted in various GeoDataCube (GDCs) implementations, standards, data formats and best practices.Interoperability between the GDC solutions has so far not been a core concern but with the increasing amount of data served as GDCs as well as its increasing uptake, it is becoming essential to understand what exactly a GDC entails, how it was created, and how different GDCs can be used together consistently. This presentation shall focus on the OGC GeoDataCube Standards Working Group (SWG) which has recently been initiated to tackle these issues proposing viable solutions defined as a standard.

The Massive Open Online Course – Earth Observation Open Data Science (MOOC EOODS) teaches the concepts of data cubes, cloud platforms, and open science in the context of Earth Observation (EO). The course is designed to bridge the gap towards the recent cloud native advancements in EO by offering a MOOC as an integrated and open learning experience relying on a mixture of animated lecture content and hands-on exercises hosted on the EO themed e-learning platform EO College.

This study aims at estimating the dry snow water equivalent (SWE) by using X-band SAR data from the COSMO-SkyMed (CSK) satellite constellation. Time series of CSK acquisitions have been collected during the dry snow period in the Alto Adige test site, in the Italian Alps, during the winter seasons from 2013 to 2015 and from 2019 to 2021. The SAR data have been analyzed and compared with the in-situ measurements to understand the X-band SAR sensitivity to SWE, which has been further assessed by Dense Media Radiative Transfer (DMRT) model simulations. The sensitivity analysis provided the basis for addressing the SWE retrieval from the CSK data, by exploiting two different machine learning (ML) techniques, namely Artificial Neural Networks (ANN) and Support Vector Regression (SVR). To ensure a statistical independence of training and validation processes, the algorithms are trained and tested using SWE predictions of the fully distributed snow model AMUNDSEN as reference data and are subsequently validated on the experimental dataset. Due to its influence on the CSK estimates, the effect of forest canopy was accounted for in the analysis. Depending on the algorithm, the validation resulted in a correlation coefficient 0.78 ≤ R ≤ 0.91, and a Root Mean Square Error 55.5 mm ≤ RMSE ≤ 87.4 mm between estimated and in-situ SWE. Further analysis and validation are needed; however, the obtained results seem suggesting the Cosmo-SkyMed constellation as effective tool for the retrieval of the dry snow water equivalent in alpine areas.

InSAR coherence is a promising parameter for land-cover classification and mapping. The ESA SEOM SInCohMap project is devised to test and analyze multi-temporal InSAR coherence potentialities exploiting dense multitemporal data from the Sentinel-1 constellation. In the framework of the project, this paper shows the first classification results using machine learning algorithms over a two-year period of InSAR coherence data. The evaluation is performed on the test site of Doñana (Seville, Southwestern Spain), mainly an agricultural area where different land covers can be identified. Classification results exploiting InSAR coherence shows accuracies around 80 % for this site.

In this paper, Sentinel-1A SAR data were evaluated for urban extent extraction using the KTH-Pavia Urban Extractor. The methodology is based on texture measures and spatial statistics as well as decision level fusion. Sentinel-1A data in Stripmap mode (SM) or Interferometric Wide Swath mode (IW) were acquired in five cities around the world. For two European cities Milan, Italy and Stockholm, Sweden, accuracy assessments were performed by comparing the extracted urban areas with the Urban Atlas in 2010. The preliminary results show that the Sentinel-1A Stripmap mode is very suitable for urban extraction, reaching an agreement of more than 83%. The Interferometric Wideswath mode shows potential as well reaching an agreement of 77%, but urban extraction over Stockholm had high omission errors in low density built up areas.

TerraSAR X data has been analyzed for its suitability of urban land cover mapping using our recently developed object based image analysis tool KTH-SEG, which is based on an edge aware region growing and merging algorithm and a support vector machine classifier. Classification results over the Shanghai International Airport area using 8 classes, Water, Grass, Roads, Buildings, Crops, Forest, Bare Crops and Green Houses have proven with an overall accuracy just shy of 84% that this is very well the case. It has further been investigated which segment sizes and image configuration yield the best results.

The objectives of this research are to develop robust methods for segmentation of multitemporal synthetic aperture radar (SAR) and optical data and to investigate the fusion of multitemporal ENVISAT advanced synthetic aperture radar (ASAR) and Chinese HJ-1B multispectral data for detailed urban land-cover mapping. Eight-date multiangle ENVISAT ASAR images and one-date HJ-1B charge-coupled device image acquired over Beijing in 2009 are selected for this research. The edge-aware region growing and merging (EARGM) algorithm is developed for segmentation of SAR and optical data. Edge detection using a Sobel filter is applied on SAR and optical data individually, and a majority voting approach is used to integrate all edge images. The edges are then used in a segmentation process to ensure that segments do not grow over edges. The segmentation is influenced by minimum and maximum segment sizes as well as the two homogeneity criteria, namely, a measure of color and a measure of texture. The classification is performed using support vector machines. The results show that our EARGM algorithm produces better segmentation than eCognition, particularly for built-up classes and linear features. The best classification result (80%) is achieved using the fusion of eight-date ENVISAT ASAR and HJ-1B data. This represents 5%, 11%, and 14% improvements over eCognition, HJ-1B, and ASAR classifications, respectively. The second best classification is achieved using fusion of four-date ENVISAT ASAR and HJ-1B data (78%). The result indicates that fewer multitemporal SAR images can achieve similar classification accuracy if multitemporal multiangle dual-look-direction SAR data are carefully selected.