ESA has launched a new activity HAPSVIEW with the goal of assessing the use of HAPS for supporting air quality and GHG applications.
The project is aimed to measure the added value of possible HAPS applications against current capabilities of air quality models integrating ground-based and satellite observations. HAPS represents a very promising technology in this field provided that perhaps the main strength of HAPS is their capability of observation with a high temporal frequency per day that can be made over an extended time period. This could lead to new insights and detailed information w.r.t. temporal variability in emissions and air pollution levels
The main outputs of this activity describe different items in the HAPS: Use Case, the definition of of the mission concepts and the formulation of the full chain of requirements: from end user needs via user requirements, data requirements, regulatory requirements and instrument and platform requirements to mission requirements.
The concept outlined in the Mission Requirements is based in one or several HAPS platforms carrying on-board one or more instruments and it shall take into consideration the possible synergies with conventional satellite missions aimed at air quality or GHG measurements. Besides, the technological and scientific gaps and limitations will be analysed in order to provide recommendations regarding the development or evolution of a HAPS-based solution.
The mission concepts derived from the Level-1 and Level-2 requirements is described, accounting for the available technological and scientific solutions at different levels and covering the specification of HAPS platforms and payload characteristic; the main elements of the Ground Segment, the integration of multi-source data and the concept of operations. Regulatory requirements for the envisaged launch and flight locations will be also investigated in the course of the project, with the aim of identifying elements for the definition of flight approval process.
Finally, the participation of stakeholders and users is fundamental in this study, since the definition of air quality and GHG applications and Use Cases will be driven by societal needs expressed by stakeholders from cities and regions. In this particular study we will count on the participation of two main users: The Regional Government of Andalusia in South Spain and the Province of South Holland together with the Municipality of Rotterdam.

In recent years, the Data Cube concept has been readily embraced by the Earth Observation community for having the potential to significantly lower the barriers of using satellite data and to overcome the limitations of the typical image-centric approach. Data Cubes are often mentioned in the same breath with Analysis-Ready Data, because for Cube generation, all data regardless of their origin have at least to be transformed into one data model, share one common grid and adhere to a common metadata convention. In the era of freely-available satellite archives on cloud infrastructures like DIAS or AWS, on-the-fly provision of Data Cubes through APIs avoids data duplication and is thus the most efficient approach for use cases involving less complex processing or one-time requests.
Many use cases, however, require complex and costly processing, the storage of intermediate products or several versions of products to be maintained, and the provision of services with these downstream products to many users. The optimum choice between on-the-fly processing of potentially the same data and permanent or at least temporary storage of redundant data obviously depends on the specific use case.
Xcube, a light-weight software package for generating xarray Data Cubes and offering tailorable downstream services, is addressing primarily the latter case. The Python software is open source and can be run in different environments from cloud infrastructures to desktop computers. It offers a rich set of configuration options for generating valid data cubes from data products, e.g. from Sentinel 2 or Sentinel 3 or model output, and can be extended for any spatial data set. Xcube uses the common data model from xarray, which has lent heavily from NetCDF, and stores the final data cubes preferably in the cloud-ready zarr format, although on-the-fly processing is also possible.
Xarray has become the quasi-standard for working with n-dimensional data sets for its capabilities in handling large data and its similarities to numpy and pandas, arguably two of the most relevant Python packages for natural scientists. For users, this means in many cases that they can readily work with large data cubes without having to become acquainted with new methods or to adapt their existing workflows.
Moreover, xcube-server offers catalogue services, a tile map service with OGC WMTS compatibility, and additional dataset services. Different clients like OpenLayers or Cesium haven been demonstrated to successfully work with xcube-server, but we are also developing xcube-viewer, which readily works with xcube-server with minimum configuration effort. Xcube-viewer offers image and time-series views, allows to pre-define regions and places, select polygons and compute on-the-fly aggregate statistics. Xcube-events is able to detect events in automatically updated Data Cubes and may inform users about it. By this means, customizable services can be implemented with reasonable effort.
Xcube is successfully used in different activities including the H2020 projects Data Cube Service for Copernicus and Cyanoalert and will also be part of ESA’s recently started EO Data Cube Facility Service. We will present here different applications of xcube and its corresponding services to demonstrate its capabilities and versatility for a large range of use cases and give an outlook to future features of the software.

In 2017 three reports published by famous investment banks predicted that the global space economy, currently valued at about € 300 billion, could reach in the most optimistic case a value of € 2.4 trillion by the 2040s. Parallelly to the space economy trend forecasts, it is more than likely that the number of new spacecrafts every year will continue to grow. The number of space segments launched indeed in the last decade averagely kept to increase at a rate of +17.5% per year. What should be added to the count for the next future is the New Space uptake and the new satellite constellations planned by private companies and agencies. Four private companies for example have announced the deployment in the next 5 years of at least 14’000 new satellites in Low-Earth orbit, forming several constellations able to provide global internet coverage. There are already six operational navigation satellites systems in orbit, and four of them have still to reach full completion. There are also four constellations in orbit for high-resolution Earth imaging, six for broadband communications, data relay and vehicles tracking, one for disaster monitoring.
Given the breeding ground, it is expected that the European space sector will follow the global trend in terms of satellites launches. ESA itself is facilitating the process preparing the new Ariane 6 and VEGA C rockets and financing the development of new micro launchers for small size payloads.

In this scenario, it is reasonable that the typical procedures of ground operations must necessarily encounter an evolution in order to support the leap in the number of on-orbit spacecrafts, especially in the case of those space segments that will be organized in large constellations and formations. The software tool named CASTeC – Context Aware Spacecraft Telemetry Checking, developed by S.A.T.E. and Planetek Group for ESA-ESOC, intends to ease the labour-intensive task of spacecraft telemetry checking, by automating the telemetry signals trend analysis and the novelty detection. It also unburdens the Flight Control Engineers from manually setting the alarm and warning thresholds over the thousands of telemetry parameters that are typically downloaded by spacecrafts, whose number, as previously discussed, is going to exponentially increase in the near future.
The telemetry checking performed by the CASTeC tool is based on Data Mining techniques, applying innovative Machine Learning methods exploiting the Big Data information made of S/C telemetries. The tool calculates a large number of statistical features over equally distributed time intervals for each telemetry parameter timeseries, both in the case of categorical and numerical signals. The values of the features are then compared to reference values that were extrapolated from telemetry dataset labelled nominal by the user. The reference set can be part of the data received from the real spacecrafts or can be generated by on-ground simulators. If the feature values deviate from the reference values above precise thresholds, that are automatically determined by a tool’s algorithm, the CASTeC software highlights the detection of a novelty or a trend anomaly in the analysed telemetry parameters. This permits the engineers of the ground operations team to obtain alarms and warnings that are based on smarter criteria than usual simple signal range check, and, in addition, with thresholds that are autonomously tailored by the software tool.
The other great advantage of CASTeC is that it works “context aware”. It means that the processed telemetry is subdivided into several sub-sets on the basis of user defined contexts. The contexts can be related for example to the spacecraft solar arrays angle with regard to the sunlight direction, or the condition of umbra and penumbra, the position of the spacecraft along its orbit, such as periapsis and apoapsis zones, or the position of the orbited planet around the Sun. Then the tool’s algorithms calculate the features in the telemetry sub-sets collected during the specific contexts and compare them with the reference part associated to the same context time intervals. In this manner, the telemetry checking is raised to a greater level of fidelity, comparing the spacecraft behaviours measured or simulated in equal conditions.
This potentiality can be particularly exploited in the case of the telemetry checking of entire constellations, because it allows to compare the states of twin spacecrafts when they enter the same orbit phase, have the same operational status, face the same environment conditions or pass over the same zones of the orbited celestial object.
Thanks to its autonomous threshold definition capabilities, the possibility of comparing long timeseries by Data Mining techniques and automatically detect the novelties, the advantage of analysing the telemetry by context intervals, CASTeC will make the telemetry checking operations effective and straightforward, reducing the costs needed to maintain large numbers of satellites while increasing spacecraft operations safety.

Our scientific understanding of Earth is based upon a combination of observational and numerical model data. Over the past twenty years, the volume of these data has increased dramatically, primarily driven by advances in satellite missions and numerical models, which now produce petabyte-scale datasets. Such large data volumes have presented new challenges and opportunities around scientific analysis and have highlighted the need for cloud-optimal data stewardship practices if the datasets are to be fully utilized. This is where organizations like the Earth Science Information Partners (ESIP) stand to play a critical role.

ESIP is a US-based nonprofit organization supported by NASA, NOAA, and USGS with a belief that the quality of life, economic opportunities, and stewardship of the planet are enhanced by regular use of scientifically sound Earth science information provided in a timely manner. ESIP members constitute many of the principal technologists charged with the stewardship of Earth science data across the public, private, and academic sectors. As such, they are also charged with updating their data systems to adapt to the current era of data- and computationally-intensive Earth science research.

While large data providers have the resources and expertise to experiment with data configuration in the cloud, other agencies, academic repositories, or individual are more-limited in their ability to develop and deploy new data systems in the cloud using best practices. This is the problem ESIP’s Lab is working to solve. Through small grant funding and community input, the ESIP (virtual) Lab supports projects working towards adoption of community-accepted best practices in scientific data management and analysis. Additionally, the ESIP Lab supports experimentation with emerging technology yet to be applied to an Earth science problem.

Most importantly, when approaching critical science and technology challenges, ESIP Lab projects work as part of an active community that provides rapid feedback and reuse opportunities, ultimately resulting in more impactful community-vetted technology. This talk will provide an overview of how small-grant funding and community input serve an important niche in the space of Earth science technology development.

Satellite Earth Observation is becoming more and more crucial these years, thanks to the increasing number of Spatial Missions and improved IT capacity of management and processing of EO-Images. In the European Context the new Sentinel missions are going to acquire petabytes of data in the next years that will be added to all the other operating and foreseen national and private missions.
The scientific community had accumulated in the last decades experience in processing EO Images with the aim to produce Value Added products useful for many fields, like hydrology, coastal or urban monitoring, geohazards mitigation or observation and so on. This trend is going to grow up and in the last period many industries started to experiment or also use some of these products in their own business.
In the near future is likely that this trend will continue to grow up, with new products, service and applications that will be based on EO-Data; to support these developments many IT challenges have been addressed and still have to be addressed.
IT tools are essential for these developments; once data arrive on the earth, it has to been stored, distributed, processed and delivered to the end-users. Many operating software and experimental projects are ongoing in this branch of the IT market.
The WASDI project lies in a complex environment of existing computer tools and services, and has been developed to work on the top of the DIAS platforms to facilitate the development and deploy of EO Data processors. As is well known, in the last years a new paradigm is driving the market: move processors to the data and not data to the processors. DIAS are cloud platforms that brings together processing capabilities and data proximity.
WASDI has been developed to become an online version of the desktop EO Image Elaboration Software like ENVI or SNAP that are ruling the market.
The concept can be compared to Google Drive: it took Microsoft Word and bring it online, directly on the Web, without the need of any desktop software. The same does WASDI: takes the features of SNAP or ENVI and brings it online. WASDI allows to manage the full EO Processor development cycle, starting from the basic features like searching the multi provider catalogue, explore images and apply pre-processing steps or mask or filters directly on the web. The key feature anyway is the possibility for EO Experts to develop their own processor and deploy it to the cloud with a simple drag and drop operation.
Nowadays, there is still a gap in the deploy of new EO processor to the market. The most available solution is the access to a Virtual Machine in a DIAS, that may have a different operating system, different tools or languages in respect to the one used by EO-Experts. Some solutions are promoting directly on line development, but this requires the choice of a specific language and the learning curve of a new development tool.
Expert Users, the actors that are able to elaborate an EO image to generate a Value-Added product, are usually expert in image manipulations, signal elaboration, classifications and segmentations and so on. They usually do not have a strong IT background, they are not generic developers. They usually use only one dedicated programming language, that nowadays seems to be usually IDL (the commercial language delivered by ESRI), Matlab or Python.
WASDI let the user develop his own processor at home, in his own environment with his own tools. The development of a new process is usually a long task, that needs time and debug, that are typically done on the developer PC.
Once the processor is tested and ready, the user should be able to drag and drop it in WASDI to let this processor been automatic deployed in the cloud. The technology is based on a smart multi language library (IDL, pyhon, Matlab, Ocave, Java) that basically simulates a virtual file system between the cloud and the developer PC, with very few and simple primitives. The user can develop using this library in his own environment, and the WASDI code takes care of accessing missing data from the cloud once is needed, during the development and testing phase. Once the processor is ready and deployed, the system recognizes that the software is running on the cloud and makes direct access to the input data in a transparent way.
The technology is based on the on-the-fly creation of a wrapped Docker Instance on the server, able to run the user-supplied code. The Docker is a template that, in reality, run an internal web-server on the container, that listen in an internal port that is mapped in the WASDI database. While an end-user wants to run one of these generic processors, WASDI public interface calls the internal web server on the known port of the right container that host the user-supplied code.
This feature is the key of the “Develop at home, deploy to the cloud” and has been used up today in real world use cases, letting EO Experts focusing on their own expertise more than on IT tasks.
WASDI is used by CIMA Foundation researchers to provide to Italian Civil Protection Operational Service for NRT flooded Area Detection, EO Based Root level soil moisture evaluation, Vegetation indexes for forest fire forecast models (www.mydewetra.org) .
WASDI is also the technological base of the eDRIFT project that is providing pre-operational EO Services developed by LIST, DELTARES and DLR for the Worldbank SEADRIF project, an IT platform able to estimate the number of people affected by a flood, by a combination of real-time models, EO data and ground observations in the South East ASIA area (www.seadrif.org) .

ο Building an ecosystem where every geospatial project can take shape, scale-up and succeed

sobloo is a cloud based solution hosted on a powerful and highly-reliable infrastructure designed to be a one-stop-shop for data users, developers, services providers & insights end-users. Through a secure cloud computing architecture, the sobloo platform provides a broad access to Copernicus Sentinel data, as well an extensive library of complementary data collections. Far beyond data, sobloo offers a wealth of value-added services to turn data into analytics, a wealth of tools & APIs and a full set of powerful GitLab features, along with a convenient desktop.

All users, public or private, are encouraged to take advantage of sobloo’s universal docking station to plug their own data, service, applications.

Building the best ecosystem possible to help developers turning data into invaluable insights is our core priority.


ο Accelerating geoprocessing operations & applications deployment thanks to Kubernetes (KS8)

The ability to process massive amounts of geospatial data, scaling-up to almost any workload that can be imagined across many machines is the biggest advantage of the cloud. It can become quickly challenging to maintain service’s performance and reliability while the number of users is rapidly growing. Furthermore, in today’s fast-evolving geospatial environment, software development cycles have been dramatically truncated.

This is where Kubernetes comes into play. Containerized applications that leverage the cloud’s extensible functionality are the best possible solution to seamlessly scale up geoprocessing operations and applications deployment.

In order to speed up project’s implementation and development, sobloo offers a state-of-the art deployment management system. The Kubernetes cluster manager is the most performant solution on the market for orchestration and management of containerized applications. It provides a container-centric management environment and allows the orchestration of computing, networking and storage infrastructure on behalf of user workload. Kubernetes allows the greatest possible simplification of containers’ management and deployment.

Our team can provide fully-configured ready-to-go Kubernetes clusters to allow every user easy and fast operation of the images Docker. Developers can spend less time in worrying about operating cloud infrastructure and more time in building performant customer-centric applications.

K8S empowers developers to launch their own containerized applications into a production-ready cluster without having the constraint to maintain the infrastructure, so they can run and scale workloads easily and quickly.

This high level framework addresses the main challenges of developers. Our team is always ready to respond to every user’s specific need with reliable & efficient solutions: they will provide you with the most stable release of K8S, combining highest level of code quality and guaranteed security.

Thanks to this effective computing environment, it becomes almost effortless to develop applications & new services on sobloo platform and to use sobloo marketplace to showcase this new offer and reach new customers.

Cloud computing offers new perspectives for the processing and reprocessing of Earth Observation data, allowing more flexibility for production and easier data access for users.
This new paradigm will change the way the quality control and performance monitoring of Earth Observation products is performed.
The purpose of this presentation is to propose an architecture for a cloud-based mission performance center and detail the expected benefits, illustrated by our recent background experiences.

A first component of the system is the data archive. For data quality and performance monitoring purposes, this archive shall comprise:
• A rolling archive of recent products, typically of 3 months, including user products and expert products.
• A long-term archive of calibration and validation products. Validation products may be stored as extraction over Areas-of-Interest
It is expected that this archive will be directly fed by the Payload Data processing system.

The second component of the system is processing resources needed by the mission performance center:
• Systematic processing for quality control purposes (generation of browser images, automatic quality control checks, data processing for performance monitoring) and QC automats;
• Ad-hoc processing for anomaly investigation, calibration purposes, or performance validation, with recourse to expert systems and artificial intelligence tools.

The system shall provide several types of interface for the experts and operators.
While an ftp interface remains a simple an efficient solution to retrieve a few products for ad-hoc analyses, virtual machines are useful to perform analyses with expert tools with a direct and fast access to a large number of products. This is critical for routine validation of products and for detailed analysis of products affected by an anomaly.

This facility is currently used by some experts of the Sentinel-3 Mission Performance Center (MPC).
For more demanding or systematic processing, a Web Processing Service (WPS) interface to access the processing resources can be an efficient solution.
ACRI-ST has developed a processor integration component to deploy a software on a remote processing system and wrap it as an asynchronous WPS service.
This component provides a simple graphical user interface as well as an API (useful to upload quickly a new version of the processing software with a script).
This component can be used by experts to install analysis software (e.g. computation of geometric performance by image correlation) and run it remotely via a manual request or as a cron job.

We will report on lessons learned and experience from the development, deployment and operation and use of these components in the framework of the Sentinel-3 MPC, using our experience of the operations of the Coastal Thematic Exploitation Platform (C-TEP), and of the data center of the Luxembourg Space Agency.

The Earth Observation (EO) industry was historically dominated by large satellites typically funded by governments and space agencies. The objectives of these missions were to produce the highest resolution imagery, in multiple spectral bands at a quality that goes beyond most application requirements. Therefore, the focus was on advanced, custom designed payloads.
In the current Space 4.0 era, a new trend is evolving. The focus is shifting towards the commoditization of remote sensing data where spatial resolution, revisit time and cost-to-company are becoming the driving parameters. This trend began in Silicon Valley and spread worldwide, arousing expectations.
These newcomers tend to position themselves in the lower end of the market and compete with the current incumbents, with disruptive business models. Betting on a large number of small and low-cost satellites with the best spatial resolution, sometimes going beyond what’s possible with physics and space mechanics. To stay competitive and differentiated in the EO market the new commercial players are revising their capital and operational expenditure models.
Simera Sense is addressing these challenges by:
• stretching the front aperture diameter to the limits of the 3U/6U and 12U CubeSat volume for the best possible spatial resolution what’s possible with physics;
• pushing the spectral and radiometric limits by using the latest CCD and CMOS technologies with on-chip filters and TDI technologies;
• incorporating advanced onboard processing with AI/ML capabilities to increase the efficient use of a limited download pipeline;
• providing relative and cross-sensor calibrated data to allow a longitudinal comparison of sensor and image products, especially when deploying a large number of small satellites;
• serial producing optical payloads to push down cost.
The xScape100 and xScape200 series of payloads for respectively 3U/6U and 12U class CubeSats are the most meaningful and tangible testimonial to the reality of this strategy. These standardized solutions are able to image the earth at a ground sampling distance (GSD) of respectively 4.7m and 1.8m.
This philosophy in developing optical payloads for the future generation of earth observation satellites do have the advantage of pushing down the total cost of ownership to satellite integrators, speed-up the time to space, while maintaining superior image quality. The optimized systems do result in more depth and spatial resolution to each pixel in multiple spectral bands. Consistent performance across a constellation is achieved by cross-sensor calibration. This results in actionable data that will enable the capacity to identify, classify and index each piece of land on the planet, daily.

In 2018, the snap2stamps open source package [1] was released in support to the science community needs for an interferometric processing tool for geohazards applications using Copernicus Sentinel-1 mission data. The snap2stamps package paved the way for the automatic DInSAR analysis using the ESA SentiNel Application Platform [2], while enabling advanced interferometric processing when combined with the Stanford Method for Persistent Scatterers (StaMPS) [3]. The utilization of these automatic tools is demonstrated in [4] and [5].

In the current work, we communicate the successive efforts to integrate the SNAP-StaMPS processor in a fully automatic way into the ESA Geohazards Exploitation Platform (GEP). The initial version of the service requires minimum interaction by users, while future implementations will allow definition of processing parameters for more tailored PSI results. The service shall be accessible via GEP for non-commercial research applications. The capabilities for wide-area processing using SNAP-StaMPS service shall be presented, in the frame of the Geohazards Lab initiative, promoting hosted processing and e-collaboration between the geohazards community.

[1] Delgado Blasco, J. M. and Foumelis, M. (2018, July 27). Automated SNAP Sentinel-1 DInSAR processing for StaMPS PSI with open source tools (Version 1.0.1). Zenodo. http://doi.org/10.5281/zenodo.1322353

[2] Foumelis, M.; Delgado Blasco, J. M.; Desnos, Y-L.; Engdahl, M.; Fernandez, D.; Veci, L.; Lu, J. and Wong, C. "ESA SNAP - StaMPS Integrated processing for Sentinel-1 Persistent Scatterer Interferometry". IEEE International Geoscience and Remote Sensing Symposium. IGARSS 2018.

[3] Hooper, A.; Bekaert, D.; Spaans, K. and Arikan, M. (2012). Recent advances in SAR interferometry time series analysis for measuring crustal deformation. Tectonophysics, 514-517, pp 1-13. Doi: 10.1016/j.tecto.2011.10.013

[4] Delgado Blasco, J.M.; Foumelis, M.; Stewart, C.; Hooper, A. Measuring Urban Subsidence in the Rome Metropolitan Area (Italy) with Sentinel-1 SNAP-StaMPS Persistent Scatterer Interferometry. Remote Sens.2019, 11, 129.

[5] Cian, F.; Delgado Blasco, J.M.; Carrera, L. Sentinel-1 for Monitoring Land Subsidence of Coastal Cities in Africa Using PSInSAR: A Methodology Based on the Integration of SNAP and StaMPS. Geosciences 2019, 9, 124.

The importance of vegetated land cover cannot be overestimated [1], especially in a context where climate change scenarios tend to range from bad to worse [2]. Vegetation provides a substantial contribution to carbon sequestration [3], and assessing the vegetation stock and its distribution is key to defining the balance and role of Ecosystem Services (ESs) in this context [4].

Although tropical rainforests represent the largest “carbon sink” available on our Planet, it is not only in the tropical band that the climate change game is played – temperate areas such as most of Europe also contain large patches of vegetation that do have a significant impact on carbon balance.

In Europe, sprawling urbanization is one among the strongest causes of loss in vegetated land cover [5]. Satellite-based monitoring of urban sprawl and inventorying of vegetated land cover can provide clues on urbanization and vegetation loss trends, in terms of both geographical distribution and speed of change. The open policy of Copernicus data greatly facilitates the implementation of large-scale monitoring actions, which is good news in terms of access to clues on vegetated land loss. There exist indeed Copernicus Land services providing data that can be used in this context [6], such as CORINE Land Cover, but their level of temporal resolution limits their usage as near-real-time (NRT) monitoring tools.

In our educational project we intend to test the possible use of pan-European, continuous assessment of vegetation cover using different vegetation indexes applied to Sentinel data as a proxy for a NRT vegetation loss monitoring service. Conceptual design and analysis will be carried out for a possible API-based system ingesting a flow of optical Copernicus space-borne data (S-2, S-3) generating a vegetation layer for the specific purpose of vegetation loss monitoring. Problems and opportunities will be discussed in the framework of an Earth observation educational exercise.

This work is being carried out as a group exercise within a Remote Sensing course at the University of Pavia, which has been selected as a new FabSpace under the H2020 “FabSpace 2.0” project [7] of the European Union, has recently accessed the status of Phi-Unet Lab under the ESA “Phi-Unet” project [8]. A “Space Communication and Sensing” graduate track is currently active within the Engineering Faculty, and the aim of these exercises is that of showing the benefits of Earth Observation and encouraging public involvement in space-borne monitoring of the terrestrial environment.

References

[1] Bounoua, L., DeFries, R., Collatz, G.J. et al. “Effects of Land Cover Conversion on Surface Climate. Climatic Change” (2002) 52: 29. https://doi.org/10.1023/A:1013051420309
[2] Richard H. Moss et al., “The next generation of scenarios for climate change research and assessment”, Nature volume 463, pages 747–756 (11 February 2010)
[3] Crossman ND, Bryan BA, Summers DM. 2011. Carbon payments and low-cost conservation. Conserv Biol. 25: 835–845.
[4] de Groot RS, Alkemade R, Braat L, Hein L, Willemen L. 2010. Challenges in integrating the concept of ecosystem services and values in landscape planning, management and decision making. Ecol Complex. 7:260–272.
[5] Maximilian V. Ehrlich, Christian A.L. Hilber, Olivier Schöni, “Institutional settings and urban sprawl: Evidence from Europe”, Journal of Housing Economics, Volume 42, 2018, Pages 4-18, ISSN 1051-1377, https://doi.org/10.1016/j.jhe.2017.12.002.
[6] Copernicus Land Pan-European information services. Online. Available at: https://land.copernicus.eu/pan-european
[7] FabSpace: An open-innovation network for geodata-based innovation – by leveraging Space data in particular in Universities 2.0. Online. Available at: https://www.fabspace.eu/
[8] “Phi-Unet - a university network powered by ESA”. Online. Available at: http://phi-unet.com/about/

The goal of the H2020 EOxposure project is to build tools to quantify the exposure of population and economic assets to multiple risks using novel information layers from current and future Earth Observation (EO) missions, as well as the growing sensor web on the ground. The project sits on three main pillars: housing condition mapping, disease spread proxies, and physical proxies to security threats. The following paragraphs provide some explanation about each pillar.

Housing condition mapping
To develop a tool for Housing Condition Mapping (HOCOM), based on open source software and able to exploit data from multiple EO sources, airborne/UAV platform and ancillary economic/social data sets.

Disease spread proxies
To develop a tool to extract maps of disease spread proxies (DISESP), based on open source software and able to exploit data from socio/economic data sets and geospatial information, assisted by additional information from multiple EO source, airborne/UAV platforms and ground sensors.

Physical proxies to security threats
To develop a tool to extract maps of physical quantities (PHYMAP) by combining and exploiting at the most accurate spatial scale observations by multiple sources, including spaceborne/airborne/UAV platforms.

The presentation will provide insights on the status of the projects and its future developments.

References

Recently, the number of organic rice producers in Europe has grown at such a fast pace that some doubts started emerging about the likelihood of such an “avalanche” transition, and about the full compliance of all actors involved with the strict regulations involved in organic crop growing [1]. In this situation, a “satellite monitoring and assessment” service could become a powerful instrument in fostering “non-compromising” organic farms and in re-assuring consumers. In this context, we conceived an innovative, Earth Observation (EO)-based service supporting the organic certification process by leveraging Copernicus data, that we called “Vialone” from a well-known traditional variety of rice. This resulted in a proposal submitted to a competitive European Space Agency (ESA) call for innovative EO applications, under the Kick-Start Activity (KSA) scheme [2]. The proposal was carefully assessed, and finally selected for funding under ESA Contract No. 4000124436/18/NL/NR; the project has just closed at the time of writing (April 2019) and we are now seeking funding for the next stages of this activity.

Thanks to the feedback collected during the KSA project, we recorded a great interest by different actors in the organic rice supply chain about the possibility to reinforce the current organic certification process, which sometimes may experience credibility issues, by exploiting spaceborne monitoring. On the technical side, we have developed a series of tools to extract information relevant to certification from time series of Sentinel and in-situ data.

Vialone can therefore play a fundamental role in the certification process: thanks to the satellite technology exploited by our platform, it is possible to collect data and generate useful information, which can also assist certifier institutions in certifying the organic product as such. In this way, some uncertainties due to “blind spots” in the certification process may be removed by providing third-party evidence. In the full paper, we will elaborate on this concept and present some of the tools developed during our project.

References

[1] Europol and the European Union Intellectual Property Office: “2017 Situation Report on Counterfeiting and Piracy in the European Union”. [online] Available at: https://www.europol.europa.eu/publications-documents/2017-situation-report-counterfeiting-and-piracy-in-european-union
[2] European Space Agency Kick-Start Activity funding scheme. [online] Available at: https://business.esa.int/news/kick-start-activities-new-funding-opportunity-for-innovative-applications-ideas

We present work done in collaboration between University of Bari and Planetek Italia, aimed at developing a novel imaging instrument, called the “space ghost imager” that exploits the intrinsic spatio-temporal correlations of astronomical light sources to bring the quantum advantage in space imaging.
This novel instrument differs from standard imaging systems in the way images are retrieved. Standard images are obtained from direct measurement of the intensity of light coming from the object of interest. Here, instead, we first register optical intensities on two disjoint sensors, and then measure correlations between them. Correlation imaging has its roots in the intensity stellar interferometer proposed and developed in the sixties by Hanbury Brown and Twiss, used to measure the angular size of stars.
Thanks to the quantum correlation properties of light, no high-resolution detector is required in the vicinity of the object of interest: instead, a single-pixel sensor, able to collect light transmitted, reflected, diffracted, or scattered by the object is sufficient for the task. Spatial resolution and high image quality are recovered by correlating the signal retrieved by such single-pixel detector with the signal captured by a high-resolution sensor that directly collects light from the source (an effect known as “ghost imaging”). Interestingly, such high-resolution sensor does not need to be in space, but can safely be a telescope on Earth.
Besides the ability to achieve high-resolution imaging of objects that are hard to access with direct imaging systems, state-of the-art research [see D. V. Strekalov et al., “Ghost imaging of space objects”, NIAC final report 2014 & J. Phys. Conf. Ser. 414, 012037 (2013)] indicates the following advantages of correlation imaging: 1) The possibility to distinguish opaque absorbers from lens-like phase and amplitude objects. 2) The increase of SNR towards the typical values of direct intensity measurement for sources with high spectral brightness. 3) An improved handling of detrimental effects such as stray light, detector dark noise, natural variation of source brightness, variation of detector responsivity and other effects that are usually omitted in SNR analyses.A lab demonstrator is being studied, built and tested to demonstrate the capability of the proposed space ghost imager to retrieve the geometry and position of opaque space objects such as exoplanets and earth-like planets, including those near bright stars. This device is expected to enable the discrimination between opaque objects from lens-like phase and amplitude objects, such as gravitational lenses, dust and gas clouds, which is a hard task for direct intensity measurements, since they can produce similar shadows.
An imaging device characterized by such features implies that the processing capabilities of On Board Computers will play a more and more important role in the satellite value chain. Among the high-performance architectures available in the Ground Segment scenario, the GPU has been recognized as one of the most promising being it designed to apply repetitive computations to a large amount of data.
Thus, one can observe that reducing the time required for acquisition, data transfer and data processing becomes crucial to make correlation imaging competitive with respect to standard imaging techniques. We study algorithms for optimizing SNR and acquisition speed through compressive sensing and machine learning techniques, optimize measurement algorithms for image reconstruction, and work toward speeding up data elaboration through high speed parallel computing (i.e. GPU).
Nevertheless, GPU usage has not yet been adopted on-board, due to the specific constraints of the space environment and of space systems and we can only rely on similar processing boards for the embedded systems world. In our case the speedup and the efficiency of the proposed pipeline for correlations exploitation is being evaluated both on a workstation and on an embedded GPU platform.
Results discussed in this paper are strengthen by a MIUR funded project (on-going) named “Close to Earth (CLOSE)”, in collaboration with the Distretto Tecnologico Aerospaziale Scarl (DTA).

The project CRUISE will face the European need for higher cyber security of Remotely Piloted Aircraft Systems (RPAS).
RPAS, during aerial operation, are totally dependent on their software systems and on inputs received from avionic sensors (GNSS receiver, SATCOM antenna, …). Payload sensors capture important and, potentially, private EO data, in example data about position of persons in hazard situation, so trustfulness of such data is mandatory for the end-user. Similarly, Very Low Level aerial operations can be associated with private data so that operator has to protect against attacks. Cyber attacks can be run by directly accessing the system (either the pilot station or the aerial platform) or can be launched by jamming or spoofing the datalink or the GNSS signal in the operation area resulting in a serious risks for the RPAS, the operation and persons and infrastructure flied over.
Project partners will design, develop and validate a CyberSec Test Range to provide RPAS cyber-vulnerability and resiliency assessment services to a wide EU community of research and industrial users. The CyberSec Test Range, that will be based in the airport Marcello Arlotta of Grottaglie (Taranto, Italy), has a technological architecture composed of an ICT infrastructure integrated with Earth Observation (EO), Navigation (SATNAV) and Communication (SATCOM) satellite platforms. Cyber attacks to Remote Pilot Station as well as to the aerial platforms launched during VLOS, RLOS and BRLOS operations exploiting communication (TELCOM, SATCOM) services and navigation (SATNAV) signals will be object of the assessment. Vulnerability tests will be realized by simulating cyber attacks — RPAS software will be in-the-loop — (in an aerial operation simulation environment), by emulating flight operations — RPAS module (hardware) in-the-loop — (integrating simulation environment and drones) and by realizing real flight tests exploiting the segregated airspace. Cybersecurity assessment will regard also payload data communication.
Satellite command and control datalink, exploited in BRLOS operations, will be realized by connecting the airport of Grottaglie with the Fucino Space Center, managed by Telespazio, partner of the project. Lastly, project consortium is also committed to create a connection between CyberSec Test Range and ESEC (European space Security and Education Centre of ESA, Redu, Belgium) with the objective to exploit ESA high level space cyber security infrastructures, knowledge and competence in the aviation and UAVs technology domain.

PROJECT context
CRUISE project is the first tangible step of the collaboration agreement among DTA, Italian Space Agency –ASI, European Space Agency—ESA and Belgian Federal Science Policy Office – BELSPO. It was signed during 2018 Global Space Economy Workshop - Space Cybersecurity for Mobility and aims at developing collaborative actions in the technology field of space cybersecurity for transport exploiting ESEC and Grottaglie Airport Test Bed infrastructure.
Furthermore, project partners are all involved in a concrete effort to develop a wide set of solutions with the objective to facilitate RPAS insertion into unsegregated airspace to provide aerial EO services. In this framework, cyber security is pre-requisite for safe, secure and effective operations and trustful EO data.

Target markets
Thanks to this integrated infrastructure the CyberSec Test Range will offer vulnerability assessment services to all classes of RPAS (from mini/micro to HALE RPAS).
Primary market: UAS & Payload manufacturers, aerial service providers
Secondary target: CyberSecurity industry, local entity, law enforcement—aviation regulators included-, civil protection, other aerial services users.

Earth Observation is day by day increasing its role in providing valuable data for a number of social tasks, from emergency management, to territory assessment, to precision farming just to mention a few of them. In the last few years, space Earth Observation has been joined by aerial Earth Observation exploiting flexibility of Remotely Piloted Aircraft System (RPAS), that are, indeed, able to overcome some of the constraints of satellite observation. RPAS can provide continuous observation up to 24 hours (or more), reach areas inaccessible to satellite due to natural or manmade obstacles, observe vertical infrastructures, and provide high resolution images. At the same time, RPAS operations are very very constrained by global flight regulations.
DTA has taken the lead to design, develop and exploit a strategic Industrial Research Infrastructure, the Grottaglie Airport Test Bed, to support development, test, experiment and demonstrate RPASs of every classes, advanced ATM, UTM as well as innovative EO applications and services.
The research infrastructure is composed of an aerial simulation system where flight tests and missions can be simulated and validate, of ground sensors through which air space is surveilled during flight test and of the Grottaglie Airport and the connected segregable airspace.
The simulation environment enable the design and simulate RPAS flight operations in a virtual 3D scenario composed of elevated ground areas and of ground and air traffic. Simulation can be live, virtual or constructive, so that, in ‘live simulation’ case, RPAS pilot, payload operators and ATCO (air traffic controlling operator) can interact and update operation accordingly. Validation of the simulated operation will take in consideration also results achieved (quality of data, reaching of the objective, …) in the EO mission.
Grottaglie Airport Test Bed will provide product test services to RPAS manufacturers and operators, to avionic module and related software developers, to aerial EO service providers and to EO end users. Services will support operators in experimenting and demonstrating the capacity of their solutions in a real scenarios thanks to the availability of a technological infrastructure for safe and secure aerial experimental operation and for aerial data gathering, processing and post-processing, of the airport facility, of segregated air space and of space services (SATCOM and EO). Grottaglie Airport Test Bed will provide access to SATCOM (in the case of BRLOS) through a secure connection with the Fucino Space Center, knowledge about quality of GNSS signal through an accurate signal assessment that will be realized soon, and space EO data through the collaboration with regional operators (Planetek Italia, Università di Bari, ReCaS, CNR).
DTA, as cluster organization, is realizing the Grottaglie Airport Test Bed in collaboration with a number of industrial and academic players with the objective to have a wide pool of knowledge, competences and technologies and to catalyse the creation of a living lab.
Grottaglie Airport Test Bed is an industrial Research Infrastructure and is expected to impact on innovation capacity of European aerospace system. It is nurturing multidisciplinary innovation be creating collaboration opportunities among space technologies developers, aviation community and academic data analysis experiences, and will contribute to reduce time to market and costs of product development.
Results discussed in this paper are related to the following projects: RPASinAir (on going), funded by MIUR, CRUISE (on going), funded by ESA, ECARO (in negotiation phase), funded by GSA and SAPERE (in negotiation phase), funded by Regione Puglia.

The reducing costs of satellite data, with the continuous availability of free and open Copernicus data, in conjunction with the power of cloud platforms, provides an unprecedented opportunity to increase the strength of different applications for infrastructures and roads monitoring.
To exploit the value of the constantly increasing availability of satellite data collected on weekly and even daily basis on Earth, Planetek Italia created Rheticus®, a cloud-based platform which provides tools for the automatic selection, processing and delivery of ready-to-use geo-information services, thus enabling the integration of vertical applications for different markets.
Thanks to Rheticus ®, users worldwide avoid a number of potential problems, such as the lack of expertise in data management. They subscribe to on-line services which continuously deliver information in the form of maps, reports, alerts and geospatial indices, which are dynamically displayed through an intuitive online dashboard, on mobile applications or in a machine-to-machine (M2M) way, directly in users’ applications.
Rheticus® represents for Planetek a disruptive innovation as it enables the transition from the traditional "project-based" to brand new "subscription-based" services, deeply transforming the traditional business model which is actually now focused on the Info@aS, Information-as-a-Service. Rheticus ® continuously processes satellite data and creates geospatial analytics; users subscribe to information content accessible through intuitive web interfaces, with dynamic indicators and automatic reports, which provides the actionable knowledge they need.
Rheticus® also delivers geo-content information via APIs. This way the continuous flow of information can be consumed, through the connection to Rheticus® APIs, by third-party software or through different installations of client applications. That represents an unprecedented opportunity for the foreign partners belonging to Rheticus' international resellers’ network, who can distribute, share and integrate Rheticus applications in their own industries. At the same time this is a tremendous boost for the internationalization process of Planetek Italia, enabled by the exploitation of Copernicus data and services available worldwide.
The current application field covered by Rheticus® are territorial changes, urban dynamics and land use changes, ground displacements (landslide and subsidence), infrastructure stability, new infrastructure and construction areas, wildfire burned areas, coastal sea waters quality and aquaculture.

CloudScout is a project financed by the European Space Agency and run by a European consortium led by cosine Remote Sensing, participated by University of Pisa, Sinergise and Ubotica technologies. The project targets the development, and for the first time ever the in-orbit demonstration, of an Artificial Intelligence based algorithm for cloud detection.
The inference is applied on images acquired by HyperScout-2, a very compact and powerful hyperspectral/thermal imager which embarks an unprecedented processing capabilities.
The specification of HyperScout-2, with its very wide FOV (31deg) and small GSD (75 m in the VNIR and 390 m in TIR), as well as the large number of spectral bands (45 in VNIR and 4 in TIR), provides very powerful capabilities for a large number of applications. However, its management at system/mission level turns to be a challenging task, especially for what concerns the data management and download. For an effective use of the instrument potentiality the information processed from the Tera Bytes of data generated needs to reach the final user in a reasonable time. This pushes the mission to the edge of feasibility, being a very challenging task even for bigger classes of satellites.
In order to maximize the quality of the information included in the downloaded data, HyperScout-2 uses its on board processing capability without relying on platform subsystems. HyperScout-2 is in fact equipped with an hybrid processing platform composed of FPGA, CPU, GPU and VPU. The VPU is a state of art Vision Processing Unit developed by Intel and will be flown in space for the first time as part of HyperScout-2. Thus, HyperScout-2 will enable experimental programs to investigate the use of Artificial Intelligence (AI) for a variety of applications in the field of object detection and data inference.
This paper presents the first application that will fly on the ESA mission PhiSat-1 / FSSCAT, part of the ESA EOP initiative to leverage small satellites to foster technology breakthrough developments.

Access premium satellite data from our multi-resolution optical archive, updated on a daily basis. Our Living Library currently proposes nearly all Pléiades images (50 cm) collected since 2016 over North America, Europe, Asia, Australia and Middle East, and since 2013 over 3,000 metropolitan areas in the world as well as selected SPOT images (1.5 m) available on a global scale.

The Autorità di Bacino Distrettuale delle Alpi Orientali is the office responsible for the implementation of the flood directive (2007/60/EC) into the Eastern Alps Water Basin District that cover over 40’000 sqkm of territory divided between several states (most of the territory belong to Italy but our competence is also on a part of Slovenia (Isonzo tributary basin and Timavo river basin), Austria, Swiss).
The office possesses a relevant expertise in hydraulic and hydrology; in 2012 we develop an innovative and accurate hydrological model coupled with an hydraulic module to produce the efficient early warning system AMICO on Bacchiglione river basin (the area of Vicenza, in the North-East of Italy).
Since we would use the most advanced technologies to provide the best prediction available, we are going to improve our early warning system and in general the hydrological model with a specific module to assimilate data from satellite and in general from space.
The EOPEN project, of which the Office is a partner, is financed under the H2020 program, provides an innovative approach to flood monitoring with the development of an intuitive platform with several dedicated services, able to elaborate satellite data and merge them with hydraulic models and information from social media (concept of Citizen Observatory and Citizen Science). The objective of this project is to provide a unique instrument for decision makers for better understand the real time situation and act properly during emergency. This merge will ensure an empowerment of forecasting system that will be more precise and accurate.
The last intense event in our territory the “VAIA” storm was important to test the effects of those new types of data. In particular we produce several elaborations from different products starting from forecast to asses the meteorological situation, to the EUMESAT and COPERNICUS products to see the “effects” of this event on the ground and the real movement of the storm with also the flood forecast of our early warning system.
This event was an important test also for the use of non EO data during the emergency; in fact, with the analysis of social network and social media in general was detect a flood not mapped for example with emergency services such as Copernicus EMS.
The EOPEN platform become very important due to the possibility to use a scalar environment to elaborate huge quantity of data without the necessity to invest into the computing infrastructure.
The impact is very relevant for our task with a reduction in terms of computational time for hydraulic model ensuring the availability of results in advice.
This event, finally, underlines unequivocally the importance of satellite data, provided with the platform, to support the hydrological reconstruction of the storm and also in general to support the entire system with distributed data that can complete the ground sensor net to provide precise, trustable and accurate flood forecasts.

Nowadays, we observe a rise in publicly accessible Earth Observation (EO) data leading to a greater number of EO data providers, each potentially having a different data access policy. This difference is visible at various levels: in the data discovery (CSW, custom OpenSearch, etc.), in the product access and storage (object storage, filesystem, etc.), in the authentication mechanisms (OAUTH, JWT, BasicAuth, etc.). These technological differences add knowledge overhead over end users and application developers wishing to take advantage of these data, causing low data usage. EODAG was designed to solve this problem.

EODAG is a command line tool and a Python SDK for searching, aggregating and downloading remote sensed images while offering a unified API for data access regardless of the data provider. The EODAG SDK provides the following features:
- List product types: list the supported products types and their description
- Search product types: search the best product types according to its characteristics
- Search products: search products according to the search criteria provided
- Store/Load results: (de)serialize the results of a search, allowing the search and download to be unravelled so that they can be used on distributed computing resources.
- Download products : download products in the form they are provided by the providers
- Get product pixels: directly access products’ pixels with band selection, re-projection, cropping, resampling, etc. The data are made available as a labelled multi-dimensional array data structure (more specifically, it is the DataArray structure of the xarray Python module) that can be exported in the NetCDF scientific data format. By now, due to some limitations in providers’ data access policies, EODAG first download the product if necessary before retrieving the pixels. In the future, it will directly stream these pixels from the provider to the user.

The architecture of EODAG is plugin oriented, with a very simple API, making it easily extensible and able to integrate new data providers. The plugin system has two levels: the first level provides an abstract interface to EODAG functionalities (called topic plugins). The second level is the implementation of one or more of the first level plugins for a provider.

EODAG exposes its functionalities through a REST API implementing OpenSearch Geo & Time extensions and OGC OpenSearch Extension for Earth Observation standards. Therefore, it can be integrated into different platforms, and used in different programming languages. To validate this API, a JupyterLab extension has successfully made available EODAG search functionality in Jupyter Notebooks.

As a result of its architecture, EODAG is almost as fast as its linked providers for search. It supports all operational DIAS platforms, the Scihub, and many more. It is now being integrated with the Opendatacube project, for easily including EO data into a Datacube, or generating Datacubes on the fly, as shown in the attached figure.
And last but not least: it is Open Source (Apache 2.0).

Geoinformation, location-based services and satellite Earth Observation science and business are playing a central role in the present and future of Digital Economy, in the twofold role of data producer and information provider. Technological investments, consistent political agendas and promising results have concurred to raise significantly the level of expectations that institutions, commercial business and scientific research have towards the Geoinformation sector to provide reliable, trusted, timely and valuable Information Products in multiple domains.

In this context, geoinformation driven services are blooming and growing in several market sectors, consolidating a trend already clear in the past decade and paving the road for the operational uptake into an increasing number of private and public business processes. National and European Space Programs such as, for example, Copernicus are closely looking at the evolution of the so-called Space Economy to take advantage of sectorial and cross-cutting opportunities to demonstrate the value of geoinformation driven services in wider contexts, ranging from Precision Farming to Fintech, from insurance to Retail and Property Management. Copernicus 2.0 is an evident evolution of the European Earth Observation flagship program in this direction, aiming at addressing needs and pain points of end users downstream in the information value chain.

Public and private investments in space technologies and assets are rapidly changing the scene of commercially available Earth Observation space capabilities. In fact, the high number of space programs announced in the previous years Is now turning into operational assets that are progressively unveiling their potential for multi-daily, high resolution and multi-sensor imaging opportunities, enabling unprecedented CONOPS through the joint exploitation of several space assets federated at different levels (space segment, ground segment, end user segment). Ideally, it is (or it will be shortly) possible to design geoinformation driven services where the access and tasking of satellite acquisitions can be fully driven by AI algorithms, based on complex rules and enabled by automated machine-to-machine interfaces with multiple Ground Segments. For example, object detection based on SAR data shortly after confirmed though object classification and recognition performed on the basis of fresh Optical data, whose tasking was automatically determined and submitted driven by object detection results from the first SAR image.

Paradigm shift from pixel to information flow and reports is a fundamental trend in this evolution. The quantity of data available from space and generated by new systems is humongous, We need to invest in new algorithms that can manage all them to extract useful information vital for many aspects of our life on the Earth to allow right access to the information in the shortest time possible from space multi sensor multi platforms generated geo spatial data. Through advanced algorithms based on deep/machine learning and AI elements information flow for a broad range of domains of interest will be delivered. Business intelligence, agriculture, infrastructure monitoring, emergency services, image intelligence and many services in support to economic activities and the UN Sustainable development Goals.

This is leading today to the power of federation of assets, stressing the importance of complementarity of observation capabilities and the central point of possible joint CONOPS to deliver a new class of geoinformation driven services opening up opportunities in currently underexploited vertical markets as well as in traditional and more consolidated ones.

The AWS Public Datasets program has proven to be an effective way to support the scaling of big data processing in the cloud. Processing can be performed close to the data by using services deployed in the same region as the data, thereby minimizing data access times. It is even better for many of the 'Earth on AWS' geospatial datasets where the data has been converted into cloud optimized formats that efficient data access by allowing users to read subsets of the data, such as a small region that is included in a much larger datafile. In some cases this can be an order of magnitude faster than traditional methods where a lot of extra needed is needlessly downloaded due to how it is packaged and distributed.

While the availability of this data has allowed for the efficient access of the data, there has generally not been a great way to search the datasets, as there is no central catalog of the datasets, or any type of standardized metadata. For geospatial data, where most users are only interested in specific regions, it can be challenging to determine which datafiles are needed as it varies by how each dataset is structured and stored.

The SpatioTemporal Asset Catalog specification is a new way to catalog geospatial data from a variety of difference sources (EO, SAR, LiDAR), and includes a specification for serving that data in an API. Element 84 is working on indexing and creating a public facing API of the Earth on AWS datasets that can be used for search and discovery, thereby making data access easier.

This talk will include an introduction of the STAC specification and how it is used to describe the most popular collections of earth data available on AWS. We will give an overview of which datasets are available through the API and examples of the fields available for the different types. We will show how to use the API to make geospatial and temporal queries of specific collections of data. Finally, a summary of STAC related open-source tools will be given that support leveraging the STAC API from within your own applications.

GRASP is one of the best algorithms in its field: Retrieving aerosol and surface properties from satellite input data. The challenging problem that GRASP solves requires a big codebase. It is therefore a bit cumbersome to install and for the processings you’re going to need a lot of storage as well as compute power:
Even though the algorithm is much faster than a few years ago, it still takes hours, days or even weeks for a complete processing pipeline - depending on the input size.

This can be fine if you only have a narrow dataset. As soon as you want to scale your processing to include a multi-year dataset and verify new findings, things don’t look so good anymore.

But first you need to get the environment in place to start processing.

The GRASP Cloud project is funded by the Austrian FFG within the “Beyond Europe” programme. Its partners span over the world from Japan (Kindai university), China (RADI), Europe (Catalysts, GRASP-SAS, EODC) and America (Airphoton, NASA/JPL, NASA/GSFC). The project scope includes the claim to deliver a user-friendly approach to process satellite data in your browser. To do so we leverage various open source solutions and combine them. This allows everybody to just get going with their most recent research ideas and not to care about the IT environment.

Three main user groups and interests have been identified:
1) Scientific researchers want to apply GRASP to already supported instruments and check for certain properties on a limited dataset. They need support in setting up the processing environment.
2) Scientific researchers want to analyze the existing GRASP dataset: Compare it to other algorithms and retrievals, compare different versions by plotting the data or using aeronet stations as ground truth data for correlation.
3) GRASP contributors want to include a new instrument to the GRASP codebase, making the algorithm more versatile. They need a collaboration tool to work together with the GRASP core dev team.

The current environment has been proven to be very helpful in various other related projects.
Another bonus: As soon as the prototyping in a small scale is proven successful, the environment can be directly transferred to a cluster and used on broader input data set.

In this talk you’re given an overview of how we connect Keycloak (user management, SSO), OpenShift (resource allocation, scaling), JupyterHub (providing the GRASP runtime environment) and Ceph (storage). Also an example notebook is shown including the necessary preprocessing steps, calling GRASP from Python and plotting some results.

Research infrastructures (RIs) and in situ data availability in Svalbard are spatially unevenly distributed, which imposes a significant gap invalidating remote sensing products. To address these challenges and gaps, our presentation mainly focuses on Earth Observation (EO) and remote e sensing (RS) activities of the Svalbard Integrated Arctic Earth Observing System (SIOS), an international collaboration of 25 institutes from 10 nations. This presentation will highlight efforts from SIOS members in the operationalization of remote sensing services. Most of the geospatial products developed by SIOS members are derived from the Copernicus Sentinels. The presentation will focus on two applications of Sentinels: understanding permafrost movement and estimation of glacier velocity over Svalbard using interferometric synthetic aperture radar (InSAR). Both of these applications are of wider interest for the remote sensing community in Svalbard to address broad Earth system science questions. (1) The observed warming of Svalbard causes thawing of permafrost, resulting in a deeper active layer. Extensive permafrost thawing will likely lead to increased risks of landslides and a potential hazard to the infrastructure in Svalbard. (2) Svalbard-wide glacier velocity products are essential for understanding the response of glaciers to the warming climate. Permafrost movement and glacier velocity derivation include extensive remote sensing applications supported by ground validation by field scientists. Therefore, SIOS is planning to make these remote sensing products available for the wider research community by collaborating with members from different nations. SIOS remote sensing services emphasize the mutual benefits to field scientists and satellite owners. International cooperation such as SIOS can provide a broader picture of Svalbard by optimizing the integration of satellite data with in situ data provided by RIs distributed in Svalbard. The limited funding available to individual institutions to conduct dedicated cal/val studies in Svalbard highlights the need for an international collaboration like SIOS, which optimizes the use of existing infrastructure and data.

Geospatial Assessments of Climate Change Impacts and Community Resilience on Coastal Degradation in West-Africa

Jejelola 1* , Akande2*, Olajire3*, Apenuwa4*

1*School of Environmental Technology, Federal University of Technology, Akure, Ondo State, Nigeria
ofjejelola097498@gmail.com

2*Department of Meteorology and Climate Sciences
Centre for Space Research and Applications, Federal University of Technology, Akure, Nigeria

3*African Regional Centre for Space Science and Technology Education-English
Obafemi Awolowo University, Ile-Ife, Osun State, Nigeria

4*Department of Urban and Regional Planning, School of Environmental Technology
Federal University of Technology, Akure, Nigeria


ABSTRACT
Climate change environmental degradation had affected the world’s natural resources, ecosystem services, socio-economy values and societies to a large extent. However, considerable differences in the range and intensity of these impacts between regions provide insight into cumulative threats to marine biodiversity. Human-induced risks along the coastal zone have lead to coastal degradation from marine pollutants, anthropogenic activities such as untreated discharges of municipal, industrial liquid and solid waste, depletion of marine species, and loss of marine habitat. The research adopted remote sensing techniques with data capabilities from ground-based meteorological station (NIMET), Landsat OLI, and Sentinel-2A satellite imageries, for monitoring the environmental impacts of marine pollutants and wastewater discharge along the Gulf-of-Guinea coast. The high-precision ALOS DSM and TANDEM-X satellite data were used to model the coastal terrain and the slope. The satellite data were processed using a web-based GIS tool known as Coastal Environmental Risk Index (CERI) was used to estimate the level of environmental damage and coastal degradation over the study area. Further geostatistical analysis were performed using the cellular automata and Support Vector Machines (SVM) methods. The results of the research suggested adaptation and protection measures through integrated management of land-ocean interaction in the coastal zone, enhancement of integrated global observation system, and coastal ecosystem-based management. Lastly, adequate information tools should be made available to policy makers and environmentalists to identify the most severe and pressing risks, and to assess and implement the most effective prevention and adaptation measures in reducing pollution of the seas.
Keywords : Ecosystem, Coastline, Degradation, Adaptation

'CREODIAS - users and data - lessons learned '
Based on our 5-year long experience of operating the Earth Observation data platform – a testbed project EO IPT (EO Cloud) and of a DIAS (CREODIAS.eu) - we propose to present our lessons learned about the users, their expectations and possible trends.
In particular we want to focus on the following areas:
 Earth Observation (EO) data
o The most often used data (satellite, sensor, geographic coverage, time coverage – recent vs historic data)
o The most popular data formats and related patterns of usage
o Statistics on how are the data usually queried, accessed and processed. What interfaces are used (download vs S3 object access vs NFS type file access vs OGC WMS).
o Automation of the search and usage via API
 “Old” physical servers vs virtual, and the move to cloud
o How difficult is it to move from the ‘desktop’ (or local “under the desk” server) processing of the downloaded data to the cloud processing? What are the advantages of the cloud model?
o What are the most common difficulties encountered by the users when moving to cloud. Best practices when changing your software. What are the user expectations and how support could help.
 The most demanded functionalities – present and emerging user’s expectations
o ‘Turnover’ of the technology (new functions are expected faster than in the traditional space industry), what is a standard minimal set of services (VMs, Storage, virtual appliances…)
o Most demanded/used and emerging new functionalities such as Kubernetes, F/GaaS (Function/Product Generation as a Service), GPU’s, Jupyter Notebook, quasi private solutions: dedicated server VMs, private hypervisors and other.
 Are EO data users different than “standard” cloud computing users? What are the additional support needs
 New model of EO data acquisition and processing chain
o More COTS approach
o Horizontal instead of vertical model – no vertical silos (one provider per one satellite: acquisition, raw data processing, dissemination, EO processing); horizontal segmentation (a provider for acquisition, a provider for raw data processing, a provider for data hub and dissemination – interfaced via open interfaces)
o Focus on the AI

CloudFerro team has almost 20 years of experience in IT and outsourcing business. The core business of the company is, from the day of its creation, providing services of public, private and hybrid cloud and Big Data repository technologies. We believe we can provide effective solutions and answers to most of the issues presented above.

River icing is typical for Poland and some regions of Europe in the winter season and can last for up to a few months. Satellite-based monitoring of the river ice is to support currently used in-situ techniques and water management authorities. Regular and continuous information on river sections affected by ice can support the planning of ice-breaking process, the prediction of floods and the validation of ongoing ice-breaking action.
River Ice Monitoring Service was created to provide user with up-to-date information about river icing in winter period for selected rivers in Poland in a form of dedicated interactive map portal allowing data review from the entire season and export the data as spatial layers and statistical report.
The cover of riverbeds is discriminated into four classes: water, fractured ice, ice cover and ice covered by snow, derived from HR Sentinel data, including both Sentinel-1A and Sentinel-1B microwave satellites. Application of Sentinel radar sensors allows to collect images, independently from weather conditions, except for intensive storms and snow falls, every 2-7 days, and process in near real time. The spatial resolution of Sentinel-1 imageries of 5x20 m in Interferometric Wide Swath mode gives the ability to effectively detect ice on rivers over 60 m wide. A fully autonomous service provides river ice coverage classification with 90% accuracy.
River Ice Monitoring Service operates in one of the regional water management authority in Poland from the winter season 2018/2019. It provides to the authorized user regular, spatially continuous information about ice events on rivers that meet the width criteria. The service enables generation of statistical report relating to 1 km of river sections through a web service, which are similar to the previously used reports.
The River Ice Monitoring Service was developed within the Earth Observation for Eastern Partnership project, financed and supported by ESA.

Information on the turbidity and optical properties of seawater is required for many different coastal applications. Examples include: protection of areas with high concentrations of fish farms (e.g. salmon farming) from algal blooms and jellyfish swarms, monitoring of dredging and dumping of dredge spoil at sea; planning and construction of wastewater treatment plants; long-term planning of port activities.

TechWorks Marine are developing a prototype web portal and app service CoastMADE (Monitoring and Assessing Dredging Environments), funded through the ESA Business Applications Kick-start Activities for the Atlantic area, combining turbidity information from satellite and in-situ data. This provides stakeholders with detailed real time and historical information on the status of relevant dredging locations (e.g. ports and harbours) and potential effects on nearby aquaculture sites. The combination of large scale monitoring via satellite and localised real time data from TechWorks Marine’s buoys provides an optimised decision support system, created via user-driven requirements gathered during the execution of this project and from our existing client base.

The service is designed for the stakeholders interested in supporting Atlantic economic development. These are a combination of industry and government organisations, including aquaculture companies, dredging companies, ports, environmental agencies and local and regional councils. Dredging and dumping at sea licences are regulated in Ireland by the Environmental Protection Agency (EPA), and they are one of the users engaged in this project.

In this talk we will present an overview of the project, including the associated end user requirements, software development and commercial feasibility of the service being developed.

In the past, infrastructure limitations mandated that if one wants to do efficient analysis on EO data, one needs to pre-process it to the common structure – create so called “DataCube”. This pre-processing usually involves re-gridding, re-projection and some more advanced steps such as atmospheric correction, orthorectification, etc. The step is costly, both from processing as well as storage perspective as it commonly required replication of data. One can leverage parallel processing in a way to eliminate additional storage, but overall costs remain. These limitations commonly result in reduced data coverage, either spatial (e.g. country instead of world) or temporal (only one year). Another disadvantage of this approach is inflexibility – whenever an improvement in the process is implemented, one has to re-process complete archive.

Nowadays, with cloud infrastructure rampant and cloud-optimized data formats appearing, the term “DataCube” is losing relevance. Or rather, it is shifting one layer down – to individual scenes. Cloud-optimized GeoTiff (COG) has all necessary elements of the data cube – internal tiling, overview layers, location data in the header, etc. Sentinel-2 native JP2000 contains a pyramid (based on wavelet transform) as well, and NetCDF/HDF format also. Match this fact with recent "standards" to process data to L2A on ground segment level (e.g. ESA) and having data available on the cloud (DIAS, AWS, GCP, etc.) and what one data analysts needs is a "data driver" rather than "DataCube". Due to different formats of the data, meta-data and other specifics, it is inconvenient for a random user to access it in an efficient manner. The task of the "DataCube operator" is therefore to adapt their data access engine to various data sources rather than to pre-process these data sources to match their processing engine.

A technical solution demonstrating above described approach is Sentinel Hub, which is around for a bit longer than two years now, more or less as long as Sentinel-2 is operational. It is used by hundreds of applications and tens of thousands of end-users, processing more than six million requests per day. Average response time is 0.7 seconds, which is more than fast enough for vast majority of users. In this time Sentinel Hub executes all the typical DataCube steps - find relevant scenes addressing user's request, read necessary data, re-project and re-grid to specified parameters, create a composite or statistical analysis, etc. Data access interfaces are compliant with commonly used OGC standards (WMS, WCS, WMTS, etc.) so that it is easy for users to integrate these in their existing environments.

We will present how Sentinel Hub is handling federation approach, running on various platforms in parallel (AWS, Mundi, CreoDIAS, ONDA), long time-series analysis, how annual cloudless mosaic can be created ad-hoc and how it is possible to engage data fusion to combine various data sources, e.g. performing on-the-fly orthorectification of Sentinel-1 GRD data.

Very important part of the service are its integration capabilities. Two SDKs have been created – one for web-apps and one for machine learning workflows. The latter one is especially important as it makes it more efficient to prepare the data in a form ready to be used by well-known ML frameworks such as TensorFlow, MX Net, LightGBM and others. Examples of large scale ML processing will be shown.

Special focus will be put on lessons learned, providing guidance on good and not as good examples of existing data structure, most commonly used functionality by end-users, limitations of existing standards and how to work around them as well as some thoughts on possible business models and how one benefits best with open-source approach.

Digitalisation is crucial in many aspects of our daily lives nowadays – and so is access to data. The concept of ‘Digital Earth’ (DE), as outlined in 1999 by the former US Vice-President Al Gore, foresees a “multi-resolution, three-dimensional representation of the planet that would make it possible to find, visualise and make sense of vast amounts of geo-referenced information on physical and social environments” (Gore, A , The Digital Earth: understanding our planet in the 21st century, Photogrammetric Engineering and Remote Sensing, 65 (5), 528).
The DE concept is quickly becoming reality, with a strong dynamic component provided by real time data, forecast and projections. The Copernicus programme provides a fundamental contribution to this concept. The challenge is now how to access and extract information from the decades of global and local environmental data generated by in-situ sensors, numerical models, satellites, and, more recently, by individuals. This implies a change in the data exploitation paradigm, moving towards a massive and continuous data analysis approach. One of the main issues to face is the inhomogeneity of data types, formats, geographic projections or, in one word, the variety of the data that describe our planet.
The Advanced geospatial Data Management platform (ADAM, https://adamplatform.eu) manages a large variety of geospatial data, with a large set of data reading features for raster, vector and point data. ADAM offers a web-based application interface for humans, a Jupyter notebook environment for experts who want to write code using all available data, and a set of APIs for developers to connect to external applications, such as GIS tools and mobile apps.
ADAM implements and provides datacubeless access and processing services, namely it exposes multi-dimensional (spatial, temporal, spectral …) subsetting capabilities as well as on-the-fly processing functions, so that the consumer (human or machine) gets only the piece of data wherever and whenever needed, avoiding transferring large amounts of useless bytes via the Internet or massive local processing.
Within this work we aim at providing an overview of the Dynamic Digital Earth created using the large availability of European-generated data, namely Copernicus data and Services (CAMS, C3S, EEA ground measurements, Sentinel 1, 2, 3, 5P data, downstream services), ESA data (various Climate Change Initiative data, Earth Explorer, heritage and third party missions), reference layers (e.g. the urban settlement layer by DLR), and the capability to connect to various data sources such as the European Datacube and Open Datacube.
Benefits of this multi-dimensional data structure will be demonstrated via thematic applications in the framework of Earth Observation for Sustainable development, health safeguard, cultural and natural heritage sites management, education, security applications.
Finally an overview of upcoming applications to optimise data exploitation through machine learning and artificial intelligence will be provided.

Images, simulations and animations derived from remotely-sensed data, not only allows us to better monitor and predict the causes and impacts of climate change, but they also play a key role in education. We present a case study from Rome, where animations and pictures from Earth Observation (EO) played a central role in supporting students to understand the complex phenomenon of climate change, take consciousness of the causes and impacts, and implement subsequent behavioural change towards more sustainable practices in their daily life. This study highlights the importance of addressing the education sector as end-user of the future of EO.

The case study is part of the Climate Lab project, which aims at raising awareness and educating high-school students on the theme of climate change. The project is implemented by volunteers of the Rome Hub of the Global Shapers Community, which is an initiative of the World Economic Forum and counts a growing number of 393 hubs in 158 countries. The project plays a complementary role in the school’s teaching programme, as the topic of climate change is often not strongly covered. The project consists in three lectures of two hours each, addressing three topics: 1) Causes and Impacts of Climate Change, 2) Climate Change and Waste Management 3) Climate Change and Energy. The lectures are followed by a visit to the European Space Research Institute (ESRIN), in order to give to students The series of lectures have the objective of providing to students an understanding of the complexity of climate change and the role of the students and civil society in mitigating its effects.

Animations and data showed through ESA and NASA websites were used as supporting material in the lectures in order to explain topics such as the temporal and spatial variation of greenhouse gases, sea surface temperature and ice sheets among others. The lectures confirmed that such EO data visualization play a unique role in helping students to get an immediate understanding and interest in the scale and mechanisms of global warming and climate change. The Climate Lab project is in its pilot phase and is planned to be replicated in other schools in Rome and around Italy in the second semester of 2019. Opportunities will be explored to further integrate EO material in the lectures, including hands-on exercises for students. This study highlights the importance of tailoring the information from EO in order to develop education material and of further designing accessible tools and games to be included in the teaching modules in Italian high-schools. The project contributes to the Sustainable Development Goals, with focus to SDG4 quality education, SDG12 responsible consumption and production and SDG13 climate action.

The ESA-NASA multi-Mission Algorithm and Analysis Platform (MAAP) is dedicated to the BIOMASS [1], NISAR [2] and GEDI [3] missions. This analysis platform will be a virtual open and collaborative environment. The main goal is to bring together data centres (Earth Observation and non-Earth Observation data), computing resources and hosted processing in order to better address the needs of scientists and federate the scientific community.

The MAAP will provide functions to access data and metadata from different sources such as Earth observation satellites data from science missions; visualisation functions to display the results of the system processing (trends, graphs, maps ...) and results of statistic and analysis tools; collaborative functions to share data, algorithms, ideas between the MAAP users; processing functions including development environments and an orchestration system allowing to create and run processing chains from official algorithms.

Currently, the MAAP is in its pilot phase. The architecture for the MAAP pilot foresees two independent elements, one developed by ESA, one developed by NASA, unified by a common user entry point. Both elements will be deployed on Cloud infrastructures. Interoperability between the elements is envisaged for data discovery, data access and identity and access management.

The ESA element architecture is based on technical solutions including: Microservices, Docker images, Kubernetes; Cloud-based virtual development environments (such as Jupyter or Eclipse CHE) for the MAAP algorithm developers; a framework to create, run and monitor chains of algorithms containerised as docker images. Interoperability between both ESA and NASA elements will be based on CMR (NASA Common Metadata Repository), services bases on OGC standards (such as WMS/WMTS, WCS and WPS) and secured with the OAUTH2 protocol.

This presentation focuses on the pilot platform and how interoperability between the NASA and ESA elements will be achieved. It also gives insight into the architecture of the ESA element and the technical implementation of this virtual environment. Finally, it will present the very first achievements and return of experience of the pilot platform.

REFERENCES
[1] T. Le Toan, S. Quegan, M. Davidson, H. Balzter, P. Paillou, K. Papathanassiou, S. Plummer, F. Rocca, S. Saatchi, H. Shugart and L. Ulander, “The BIOMASS Mission: Mapping global forest biomass to better understand the terrestrial carbon cycle”, Remote Sensing of Environment, Vol. 115, No. 11, pp. 2850-2860, June 2011.
[2] P.A. Rosen, S. Hensley, S. Shaffer, L. Veilleux, M. Chakraborty, T. Misra, R. Bhan, V. Raju Sagi and R. Satish, "The NASA-ISRO SAR mission - An international space partnership for science and societal benefit", IEEE Radar Conference (RadarCon), pp. 1610-1613, 10-15 May 2015.
[3] https://science.nasa.gov/missions/gedi

Italian Space Agency (ASI) invests more than 50% of its budget in the Earth Observation (EO) field, considering both National and ESA contribution. In the framework of Space Systems for EO, the COSMO-SkyMed (COnstellation of small Satellites for Mediterranean basin Observation) Program represents the main Italian investment. The program was funded by Italian Ministry of Research (MIUR) and Italian Ministry of Defence (MoD) and developed by the National Industry, with the involvement of national small and medium enterprises (SMEs).
The most important EO Italian space asset was designed to satisfy a wide variety of user needs in both civilian (institutional and commercial) and defense domains exploiting four satellites that host on board a Synthetic Aperture Radar (SAR) operating in X-band.
In this scenario e-GEOS (20% ASI, 80% Telespazio) is the company which has the exclusive rights to sell COSMO-SkyMed data to the commercial community.
To ensure the continuity of SAR data for the Earth Observation market, providing at the same time a significant improvement of performances and to consolidate and enlarge the Italian leadership on SAR observation from space keeping a strategic prevalence on an increasing market of high technology, a second generation of satellites and a significant Ground Segment upgrade have been designed and developed. This is known as COSMO Second Generation (CSG) program and it has been taking place since 2010 including two additional satellites, planned to be launched starting from Q3 2019. A smooth transition from first to second generation of satellites will be ensured through an integrated system accounting of both capabilities (CSK and CSG), meaning that COSMO-SkyMed services provided to the users will be granted as if a unique constellation is operating. More than 10 years of operative servicing constitute a great return of experiences and lesson learned that have been capitalised in defining the new CSG system architecture and operative concepts aimed at greatly improving efficiency and reliability of the on-demand services that will be provided to a large variety of users, in a complex context as a dual-use priority-based system, as COSMO system actually is.
This presentation aims to present an overview of the COSMO Second Generation scenario, detailing capabilities and improvements for the users community, including synergies with the current first generation.

Earth Observation (EO) satellite missions provide an unprecedented increase volume of data production leading to new EO mission’s challenges especially the ability to benefit from greater efficiency Big Data solutions.
To support the actors of the Earth observation world CS SI has developed innovative and powerful open source tools: SafeScale, an open source multicloud DevOps solution to seamlessly deploy cloud applications on any combination of cloud platforms and EODAG (Earth Observation Data Access Gateway), a Python SDK (Software Development Kit) for searching, aggregating and downloading remote sensed images using a unique API for any EO data sources.
SafeScale is a self-service Web Portal developed for on-demand deployment of versatile and scalable platforms adapted to distributed computing frameworks, offering simplified access to the Earth Observation Data and facilitating the creation and the management of data processing platforms. The innovative aspect of SafeScale platforms lies in their capacity to offer a combined usage of a large variety of frameworks and technologies (Big Data, HPC, Deep Learning) without having to manage resources allocation (Node, RAM, CPU, and GPU). SafeScale platforms resource management is centralized by Apache Mesos which guarantees a fair and efficient distribution of resources for all components hosted by the platform. This particularity enables SafeScale users to run concurrently services and compute loads of data without worrying about their partitioning over the nodes of the cluster and thus significantly accelerate the implementation of complex distributed services.
Such scalable cloud working environments have been successfully deployed in the frame of the Copernicus Reseach and User Support (RUS) Service and the BIg Data From Space – Reference Architecture Framework (BIDS RAF) projects identifying the overall economic benefits (including Cloud costs, development and adaptation of processing pipelines, maintenance effortsetc..) and offering a maximal reactivity & security of the platforms in terms of:
- Combined usage of a large variety of Big Data frameworks (and/or on-demand applications) based on optimized computing infrastructures
- Easily up-scale and down-scale to adapt to load fluctuations and therefore to optimize IT costs
- Secured services and user data (encryption, authentication and authorization, application log, security information and events) to protect data & services

Summary:

The Earth's atmosphere is defined as the environment in which we live. If there is no clear limit to it, we consider that beyond 100 km (Karman limit), the space begins. However, the nomenclature of the Earth's atmosphere is more complicated. Several layers describe the different physical properties encountered. For example, at each temperature gradient corresponds a layer of the atmosphere, starting with the troposphere on the ground, up to the thermosphere exceeding 600 km altitude. Similarly, the gravitational links separate the atmosphere into two layers, the vertical transport into three, and so on. We are interested in this study more particularly in the average atmosphere, between 15 and 100 km of altitude.
Nightglow infrared emission
In order to understand the interests of this study, we will introduce here the notion of infrared radiation called the nightglow, which can be translated by night glow. The system is then entirely devoted to a detailed description of the phenomenon. It is a radiation mainly emitted in the near infrared whose main contributor is the OH molecule. This, excited on vibrational states following a reaction between hydrogen and ozone, will return to its stable state by emitting a photon, the airglow.

Dynamic at high altitude

Specialists in the upper atmosphere are also strongly interested in observables that can be measured through this radiation [Khomich et al, 2008], such as gravity waves, Rossby waves, or sudden stratospheric warming (SSW). . There are many methods for determining the atmospheric temperature. For tropospheric altitudes, balloon probes and LIDARs (LIght Detection And Ranging) are the main tools used, reaching altitudes ranging from 30 to 80 km with good vertical resolution, but poor temporal resolution. The rockets-probes make it possible to gain altitude (of a hundred kilometers to the thousand), but they keep a temporal resolution low because of the cost of launching. Satellites, meanwhile, provide information on the average temperature of an area whose size depends on the instrument. There is therefore a lower spatial resolution but a very good temporal resolution.

Night vision: How night vision works
Most night vision systems currently work on the light amplification method. The principle here is to collect the photons emitted by the Moon-type light sources, urban lighting, stars, and to amplify them using a photomultiplier. The principle of operation of the photomultiplier is to use the photons to extract electrons from a photocathode and then to amplify the resulting current by cascade of electrons. These photomultipliers are usually mounted on binoculars and are used in the military and civilian field. There are several generations of binoculars, all based on sensors whose technology varies.

Interest of nightglow radiation
During nights without a moon, the amount of radiation at ground level is insufficient in the spectral range of night vision systems (operating in the visible band 0.4- 0.7 μm) to allow a good recognition of the terrestrial scenes as the observation of a military scene using glasses by light amplification. U.S. Department of Defense http://www.defense.gov/.Thermogram observed, representing cats on a roof. Institute of Thermography http://www.institut-thermographie.com/.

Nightglow radiation

Origin of radiation: History

Besides the observation of the stars, the night sky was subject to many studies in the early twentieth century. Most scientists thought that light came mainly from starlight. However, some spatial and temporal variations observed by photographic plates did not correspond to expectations. Studies evoked a possible permanent aurora to explain for example the green radiation, but it was not until the first spectroscopic studies of Lord Rayleigh [Rayleigh, 1921] using filters to characterize this radiation, in particular the green line of the 'oxygen. Comparisons to the aurora and sun spectra were then made using a Fabry-Perot interferometer [Babcock, 1923], from the UV to the infrared (IR) band.

Then, the vibrational and rotational OH bands are discovered by Meinel who used a spectrometer to study the sky between 700 and 900 nm Meinel [1950a, b]. It is then Bates and Nicolet [1950] who propose the first mechanisms to explain the OH radiation. Jones and Gush [1953] complete the observation up to 2μm. At the same time, observations are compared [Dufay and Dufay, 1951], laboratory studies are carried out [Herman and Horbeck, 1953], and new observations are made [Jones and Gush, 1953]. The name of airglow was proposed by Otto Struve [Elvey, 1950], as well as declensions dayglow, nightglow even twilightglow, in opposition to the non-polar appellation aurora or light of the night sky [Petitdidier and Blamont, 2006].

sources
The sources of nightglow radiation are numerous. Leinert et al. [1997] list these sources as well as their spectral distributions and intensities. The main source in the near IR (PIR) remains the de-excitation of the rovibrational OH transitions whose spectral band extends from 0.6 to 4 μm (or in cm-1: 2500 to 16 666). Then, the oxygen emits from its electronic transitions over a broad spectral band, including UV, visible and PIR. In the visible are O, O2 and Na. We can see the atomic oxygen emitting in the red from 150 km. The green glow corresponds to the green line of oxygen (electronic transition, identical to the aurora). It is issued around 90 km. Around 80 km, it is the turn of sodium Na to emit in a yellow tint. There is also ozone radiation, measured by Zhou et al. [1998].
The atmospheric radiation called nightglow is a phenomenon occurring at high altitude (about 90 km). It consists of the emission of radiation following the de-excitation of certain molecules and atoms (OH, Na, O2 and O). It is spread over a wide spectral range, especially in the infrared and spreads to the ground level. Nightglow radiation is an important marker for the upper atmosphere, making it possible to go back to temperature, but also to many dynamic phenomena such as atmospheric tides or gravity waves.

Its propagation at ground level allows terrestrial scene lighting and thus night vision using near-infrared cameras. In order to better understand the fluctuations of these emissions as a function of time at different scales and in different places on the planet, the theme focused on an observational study and a theoretical study. The observational study produced a large-scale climatology by extracting the radiation from GOMOS data. The measurement campaigns carried out on the ground have for their part highlighted some important dynamic aspects such as tides and gravity waves. To reproduce the nightglow radiation, it was necessary to model the chemical reactions of the many species present at high altitude, the heating, the photodissociation of certain molecules by the solar radiation and the propagation of the radiation towards the ground. Some dynamic processes have been included such as molecular diffusion, turbulent diffusion and tidal parameterization. Finally, the results of the model are compared with satellite observations as well as at ground level and sensitivity tests are performed to estimate the response of radiation to the different modules of the model.
Derivation of nightglow intensity and estimation of error

To calculate the intensity of nightglow radiation at a given altitude z, a few steps are necessary. Measurements from GOMOS are provided in electrons / 0.5s / pixel. Radiometric conversion is used to obtain the physical unit photons / cm2 / s / nm / sr (spectral luminance). The number of electrons N (e) created per pixel during dt is written according to

N (e) = I (λ) dωSeff (λ) dλdt.

Observational study from the ground

We are talking about an observational study of radiation from the GOMOS instrument. We will now study radiation from observations at ground level, produced during measurement campaigns. We begin by introducing the instruments used. The preferred spectral band is the near infrared due to the low absorption of radiation by the atmospheric layers. We explore the evolution of the radiation during the night by pointing the cameras directly to the sky but, we also characterize the terrestrial scenes to check the capacity of the radiation to illuminate the ground. The part is devoted to the Gadanki measurement campaign with the NARL corresponds to the article that was published and included in the manuscript entitled Coordinated night airglow measurements of gravity waves during pre-monsoon period over a low latitude Indian station. Bellisario et al. [2016].
BIBLIOGRAPHY

W.R., Sica, R.J., and Neal, P.C. (1991). Simultaneous intensity, temperature and imaging measurements of short period wave structure in the OH nightglow emission. Planetary and Space Science, 39: 1171-1188. Taylor, M.J. and Hapgood, M.A. (1988). Identification of a thunderstorm as a source of short period gravity waves in the upper atmospheric nightglow emissions. Planetary and Space Science, 36: 975-985. Taylor, M.J. and Hapgood, M.A. (1990). Ripple-type wave structure in the oh nightglow emission. Planetary and Space Science, 38: 1421-1430.

Taylor, M.J., Turnbull, D.N., and Lowe, R.P. (1995b). Spectrometric and imaging measurements of a spectacular gravity wave event observed during the ALOHA-93 Campaign. Geophysical Research Letters, 22: 2849-2852. Thirupathaiah, P. and Singh, V. (2014). An updated model of the atomic oxygen redline dayglow emission. Advances in Space Research, 54: 939-945. Thomas, L. and Bowman, M. R. (1972). The diurnal variations of hydrogen and oxygen constituents in the mesosphere and lower thermosphere.

Journal of Atmospheric and Terrestrial Physics, 34: 1843-1858. Toon, O.B., McKay, C.P., Ackerman, T.P., and Santhanam, K. (1989). Rapid calculation of radiative heating rates and photodissociation rates in inhomogeneous multiple scattering atmospheres. Journal of Geophysical Research, 94: 16287-16301.

Cordially

The availability of open satellite imagery has grown rapidly in the last years, and so are the number of studies and applications employing thereof for different purposes. Nevertheless, choosing suitable dataset for a remote sensing project often results in a tedious and time-consuming operation. In view of the above, having access to complete and user-friendly tools -dedicated to data overview as well as metadata harvesting- is key to facilitate this task. Emerging technologies -such as the distributed computing- pose a terrific opportunity for tackling any satellite data planning issue. Among the cutting-edge solutions, the Google Earth Engine (GEE) represents a reference in terms of capabilities and usability for satellite imagery operations. GEE is a cloud-based geospatial processing platform that provides catalogues of open satellite imagery, and a set of algorithms that can be used directly on these data. Its functionalities can be accessed via a dedicated online code editor as well as using JavaScript and Python API.
In this project, we aim at developing a user-friendly web application -called EO Hound- to query catalogues and generate summary statistics about satellite imagery availability and quality, bound to an user-selected area of interest. The access to the catalogues is accomplished using GEE API that guarantees access to both the historical and the most up-to-date catalogues information. A map-based web application is developed exploiting the Node.js Express for the backend to embed the GEE API. The frontend provides with dynamic visualisations using OpenStreetMap as background layer. An embedded dashboard allows users to specify parameters of interest, such as the imagery collection source, date ranges, the cloud cover percentage, and the area of interest by means of a GeoJSON file in a specific reference system. After the parameters set-up, the different granules/tiles intersecting the area of interest are rendered on the map, as well as a sample image. In addition, a summary of statistics over the collection is shown to the user. Selecting one of the granules/tiles will lead to specific information about it (e.g. data availability and cloud coverage evolution over time, revisit time, etc). For a better interpretation of the results, the graphs such as barplot are generated and displayed to users. Currently, the application focuses on the analysis of Sentinel-2 and Landsat8 metadata. Since GEE makes many datasets available, the application may easily include other satellite imagery catalogues. Therefore, future work will focus on extending the summary statistics set as well as allowing users to load the area of ​​interest with different formats and reference systems.

Monitoring of the coastal zone is necessary in order to quantify human activities. Offshore wind farms, tidal and wave power stations need marine and atmospheric environmental information to select optimal sites and to forecast power production. The large-scale construction of offshore renewable energy sites may also have environmental impacts which are still unknown in the long term. These impacts might not be only produced by constructed power devices, but also by the following increase in marine traffic around the area. Some natural coastal processes have also an important impact such as the dragged sediments and organic matter by rivers. The collection of samples for analysis and in situ measurements at specific locations have been the most common conventional methods for environment and resource monitoring. They can provide accurate information with high temporal frequency, but just for particular points in terms of space.
Earth Observation satellites have the potential to help overcome these limitations, by greatly increasing the spatial coverage of samples. Satellites are a most cost-effective device for ocean observation, providing water quality and sea parameters related information over extensive spatial and temporal scales. The introduction of remote sensing into environmental monitoring provides a novel capability to evaluate, analyse and forecast changes of specific ocean products.
Examples of such value-added water quality products, which come from thematic processing of SAR and optical images are: remote sensing reflectance, turbidity, chlorophyll-a concentration, algal bloom detection products, man-made object detection, wave height and many others. These represent an important set of parameters to control the quality of the water in coastal and inland waters. Furthermore, ocean resource products can be also obtained from satellite remote sensing such as sea level, significant wave height, sea surface winds, currents, salinity or surface temperature among others.
Making the best use of the vast quantities of earth-observing data arising from the Sentinels and other satellites requires new methodologies to merge all the different environmental parameters. A solution in development by different teams for different locations is the Datacube. This new technique will allow time-series analysis and at the same time help find correlations between parameters which will lead to a better understanding of physical processes and the human impact on the environment.
The Humber region in UK has the highest density of neighbouring offshore wind farms in the world; with five sites fully commissioned, and others under construction or at the planning stage. Other maritime users in the regions are shipping, ports and harbours, tourism, fishing and aquaculture, all with their own impacts on the environment. Hence, a Datacube is being created to store and visualise processes along the Humber coastline and open North Sea. The first target of this Datacube is to support offshore wind farm activities and quantify possible impacts of a changed distribution of suspended sediments.

The scope of this presentation is to feature the G-POD SARvatore service to users for the exploitation of CryoSat-2 and Sentinel-3 data, which was designed and developed by the Altimetry Team at ESA-ESRIN EOP-SER. The G-POD service coined SARvatore (SAR Versatile Altimetric Toolkit for Ocean Research & Exploitation) is a web platform that allows any scientist to process on-line, on-demand and with user-selectable configuration CryoSat-2 SAR/SARIN and Sentinel-3 SAR data, from L1a (FBR) data products up to SAR/SARin Level-2 geophysical data products.
The G-POD graphical interface allows users to select a geographical area of interest within the time-frame related to the Cryosat-2 SAR/SARin FBR and Sentinel-3 L1A data products availability in the service catalogue. The processor prototype is versatile, allowing users to customize and to adapt the processing according to their specific requirements by setting a list of configurable options. Pre-defined processing configurations (Sentinel-3 official, CryoSat-2 official, Ocean, Costal, Inland Water, Ice and Sea-Ice) are available to ease the use of the service. After the task submission, users can follow, in real time, the status of the processing. The output data products are generated in standard NetCDF format, therefore being compatible with the Multi-Mission Radar Altimetry Toolbox (BRAT, http://www.altimetry.info/toolbox/) and typical tools.
Initially, the processing was designed and uniquely optimized for open ocean studies. It was based on the SAMOSA model developed for the Sentinel-3 Ground Segment. However, since June 2015, the SAMOSA+ retracker is offered as a dedicated retracker for coastal zone, inland water and sea-ice/ice-sheet. This will be soon followed by the new SAMOSA++ retracker, which further improves the retracking capabilities in these complex scenarios. The scope is to maximize the exploitation of CryoSat-2 and Sentinel-3 data over all surfaces providing user with specific processing options not available in the default processing chains.
Recent improvements include: 1) A Join & Share Forum to allow users to post questions and report issues; 2) A data repository to better support the growing Altimetry Community avoiding the redundant reprocessing of already processed data; 3) New Tide Models (FES2014b & TPX09) and Mean Sea Surface (DTU18, DTU15) selectable among the L2 processing options; 4) the inclusion of Sentinel-3B data.
The service is open, free of charge (supported by ESA Earth Observation Envelope Programme) for worldwide scientific applications and available at https://gpod.eo.esa.int/services/CRYOSAT_SAR/.

Information on mass movements in remote mountainous areas are often either lacking or inhomogeneous especially when they do not affect people or settlements. Geohazards in mountainous regions comprise debris or mud flows, landslides, avalanches, rock falls, and (flash) floods. These can be either water or cryosphere-related and because they have different velocities they leave identifiable footprints on the landscape. Satellite data can be used to record these events and complement human observations that allow the natural hazard triggers (e.g. climate phenomena) to be determined as well as the processes involved, their location and magnitude. However, so far, Earth observation data have been barely exploited to complete existing natural hazard inventories or investigate the influence of climate perturbation on potentially dangerous natural phenomena.

Synthetic Aperture Radar (SAR) techniques, such as those provided by the ESA Geohazards Exploitation Platform (GEP), allow spatiotemporal monitoring of natural hazard hotspots. The spatiotemporal behaviour of slow Earth surface deformation can be detected using images from the Copernicus Sentinel-1 SAR and these can be used for checking and completing actual mass movements inventories. FASTVEL is an automated Persistent Scatterer Interferometry (PSI) based service at GEP which uses a stack of SAR images and provides a velocity displacement map of the Area of Interest. For instance, the FASTVEL algorithm enables the recognition of active Earth surface deformations corresponding to slow landslides and rock glaciers. Other radar satellite sensors (such as the European remote sensing satellite, ERS and the Environmental Satellite, EnviSat) can also be used to measure changing slope dynamics in the past.

This study aims to examine the benefit of GEP for monitoring slope deformation and hazard-prone areas as a precursor to its use in systematic regional hazard and risk assessment. Findings would help local stakeholders to recognise hot-spot areas where early-warning systems are possibly needed, enabling disaster risk reduction in remote and low-density hazard-prone regions. The test has been performed on hazard-related changes in the South Tyrol region (Italy) where, in 2018, around 160 natural hazard events, causing actual damage, were recorded by the Civil Protection. A further objective is to examine these changes in relation to particular meteorological events at regional and local scales and thus investigate trends in current and past activity.

Within the advent of digital societies existing and emerging heterogeneous data streams call for their conflation. OSM data was found useful to automatically derive labeled training data for remote sensing image analysis problems. OSM object footprints can be injected into information abundant feature spaces such as provided by remote sensing for semantic segmentation, land use mapping and target detection. We propose to identify OSM and EO data suitability when varying tags (anthropogenic, natural), deep learning model (convolutional neural network (CNN), fully connected network (FCN)) and EO data (Sentinel 1 and Sentinel 2, Pleiades, Landsat) for data rich dense time series in Toulouse.
Commonly training data is digitized, we propose using existing OSM data instead. Simplicity of OSM footprint training data is insensitive to spectral responses and provides the ability to serve as training regardless of spectral response and data type. Variety of OSM elements allows the construction of different sub- and super level training features accounting for existing feature space and desired scale of classification. OSM based model training can unfold full potential when EO feature spaces of multiple sensors are combined and fused into hyper complex feature spaces allowing to harness deep learning to its full potential, avoiding existing VHR RGB limitations.
Millions of humans voluntarily generate spatial data in continuous flows of information. OSM offers rich geometric and semantic information that goes beyond land use tags, a library of machine training data. Satellites monitor land cover directly allowing us to draw conclusions of bio-geochemical dynamics and trends, where human knowledge allows us to relate these measured spectral responses to specific land usages. Combination of OSM and remote sensing data streams, people and pixels is an ongoing and increasing inevitable logical reality. Our work highlights developments, when varying OSM semantics, sensors and classification models. Comprehensive quality and performance estimates allow insights of methods strength and limitations and therefore reveal upcoming development trends.

In short:
• Train deep learning models using OSM for semantic segmentation
• Fuse EO data to construct hyper complex feature spaces to solve classification problems
• Identify methods suitability for time series analysis semantic object evolution monitoring

Having an enormous amount of data available can be difficult to handle. At EOX we are used to create global satellite basemaps, therefore we understand that saving resources increases reliability of any tool, product or data. Having tested various approaches on a global scale it is time to accommodate local needs, meaning a country- or state-wide preparation of satellite data.

A good extraction method and data bundling will become more and more important to make it easier for projects to access Earth observation data without having to dig through the archives. Such dissemination options allow everyone to easily access large datasets which are reduced and prepared for instant analysis, machine learning, validation, etc.

There are some guidelines which try to define analysis ready data (ARD), however with no clear definition at hand. We are utilizing the experience we have gathered while working closely with our customers. These range from scientists, industry and various national agencies. Every single one of these have their own specifications and preferences. With this information we have found and tailored the refinement of satellite data to operational production.

Our general aim is to enable land monitoring with primarily agricultural applications in focus. For this we use Sentinel-1 and Sentinel-2 data.

By default we process Sentinel-1 data to normalized flattened scattering cross section gamma0 (γ⁰), parallel to the backscatter we deliver the coherence mosaics as well. We can even simulate the normalized scattering cross section sigma0 (σ⁰) for any incidence angle if needed, for this we use a simple linear regression by default.

For Sentinel-2 processing, we are using the provided sen2cor data (L2A) to make homogeneous monthly composites of individual countries. We have 4 major mosaicking methods, all of which are single pixel time series (stack) analysis. Additionally, custom masks are used to detect clouds, shadows, water, urban areas, etc. when needed. Two of the methods are selective and two are statistical. The first selection algorithm sorts the brightness of input bands and gives the appropriate measurement, the other searches for maximum NDVI (and CVI) values to give the "greenest" data output. The first statistical method is weighted average, where the weights of the individual measurements are evaluated and averaged, the evaluation prefers high NDVI, CVI and NDWI. Second statistical method is an implementation of the linear regression from Sentinel-1. Here the brightness values are simulated for the corresponding NDVI value. This last method is rather experimental. It leads to correct values for most of vegetation types, yet for non-vegetated areas the results are failing due to the lack of variation.

We used custom developed software for geospatial analysis, which is based on open-source tools like GDAL. For the processing of Sentinel-1 data the ESA SNAP Toolbox jython integration has been utilized.
This EOX software bundle consists of several of packages, where one is responsible for (remote) data handling (cloud reading/writing, currently maintained for Amazon Web Services (AWS) and DIAS mundi Open Telekom Cloud (OTC)), other is responsible for SAR processing, the next for optical data analysis and the last ones are handling the geospatial data in general, like projections, grids, data formats (raster, vector), ...

Such a processing interface is similar to some datacubes concepts. It is designed to work with very high temporal and spacial dense data as time efficiently as possible.

Only a relative small portion of the customers who aim to buy maps is interested in the processed data and just wants a nice basemap for their GIS interface. For these scenarios we render the data to deliver the beautiful imagery so that everyone can enjoy its visual representation to the fullest. The actual (not rendered) data are mostly used by the corresponding project or customer group. For instance EOX global mosaic data 2018 (s2maps.eu) were used to determine land cover to update a global virtual 3D render which is used for simulations.

Our next generation mosaics which are based on the Level-2 products are meant to be used with auxiliary data to examine and verify studies and experiments over large areas, providing countrywide uniform datasets with easy access for everyone.

For evaluating in quantitative manner a process of achieving the sustainable development goals (SDG) and how big its progress is, a global indicator framework was developed. We propose the approach to the comprehensive Air Quality (AQ) monitoring in the framework of the HORIZON 2020 Project. It’s well-known, that existing smart cities do not fully exploit the capacities provided by EO. It is especially relates to air pollution, which impacts on health straightly. It causes the respiratory diseases (including asthma), cardiovascular diseases, adverse pregnancy outcomes. Moreover, the air pollution has a carcinogenic effect on humans. A high level of air pollution in Ukraine was confirmed by anti-rating list of the countries where our state ranks ninth place due to a high level of industrial and transport pollution. Within this survey we describe conceptual architecture of the system to determine the air pollutants’ concentrations (in particular, PM2.5 and PM10) over the Kyiv using up-to-date monitoring instruments; the requirements for information technologies for making public the state of AQ in Kyiv was reported. The described infrastructure will be created on the base of existing AQ monitoring stations and free satellite data. Both in-situ and satellite data will be harmonized providing a new single standard for AQ Estimation in Kyiv. The comprehensive review of in-situ and satellite measurements of atmospheric pollutants is provided as well as the description current state-of-the-art for AQ monitoring with intelligent sensors and systems in Kyiv to study the basis for building a large-scale AQ Monitoring system. Currently, the prototype of client part of SMURBS AQ SYSTEM is being developed. Its functionality includes the visualization of the AQI and the pollutants it depends of (NO2, SO2, O3, PM2.5, PM10) as separate layers; viewing the history of values at a point and, accordingly, the choice of dates for viewing information with the possibility of a 3-day forecast (based on CAMS data).

A core technology for ocean and atmospheric sensing is passive imaging of reflected solar radiation, typically detected with multispectral CCD array based systems. General challenges in adapting such imaging technology to CubeSat platforms over the littoral environment include 1) sufficient signal-to-noise for adequate algorithm retrieval of environmental properties, 2) acceptable photon efficiency, 3) efficient information transmission to the ground station that enables the rendering of high quality images given severe data transmission limitations, and 4) saturation, blooming and edge effect problems with water adjacent to bright land and clouds.
We are developing a novel optical acquisition hardware architecture for a pushbroom-type CubeSat imager based on a Digital Micromirror Device (DMD) and an improved backend compression processing scheme to optimize information transmission given severe data bandwidth limits from CubeSats. SNR will also be substantially improved relative to CCD arrays. A DMD consists of millions of electrostatic-actuated micro-mirrors that can be used to control light collection dynamically for each individual pixel equivalent. The imager will allow for adaptive optimization of spectral resolution, spatial resolution, and SNR based on a particular scene being imaged. The spectral range of the DMD imager covers 400 to 900 nm over 1600 bands, capable of resolving a host of ocean parameters, including chlorophyll, turbidity, phytoplankton functional groups, productivity, and bathymetry, from published algorithms. Projected spatial resolution will be as fine as 20 m over a 50 km swath. The concept would have tremendous research potential as an imaging system dedicated to monitoring European coastal waters over the short time and spatial scales that none of the global imagers from the Copernicus constellation can currently resolve. The CubeSat program at SSC Pacific is developing our satellite operations program plan. This imaging system has been prototyped in our lab with several design improvements that have been pursued to maximize optical throughput.

With the launch of Sentinel-1, C-Band SAR data is freely available over the whole planet for the first time. Due to its capability to sense the earth independent of day and nighttime as well as cloud cover, it serves a wide spectrum of applications. However, before being able to run data analytics on the SAR measurements for the actual information retrieval, specific pre-processing steps need to be applied in order to get the data “analysis ready”. On the one hand, this hinders non-SAR experts to make use of the data and, on the other hand, those steps are computational expensive, so that the generation of time-series or large-scale mosaics can take up several days.
We address both issues with the Open SAR Toolkit (OST). OST is a Python 3 API that provides smart data access from online archives and the DIASes as well as fully automated wrappers around workflows of ESA’s SNAP Sentinel-1 Toolbox for the generation of Analysis-Ready-Data (ARD). Some basic parameters are customisable in order to address different definitions and scenarios of “analysis-ready”. For this, single-scene processing is possible as well as the automated generation of time-series and timescan image stacks. Input data can be either GRD or SLC data, whereas for the latter short-term coherence (i.e. from consecutive images) as well as the dual-polarimetric H-A-alpha decomposition can be added to the standard backscatter products.
OST ships with self-explanatory Jupyter notebooks and provides ready-made functions to run on clusters of cloud machines to process multiple images at once on various computing nodes. This enables to drastically reduce the time, without increasing the resource costs used for processing.
We will present some example applications that build on the processing scheme of OST and demonstrate new colour composites based on multi-temporal timescan features, combining backscatter with interferometric coherence and polarimetric decompositions that can serve a wide range of land applications.

Jupyteo - Jupyter-Oriented Cloud Service for EO Data Processing

Jupyteo is an IDE environment based on the popular Jupyter Notebooks technology, focused on processing of EO data in the cloud. By using the service a user gains access both to data sets and to functions useful for their processing, visualization and results sharing. Jupyteo is intended for advanced users since, due to the programming-based approach, it offers immense opportunities to work with data, to process and modify it. Thanks to an access to the shell it is also highly flexible, allowing free configuration. Although it is an online tool placed in a virtual operating system and accessible via a web browser, working with it is similar to working with a desktop application.

The main advantage that distinguishes this product from other tools is, apart from using ready-made algorithms, the ability to quickly and easily create users’ own algorithms or scripts. As a result, the data processing can be controlled strictly at every stage and in a formula adapted to any user needs. Although the basic programming language in Jupyteo is Python, it is possible to create code in other languages by launching additional engines offered by the integrated Jupyter Notebooks functionality. Jupyteo by default also offers the ability to create scripts in R, Scilab, Bash and it gives access to JavaScript programming capabilities.

Each user gains access to his own separate virtual environment based on Docker virtualization technology. There is a fully functional application at the user’s disposal along with the disk space for data storage. The result of processing from each cell is visible directly below it and is presented in a text or graphic form, depending on what value is given as an output. It will be saved to user's account and can be launched at any time or downloaded to home computer. Another useful feature of Jupyteo is the ability to create code snippets that can be reused at any time after saving. A user can find a set of ready-made snippets in the "Snippets" tab on the side panel.
The feature that distinguishes Jupyteo and adapts it to the needs of community interested in processing EO data is the map component. The side panel with map allows to search EO data and to automatically present the results of the processing. Its functionality has been taken from popular GIS systems. Thus, any result of the algorithm in a form of spatial data can be added to the map on an ongoing basis. In addition, the user has an option to create, launch and share his own WPS service that processes data and sends results to a specific location or to use the results of data processing services from other providers. Jupyteo also provides the ability to search spatial data services by using functions placed in the side panel toolbar. Each object representing the single range has attributes that allow to read metadata of the material and download or connect it to one’s own algorithm.

The user of Jupyteo gets his own virtual environment, which gives the possibility of quite extensive interference and control over the installed software components. Jupyteo user can add or remove libraries on which Jupyteo is based and customize the system's functionality to suit user's needs. It also has possibility to make other changes within the installed applications or additions of the system platform itself and configure it for strictly defined needs. All files with results can be freely downloaded, placed on the server or shared with other users.