TEP - Exploitation Platforms

The EO Broker platform is designed to encourage the uptake of earth observation by the energy sector by supporting users through stages of pre-procurement including product discovery, feasibility and supplier interaction. The open source platform has been developed through support from ESA's GSTP programme and as such includes some innovative brokering capabilities for products and services, but at the same time is aimed to support relatively easy uptake by both users and suppliers. The design has been influenced strongly by user cases derived from the energy sector, including from ESA's "EO4OG" projects, but has benefitted also from guidance by a project Steering Committee consisting of oil and gas industry professionals, The platform is at a mature stage of development, but is not operational. Its capabilities will be demonstrated through the presentation, along with an update on planned exploitation beyond the development phase of the project, and interested users and suppliers are invited to get in touch with the team.

Exploitation Platforms (EPs) have been becoming the preferred way to distribute EO data and derived information since their concept introduction in 2013. They have been declined with several approaches (e.g. Thematic, Mission EP…) according to the intended users community. A new one, specifically designed to satisfy the needs of the scientific community is described in this paper, demonstrating how EP concept perfectly fits and opens up to new possibilities. In fact, aside with the already assessed advantages this EP defines a set of thematic data coming from different domains, provides their easy centralized availability and provides algorithms and processing capabilities in order to test, tune and assess scientific outcomes. The study case is the SAFE (SwArm For Earthquake study) project, funded by the European Space Agency (ESA), in the framework of "STSE Swarm+Innovation”, and coordinated by the Istituto Nazionale di Geofisica e Vulcanologia (INGV), that deals with the integrated analysis of several physical parameters whose abnormal variations have been found to be possibly associated with impending earthquakes (EQs). In the frame of the project Planetek has developed the so-called SAFE Exploitation Platform with the aim of sharing its outcomes and results, and of demonstrating the implemented techniques and scientific algorithms. Main purpose of the SAFE project is the investigation of the phase preceding large earthquakes with the aim to identify any electromagnetic signal eventually related to the forthcoming seismic events. In particular, the project has been intended to assess the Lithosphere-Atmosphere-Ionosphere Coupling (LAIC), using both space-borne data coming from the ESA’s Swarm Constellation satellites together with ground geophysical data from different sources (e.g. in-situ instruments, seismic catalogues and other archives). The platform is specially designed to integrate both Swarm-satellites dataset with these ground-based geophysical data; it allows for accessing and visualizing data, performing online user-adjustable analyses and disseminating the project’s results to the scientific community. At a functional level, the SAFE Exploitation Platform provides means to properly configure and perform the collection and automatic update of data from external catalogue sources, to make them available to the user for browsing, visualization of their geographical information on the world map, and to perform a set of customized analyses, which can be configured according to the scientific findings and can then contribute to their assessment at a global scale. The SAFE Platform is able to integrate both engineered software (in C/C++) and prototypal implementations (in Matlab® and Python). Among the advantages we can mention: direct (and simplified) web access to data from different sources, to global results and to algorithms configuration and execution environment; unique reference deployment so to always share the latest version of the scientific algorithms, that have fast and flexible evolution maintaining a prototypal implementation approach. Finally it includes also integrated dissemination instruments, as a Web Portal intended to share the project’s results with the international scientific community, collect feedbacks on algorithms and results, and to highlight all the related events/ initiatives. This paper presents the scientific Exploitation Platform approach, with its architecture and functionalities, with a special focus on the SAFE case study in which its effectiveness has been demonstrated, with scientists and users involved in geophysical studies, for the multiple analyses in geomagnetic, ionospheric and seismic data domain and the contribution to the investigation of EQ preparation process.

Recently, a great attention has been given to the exploitation of Earth Observation data, as a mean of industrial innovation and source of potential societal benefits. EU is at forefront in Earth Observation technologies: ESA launched the Sentinel Constellations, a set of redundant satellites that will offer high availability and resiliency as required by industries to run businesses. Several attempts and approaches have been experimented with alternate success in terms of self-sustainability, easy-of-use and scalability: from Thematic Exploitation Platforms , Business incubators to a Marketplace for EO services and data . All these approaches seem to enable a pipeline business model, i.e. where business is based on the acquisition of resources (a product and/or service) that are pushed to the consumer through the value chain in a unidirectional way. However, the Digital Transformation is radically changing the market landscape: ubiquitous connectivity, hands-held technology and user interactions are enabling elements of the platform business model, as successfully exemplified in various markets such as AirBnB, Uber, Google, etc. The platform model, instead, focuses on the creation of value through establishing an intelligent networking among users: where pipelines create value “on-top” of managed resources, platforms (that usually don’t even own such resources) create value by linking producer and consumer of resources. Platforms, as analysed in depth by S.P. Choudary (Choudary et.al. 2015), execute as a content aggregator that can simultaneously satisfy different type of interests. Platforms exploit also the phenomenon of network externalities, i.e. a service increases in value whenever increases the number of interacting individuals. Externalities could be of two types: same side (e.g. as in the telecommunication networks) or cross-side. Platform model exploits cross-network externalities, which are linked to the diffusion of the product, not among members on the same side of the market, but to the diffusion of the product on another network (or side). For example, Amazon marketplace bridges producers and consumers, while Uber helps drivers and riders to meet or AirBnB links housekeepers and tourists. These features also allow the business model to scale faster, thanks to the nature of the user, who could be both resource provider as well consumer (prosumer): this creates new opportunities, but also new challenges to face with. The breakthrough work of Choudary on the platform analysis models highlights the distinctive features of this new approach as well as the best practices that facilitate its understanding and implementation. This presentation aims to open a debate among stakeholders and to illustrate the initial hypothesis about how to implement the platform model in the EO sector for the benefit of all actors involved.

A prototype platform for Earth Observation data in the Atlantic region (EO4Atlantic) is being developed in Ireland, and is funded by the European Space Agency. The project is led by Parameter Space Ltd., an Irish SME with expertise in platform development for a number of scientific missions, in collaboration with a number of SMEs including Treemetrics, Techworks Marine and iGeotec. Trial users are located in relevant institutions and organisations located along the Atlantic coastline, with an interest in satellite data. The EO4Atlantic platform will allow users to analyse large data sets in a high-performance capacity, without the need to download large data volumes to their computers. The users can make use of existing available software (e.g. open source tools currently being used for analysis of satellite or sensor data) or upload their own tools to the platform and run their analysis close to the data. The objective of the EO4Atlantic pathfinder platform is to support the development and delivery of EO based information services, based on high volume EO data access and processing, with a focus on expert and non-expert users of EO data from the European Atlantic region. The platform is designed to demonstrate how best to use and integrate existing capabilities and infrastructure in the Atlantic region, and how a full regional exploitation platform could operate in this area with potential new capabilities and infrastructure which will be implemented in the medium to long term. Customised toolkits are made available through the platform for testing. These include open source toolboxes for analysis of satellite data and workflows allowing the chaining together of different tools to create multiple processing steps. End to end services also provide products such as sea surface wind speed evaluations for renewable energy exploitation, cloud-free pixel databases for high-cloud areas, forestry monitoring and management tools, flood risk assessment in coastal regions, and ecological and physical observations of inland and coastal waters. These services are designed to complement the common issues and goals of the regional and European initiatives specific to the Atlantic area.

SPOT 1-to-5 satellites have collected more than 15 million images all over the word during the last 30 years from 1986 to 2015 which represents a unique historical dataset. Spot World Heritage (SWH) is the CNES initiative to preserve and promote this SPOT archive by providing new enhanced products on an open Web platform. A first step has begun in 2015 with the start of the repatriation of the last SPOT data hosted in the Receiving Stations spread across the world in the CNES central archive system. Meanwhile, some preliminary SWH processing chains have been developed with the production of more than 100.000 orthorectified SPOT products, provided by CNES through the French Land products data Centre THEIA. From mid-2017, the SWH initiative will move into another phase with the development of operational SWH processing chains in line with Sentinel-2 “standards” to allow deeper time series analysis. The first one is based on the current SPOT operational processing chain and will provide SWH-L1A products which are first exploitable images with radiometric corrections. The following ones will provide SWH-L1B and SWH-L1C products, geometrically compatible with ESA Sentinel-2 products, SWH-L1C being the orthorectified product. SWH processing will take place on CNES High Performance Computing Centre to take advantage of SPOT archive proximity. Dedicated means are being put in place to process the whole SPOT archive using 24-core processors and optimized solutions for file sharing (GPFS), deployment (Docker) and cataloging (Elastic Stack). Finally, all the generated SPOT products will be accessible free of charge to registered users via the Web. First SWH products are expected to be distributed in 2018 while the whole archive is expected to be processed within 2 years until 2020, depending on the timing of SPOT data retrieval from the reception stations.

Earth observations from satellites produce vast amounts of data. In particular, the new Copernicus Sentinel missions are playing an increasingly important role as a reliable, high-quality and free open data source for scientific, public sector and commercial activities. Latest developments in Information and Communication Technology (ICT) facilitate the handling of such large volumes of data, and European initiatives (e.g. EOSC, DIAS) are flourishing to deliver on it. In this context, Terradue is moving forward an approach resolutely promoting an Open Cloud model of operations. With solutions to transfer EO processing algorithms to Cloud infrastructures, Terradue Cloud Platform is optimising the connectivity of data centres with integrated discovery and processing methods. This is for example the case with the Geohazards Exploitation Platform initiative, an R&D activity funded by ESA. Implementing a Hybrid Cloud model, and using Cloud APIs based on international standards, the Platform fulfils its growing user needs by leveraging capabilities of several Public Cloud providers. Operated according to an “Open Cloud” strategy, it involves partnerships complying with a set of best practices and guideline: • Open APIs. Embrace Cloud bursting APIs that can be easily plugged into the Platform’s codebase, so to expand the Platform offering with Providers offering complementary strategic advantages for different user communities. • Developer community. Support and nurture Cloud communities that collaborate on evolving open source technologies, including at the level of the Platform engineering team, when it comes to deliver modular extensions. • Self-service provisioning and management of resources. The Platform’s end-users are able to self-provision their required ICT resources and to work autonomously. • Users rights to move data as needed. By supporting distributed instances of its EO Data management layer, the Platform delivers the required level of data locality to ensure high performance processing with optimized costs, and guarantees that value added chains can be built on top of intermediate results. • Federated Cloud operations. The Platform’s collaborative environment and business processes support users to seamlessly deploy apps and data from a shared marketplace and across multiple cloud environments. As a recent case, thanks to the integration within the Platform of the Open Cloud Computing Interface (OCCI), and the close partnership between EGI and Terradue (a Service Level Agreement signed in September 2016 allowing Terradue to access a set of EGI federated Cloud providers in Europe), our provisioning of ICT resources supports ever more demanding exploitation scenarios. At this stage, EGI compute and storage resources from GOEGRID-GWGD (Germany) are used to support the ESA SNAP Sentinel-1 COherence and INtensity (COIN) Service, an on-demand processing model accessed by GEP users. Also, EGI compute and storage resources from ReCaS Bari (Italy) and BELNET-BEGRID (Belgium) are used to support the global scale systematic production of InSAR Browse Services integrated on GEP by the DLR. As Platform services, they automatically produce interferograms out of Copernicus Sentinel-1 acquisitions, over a subset of the global strain rate model, where volcanoes eruptions and earthquakes are most likely to impact society.

With launch of Sentinel-1 and Sentinel-2 missions new era of geospatial science in agricultural domain based on open data has started. High spatial (10 m) and temporal resolution (6 days of revisit period) leads to big amount of satellite data that couldn’t be processed within local computational infrastructure due to timely download, difficulties in data storage and processing. On the other hand such new quality of satellite imagery is very efficient for crop type mapping, area estimations etc [1-3]. These issues can be solved with use of modern satellite platforms for dealing with big amounts of geospatial data. Among them Google Earth Engine (GEE) and Amazon Elastic Compute Cloud (Amazon EC2) with Amazon Simple Storage Service (Amazon S3) are good examples of different ways of processing geospatial data in the cloud. GEE provides high-performance computation on free of charge basis with some amount limitations (in test mode only). With GEE geospatial analyst can use wide range of already developed methods for different modern tasks that occur in daily routines quite easy and very intensively. On the other side users are forced to use data catalog with already preprocessed wide range of satellite data. In case of SAR data this leads to some difficulties in further data exploitation. Within exploitation of Amazon EC2 and S3 services users are allowed to download and preprocess data in timely efficient way with use of scalable computational, networking and storage infrastructure. Amazon already hosts Sentinel-2 data for ESA (within official bucket) and Sentinel-1 for Alaska Space Facilities. Actually, Amazon Web Services and GEE demonstrate different ways to organization of cloud based geospatial data processing – more flexible with requirements of higher competence from end-users or restricted with easier daily exploitation. References 1. Kussul, N., Lavreniuk, M., Skakun, S., & Shelestov, A. (2017). Deep Learning Classification of Land Cover and Crop Types Using Remote Sensing Data. IEEE Geoscience and Remote Sensing Letters, 14(5), 778-782. 2. Shelestov, A., Lavreniuk, M., Kussul, N., Novikov, A., & Skakun, S. (2017). Exploring Google Earth Engine Platform for Big Data Processing: Classification of Multi-Temporal Satellite Imagery for Crop Mapping. Frontiers in Earth Science, 5, 17. 3. Kussul, N. N., Lavreniuk, N. S., Shelestov, A. Y., Yailymov, B. Y., & Butko, I. N. (2016). Land Cover Changes Analysis Based on Deep Machine Learning Technique. Journal of Automation and Information Sciences, 48(5).

Urthecast's vision to democratise the access to Earth Observation touches ground with our data distribution and exploitation ecosystem, the UrthePlatform. This platform is putting together: - Data sources - an extensive and rich data sources offer (including our current operational satellites Deimos-1 and Deimos-2, public satellite data sets like Landsat and Sentinels, our partner satellites of the Pangeo Alliance, and the future UrtheDaily and OptiSAR constellations); - Geo-analytics - an ever-growing set of EO services and applications specifically designed for quick combination, analysis and extraction of information (including bring-your-own-algorithm capabilities); - Visualisation - complete portfolio of visually captivating presentation options (including graphs, 2D- and 3D-maps of both raster and vector layers); - GIS - close integration with unmatched Geographical Information Systems (GIS); - Marketplace - the possibility to monetise developed services and products; - Cloud - securely running over the world-biggest and most advanced cloud infrastructure; - Community - a massively big community of remote sensing experts, engineers and data scientists developing algorithms and final users consuming Earth Observation imagery and analytics; - Vertical integration - seamless compatibility with technologies used for companies in much larger industries (b2b and b2c model) and sectors like forestry, agriculture, infrastructures, urban planning, defence and intelligence. UrthePlatform provides unhindered and near universal access of EO imagery and data at affordable price point, in formats and on platforms that do not require expertise, within an eco-system that attracts third-party investment and innovation, that significantly broaden the utility of the data for organisations and individuals. This combination of data wealth, engineering effectiveness, expansion possibilities and massive community makes the UrthePlatform a perfect place for both EO and data scientists to deploy their work and make their work truly useful for their institutions and the society This presentation will introduce this UrthePlatform and discuss how its user could benefit from it.