Energy SCOUT is a commercial business intelligence service developed by Terrabotics in collaboration with the European Space Agency (Business Applications Centre), following a successful kick-start project in 2018/2019. Energy SCOUT stands for Energy Supply Chain Observation & Understanding Tool and uses cutting-edge satellite remote sensing & advanced image data analytics to shine a new light on oil & gas and energy supply chain activities around the world.
Energy SCOUT information products are built around user needs and requirements captured thorough rigorous engagement with a representative selection of end users. End users were chosen from identified customer segments and actively engaged through a pilot program. Energy SCOUT products are intended for commercial application by end users such as oil & gas operators, private equity firms, ESG investment groups, and commodity traders. The ultimate goal of the Energy SCOUT products are to enable better decision making based timely and unbiased information, underpinned by Earth Observation data.
Energy SCOUT products have been built by leveraging open-source satellite data. The products provide reports on a weekly basis over areas on the order of tens or even hundreds of thousands of square kilometres. By combining raw satellite data of various spatial, temporal, and spectral resolution together with technologies like Artificial Intelligence, Computer Vision and Machine Learning, activities and changes can be detected fast and information can be provided routinely. The radar, optical, and thermal satellite sensor that were used to develop the modules are the Copernicus Sentinel 1, Sentinel 2 and NOAA VIIRS respectively.
Energy SCOUT is a modular set of data services, employing feeds of open-source satellite and reported in-situ data and transforming them into value-added information. The information services target different activities along the oil & gas supply chain. The Energy SCOUT modules have been released in stages since 2017.
Currently available modules include ShaleAnalystTM and FlareAnalystTM, which launched as a commercial pilot in July 2021. The ShaleAnalystTM module is a shale well-level activity and event tracking tool, that provides information about well pads, water ponds and lease roads, as well as monitoring oil & gas well pad lifecycle on a weekly basis using optical and radar satellite data. The FlareAnalystTM module is an Oil & Gas flaring activity monitoring tool, that delivers reports about flare activity detected by satellites grouped by operators on a monthly basis.
Energy SCOUT roadmap modules include various tools, for example TBX Basin Production Index as an oil & gas production proxy index, EmissionsAnalyst for field/facility emission monitoring, RefineryAnalyst for oil refinery activity monitoring, TBX Pipes for pipeline activity monitoring or ESG Observer & LandMetric for rapid site & biodiversity evaluation.
With funding from the European Space Agency’s Business Applications programme, Earth-i is exploring the integration of multi-sensor and multi-operator satellite data and the application of advanced AI technologies to benchmark, monitor and forecast supply chain activity. A range of satellite and in-situ data are being routinely collected and processed to understand production levels and environmental impact of global commodity supply chains including copper, nickel and steel production. Early results have been extremely promising, and Earth-i is now actively working to develop the operational proposition and to commercialise the capability.
The products are being developed initially for the financial services markets, and for this Earth-i is working with Marex, a world leading UK-based commodity broker who provides clients with an unrivalled trading platform and breadth of coverage in metals, energy and agricultural markets. Initial customers of the service range from physical traders to hedge funds, quantitative funds and commodity research organisations. For non-earth observation companies like these, the challenges of identifying, procuring, processing, analysing and integrating satellite data are too significant, and so the benefits of satellite derived datasets are failing to be taken up, despite their significant value in answering complex supply chain questions.
The supply chain data does not need to be 100% accurate, as they are not the sole information source used for decision-making: sometimes only a notification of change of direction or speed is enough to trigger further investigation. Conversely, significant events of this kind can only be detected by monitoring hundreds of locations at a high temporal frequency to identify significant changes or abnormal behaviours, be it at global, regional, countrywide or individual site level. They key is to be able to detect changes over time at key nodes of each supply chain and in the environment surrounding those nodes, and to create absolute and relative measurements of those changes.
With a multi-year historic time-series of high-revisit images, such as those now available from satellites including the Copernicus Sentinels, it is possible to gain a deeper understanding of supply chain activity, productivity and environmental impact over multi-annual periods. A multi-sensor approach also minimises the limitations of relying upon a single data type, by using optical satellites for object identification and change detection, SWIR and IR for thermal feature and change detection, complemented by SAR and InSAR to ensure coverage during cloudy periods or for night-time observation. The combination of sensor types gives good primary and secondary indicators of activity to a high level of confidence, while the multi-operator strategy limits the reliance upon any single source of data whilst increasing the data capture opportunities.
With the application of machine learning techniques, it is possible to set up automated processing chains to detect change over time to a high level of accuracy, whilst keeping the manual quality control burden down to minutes on a daily basis. Such automated detection helps to provide a consistent level of accuracy that allows hundreds of sites to be analysed and published daily within a couple of hours of data receipt. Machine learning also unlocks the value of data not visible to human eye within high resolution and medium resolution data sets. This has a significant commercial benefit in keeping the service affordable, unlike services based upon VHR data only.
The products generated are of significant interest to the financial services sector not only because of their ability to help inform of likely movements in commodities and equities market prices, as well as acting as leading indicators for macro-economic performance, but also because they act as an independent evidential benchmark of the commodity asset owners on their net zero and environmental sustainability targets and provide a means of measuring the impact of their ESG investments over time.
In this presentation Earth-i will demonstrate the latest developments from the project, including several examples of the high-revisit, multi-sensor approach, the machine learning analytics applied, and the commercial products generated for financial and regulatory institutions to exploit, and how the work is set to drive transformational change in the industries being monitored.
The goal of the AquaLedger project is to demonstrate the integration and connection between Distributed Ledger Technology (DLT)/blockchain and earth observation (EO) technologies in the sector of supply chain management. More specifically, automated earth observation based analytical processes will be employed to provide valuable information about the tracking and tracing of fishery and aquaculture commodities and/or the verification and appending of transactions recorded in the blockchain. Thus, the activity aims to advance current knowledge and expertise in DLT and EO convergence resulting in an integrated system that will be demonstrated in real life conditions while also addressing the needs of the relevant actors in the supply chain management of the aquaculture sector. The implemented system is validated in the context of near-shore aquaculture activities.
A sustainable supply chain can be defined as a process of providing services to the aquaculture users during a way that fosters continuous and reliable water quality information and indicators. Water quality remote sensing constitutes a fast-growing field spurred by major investment into satellite observation capabilities. In aquaculture, water quality monitoring is required not only for sea farms but also for hatcheries that pump water from the ocean, especially because of certain species (i.e. fish larvae) that are very sensitive to water quality issues. The quality of the water where the aquaculture activities take place, affects the final products, jeopardizing the security of the food that reaches the consumer. However, despite its undeniable value, this type of information is not usually tracked and monitored during the supply chain of the food.
Thus, the main use cases of the implemented system involve i) employing EO based services towards the provision of accurate information about the water quality of aquaculture assets as well as the farming productivity ii) showcasing how EO data can connect the physical environment to the distributed ledgers and be utilised for validating the conditions of transactions between actors in the supply chain and iii) monitoring the compliance of the food movement in the supply chain
Among the abovementioned water quality parameters, sea surface temperature is a key factor because it influences the variation of many physical and biochemical seawater properties as well as the entire life cycle of the fish. Sea surface temperature is monitored through the use of the Sea and Land Surface Temperature Radiometer (SLSTR) of Sentinel 3 with a spatial resolution of 1 km. Additionally, turbidity, chlorophyll-a and nutrient concentration are very important parameters for health and growth whilst the monitoring of pollution factors is essential for preventing their catastrophic impact (i.e. high mortality rate) on fish farms. Various algorithms (empirical, semi-empirical and analytical) were tested to generate the essential water quality information from atmospherically corrected Sentinel 2 MSI and Sentinel 3 OLCI data which have been widely used in the context of water quality monitoring studies. The most appropriate algorithm was selected through the use of field observations collected at the fish farming site. In the context of this application case, the Case 2 regional Coast Colour (C2RCC) processor and empirical algorithms were utilized for the acquisition of water quality parameters from Sentinel 2 MSI. The sea skin temperatures values extracted from Sentinel 3 SLSTR were also considered accurate if prior outlier detection and removal is performed. It must be noticed that the Sentinel 2 data were able to extract valuable water quality information both with analytical and empirical algorithms. Sentinel 3 OLCI data did not achieve competitive results given that it has a limited performance when it comes to coastal areas or similar ecosystems laying within a short distance from the land.
Last but not least, in order to serve the AquaLedger application cases for seafood supply chain, different DLT/Blockchain platforms have been analysed towards the realisation of the most effective solution. External events being provided by EO component and other external data sources (i.e. Enterprise Resource Planning software) are incorporated in the DLT system through the realisation of web-services.
Fisheries are the major activity exploiting marine resources and have a significant impact on marine life as well as on the economic development and food security of nations. Nowadays, more and more consumers are demanding sustainable and responsible fish products and the comprehensive monitoring of fishing activity provides new insights into improving the transparency of the fish supply chain. While policies and regulations attempt to control and manage the origins of fish, however, the traceability of the fish supply chain presents many obstacles concerning the data availability.
Deimos and IPMA (Instituto Português do Mar e da Atmosfera), with the support of the Portuguese Directorate-General for Natural Resources, Safety and Maritime Services (DGRM), are developing the pilot application “Monitoring Fishing Activity” for the e-shape project (Horizon 2020, Grant agreement ID: 820852). This is a unique initiative that brings together decades of public investment in Earth Observation (EO) and cloud computing resources to develop services for decision makers, citizens, industry and researchers, always in close collaboration with key users.
The main goal of this pilot is to strengthen the knowledge on the fish supply chain by developing an operational EO service that monitors the dynamics of vessels operating in the Northeast Atlantic waters, focusing on two fleets involved in pelagic fisheries of highly migratory oceanic species: the pole and line and the drifting longline fisheries targeting tuna and swordfish respectively. Using data from the 2012-2018 period recorded within the Portuguese ZEE, the results aim to characterize and quantify the fishing pressure of these fleets on marine ecosystems, and to relate fisheries with environmental parameters. A further objective is to raise awareness among key users of human footprint in marine ecosystems and motivate them towards more efficient, environmentally friendly and sustainable fishing strategies and practices, complying with international, European and national regulations.
Key organizations involved in the fisheries monitoring at different levels have been providing different sets of requirements based on their needs. These organizations range from public institutions as DGRM, the central service of the Portuguese State for the administration and implementation of policies on fisheries and the Regional Directorate for Fisheries for the Azores region (DRPA), academic centres as the Centre of Marine Sciences (CCMAR, one of the foremost marine science research centres in Portugal, and non-governmental organizations as Sciaena, whose mission is to promote the implementation of the sustainability of fisheries.
Fisheries management is continuously frustrated by the lack or poor quality of critical data on fish catches/ sizes, fishing locations, and relevant environmental conditions. While quantitative methods for managing fisheries have developed considerable complexity, the quality and availability of data remains an obstacle to meaningful advances in fisheries management. The integration of information technology will play a major role in fisheries management and end-to-end traceability of the fishing activities. To that end, the application currently being developed integrate the following key datasets:
- Automatic Identification System (AIS) data for fishing vessels. AIS is a surveillance solution that uses transceivers on ships which can be tracked by terrestrial stations across the coastline or with satellite receivers. Information provided by AIS, such as unique vessel identification, position, course, and speed, can be displayed on an electronic chart and/or ingested into an information system.
- eLogBooks.Electronic records of catch and effort registered at the time of the catch operation that need to be reported to National Fisheries Control Authorities. Data transmission to authorities can be at the time of landings or immediately after the fishing operation has been concluded and catches recorded. Logbooks are widely used as a method of collecting statistical information on commercial activities.
- Sales. Often referred to as landing declarations, reporting the weight and value of the vessel sales by year and fishing port.
Based on those key datasets, three different products are provided to users within the Monitoring Fishing Activity pilot:
- Fishing Trips. Trajectories of fishing vessels from the Portuguese pelagic longlining and pole and line fishing fleets based on satellite AIS points datasets. From the analysis of the speed profiles of vessels, additional information is provided for the possible activity status in each AIS emission data point - fishing and non-fishing.
- Fishing Footprint. Geographical distribution of the fishing activity in a period of time. The results are based on the aggregation of the AIS data points that were identified as fishing over a certain grid.
- Fishing sales. Yearly sales in quantity (ton) and percentage rate of target species per port. Sales reports of the anonymized vessels and the weight amounts are aggregated according to the fishing fleet, year, fishing port, and target species.
During the first half of 2022, Deimos and IPMA will add environmental characterization of fishing areas to the service portfolio. The proposed service takes advantage of state of the artf EO-based datasets that provide key parameters such as Chlorophyll concentration, Sea Surface Temperature and others, correlating those with tuna and swordfish catches reported in elogbooks. This shall allow a better understanding of the behaviour of those fishing fleets and how they align with the environmental conditions of fishing grounds.
Moreover, a major purpose for this project is to disseminate the outcomes of these services through a web-based tool following rigorous access criteria for the different users, to be established by IPMA and DGRM. In that regard, Deimos is currently working on integrating the services presented here into the NextGEOSS platform, an integrated ecosystem for users to access EO data and deploy EO-based applications.
The United Nations Economic Commission for Europe (UNECE) has led a process to develop a free opensource standard and system to track and trace products through the entire garment and footwear supply chain – from field to factory to shop floor, enabling companies to make verifiable sustainability claims that consumers, governments and regulators can trust. This short presentation will focus on the UNECE Toolbox, including Policy, Standard and Guidelines as well as it’s work on Blockchain Pilots for enhance transparency and traceability in the garment and footwear sector.
There is an increasing need to track food from the source to the end stages of consumer outlets, primarily in order to ensure the safety and quality of the produce. Implementation of the European Green Deal strategy will require improved due diligence of deforestation-free commodity production, sustainable fisheries, fair trade, organic foods, carbon emission reduction from supply chain activities, etc.
To date, knowing whether or not a company’s supply chain is sustainable, and is not related to e.g. deforestation, remains a challenging task, because of either the lack of supply chain data on environmental factors, or because of limited visibility into supply chain asset data relations. Lastly, an equally important factor are the linkages between the environmental data and the supply chain data.
In terms of traceability, there are significant new technological opportunities in the context of “tracking and tracing” of agriculture commodities, in particular by the convergence of traditional EO information services with emerging distributed data, platforms and IoT sensor networks.
The lack of data on environmental factors has been largely solved due to advances in cloud computing, allowing the low cost processing of satellite imagery at scale. Satellites acquire petabytes of earth system imagery every day. This data contains valuable information on supply chain sustainability and transparency waiting to be uncovered.
In this session we highlight the possibilities of using new traceability technology together with satellite data for end-to-end supply chain due diligence and supply chain management. We will provide case study results with a special focus on deforestation due to the pending EU Regulation to mimimise EU-driven deforestation and forest degradation. Specifically, we focus on the latest practical challenges and opportunities of Distributed Ledger Technologies as a typical hyped "newtech", OCG standard setting and encrypted analysis on sensitive commodity sourcing geodata, in relation to its use in agricultural commodity supply chains in combination with satellite derived business insights.