Norway has used SAR satellites operationally since 1998 for open ocean Maritime Domain Awareness in the High North. RADARSAT-2 and the Sentinel-1A and -1B satellites provide valuable data for Norway on a daily basis. Norway has its own national ground segment to be able to download data directly from the satellite to a ground station in Norway with reduced time delay. The user can have the SAR image in hand around 15-30 minutes after acquisition. The ground station downloads data from the areas within the Norwegian Area of Interest, i.e. the Norwegian Sea, the Barents Sea and the Greenland Sea. It is possible for Norwegian authorities to get information from the satellite images to detect vessels, give positions and size estimates.

The Norwegian Defence Research Establishment (FFI) receives both full resolution level-1 GRD (Ground Detected) and high resolution level-1 GRD images in the EW (Extra Wide) and IW (Interferometric Wide) modes of all images received through the national ground segment. The resolution in the EW mode in full resolution is 50x50 m, while it is 93x87 m in high resolution mode. The pixel spacings are 25x25 m and 40x40 m, respectively. The resolution in the IW mode in full resolution is 20x22 m, while it is 88x87 m in high resolution mode. The pixel spacings are 10x10 m and 40x40 m, respectively.

The paper will present an evaluation of the Sentinel-1 images received in the full and high resolution modes. The ship detection capability is analyzed to see how suitable the different modes are for national operational requirements for monitoring of Norwegian waters in the High North. The modes with lower resolution are faster to process and analyze. It is interesting to see if we lose many vessels in the open ocean if we use the modes with lower resolution. We also look into the same situation closer to the Norwegian coast. A comparison of the modes in the open ocean and in coastal waters will be given in the paper. The pros and cons will be weighed against each other. How important is NRT (Near Real Time) delivery, analyzing the high resolution modes, compared to using more time on processing and analyzes with the full resolution modes and then be able to possibly detect more vessels and smaller vessels.

[1] https://sentinel.esa.int/web/sentinel/user-guides/sentinel-1-sar/resolutions/level-1-ground-range-detected

Area D - Board 357

Operational services in the security domain (e.g. counter terrorism, border control, military) need timely and focused information to guide their operational work. The focus is obtained by selecting relevant intelligence questions which can be answered by using sets of supporting indicators. Intelligence officers may be alerted when sufficient evidence is gathered for these indicators. The evidence may be obtained by using dedicated monitoring assists, for example areal imagery (IMINT) and signal intelligent (SIGINT). Another source which is increasingly considered as valuable for this purpose is open source intelligence (OSINT).
Continuous monitoring with open information sources produces enormous amounts of data containing sparse relevant information. Available sources on the internet contain unstructured text (e.g., social media, blogs, news sites), imagery (e.g., photo, video, satellite), or structured data (e.g., traffic information, geospatial information) and more. It is not feasible using human operators for the processing all of the data. Therefore, including these sources in the intelligence analysis requires automatic processing producing answers to the intelligence questions. These answers should be explainable, such that they can be a starting point for further and more directed analysis and intelligence gathering.
We present a conceptual framework for autonomous big-data monitoring that brings together top-down intelligence questions with bottom-up data processing. Autonomous information collectors with scraping, crawling and explaining capabilities are responsible for monitoring specific sources. They launch triggers that can be translated to search cues for other collectors to direct their searching process. Via confirmations or contradictions, the data from multiple heterogenous resources are combined and the resources reinforce each other. Artificial intelligence techniques are applied here to fuse the information of the various sources and to derive evidence for the indicators to eventually produce the possible alerts. As such open big data sources can be included in intelligence analysis.
For above mentioned approach the spatio-temporal context is an essential part of the intelligence analysis and geospatial intelligence (GEOINT) is therefore an indispensable component. Satellite images form an efficient manner to collect such information. The satellite data allow to search for changes which are indicators for activities on the ground. With the increasing availability of high and medium resolution SAR systems in space, persistent monitoring of these activities becomes feasible. The use of SAR time series enables change detection with regular time periods unimpeded by weather influences and therefore allows determination of normal patterns and detection of possible anomalies. The changes are combined with existing geographical information for change characterization. In this way increasing or decreasing activities at airfields, harbors, and trains stations can be monitored. To support the multi-temporal change analysis we describe here an analysis workflow and present results using a test bed using prototype implementation of the algorithms.
With the successful launches of Sentinel 1A/B in 2014 and 2016 a continuous open source for medium resolution SAR data has become available. Vast areas are monitored with revisit times of 6-12 days with comparable imagery in the order of 10 meter resolution. The Sentinel data are in this way a valuable source for geospatial information to be used as an important contributor in the autonomous multi-source security application described above.

Area D - Board 358

The protection of the European and national Critical Infrastructure is of utmost importance to the EU and its Member States, as it considers the security and safety of European citizens, the economy, the ecology, and the freedom to govern. Networks and systems operating in energy, transport, telecommunications, water and waste management are all examples of critical infrastructure.
The European Commission has issued the Directive 2008/114/EC, with the goal to foster the implementation of protection measures in all Members States and to protect the European CI (ECI) that span multiple Member States. The implementation of these recommendations can be done by following several directions, which have been recalled in a recent ESA call for a feasibility study, and are already pillars on which the EU Horizon 2020 programme has pushed the attention of Research, Academy and Industrial players. Among the most relevant priorities, we can quote:
1. monitoring Critical Infrastructure and producing a relevant and continuous risk assessment, by using a manifold of technologies and adopting an “holistic” approach;
2. improving resilience through (a) enhancing situational awareness - (b) improving the emergency management - (c) improving and enhancing the quality of Scenario analysis;
3. adopting maximal CI protection;
4. protecting CI, assets and citizens toward new threats e.g. coming from the cyber sphere.
According to these priorities, it is evident that Earth Observation from satellite, with its spatial resolution and revisit time, can significantly contribute to increase the resilience of the CI; among them, Planetek Italia, ENEA and INGV selected the integrated water cycle management as object for a feasibility study for innovative EO services recently rewarded by ESA: the project SCIRES - Supporting Critical Infrastructures Resilience from Space.
The integrated water cycle management relies on different infrastructures: water basin and impoundment, transport and distribution pipelines, wastewater collection, and as interconnected infrastructures, power distribution networks on which the infrastructures rely on for energy supply, and power plants that exploits the path of the water to produce energy. Two innovative EO services, eventually integrated into the CIPCast system (an existing Decision Support System developed by ENEA) are object of the study: (1) pipeline monitoring service and (2) impoundment stability service.
1. Pipeline monitoring by using EO data, IoT (in-situ sensors) and AI technique
The service aims at identifying precursor phenomena related to the breaking of pipeline, responsible for the losses of water, and subsequent water infiltration in the ground and potentially causing damages to the surrounding infrastructures. The goal is to prioritise the maintenance of the pipelines, basing the priority on the computed vulnerability and on the subsequent risk computed in the CIPCast DSS.
Damages to pipelines may be caused by several factors, coming from the environment (weather, subsidence-causing stress on the ground…) or from the working condition of the network (pressure and velocity of the fluids and relevant gradients…).
Many statistic models have been implemented to correlate the probability of breaks, causing losses, to environmental and endogenous parameters. In this scenario, it is possible to insert an element of innovation based on the concept of Artificial Intelligence and Machine Learning. These technologies will allow, if adequately trained using big data, which the precursor can be. The bigger is the database, the more positive results are reachable. With the support of the user, the number of potential parameters can be reduced, enhancing the likelihood of success of the technical feasibility.
2. The Impoundment Stability service
The service is focused on the monitoring by means of mean ground velocity and time series of deformation (Persistent Scatterers) computed by Rheticus®’ services, integrated with in situ geological and geomorphological data. The objective of the service is to provide a comprehensive analysis of horizontal and vertical surface and near sub-surface deformations and dislocations that affect hydraulic works, such as water impoundments and distribution networks that could be damaged by hazardous events. In particular, landslides (LS), deep-seated gravitational slope deformation (DGSD), and differential ground subsidence (GS) phenomena are taken into account. Both LS, DGSD and GS can damage or threaten the functionalities of the infrastructures itself (e.g. excessive pipeline deformations, differential subsidence close or below hydraulic works), especially if triggered by heavy rains or earthquakes.

Area D - Board 359

Recently, the concept of datacubes has gained substantial traction as a means for presenting massive amounts of data homogenized and aggregated along space and time, thus easing access and analysis substantially for experts, non-experts, and tools alike. Transparent federations across data centers establish planetary-scale integrated access and data fusion on datacubes. However, these new capabilities also harbour some new security challenges, such as:

- The sheer size of datacubes, which typically reaches intot he Terabyte and Petabyte range, makes it unsuitable as a unit of access control; rather, it must be possible to protect parts of a datacube selectively.

- While users enjoy seeing a uniform single information space when accessing a federation, each federation node (i.e., data center) must retain full autonomy about its data offering, quota, billing capabilities, etc. This includes, but is not limited to, configuring which data are accessible to whom, which federation nodes are allowed to send or receive queries (both written by users or as automatically generated sub-requests during federated processing), etc.

- All such access control and billing mechanisms should not impose new complexities to administrators, but should be easy and intuitive to define and automatically monitored by the system without substantial loss of efficiency.

Notably, such requirements apply to both security, public administration, and research environments. For example, ECMWF offers the long tail of its climate variable timeseries for free while access to the most recent two weeks get charged.

In our talk we discuss these challenges and present an approach for enhancing role-based access control in a manner suitable for datacube federations, and look at its implementation in the "worldwide leading" (ESA) datacube engine and OGC datacube reference implementation, rasdaman.

Area D - Board 360