Understanding what is occurring on our oceans is of critical importance globally. Whether it be monitoring shipping on behalf of global markets, timely awareness of emergencies at sea, determining environmental hazards, tracking illegal activities, or gathering intelligence regarding potential threats to national security, access to timely and accurate understanding of what is happening in our maritime environments – known as Maritime Domain Awareness (MDA) is essential.
However, the oceans remain one of the last ‘wild west’ frontiers. Once ships leave port, our ability to track their activities becomes severely limited.
Today’s MDA capabilities are heavily reliant on shore-based surveillance, ship self-reporting, an Automatic Identification System (AIS), and space-based AIS (S-AIS) data sources. While AIS and S-AIS have been extremely beneficial in providing improved MDA capabilities on a global scale, these technologies suffer from a variety of issues including gaps in coverage, data latency, errors in track association, and increasingly deliberate AIS spoofing or interference. Also, the majority of small vessels do not carry AIS transponders and are therefore invisible to AIS-based systems.
As a result, current MDA systems cannot provide timely, complete, and trusted awareness of maritime activity off our coasts. To improve MDA, commercial and Government maritime stakeholders need to leverage new and emerging technologies to improve persistent, prompt, and accurate ship detection, identification, and tracking.
To these ends, several technologies hold significant promise for improving MDA, namely commercial Earth Observation (EO) systems, vast quantities of open-source data, and current trends in big data analytics.
Commercial Space-Based EO Data
Commercially available EO data shows great promise in improving overall MDA. Combining S-AIS data with space-based Synthetic Aperture Radar (S-SAR) data has already been shown to improve MDA. SAR satellites are known for their ability to image large swaths of the oceans, and can “see” day and night, through clouds, and during inclement weather. Canada is at the forefront of space-based MDA. Data provided by Canada’s RADARSAT-2 satellite is routinely used to characterize sea ice, improve ship and wake detection, discriminate icebergs from ships, provide pollution monitoring, and estimate wind and sea states.
However, SAR satellites are few in numbers and hence lack persistence. But as constellations grow, so does area coverage and re-visit rates. The key strength of SAR sensors is their ability to detect non-cooperative vessels, namely those not transmitting on AIS or not self-reporting through other means, making it the “gold standard” for vessel detection. Although Canada’s RadarSat Constellation Mission is not a commercial asset, its capability to conduct MDA is poised to dramatically change our understanding of global ocean traffic as we know it.
Similarly, large optical systems (e.g. Digital Globe) have already demonstrated their utility for MDA. They bring a data set that complements other information sources and helps build a much more comprehensive understanding of vessels and other objects of interest.
The Earth Observation Explosion
There are hundreds of optical SmallSats planned for launch in the next decade. As examples, Planet (Terra Bella / Skybox14) currently has a constellation of over 200 optical satellites in orbit, BlackSky Global plans for a constellation of 60 optical, 1m resolution satellites by 2020, and China’s JiLin plans to launch 60 optical satellites by 2020 (16 of them between 2018-2019) and 137 by 2030. These large constellations of small optical satellites could dramatically transform the nature of Maritime Domain Awareness. These systems will provide excellent timeliness, some claiming to revisit rates as low as 10 minutes – and may (in effect) enable near-continuous, tactically-relevant coverage for MDA decision making, despite the current limitations imposed by daylight and local weather conditions.
Traditionally, S-SAR systems are not classified as Smallsats, however, the initiative by ICEYE to develop small EO radar satellites has the potential of transforming the MDA world. If they are successful, large constellations of radar satellites that can operate effectively at all times could provide literally continuous coverage of vessels of interest – an asset never before seen in MDA systems. Other entrances into the commercial SAR arena include companies such as Capella Space, with a planned constellation of 36 S-SAR satellites.
Evaluation of electro-optical (including infrared and hyperspectral sensors), combined with S-AIS and S-SAR data are of particular interest because the addition of electro-optical information is expected to greatly reduce false alarms, improved ship classification and identification, as well as improved pollution monitoring, detection of hazards to shipping, wakes, and small vessels (including those from thermal signatures).
In addition to traditional single-sensor missions, the near future may see multi-sensor and integrated mission concepts transforming MDA, with Canadian companies leading the way. For example, NorthStar is in the planning stages of developing a 40-satellite constellation hosting both hyperspectral and infrared sensors. Urthecast plans to launch a 16-satellite constellation of optical and multi-band radar sensors using two orbital planes. Both of these systems promise to provide unique capabilities for MDA.
However, space-based MDA does not end with satellites and the data they produce. Valuable MDA data is available through a plethora of open sources, and vast amounts of data need to be integrated into a coherent MDA picture.
Open Data Sources
Open Data is any data that can be freely used, reused, and redistributed by anyone – subject only, at most, to the requirement to attribute and share alike. Open data sources (such as social media) may also be able to provide the information necessary to improve MDA. Mobile applications are being developed to allow and encourage private citizens to monitor AIS feeds and report illegal fishing and other suspicious maritime activities to the local authorities.
Open data can be a significant source of information to maritime authorities. It can assist in framing the operational picture by revealing suspicious activities, behaviours of interest, and nascent or evolving situations. It can also serve as valuable information, imagery, and audio reporting to corroborate and/or reconcile initial reports. Examples of this could be YouTube videos of fishing in protected zones or Facebook postings of improper hazardous waste disposal.
Sources of open data need to be evaluated for their possible contribution to improving MDA. To date, the use of open data sources has not been looked at in a systematic way. Research and analysis of the utility of Open Source Data as a means of improving MDA shows much promise. The unstructured nature of this data makes its integration particularly challenging. Methods of ingesting, validating, and fusing these sources into one useful and effective MDA data feed is an area of important yet nascent research.
Big Data Analytics
Big Data is often characterised as large in volume, is collected and distributed at high velocity, and usually comes in a variety of formats (known as the 3Vs). From this perspective, MDA data derived from EO, S-AIS, and open sources certainly qualifies as Big Data.
Current EO satellites can process enormous quantities of data, yet the operational and scientific communities have only begun to scratch the surface of this potential. Although satellites are capable of collecting vast amounts of information that could provide an integrated and highly accurate MDA picture – practitioners simply cannot find the data they need, often because there is too much to sort through. From this perspective, having too much data that cannot be efficiently exploited is just as useless as having too little data, perhaps even more so.
In the near term (1-5 years), the explosion of EO systems (hundreds of satellites) will dramatically expand the universe of data available well beyond its already astronomical proportions. To take full advantage of the potential that Big Data can bring, and draw out the benefits of new space applications, there is an urgent requirement for the application of Big Data technologies that can extract useful MDA information from large pools of raw data. The emerging generic best practices for Big Data implementations can be usefully applied to Maritime Domain Awareness.
The application of Big Data techniques to MDA comes with some challenges. In particular, making disparate data sets (such as government records, personal data, social media, etc.) available to Big Data applications while also protecting sensitive or proprietary information from misuse or exposure will be critical and must be addressed.
Canadian Concepts of Operations for MDA
The exploitation of current and planned data sources, combined with rapidly advancing space system technology and Big Data analytics, presents a significant information management challenge. The merging of large, disparate data sets into a coherent approach to MDA is far from trivial. It promises a powerful surveillance capability, yet it contributes to the most challenging MDA problem, the ability to extract accurate, complete, and trusted information through the management, analysis, and exploitation of potentially beneficial yet massive data sets. In order to accomplish this, there is an urgent need for a Concepts of Operations (CONOPS) that recognizes the paradigm shifts occurring in data acquisition and analytics.
Little exists in open literature on Canadian (or allied) CONOPS for MDA. However, from our research three priorities emerge: the need to exploit new technologies (especially space-based sensors); more integrated and collaborative information-sharing capabilities within the Government of Canada and with allies; and more timely processing of information to produce a trusted MDA picture. Within these priorities, the following gaps were specifically identified: the lack of persistent wide-area surveillance; the lack of awareness of the location and identification of small vessels; and the poor existing infrastructure for information sharing within the Government of Canada and its allies. If one can find comfort in numbers, these priorities and gaps are echoed in many of the MDA discussions amongst Canada’s allies and trading partners.
It was clear in our research that monolithic, stand-alone MDA systems could not be made flexible or resilient enough to keep pace with technological advances. Any future successful CONOPS must be enabled by High-Level Systems Architectures (HLSAs) capable of ingesting multiple data sources and adapting to advances in technology as required. We believe an appropriate HLSA would consist of an enterprise-level, cloud-based, service-oriented architecture with associated application services. New data sources and technical capabilities can be easily integrated, keeping the system “evergreen.” The HLSA would also support the information-sharing environment necessary for government and international MDA collaboration. The growing trend of Software as a Service (or perhaps in our case MDA as a Service) is a model worth exploring.
The Future
We are on the threshold of the Golden Age of space-based surveillance. A casual review of earth observation companies’ websites finds references to the democratization of space data, unprecedented insight into the health and wealth of our planet, and the ability to understand and save our planet for our children and the generations to follow. These are lofty, and some would say idealistic goals. Our study of future trends in earth observation to support Maritime Domain Awareness suggests they might be right.
The forthcoming explosion in earth observations systems, Big Data analytics, and concepts around HLSAs can and will transform MDA. Canada has a unique advantage as an early innovator of space-based MDA capabilities. Will we leverage our unique knowledge and advanced industrial base to make the next great innovative leap to enable global, persistent Maritime Domain Awareness?
This article first appeared in the Feb/Mar 2018 issue of Vanguard.
Andre Dupuis is President of Space Strategies Consulting Ltd and Maria Rey is Vice-President and Chief Science Officer of Space Strategies Consulting Ltd.