Data e-Infrastructure Initiative for Fisheries Management and Conservation of Marine Living Resources iMarine

Scientific and/or technological quality, relevant to the topics addressed by the call

1.1Concept and objectives

Marine life plays a vital role in the well being of humanity and the entire Earth ecosystem. Wise and judicious management of all relevant resources is of paramount importance to ensure that biodiversity in the oceans of the world remains rich and that all forms of marine life remain sustainable at the necessary levels. Any effort in this direction, however, is severely hindered by extreme compartmentalization and heterogeneity at all levels and stages: the global, national, and local organizations involved; the scientific disciplines enlisted in formulating the relevant problems and participating in solving them; the accuracy of data collected and reported; the formats in which the data is presented; the models and other software used to analyze the data; and several others. The main goal of the iMarine project is to support the application of the principles of the Ecosystem Approach (EA) to fishery management and conservation of marine living resources through the establishment and operation of a data infrastructure and, in so doing, to facilitate the emergence of a unified Ecosystem Approach Community of Practice (EA-CoP). This will contribute to sustainable environmental management with invaluable direct or indirect benefits to the future of our planet, from climate change mitigation and marine biodiversity loss containment to poverty alleviation and disaster risk reduction. To reach its goals, iMarine has established three main objectives and has organized its efforts around them: First, establishment of a project board of community representatives that will define a sustainability-driven data infrastructure governance model and will formulate organizational and technological policy recommendations. Second, management and operation of a data infrastructure that will offer user-level and application-level services that support the above policies and realize functionality (especially data and knowledge integration functionality) that is required by EA-CoP to step up its EA implementation in line with the deadlines and targets adopted by the 10^th Conference of the Parties of the Convention on Biological Diversity in Nagoya, in October 2010. Third, extension, adaptation, and deployment of a rich set of software components that implement the above services. The results of several European and other projects will be the starting point for iMarine.

Europe will emerge as a two-fold beneficiary of this effort. First, European scientists, managers, policy makers, and other stakeholders will play a central role in the EA-CoP to emerge, e.g. involving members of the International Council for the Exploration of the Sea (ICES) and the Regional Advisory Councils (RACs). They will exert significant influence on the data-related policies to be formed and recommended, and will be the initial users of the resulting data infrastructure, taking advantage of unprecedented computational power and information richness and addressing their problems with several new methodologies thus afforded. Second, their counterparts around the globe are likely to adopt the policy recommendations made, will be drawn to the data infrastructure for their own work, and will form other such infrastructures locally based on the techniques developed in the project, if not the iMarine data e-Infrastructure and software itself. This will increase and reinforce the European leadership in the relevant political and technological arenas as well as the impact and global relevance of the iMarine data infrastructure (and its ancestor infrastructure) as a unique platform for supporting science as well as fisheries and environment policy development.

Key Concepts

There are four main conceptual pillars on which the iMarine project is built:

(Data) e-Infrastructures: Electronic infrastructure are environments where various resources (hardware, software, middleware, networks, and data) are integrated and can be readily shared and accessed by user communities for effective collaboration in scientific research, policy development, and many other societal activities. Data infrastructures are special cases that focus on their data resources and offer advanced services dealing with data integration, management, discovery, analysis, processing, presentation, etc.

(Data) e-Infrastructure Federations: Interoperation of existing, autonomous, heterogeneous (data) e-Infrastructures lead to the creation of (data) e-Infrastructure federations, where the individual resources are harnessed and aggregated for a multiplicative effect on information availability, processing power, and scientific perspective.

Ecosystem Approach (EA): According to the Fifth Conference of the Parties of the Convention on Biological Diversity (CBD) in 1992, the Ecosystem Approach is a strategy for the integrated management of land, water, and living resources that promotes conservation and sustainable use in an equitable way. Its operational meaning is better specified in the 12 “Malawi Principles” adopted by the CBD in 1998.

Community of Practice (CoP): According to cognitive anthropologists, a community of practice is a group of people of various backgrounds (e.g. scientists, practitioners, managers, fishers) who share an interest (i.e. the implementation of the EA), a craft, and/or a profession. The group can evolve naturally because of the members' common interests, or it can be created specifically with the goal of gaining specialized knowledge. Through sharing information and experiences, the members increase their knowledge, improve their performance, develope a body of best practices, and develop themselves personally and professionally.

Background and Detailed Motivation

ong tradition. It rests on a solid scientific foundation and a network of nested national, regional and global institutions. The people that interact in the assessment and decision process come from a broadening set of horizons (scientists, managers, lawyers, industry leaders, unions, fishery operators). Interconnecting these actors around common knowledge-building, fostering e-meetings, developing of knowledge repositories (case studies) and e-training facilities would accelerate its emergence.

An EA-CoP may include, intersect with, a number of smaller groupings of interests such as ecosystem modelling, ecological economics, small-scale fisheries, spatial planning, fisheries assessment, coral reefs,, MPAs, marine debris, ecolabelling, large pelagic fish, seamounts, fishing rights, community-based management, common property, etc., presently loosely connected or unconnected and uncoordinated. These may have developed specialized websites but they are generally not supported by any formal infrastructure for data and information sharing and integration. The fishery and environmental scientists interact quite often, use common concepts, approaches and methods and are developing some consensus about the state of the fisheries and their environment, the problems and the solutions potentially available to resolve them. Formal publications represent a powerful but still too slow instrument of exchange and “integration” and the important grey information (including policies, plans, legislations) is hardly accessible.

The various stakeholders of fisheries and conservation meet at national, regional and global levels, e.g. in inter-ministerial mechanisms, Regional Fishery Management Organizations (RFMOs), Regional Seas Organizations (RSOs), the European Advisory Committee on Fisheries and Aquaculture (ACFA), the Regional Advisory Councils (RACs), the FAO Committee of Fisheries (COFI) or the UN General Assembly and related processes. None of these components is homogenous and the level of exchange within and between the potential components of an EA-CoP varies greatly between countries and between the developed and developing worlds. Constituting an EA-CoP is therefore an opportunity, a challenge and a necessity.

Motivation for the Data e-Infrastructure: To progress in its implementation, the EA needs innovative technological solutions to foster cohesion among its members, support the multidisciplinary tasks required, and greatly improve the process of social learning while reducing costs. The requirements are particularly challenging. The implementation relies on a set of knowledge and data sources much broader than that used respectively in conventional and fishery management and conservation. In both arenas, the EA requires that monitoring and assessment of target, emblematic, or vulnerable species be broadened to cover species assemblages, communities, habitats, and ecosystems, and that socio-economics be also broadened to cover fisheries’ impacts on all goods and services offered by those ecosystems.

The EA also demands better availability of close to real time up-to-date information, much more easily analysed, and in shorter timeframes. This information is maintained in a large number of repositories, archives, and databases, either institutional or personal. Thus, these data resources have different histories, and respond to different policies, practices, and standards. From the technological point of view, they are represented in a large variety of formats, which range from simple textual documents, to images and other multimedia content, time series, geospatial data-products, code-lists/hierarchies, etc. They are processed using a range of specialised tools for data management, analysis, publication, mapping and plotting, reporting, etc. The overall result is that these multi-disciplinary data sources are scarcely compatible and can be made interoperable with difficulty, hence their combination remains an ad hoc and painful process.

In recent years, several efforts aimed at aggregating and integrating existing data sources of interest to EA have been launched. Particularly significant in this context are some initiatives to document European and global biodiversity (several of which supported by the EU), like Species 2000 Europa and i4Life, aimed at building the ultimate register of taxonomic names, and PESI, providing standardised and authoritative taxonomic information. Several initiatives collate biodiversity information on a global scale: OBIS is a global network assembling and integrating biogeographic data; WoRMS and the Catalogue of Life aim to compile reference lists of taxonomic names; the Encyclopedia of Life is a portal pulling togetherall freely accessible information on species. Regarding Fisheries data, one can cite the FAO Fisheries Global Information System (FIGIS), and IRD’s Ecoscope. The EA process, however, is not benefiting as it should from these excellent initiatives. Its comprehensive application in both fisheries and conservation is still slowed down by the limited amount of information exchange between disciplines, institutions and areas. In addition, data heterogeneity is too high for single applications and data quality is often inadequate to produce reliable analyses and assessments. These resources are often incomplete, show discrepancy or distortion introduced by various actors along subsequent layers of data aggregation. Finally, in many cases the information production process is too slow and decision makers do not have access to up-to-date information.

For all the above reasons, it is imperative that an open, dynamically-expandable, and well-regulated data infrastructure is established, operated, maintained, and enriched to provide the necessary support to a concrete implementation of the EA through an active EA-CoP.

iMarine Objectives

Main Objective

To support the application of the principles of the Ecosystem Approach (EA) to fishery management and conservation of marine living resources through the establishment and operation of a data infrastructure and, in so doing, to facilitate the emergence of a unified Ecosystem Approach Community of Practice (EA-CoP).

Relevance to the call: “Progress towards the vision of open and participatory data-intensive science.”

Instrumental in the work of iMarine towards the above goal will be the establishment of a rich set of collaborations with other EU and international organizations, initiatives, and projects as well as national centres of excellence. Existing policies, technologies, and infrastructures will be reused and rendered interoperable by iMarine. By leveraging these collaborations and by taking advantage of additional funds that some of these organizations are willing to invest in the project activities, the number of resources available to the entire effort will be maximized, the e-Infrastructure deployment process will be speeded up, and the overall cost of building and operating the infrastructure will be reduced. To achieve its main goals, iMarine has established three main detailed objectives and has organized its efforts around them.

Detailed Objective #1: Policy Development

To develop community-driven data-related policies on governance and operation of a data infrastructure, sharing of data and other resources, and processing data to support the EA-CoP in implementing EA to fishery management and marine living resource conservation.

Relevance to the call: “Progress towards the vision of open and participatory data-intensive science.” “Increase of the user base and bridging across disciplines, enabling of cross fertilization of scientific results and stakeholders knowledge and favouring innovation, improving the knowledge elaboration processes in support to decision-making”.

The main instrument for policy development will be the iMarine Board, composed of two bodies:

• 
  a main body called “Steering Board”, comprising info system specialists, managers-“mediators” between users and technologists, and representatives of data-holding institutions involved in EA, e.g., FAO, FIN, GENESI-DEC, ICES, UNESCO (IOC-OBIS), IRD, NEAFC, VLIZ, Species2000.
  
• 
  a subsidiary body called "Advisory Council", comprising high-level representatives of selected EA initiatives, as the users of the infrastructure, e.g., FAO’s EAFNet, the 2010 Galway Conference follow-up group, the “High Seas, Deep Seas Fisheries” discussion Group, the Global Oceans Biodiversity Initiative (GOBI), or the NSERC Canadian Capture Fisheries Research network.
  

The primary goal of this Board will be to define the data infrastructure governance model, with a sustainability focus, and to formulate a set of organizational and technological policy recommendations regulating the resources shared and services provided by the infrastructure. In particular, the policies developed by the iMarine Board will define plans, methods, and guidelines regulating the management and operation of the infrastructure and the development and deployment of its services with the aim of promoting the sharing of data, application and other resources, and the development of trust. These policies will be identified by analysing the needs of the community and the recommendations issued by leading organisations at international level. Particular attention will be dedicated to policies related to the adoption of emerging standards (e.g., SDMX for statistical data, OWL-RDF for concepts, COST for fishery sampling data, Darwin Core for species distribution data) and their possible application to the representation of observation records, provenance information, quality assessment, harmonisation, etc. that promise to largely facilitate iMarine data exchange and usage. Policies will also tackle issues of sharing within data confidentiality requirements in an attempt to promote use of sensitive data at acceptable aggregation levels.

The iMarine Board will collaborate with other major boards in the sector to validate and promote its recommendations within the community at large. Given the leading role of the iMarine Board member organizations it is expected that the policies defined will have a large impact not only on the services made available by the iMarine data infrastructure, but also on those developed by all those interacting with it to form an even further expanding iMarine data infrastructure federation.

Detailed Objective #2: Data e-Infrastructure Management and Operation

To manage and operate a data infrastructure that will serve the application-building and application-using needs of the EA-CoP members. To establish this data infrastructure by largely leveraging on resources and services from existing infrastructures.

Relevance to the call: “Increase of the scale of federation and interoperation of data infrastructures, better exploitation of synergies with the underlying e-Infrastructures, reduction of costs”

The iiMarine data infrastructure will operate as a facilities/commodities/services provider to support the analysis and decision processes in the entire data production-consumption life-cycle of the EA-CoP, simplifying and expediting the exchange, consumption, and analysis of cross-domain data collections and improving the quality of decision-forming data. In particular, through its long-standing collaboration with Eurostat and in coordination with other stakeholders of the EA-CoP, FAO has established several business cases that will be supported by the iMarine data e-Infrastructure. These range from statistical management of socio-economic data and specialized portals for information management of aquatic ecosystems and monitoring of vessels, fleets, and activities, to specialized data management, transformation, and monitoring. Each business case will benefit from the applications made available for the others, and the EA-CoP as a whole will inherit a rich set of facilities to support its business. Third-party users (members of the EA-CoP) will be exploiting these facilities programmatically or through innovative applications on-demand, reducing the effort and investment in ad-hoc implementations, operational and maintenance tasks, and provision of the necessary resources. The e-Infrastructure will be operated, monitored, and maintained as a 24/7 service based on policies established by the iMarine Board.

As an open environment, the iMarine data e-Infrastructure will profit from existing infrastructures and available services & other resources. It will be based on the technology developed for the D4Science infrastructure federation and will actually interact with it as well, thereby accessing resources managed and orchestrated by EGI, GENESI-DEC. VENUS-C, FARM, AquaMaps, FIGIS, Geonetwork, OpenAIRE, and Driver,It will also interoperate in a systematic fashion with other information sources maintained by major EA authoritative organisations, e.g., UNESCO (IOC-OBIS), FAO (TechDataCDR), ICES, NEAFC, IRD (Ecoscope) and projects, e.g., i4Life and PESI, but also commercial computing and storage resources provided by private data centers (ENG at Pont Saint Martin, Italy) or even public clouds (Amazon AWS or Microsoft Azure). The breadth and depth of the data infrastructure thus created will ensure exploitation and broad take up of the iMarine e-Infrastructure and its long-term sustainability.

Detailed Objective #3: Service Enrichment and Deployment

To extend, enrich, and deploy a broad set of services that exploit, harmonize, improve the quality, and integrate the large variety of heterogeneous data required in applying EA to fishery management and conservation of marine living resources, .enable new cost-effective applications aidng the production of knowledge required by policy makers, and support the policies established by the iMarine Board.

Relevance to the call: “The generic tools and services developed under the infrastructure part of the programme can be used for the further development of research infrastructures in Europe and in particular for the implementation of clusters of ESFRI projects”. In particular, this proposal is particularly relevant for the ESFRI Environmental Cluster. This suite will include support for data generation, provenance, quality assessment, certification, curation, annotation, navigation and management of data. Mechanisms for facilitating interoperability across- infrastructures, not only at the level of metadata but also at the level of data, will be introduced. These will exploit standards, ontologies and new mediation and harmonization approaches.

To support the business cases established, the project members as well as external members of the CoP will deploy all the necessary services for the iMarine data infrastructure. These will be obtained primarily by relying on, integrating, and enriching open software developed in the framework of other projects funded by EU and EA-CoP Relevant examples are gCube and OpenSDMX. Developed in the context of the Diligent, D4Science, and D4Science-II projects, the former makes available a large variety of services for managing, manipulating, and processing data and metadata within an autonomously-managed infrastructure. It also provides a suite of generic service frameworks for supporting data production-consumption life-cycle. An initiative of FAO, OpenSDMX is backed by the FIGIS and FAOSTAT infrastructures and offers components and a web application for producing SDMX (Statistical Data and Metadata Exchange) data and metadata in a RESTful way. Furthermore, exploiting the synergies between FAO, IRD, NEAFC, and ICES, additional services will facilitate the export, use and diffusion of statistical data according the new SDMX standard, completed by services for heterogeneous source harmonisation and discovery, for observation records with confidentiality requirements (e.g., COST standard), and for interactive mapping.

New services and specific process workflows will be designed to solve major pending issues, like heterogeneity, low quality, lack of standardization, and lengthy and demanding production phases, especially services facilitating the life-cycle of the socio-economic data resources of the EA-CoP that are used today in producing reports for policy makers. Despite their importance in the management and conservation processes, the tools available to operate on such data types are often limited in functionality, isolated, and not interoperable. By overcoming these limitations for data collections organised as statistical time series, registers, and geo-referenced records, new forms of automatic data quality improvement will be enabled.

Bringing high-resolution environmental data within the same framework as biogeographic, distribution data on species will facilitate better modelling of species distributions, and novel approaches to quality control of the existing bio-geographic information. Algorithms will be developed to automate the detection of outliers in environmental space. Several algorithms modeling species’ distribution ranges will be made available and will operate on the same data; this will constitute both a refinement and an expansion of the Aquamaps Virtual Research Environment operated within the D4Science data infrastructure federation. Tools developed by OBIS and the PESI and i4Life projects will be used to harmonise taxonomy and nomenclature with global registers, such as Catalogue of Life and WoRMS. Additional data flow from IRD Ecoscope will enrich the above knowledge about species distribution.

The services developed aim beyond the goals of the project. They will also enrich the exploited open source software initiatives with enhanced versions and new software packages. In so doing, they will be also enrich the suite of open source data services available to ESFRI projects, e.g., ESFRI environmental projects, which often have to deal with a level of data heterogeneity similar to the one addressed by the iMarine project.