Search

Interoperability has become a key concept amongst scientific communities, in particular with the focus of heterogeneous data discovery and access as well as services discovery. Our work is developed in the context of the Global Earth Observing System of Systems (GEOSS) Interoperability Process Pilot Project and it demonstrates the feasibility of interoperability between different communities. The Global Biodiversity Information Facility (GBIF) provides through its web portal and services a huge amount of primary species occurrence data, acquired and uni- fied from different data providers. GBIF uses standards relevant to the biodiversity community (e.g. Darwin Core, Taxon Concept Schema), as well as newly introduced concepts and interfaces. The Open Geospatial Consortium (OGC) Catalogue Service for the Web (CSW) is a service that can perform discovery of geospatial data and services, returning results in the form of profiled metadata. In particular the ISO Application Profile (AP) defines a particular set of metadata well known in the geospatial community (ISO 19115/ISO 19119). We make GBIF biodiversity data and services available through a CSW ISO AP catalog; this is done by means of a mapping of the GBIF data model and GBIF services to their CSW ISO AP equivalents. Regarding the data model mapping, the main work was to design the mapping be

In this paper the concept of the Siberian Earth System Science Cluster (SIB-ESS-C) being established at the University of Jena (Germany) is presented. SIB-ESS-C is a spatial data infrastructure for remote sensing product generation, data dissemination and scientific data analysis. The prime objective is to enable researchers to extract information on the state of the Siberian environment and the changes that are occurring using Web-based tools. The region under study covers the entire Asian part of the Russian Federation from the Ural to the Pacific Ocean including the Ob-, Lenaand Yenissey river catchments. Taking into account the large extent of this region the focus is on remote sensing data as the primary data source. A key aspect is to create and analyze long-term time series of various environmental parameters derived from Earth Observation data. The SIB-ESS-C infrastructure is to provide the technical means by which remote sensing time series can be created, distributed and analyzed through Web-based tools. The development of the SIB-ESS-C system follows a service oriented architecture (SOA) approach. Interoperable Web services are being implemented based on standards published by the Open Geospatial Consortium (OGC) and the International Organization for Standardization (ISO). The main components of the system comprise a Catalogue Service (CSW) publishing metadata on available data products and services, Feature and Coverage Services (WFS, WCS) providing direct access to existing datasets and Web Processing Services (WPS) for spatio-temporal analysis and visualization.

In the last years REST (Representational State Transfer) has emerged as an alternative to existing approaches for Web applications development. Indeed, its characteristics of scalability and simplicity make it appealing for building complex Web services systems. Anyway an investigation is required to verify if and how it can be applied to specific domains. We analyzed the case of Geospatial services also building a prototype of a RESTful OGC WCS implementation. The WCSplus community discussions were considered as a starting point for our work. REST is an architectural style for distributed systems defined to describe the Web architecture and to guide its future evolution. It allows to describe a set of ResourceOriented architectures which share six constraints: client-server and stateless interaction, uniform interface, caching and code-on-demand (optional) support, multi-layer distribution. The fundamental characteristics of REST is the uniform interface which is defined by four interface constraints: identification of resources; manipulation of resources through representations; self-descriptive messages; and, hypermedia as the engine of application state. It is a generic (not resource-specific) interface, in the sense that it allows to perform the same set of actions on all the resources. Thus it exposes very basic operations for retrieving and sending representations of resources. Different application-level protocols can be used, but the most common implementation makes use of HTTP with GET/POST/PUT/DELETE verbs detailing the four basic operations mapping the CRUD (Create, Retrieve, Update, Delete) pattern. Since representations are generally encoded in XML, REST applications looks much like the traditional Web applications, usually referred to as POX-HTTP (Plain-Old-XML over HTTP), which implements Web Services using HTTP-GET and HTTP-POST operations. The uni

The CYCLOPS project is an FP6 Specific Support Action which aims to bring together two important Communities: GMES and Grid, focusing on the operative sector of European Civil Protection (CP). Recently RISICO, an operative Italian Civil Protection application for wild fires risk assessment, has been ported to gLite by the CYCLOPS working group. RISICO presently runs in Grid accessing data stored using various proprietary formats in a gLite Storage Element. GLite is the Grid middleware developed by the EGEE European project (Enabling Grids for E-sciencE). As a further step we discuss which benefits could be granted to RISICO, realising an intermediate layer of geospatial web-services between the CP application and gLite. All the web-services implement standard interfaces specified by the Open Geospatial Consortium (OGC). In this case the georeferenced data will be stored in standard formats (GRIB/NetCDF) and will be accessed through standard interfaces. The workflow will be as follows: o The CP user selects an area in which the model should be run, selects the input data URIs and indicates an appropriate priority for the action. o Our application, exposing the Web Processing Server (WPS) interface, receives the request, evaluates the input size and the priority, and then activates various independent data access services. These, implementing the Web Coverage

Over the past decade visualization and analysis of geospatial data followed the general trend in information technology from monolithic desktop applications towards loosely coupled Web Services. In order to achieve interoperability among these services standard interfaces are required. The standards and specifications published by the Open Geospatial Consortium (OGC) and the International Organization for Standardization (ISO) form the basic set on which such service-oriented architectures (SOA) can be build. With respect to online data visualization a map like representation of spatial content has found widespread use, especially as part of Spatial Data Infrastructures (SDI). In combination with metadata catalogues the primary aim of these map services is data publication and distribution, hence the capabilities are limited to viewing or browsing (e.g. zoom, pan, identify). Only a few examples exist enabling users to analyse data (e.g. calculating statistics or merging different data layers) through a web interface or Web Service. In this paper first results are presented from research conducted on the implementation of a OGC Web Processing Service (WPS) for online analysis of Earth observation time series data. Earth observation data at regional to global scale has been collected with various sensors and satellite systems for more than three decades. The amounts of data acquired seem to have outpaced our ability to exploit and analysis it. With aid of consistent data products (e.g. MODIS suite of land surface products) and the advancements in information technology and in particular the OGC/ISO standards for Web services the basis to overcome this shortfall is avail

At the Boulder OGC (Open Geospatial Consortium) Technical Committee meetings, Unidata hosted a special Interoperability Day to address the use of standard interfaces (CS-W/ebRIM, WFS, WCS, SOS, GML) for providing access to data currently served via THREDDS, OPeNDAP, netCDF-CF and IDD/LDM technologies. The primary data served is Weather, Climate and Ocean data from the community, sometimes referred to as Fluid Earth Sciences (FES). Subsequent discussions have led to a subset of participants in the OGC GALEON (Geo-interface for Air Land, Envrirohment, Ocean NetCDF) Interface Experiment to embark on a revised approach to contributing to the evolution of coverage-related standards. The overall objective remains the development of practical and concrete ideas for

Ecological forecasting-making predictions about the ecological consequences of various types of change-is constrained by a variety of factors, including computer model interoperability. This, in turn, is limited by technical barriers, such as semantic and format consistency, and non-technical ones, such as model isolation that results when sponsors use a "stovepiped" funding approach rather than an integrated one. Because these constraints limit the types of ecological questions that can be addressed, they need to be overcome if decision makers are to obtain the information they require to make informed decisions. The problem of limited model interoperability exists, to varying degrees, in all disciplines that rely on computer models, including hydrology, air quality, and the ocean sciences, among others. One solution is being called the Model Web, a concept for an open-ended system of interoperable computer models and databases communicating via Service Oriented Architectures. The Model Web would consist of a distributed, multidisciplinary network of independent, interoperating models (plus related datasets and sensors). Like the World Wide Web it would grow organically, without central control, within a framework of broad goals and data exchange standards or guidelines. These are under discussion but they should emerge naturally from the modeling community; no single

Coverage concept represents the mapping from a domain (e.g. spatio-temporal domain) to parameter values where parameters value types are common to all the locations within the domain. For example, a spatio-temporal domain consists of a collection of direct positions in a coordinate space that may be defined in terms of up to three spatial dimensions as well as a temporal dimension. Coverages are the prevailing data structures in a number of application areas, such as remote sensing, meteorology, and mapping of bathymetry, elevation, soil, and vegetation. Examples of coverages include rasters, profiles, trajectories, triangulated irregular networks, gridded data, and polygon coverages. Hence, coverage is a general geospatial information concept representing space and time-varying phenomena; virtually, any geospatial data may be viewed as an instance of a coverage type. Most of the different Geospatial Information Communities (GICs) are characterized by acquiring, managing, processing and serving diverse coverage types which are described using specific metadata. In order to support interoperability, the implementation of standard services to access the different coverage types is becoming more and more important. The presentation will discuss the coverage access services as far as simplicity, effectiveness and interoperability are concerned. The Earth and Space Sciences, GIS, and Society perspectives will be introduced. The need for a robust abstract model underpinning different implementation models is addressed. These specific implementation models are important to accommodate heterogeneous GIC needs and use scenarios. In fact, access services may play an important role in order to map complex datasets to simpler coverage types, introducing explicit semantics.

Earth scientists are engaged in integrating knowledge stemming from different disciplines concerned with the constituent parts of the complex Earth system. The goal is to understand Earth's properties as a whole and thereby deliver benefits to society. The scope and complexity of Earth system investigations demand the formation of distributed, multidisciplinary collaborative teams, which presents both scientific and technological challenges. The information systems used by the different disciplines are characterized by heterogeneous and distributed data and metadata models, different semantics and expertise, diverse protocols and interfaces, and different data policies and security levels. We describe how the Global Earth Observation System of Systems (GEOSS) infrastructure facilitates the formation of these distributed, multidisciplinary and collaborative teams. Initiated in 2005 by the GEO Architecture and Data Committee, the GEOSS Interoperability Process Pilot Project (IP3) is a medium for reaching interoperability between disparate components contributed to GEOSS from multi-disciplinary Earth and Space Science communities. In 2007 one of IP3's key successes was an end-to-end demonstration that combined biological species data, accessed through the GBIF (Global Biodiversity Information Facility) contribution to GEOSS, and climate model output data, part of the WMO's Information System's contribution to GEOSS, accessed through NCAR, and seamlessly fed the data stream into an ecological niche model (ENM). The ENM runs in an open modeling framework to produce a new product displaying the impacts of climate change on the geographic distribution of selected