(http://inspire.ec.europa.eu/data-model/approved/r46
18-ir/html/), consider the cartography, but are
insufficiently able to manage the data of particular
application fields in a GIS. For this reason the need
for ontology integration arises.
In this paper, the environmental data (geometric
and dynamic data) studied during the European
project ALCOTRA ALIRHYS have been managed in
a GIS and in an external database structured
according to a data model designed by integrating
existing ontologies. The project aims to study
subterranean water resources through several
monitoring activities (www.polito.it/alirhys). In this
way the data produced are codified to be uniquely
interpreted, and to be effectively shareable on the web
without losing part of their meaning. Moreover, when
a GIS or a database is constructed in the traditional
way, the conceptual model used is shared, together
with other metadata, in order to limit interpretation
ambiguity. However, sharing of implemented
structures is not common. In this project, both the
models and the XML schema of the constructed
geodatabase in ESRI ArcGIS and the related database
tables in MS Access are published on the project's
website (www.polito.it/alirhys) in order to foster the
reproduction and the reuse of the method for similar
studies, even in a stand-alone environment.
The advantages of data management in GIS have
been exploited by running different analyses on the
data archived in order to acquire further information,
useful for the monitoring of springs; this work is
described in the final part of this paper.
1.1 Semantic Data Management:
Ontologies and Standards
Representation in many fields of application has been
influenced in recent years by new paradigms
established by the web and informatics technologies.
The Semantic Web is intended to realize a common
framework that allows the meaningful content of data
to be shared, beyond the boundaries of applications,
and to be easily reused for automatic processing.
Permitting an effective semantic interoperability,
documents are associated to metadata that specify the
semantic context in standardized language, and data
are structured according to pre-formatted and known
data models. One of the most powerful tools for
reaching these objectives is the formulation of
domain ontologies, primary conceptual models that
provide completely solution-independent schemas in
order to support modelling in producing suitable
representation structures. A set of standard languages
is available for defining interoperable ontologies
(http://www.w3. org/). Almost all these languages
have as their origin structure the eXtensible Markup
Language (XML), which is then specified in a set of
more evolved structures developed to better express
the logic structure and correcting semantics of data.
W3C (World Wide Web Consortium) is the
organization developing the structures to enable the
Semantic web. It published some specific languages
to model the data in domain ontologies, in order for
the data to be effectively shared. These evolve from
the original RDF (Resource Description Framework)
(http://www.w3.org/RDF/) to more suitable ones for
representing knowledge structures: OWL (Ontology
Web Language) (http://www.w3.org/OWL/), and
some evolutions of it (OWL 2, DAML+OIL, OWL1
DL). These languages are best suited for representing
information in the form of alpha-numeric data, but
they have some lacks in the suitable management of
geometric information.
In this scenario, many domain ontologies are
being developed. Among these, we find the NASA
(National Aeronautics and Space Administration)
SWEET (Semantic Web for Earth and Environmental
Terminology) ontology, intended for the
representation of Earth and Environmental Sciences
(EES) (Raskin, 2004). This ontology is in modelled
in the OWL language, and includes a very high
variety of concepts about both the object studied and
the methodology aspects (research, analysis,
measurements, residuals, etc.); these could be
effectively transposed to other application fields.
SWEET has become the de facto standard for data
management in EES (Di Giuseppe, 2014).
Concurrently, some particular formal languages
have been developed for spatial data. The most
widespread of these are published by the OGC (Open
Geospatial Consortium), an organization founded to
develop publicly available geo-enabled standards. In
particular, it encoded GML (Geography Markup
Language, ISO 19100 series of International
Standards and OpenGIS Abstract Specification),
which is an XML schema for the description of the
application, transport and storage of geographic
information. This has been applied to the formulation
of some standard models - such as CityGML, the
common semantic information model for the
representation of 3D urban objects – and allows these
data to be shared over different applications. Another
fundamental application of GML is the INSPIRE
conceptual model (http://inspire-twg.jrc.ec.europa.eu
/data-model/draft/r4530/), which is the pre-specified
ontology for formulating harmonic and homogeneous
maps in the European Union.
A point of contact between these two fields can be
GISTAM2015-1stInternationalConferenceonGeographicalInformationSystemsTheory,ApplicationsandManagement
214