Handbook of Research on Geoinformatics - Hassan A. Karimi Part 5 potx

172
Geospatial Image Metadata Catalog Services
1. Introduct Ion
As earth observation continues worldwide, large
volumes of remotely sensed data on the Earth’s
climate and environment have been collected and
archived. In order to maintain the data archives
efciently and to facilitate discovery by users of
desired data in the holdings, each data provider
normally maintains a digital metadata catalog.
Some online catalogs provide services to users
for searching the catalog and discovering the data
they need through a well-established Application
Programming Interface (API). Such services are
called Catalog Services. The information in the
catalog is the searchable metadata that describe
individual data entries in the archives. Currently
most Catalog Services are provided through Web-
based interfaces.
This chapter analyses three open catalog
service systems. It reviews the metadata stan-
dards, catalog service conceptual schemas and
protocols, and the components of catalog service
specications.
2. rev Iew of geosp At IAL IMAge
cAt ALog ser vIces
2.1 Pilot Catalog Service Systems
The Federal Geographic Data Committee (FGDC)
Clearinghouse is a virtual collection of digital
spatial data distributed over many servers in the
United States and abroad. The primary intention

of the Clearinghouse is to provide discovery
services for digital data, allowing users to evalu-
ate its quality through metadata. Most metadata
provide information on how to acquire the data;
in many cases, links to the data or an order form
are available online.
The NASA Earth Observing System Clear-
ingHOuse (ECHO) is a clearinghouse of spatial
and temporal metadata that enables the science
community to exchange data and information.
ECHO technology can provide metadata discovery
services and serve as an order broker for clients
and data partners. All the NASA Distributed Ac-
tive Archive Centers (DAACs), as data providers,
generate and ingest metadata information into
ECHO.
The Open Geospatial Consortium (OGC) has
promoted standardization and interoperability
among the geospatial communities. In catalogue
service aspect, OGC has dened the Catalog
Service implementation standard (OpenGIS,
2004) and published two recommendation papers
(OpenGIS, 2005a; OpenGIS 2005b). The George
Mason University (GMU) CSISS Catalog service
for Web (CSW) system is an OGC-compliant
catalog service, which demonstrates how the
earth science community can publish geospatial
resources by searching pre-registered spatial and
temporal metadata information. In particular, the
GMU CSISS CSW catalog service is based on

the OpenGIS implementation standard, and the
ebRIM application prole (OpenGIS, 2005). It
provides users with an open and standard means
to access more than 15 Terabytes global Landsat
datasets.
2.2 Conceptual System Architecture
Since these geospatial catalog services address
similar needs, it is not surprising that they have
almost the same conceptual system architecture,
as shown in Figure 1.
From the point of view of metadata circula-
tion, a catalog service usually consists of three
components: metadata generation and ingestion,
a conceptual schema for catalog service, and a
query interface for catalog service.
Metadata generation and ingestion is always
based on applicable metadata standards, such
as the Dublin Core (DCMI, 2003), Geographic
information – Metadata (19115) from Interna-
tional Organization for Standard (ISO, 2003),
Content Standard for Digital Geospatial Metadata
(CSDGM) from Federal Geographic Data Com-
173
Geospatial Image Metadata Catalog Services
mittee (FGDC, 1998), or the ECS Earth Science
Information Model from National Aeronautics
and Space Administration (NASA, 2006).
Metadata structures, relationships and deni-
tions, known as conceptual schemas, play a key
role in catalog services. They dene what kind

of metadata information can be provided and
how the metadata are organized. The concep-
tual schemas are closely related to those of the
pre-ingested metadata information, but are not
necessarily identical. Catalog service conceptual
schemas are always oriented toward the eld of
application and may be tailored to particular ap-
plication proles.
The query interface for a catalog service
denes the necessary operations, the syntax of
each operation, and the binding protocol. To
facilitate access and promote interoperability
among catalog services, the interface denition
may be kept open.
2.3 Metadata generation
In this section, the three open catalog services
identied in Section 2.2 are analyzed on the follow-
ing two aspects regarding metadata generation.
2.3.1 Base Metadata Standard
The base metadata standard is the public geospatial
metadata standard on which the catalog service
is based and to which the catalog service is tai-
lored, to meet a given agency’s requirements. In
addition to international and national geospatial
metadata standards, such as ISO 19115 and FGDC
CSDGM, several agencies may have de-facto
standards in their production environment, such
as NASA ECS.
The metadata used by the FGDC Clearing-
house follows FGDC CSDGM. Each afliated

catalog service site must organize their metadata
information following the CSDGM standard
before they join the clearinghouse.
The ECHO Science Metadata Conceptual
Model has been developed based on the NASA
Earth Observation System Data and Information
Core System (EOSDIS) Science Data Model, with
modications to suit project needs.
GMU CSISS CSW builds up its metadata con-
ceptual model by combining the ebRIM informa-
tion model and the ECS science data model.
2.3.2 Automatic Generation of
Metadata
As the volume of spatial datasets keeps growing,
generation of metadata becomes increasingly
time-consuming. An automatic mechanism for
generating metadata will facilitate the generation
and frequent update of metadata.
Metadata information needs to be organized
as TXT or SGML or HTML les before a node
Figure 1. Conceptual Architecture of Catalog Service
Catalog Service

Client
Catalog Service

Metadata

Holdings

Data

Holdings

Query Interface

Conceptual Schema
User

174
Geospatial Image Metadata Catalog Services
joins the FGDC clearinghouse. Some metadata
generation tools are available in addition to the
commercial software packages. These tools are
advertised on the FGDC website. To help the user
set up a clearinghouse node easily, a software
package, ISite, is provided. With this software,
a qualied clearinghouse node server can be set
up in minutes.
All the ECHO metadata holdings are obtained
directly from the data providers. DAACs can
use some ECS tools to automatically generate
metadata information.
GMU CSISS is developing Java-based tools
to automatically extract metadata information
from each granule. The Hierarchical Data Format
(HDF), Hierarchical Data Format - Earth Observ-
ing System (HDF-EOS), GeoTIFF and NetCDF
data formats are currently supported.
2.4 Metadata Ingestion

2.4.1 Metadata Distribution
This function deals with the physical distribu-
tion of metadata information within the catalog
service.
The FGDC Clearinghouse is a decentralized
system of servers that contain eld-level meta-
data descriptions of available digital spatial data
located on the Internet. The metadata informa-
tion is physically managed within the afliated
server node.
Even though in ECHO scenario, the metadata
information is periodically generated by those
distinct data centers, they are centrally managed
by the ECHO operation team. That is, in the
design time, metadata information in ECHO is
distributed; while in the run time it is managed
centrally.
The GMU CSISS CSW maintains more than
15 Terabytes of global Landsat images. All the
metadata information for these images has been
registered into a centralized metadata database.
2.4.2 Ingestion Type
This section examines how each catalog service
ingests metadata. It focuses on two aspects: remote
vs. local and automatic vs. manual.
In the FGDC Clearinghouse, all the metadata
information is manipulated only in the afliated
server node. Remote ingestion is not supported
in server nodes. The ingestion has to been manu-
ally.

Due to a centralized metadata information, a
database approach is taken. Metadata ingestion in
ECHO involves two steps. Data centers need to up-
load their current metadata information remotely
to a dedicated File Transfer Protocol (FTP) server,
and the ECHO operation team is responsible for
ingesting these metadata information into the
ECHO operational system.
GMU CSISS CSW provides published inter-
faces. As long as the metadata information is well
organized, it can be remotely ingested into the
GMU CSISS CSW metadata database. All the
metadata information in that database is online
and ready for client’s query.
2.5 Conceptual Schema
We examine how the metadata conceptual schema
is dened in each catalog service.
In each FGDC Clearinghouse collection, all
the metadata information is organized according
to the FGDC CSDGM. The conceptual schema
of FGDC Clearinghouse collection is exactly the
same as that of the FGDC CSDGM.
In ECHO, all the metadata information col-
lected in the NASA DAACs is based on the ECS
science data model, with some modications
necessary to suit project needs.
GMU CSISS CSW denes its conceptual
schema based on the ECS science data model
combined with ISO 19115. Since GMU CSISS
CSW supports metadata queries and data retrieval

(through the OGC services), an ebRIM-based
prole has been selected to support dening the
175
Geospatial Image Metadata Catalog Services
association between a data granule instance and
applicable geospatial service instances.
2.6 t ransfer protocol
A catalog service usually provides a standard,
API-based interface to support the client’s query.
This “design-by-contract” mechanism promote
third party members’ contribution to develop new
query interfaces, besides those web-based query
interfaces provided by the catalog server itself.
The backbone of the FGDC Clearinghouse is
Z39.50 (ISO, 1998). This protocol was initially
developed by the library community to discover
bibliographic records using a standard set of attri-
butes. To guide how to implement FGDC metadata
elements within a Z39.50 service, the FGDC has
developed an application prole for geospatial
metadata called "GEO," which provides sets of
attributes, operators, and rules of implementation
that suit geospatial needs. In fact, the node server
is a Z39.50 server, which enables FGDC query
utilities to search its metadata holdings on the y
through Z39.50 protocol and GEO prole.
ECHO exposes the Session Manager and a lim-
ited set of the ECHO services as Web Services de-
ned via the Web Services Description Language
(WSDL). ECHO also provides two client packages,

Façade and EchoTalk, for client developers. The
syntax of the communication protocol between
client and ECHO is based on the Web Services
Interoperability (WS-I) Basic Prole. However,
the semantics of the communication protocol are
dened by ECHO itself. Specic query syntax,
in Extensible Markup Language (XML) format,
has been proposed and implemented.
GMU CSISS CSW’s communication protocol
is based on the OGC Catalog Service Implementa-
tion Specication, which species the interfaces
and several applicable bindings for catalog ser-
vices. Operations, core information schema and
query language encodings are included. The
transportation-related communication protocol
follows this specication.
2.7 System Distribution
This section examines the physical distribution
of catalog service systems.
The FGDC Clearinghouse has 400 worldwide
registered nodes as of March 22, 2006. FGDC
maintains several Web-based search interfaces
to carry out distributed searches across multiple
clearinghouse nodes.
ECHO acts as an intermediary between data
partners and client partners. Data partners provide
information about their data holdings, and client
partners develop software to access this informa-
tion through ECHO Query and Order Web Service
interface. End users who want to search ECHO's

metadata must use one of the ECHO clients.
Although ECHO has close connections with the
DAACs and ECHO Clients, ECHO itself is not
a distributed system. It does not need to build a
distributed search across multiple agencies and
nodes at run time.
GMU CSISS CSW is a standalone service.
Like ECHO, it is not a distributed system.
2.8 Review Summaries
Table 1 summarizes the results of the analysis.
3. conc Lus Ion And dIscuss Ion
We have reviewed three public catalog services
— FGDC Clearinghouse, NASA ECHO and GMU
CSISS CSW— considering the following aspects:
metadata generation, metadata ingestion, catalog
service conceptual schema, query protocols and
system distribution. This review shows how it
is becoming possible to query metadata hold-
ings through public, standard Web-based query
interfaces.
The review results also show that the catalog
service providers still must dene a catalog service
schema that meets their particular needs. These
application-oriented approaches can meet projects
176
Geospatial Image Metadata Catalog Services
requirements, but they will make it more difcult
to create future cross-federation multi catalog
services. We recommend that a standard, common
and discipline-oriented-metadata based schema

be used for future implementations of catalog
services in the same and/or related elds.
r eferences
DCMI. (2003). DCMI Metadata Terms. Retrieved
March 8, 2007, from />ments/dcmi-terms/ß
ECHO. (2005). Earth Observing System Clearing-
house. Retrieved March 8, 2007, from http://www.
echo.eos.nasa.gov/
FGDC. (1998). Content Standard for Digital
Geospatial Metadata (CSDGM). Retrieved March
8, 2007, from />contstan.html
FGDC. (2005). FGDC Geospatial Data Clear-
inghouse Activity. Retrieved March 8, 2007,
from />inghouse.html
ISO. (1998). ISO 23950: Information and
documentation - Information retrieval (Z39.50)
- Application service denition and protocol
specication.
ISO. (2003). ISO 19115: Geographic Information
- Metadata.
LAITS. (2005). LAITS OGC Catalog Service
for Web - Discovery Interface. Retrieved March
8, 2007, from />discovery/
NASA. (2006). EOSDIS Core System Data Model,
Retrieved March 8, 2007, from c.
nasa.gov/standards/heritage/eosdis-core-system-
data-model
OpenGIS. (2004). OpenGIS Catalogue Service
Implementation Specication. Retrieved March
8, 2007, from />specs/?page=specs

OpenGIS. (2005a). OGC Recommendation Pa-
per 04-17r1: OGC Catalogue Services- ebRIM
(ISO/TS 15000-3 profile of CSW. Retrieved
March 8, 2007, from ngeospatial.
org/specs/?page=recommendation
Tables 1. Review summaries
Evaluation Points FGDC Clearinghouse NASA ECHO GMU CSISS CSW
Metadata generation –
Base standard
FGDC CSDGM ECS Core ECS Core/ISO 19115
Metadata generation –
Generation automation
manually with tools manually with tools automatically
Metadata ingestion –
Metadata Distribution
distributed centralized centralized
Metadata ingestion –
Ingestion Type
N/A Remotely and
automatically
Locally and automatically
Conceptual Schema FGDC CSDGM Based on ECS Core Based on ISO 19115 and
ebRIM
Transfer Protocol Z39.50 and GEO profile Proprietary and based
on Web Service
OGC Catalog Service and
HTTP binding
System distribution Distributed Centralized Centralized
177
Geospatial Image Metadata Catalog Services

OpenGIS. (2005b). OGC Recommendation Pa-
per 04-038r2: ISO19115/ISO19119 Application
Prole for CSW 2.0. Retrieved March 8, 2007,
from />?page=recommendation
key t er Ms
Catalog Service: A set of information, con-
sisting of some or all of directory, guide, and in-
ventories, combined with a mechanism to provide
responses to queries, possibly including ordering
data. (Source: Earth Science and Applications
Data System)
Catalog System: An implementation of a
directory, plus a guide and/or inventories, inte-
grated with user support mechanisms that provide
data access and answers to inquires. Capabilities
may include browsing, data searches, and placing
and taking orders. A specic implementation of
a catalog service. (Source: Earth Science and
Applications Data System, Interagency Working
Group on Data Management for Global Change,
European Patent Organisation).
Service: A distinct part of the functionality
that is provided by an entity through interfaces.
(Source: ISO 19119: Geographic information
– Services)
Interface: A named set of operations that
characterize the behavior of an entity. (Source:
ISO 19119: Geographic information – Services)
Operation: A specication of a transformation
or query that an object may be called to execute.

(Source: ISO 19119: Geographic information
– Services)
Transfer Protocol: A common set of rules
for dening interactions between distributed
systems. (Source: 19118: Geographic information
- Encoding)
178
Chapter XXIII
Geospatial Semantic Web:
Critical Issues
Peisheng Zhao
George Mason University, USA
Liping Di
George Mason University, USA
Wenli Yang
George Mason University, USA
Genong Yu
George Mason University, USA
Peng Yue
George Mason University, USA
Copyright © 2009, IGI Global, distributing in print or electronic forms without written permission of IGI Global is prohibited.
Abstr Act
The Semantic Web technology provides a common interoperable framework in which information is
given a well-dened meaning such that data and applications can be used by machines for more ef-
fective discovery, automation, integration and reuse. Parallel to the development of the Semantic Web,
the Geospatial Semantic Web – a geospatial domain-specic version of the Semantic Web, is initiated
recently. Among all the components of the Geospatial Semantic Web, two are especially unique – geo-
spatial ontology and geospatial reasoning. This paper is focused on discussing these two critical issues
from representation logic to computational logic.
179

Geospatial Semantic Web
Introduct Ion
Inspired by Tim Berners-Lee (Berners-Lee, 1998;
W3C, 2006), inventor of the Web, a growing
number of individuals and groups from academia
and industry have been evolving the Web into
another level - the Semantic Web. By represent-
ing not only words, but their denitions and
contexts, the Semantic Web provides a common
interoperable framework in which information is
given a well-dened meaning such that data and
applications can be used by machines (reason-
ing) for more effective discovery, automation,
integration and reuse across various application,
enterprise and community boundaries. Compared
to the conventional Web, the Semantic Web excels
in two aspects (W3C, 2006): 1) common formats
for data interchange (the original Web only had
interchange of documents) and 2) a language for
recording how the data relates to real world objects.
With such advancements, reasoning engines and
Web-crawling agents can go one step further – and
inductively respond to questions such as “which
airelds within 500 miles of Kandahar support
C5A aircraft?” rather than simply returning Web
pages that contain the text “aireld” and “Kan-
dahar”, which most engines do today.
Figure 1 shows the hierarchical architecture
of the Semantic Web. At the bottom level, XML
(Extensible Markup Language) provides syntax

to represent structured documents with a user-
dened vocabulary but does not necessarily
guarantee well-dened semantic constraints on
these documents. And XML schema denes the
structure of an XML document. RDF (Resource
Description Framework) is a basic data model with
XML syntax that identies objects (“resources”)
and their relations to allow information to be
exchanged between applications without loss of
meaning. RDFS (RDF Schema) is a semantic
extension of RDF for describing the properties
of generalization-hierarchies and classes of RDF
resources. OWL (Web Ontology Language) adds
vocabulary to explicitly represent the meaning of
terms and their relationships, such as relations
between classes (e.g. disjointness), cardinality
(e.g., “exactly one”), equality and enumerated
classes. The logic layer represents the facts and
derives knowledge, and deductive process and
proof validation are deduced by the proof layer.
A digital signature can be used to sign and export
the derived knowledge. A trust layer provides
the trust level or a rating of its quality in order
to help users building condence in the process
Figure 1. Semantic Web architecture (Berners-Lee, 2000)
180
Geospatial Semantic Web
and quality of information(Antoniou & Harmelen,
2004).
Parallel to the development of the Semantic

Web, the Geospatial Semantic Web – a geospatial
domain-specic version of the Semantic Web, is
initiated recently. Because geospatial information
is heterogeneous, i.e. multi-source, multi-format,
multi-scale, and multi-disciplinary, the impor-
tance of semantics on accessing and integration
of distributed geospatial information has long
been recognized (Sheth, 1999). The advent of the
Semantic Web promises a generic framework to
use ontologies to capture the meanings and rela-
tions for information retrieval. But this framework
does not relate explicitly to some of the most basic
geospatial entities, properties and relationships
that are most critical to a particular geospatial
information processing task. To better support
the discovery, retrieval and consumption of
geospatial information, the Geospatial Semantic
Web is initiated to create and manage geospatial
ontologies to capture the semantic network of
geospatial world and allow intelligent applications
to take advantage of build-in geospatial reasoning
capabilities for deriving knowledge. It will do so
by incorporating geospatial data semantics and
exploiting the semantics of both the processing
of geospatial relationships and the description
of tightly-coupled service content (Egenhofer,
2002; Lieberman, Pehle, & Dean, 2005). The
Geospatial Semantic Web was identied as an
immediately-considered research priority early in
2002 (Fonseca & Sheth, 2002) by UCGIS (Uni-

versity Consortium for Geospatial Information
Science). As an international voluntary consensus
standards organization, OGC (Open Geospatial
Consortium) conducted the Geospatial Semantic
Web Interoperability Experiment (GSW-IE) in
2005 aiming to develop a method of discovering,
querying and collecting geospatial content on the
basis of formal semantic specications.
The architecture of the Geospatial Semantic
Web is similar to that portrayed in Figure 1. The
Geospatial Semantic Web and the Semantic Web
share top level (general) ontology, ontological lan-
guages, and general reasoning mechanisms. The
Geospatial Semantic Web extends the Semantic
Web with domain-specic components. Among
all the components of the Geospatial Semantic
Web, two are especially unique – geospatial
ontology and geospatial reasoning. The former
aims at expressing geospatial concepts and re-
lationships, specialized processing of geospatial
rules and relationships, and self-described Web
service with its highly dynamic geospatial con-
tent beyond the purely lexical and syntactic level
(Egenhofer, 2002; Lieberman et al., 2005; O'Dea,
Geoghegan, & Ekins, 2005). The latter embraces
sets of geospatial inference rules on the basis of
geospatial ontologies and techniques to conduct
automated geospatial reasoning by machine with
less human interaction for deriving geospatial
knowledge. These two are the foci to be elaborated

in the following two sections in this paper. Two
application cases are presented to show the syn-
dicated achievements of the Geospatial Semantic
Web. A short summary is given at the end.
geosp At IAL ont oLogy
It is widely recognized that ontology is critical for
the development of the Semantic Web. Ontology
originated from philosophy as a reference to the
nature and the organization of reality. In general,
an ontology is a “specication of a conceptual-
ization” (Gruber, 1993). In the computer science
domain, ontology provides a commonly agreed
upon understanding of domain knowledge in a
generic way for sharing across applications and
groups (Chandrasekaran, Johnson, & Benjamins,
1999). Typically, ontology consists of a list of
terms (classes of objects) and the relationships
between those terms. Moreover, ontology can also
represent property information (e.g., an aireld
has runways), value restrictions (e.g., aircraft can
only take off at an aireld), disjointness statements
(e.g., aircraft and train are disjoint), and speci-
181
Geospatial Semantic Web
cation of logical relationships between objects
(e.g., a runway must be at least 400 meters long
for supporting C5A aircraft).
In the geospatial domain, a specic range of
geospatial ontolgoies are needed to dene a formal
vocabulary that sufciently captures the semantic

details of geospatial concepts, categories, rela-
tions and processes as well as their interrelations
at different levels. A geospatial ontology does
not simply give a denition, but also represents
relationships between concepts. For example, an
ontological denition of “surface water” describes
its properties and characteristics but also carries
relationship meanings to other entities, such as
“surface water” belongs to “hydrosphere”, and
“river” is a kind of “surface water”.
A well-formatted geospatial ontology is very
useful in the following areas:
•
Int
eroperability. Since the geospatial sci-
ences deal with phenomena across a variety
of scales and disciplines, the semantics of
geospatial information is essential for the
development of interoperable geospatial
software and data formats. Geospatial ontol-
ogy provides a common understanding of
not only general geospatial concepts but also
complex geospatial scientic computing.
Through geospatial ontology, the different
geospatial data models and representations
can be integrated.
•
Spa
tial reasoning about geospatial associa-
tions and patterns, e.g., topological relations

(connectivity, adjacency and intersection
of geospatial objects), cardinal direction
(relative directions among geospatial ob-
jects, such as east, west and southwest), and
proximity relations (geographical distance
between geospatial objects, such as A is close
to B and X is very far from Y, and contextual
relations, such as an obstacle separates two
objects that would be considered nearby
space, but are considered far because of the
obstacle) (Arpinar, Sheth, & Ramakrishnan,
2004).
•
Reu
se and organization of information,
such as standardizing libraries or reposi-
tories of geospatial information and work-
ows.
Compared to general ontologies, geospatial
ontologies specically encode 1) spatial concepts,
e.g., location and units, 2) spatial relationships,
e.g., inside, near and east, 3) physical facts, e.g.,
physical phenomena, physical properties and
physical substances, 4) geospatial data, e.g., data
properties, such as instruments, platforms and
sensors, and 5) geospatial computing processes,
e.g., disciplines, parameters and algorithms.
According to the interactions and the role
within the context of the Geospatial Semantic Web,
geospatial ontology can be classied into several

large groups with hierarchical relationships as
Figure 2 in which the ontologies at upper levels
are consistent to the ontologies at lower levels.
General ontology is the core upper level vo-
cabulary representing common human consensus
reality that all other ontologies must reference. It
is domain independent. The widely used Dublin
Core Metadata (Dublin, 2006) provides a standard
for metadata vocabularies to describe resources
that enable the development of more intelligent
information discovery systems. OpenCyc (Open-
Cyc, 2006) is the world's largest and most com-
plete general knowledge base and commonsense
reasoning engine dening more than 47,000 upper
level concepts and 306,000 assertions about these
concepts.
Geospatial feature ontology, dening geospa-
tial entities and physical phenomena, provides
the core geospatial vocabulary and structure, and
forms the ontological foundation of geospatial
information. It should be coordinated with the
development of geospatial standards to dene
its scope and content, such as the ISO 19100
series and the OGC specications. Geospatial
factor ontology describes geospatial location,
unit conversion factors and numerical exten-
sions. To enable geospatial topological, proximity
182
Geospatial Semantic Web
and contextual reasoning, geospatial relation-

ship ontology represents geospatial and logical
relationships between geospatial features. The
RDF geo vocabulary (Brickley, 2004) provides
a basic RDF vocabulary with a namespace for
representing lat(itude), long(itude) and other
information about spatially-located things using
WGS84 as a reference datum. The OWL-Space
initiative (formerly DAML-Space) provides the
ontologies of comprehensive spatial properties
and relations including topology, dimensioin,
orientation and shape, length and area, lat/log
and elevation, geopolitical subdivisions, granu-
larity, aggregate and distributions(Hobbs, 2003).
By incorporating Region Connection Calculus
(RCC), the CoBrA (Harry Chen, Finin, & Joshi,
2003) and the SWETO-GS (Arpinar et al., 2004)
dene basic relations between two-dimensional
areas and relevant rules to enable the reasoning of
spatiotemporal thematic proximity. The ontologies
described above facilitate the information com-
munication between geospatial applications.
Geospatial domain-specic ontology rep-
resents the specic concepts in one domain by
using proprietary vocabularies over which the
user will query. Sometimes there exists using
different terms to represent a geospatial feature
in different domains. To achieve interoperability,
there must be a link between domain ontology
and feature ontology, either by subclassing feature
ontology concepts or by mapping from feature

concepts to domain concepts. To provide formal
semantic descriptions of geospatial data collec-
tions and scientic concepts, several projects are
underway to develop a semantic framework. The
ontologies within the Semantic Web for Earth and
Environmental Terminology (SWEET)(Raskin,
2006) contain several thousand terms spanning
a broad extent of Earth system science and re-
lated concepts in order to provide a high-level
semantic description of Earth system science.
To facilitate data register and data discovery, the
GEON (Geosciences Network) project develops
a series of geological ontologies to standardize
the description of geological map including data
structure and content(Lin et al., 2004).
Geospatial data ontology uses the ontologies
that fall below it in Figure 2 to provide a dataset
description which includes representation, stor-
age, modeling, format, resources, services and
distributions. It ensures that the data are discov-
erable, usable and interoperable in a standard
way. The ontologies of geospatial metadata for
ISO 19115 and FGDC (Federal Geographic Data
Committee) being developed in (Islam, Bermudez,
Beran, Fellah, & Piasecki, 2004; Zhao, 2004)
add semantic meanings and relationships to the
metadata terms by which data sets are explicitly
associated with providers, instruments, sensors
and disciplines.
Figure 2. The hierarchy of geospatial ontology

General Ontology
Geospatial Feature Ontology
Geospatial Factor Ontology Geospatial Relationship Ontology
Geospatial Domain-Specific Ontology
Geospatial Data Ontology
Geospatial Service Ontology
183
Geospatial Semantic Web
Geospatial service ontology describes who
provides the service, what the service does and
what other properties the service has that make
it discoverable by incorporating the lower level
ontologies. It also states the service inputs, outputs,
preconditions and effects to ensure the service
measurable by using data ontology. To allow
dynamic invocation, this ontology also includes
the concrete service ports, protocols and encod-
ings. Geospatial service ontology enhances and
extends the current offerings of web services by
enabling full semantic queries against service
offerings. Since the OWL-S provides a semantic
framework to describe web services, it seems like
a good choice to follow its logical structure for the
denition of geospatial service ontology.
geosp At IAL seMAnt Ic
r eAson Ing
Geospatial semantic reasoning can deduce (infer)
conclusions from given geospatial knowledge by
uncovering implicit ontological knowledge and
unexpected relationships and inconsistencies. For

example, suppose area Y is inside area X and area
Z is within area Y. We can deduce that Z is inside
X if the meanings of inside and within are well
dened as following, i.e., inside is same as within
and both of them are transitive properties.
Logic, particularly in the form of predicate
logic (rst order logic), is the foundation of reason-
ing by offering a formal language for expressing
knowledge in well-understood semantics. The
proof system can automatically drive statements
syntactically from a set of premises and provide
explanations for answers by tracing logical conse-
quences. In general, the more expressive a logical
system is, the more computational complexity it
has for drawing conclusions. Description logic
is a particular decidable fragment of predicate
logic with desirable computational properties for
reasoning systems in which correspondences are
illustrated by the axiomatic semantics in the form
of logical axioms. RDF(S), OWL Lite and OWL
DL can be viewed as description logics. Rule
systems (Horn logic), known as A
1
, … A
n
 B, is
another subset of predicate logic with an efcient
proof system that is orthogonal to description
logic. In geospatial domain, RDF and OWL al-
low ontology reasoning by dening the geospatial

representation of real-world features, and rules
allow default reasoning and fuzzy reasoning by
dening how spatial features relate to each other
and how they process geospatial data (H. Chen,
Fellah, & Bishr, 2005; O'Dea et al., 2005).
The subsumption is the basic inference in on-
tology, typically written as
XY ⊆
, i.e.,. checking
whether the concept denoted by X is considered
more general than that denoted by Y. For example,
FeatureRunway ⊆
iff AirportFacility ⊆Fea-
ture

and
ilityAirportFacRunway ⊆
. Such a
concept hierarchy provides useful information on
the connection between different concepts. The
<owl:ObjectProperty rdf:ID=”inside”>
<rdf:type rdf:resource=”&owl;TransitiveProperty”
/>
<rdfs:domain rdf:resource=”#area” />
<rdfs:range rdf:resource=”#area” />
</owl:ObjectProperty>
<owl:ObjectProperty rdf:ID=”within”>
<rdf:type rdf:resource=”&owl;TransitiveProperty” />
<rdfs:domain rdf:resource=”#area” />
<rdfs:range rdf:resource=”#area” />

</owl:ObjectProperty>
<Spatial rdf:ID=”within”>
<owl:sameAs rdf:resource=”#inside” />
</Spatial>
Box 1.
184
Geospatial Semantic Web
instantiation inference can check if individual
i is instance of class C, i.e.,
I
Ci∈
. Moreover,
this instance relationship may trigger the appli-
cation rules to get additional facts. The retrieval
inference can thus retrieve set of individuals that
instantiate the class. The other basic inference is
satisability, which detects whether a new concept
makes sense or whether it is contradictory to the
existing one. For example,
DC ⊆
iff
CD ¬
is not satised.
In the Geospatial Semantic Web applications,
rules are needed to deal with the situations where
reasoning moves from representation logic to com-
putational logic. For example, to answer “which
airelds near Kandahar support C5A aircraft”,
the following rules are necessary.
IF isTypeOf (?airport, Airport:airport) AND

Locate (?location, gazetteer:Kandahar) AND
Buffer (?airport, ?location, 500, “miles”)
AND
HasRunway (?runway, ?airport) AND
GreaterThan (?runway, 400, “meters”)
THEN
supportC5A (?airport)
The above example expresses that an airport
supports C5A landing near Kandahar if it has a
runway at least 400 meters long, and is located
within 500 miles of Kandahar (by nding the
location of Kandahar and calculating its circle buf-
fer). At present, there are a number of proposed
standard Semantic Web rule languages. SWRL
(Semantic Web Rule Language) (Horrocks et al.,
2004), based on a combination of the OWL with
the RuleML, is a good candidate.
On the Semantic Web, “theory” reasoning is a
desirable complement to “standard” reasoning (F.
Bry & Marchiori, 2005). The Geospatial Semantic
Web requires specic geospatial reasoning meth-
ods to take advantage of any particular properties
of geospatial entities. Region Connection Calculus
(RCC8) (Cohn, Bennett, Gooday, & Gotts, 1997)
uses regions as the primary spatial entity and
provides a rich vocabulary of qualitative shape
descriptions and a logical calculus for qualitative
reasoning with a rst-order and propositional
sub-variant. In (Francois Bry, Lorenz, Ohlbach,
& Rosner, 2005), a geospatial world model is

developed for Semantic Web applications. In
this model, geospatial data is represented as a
hierarchy of graphs that combine very low-level
coordinate-based computations with abstract
symbolic reasoning and the ontology of transport
networks. Based on this model, two-dimensional
fuzzy reasoning for geospatial data is developed
(Ohlbach & Lorenz, 2005).
AppLIcAt Ions
The Geospatial Semantic Web promises an
“intelligent” method to discover, retrieve and
integrate large and diverse geospatial information
and services. Numerous efforts to approach this
“intelligence” are currently active.
geon Intelligent geologic data
Integration
GEON project () develops
an interoperable framework and system that al-
lows the intelligent integration of geologic data
from different resources. In this framework, data
providers register a data set with one or more
“mediation ontologies”, i.e., standards for data
structure and content, and subsequently query
the different data sets in a uniform fashion. Thus,
heterogeneous source vocabularies are made
compatible via the ontologies, and multiple con-
ceptual dimensions become queryable simultane-
ously, e.g., nd regions with igneous rocks from
geologic period P having composition C, fabric F
and texture T(Lin et al., 2004). In order to answer

semantic queries, such as synonymous, more
specic and less specic, the system embedded
in the GEON grid environment reasons with the
ontologies and builds the distributed query plans
accordingly.
185
Geospatial Semantic Web
Intelligent geospatial web services
A geospatial Web service is a modular application
representing a real geospatial computing process
over the network. In order to solve real-world
geospatial problems through Web services, an
“intelligent” mechanism is required to facilitate
service discovery and integration and automate
the assembly of service chains. An approach to
intelligent geospatial Web service is presented in
(Di, Zhao, Yang, Yu, & Yue, 2005). This approach
uses ontologized “Geo-Object”, a component of
“Geo-Tree”, to integrate the views of geospatial
services and make them understandable and in-
ferable. By using semantic-enabled OGC catalog
service, a “Geo-Object” can be easily found based
on the exible semantic match on its inputs and
outputs. Actually, a “Geo-Tree” describing a
geospatial modeling process can be easily rep-
resented as a service chain. The construction of
such a tree is backward (from goal to source) in
which all relevant “Geo-Object” are discovered
and linked automatically based on the user’s goals.
Once all the components have been found, the tree

is initiated as a service chain in BPEL (Business
Process Execution Language) and then be sent to
the workow engine for being executed.

conc Lus Ion
The Semantic Web technology is concerned
with preserving the concept meanings and links
between concepts in order to make information
machine-accessible. It is an efcient way of
representing data and applications. Geospatial
ontologies provide semantic descriptions of the
geospatial world, and geospatial reasoning derives
additional facts from them. From representation
logic to computational logic, the Geospatial Se-
mantic Web enhances our ability to express and
deduce geospatial concepts and relationships for
achieving interoperability among distributed
heterogeneous geospatial data and applications.
r eferences
Antoniou, G., & Harmelen, F. v. (2004). A Se-
mantic Web Primer: The MIT Press.
Arpinar, B., Sheth, A., & Ramakrishnan, C.
(2004). Geospatial Ontology Development and
Semantic Analytics. In J. P. Wilson & A. S.
Fotheringham (Eds.), Handbook of Geographic
Information Science: Blackwell Publishing.
Berners-Lee, T. (1998). Semantic Web Road Map.
Retrieved March 21, 2006, from http://www.
w3.org/DesignIssues/semantic.html
Berners-Lee, T. (2000). Semantic Web on XML.

Paper presented at the XML 2000, Washington
DC.
Brickley, D. (2004). Basic Geo (WGS84 lat/long)
Vocabulary. Retrieved March 25, 2006, from
/>Bry, F., Lorenz, B., Ohlbach, H. J., & Rosner,
M. (2005, Septemper 11-16, 2005). A Geospatial
World Model for the Semantic Web. Paper pre-
sented at the Third Workshop on Principles and
Practice of Semantic Web Reasoning, Dagstuhl,
Germany.
Bry, F., & Marchiori, M. (2005, 27th 28th
April, 2005). Ten Theses on Logic Language for
the Semantic Web. Paper presented at the W3C
Workshop on Rule Languages for Interoperability,
Washington D.C., USA.
Chandrasekaran, B., Johnson, T., & Benjamins,
V. (1999). Ontologies: What are they? why do we
need them? IEEE Intelligent Systems and Their
Applications, 14(1), 20-26.
Chen, H., Fellah, S., & Bishr, Y. (2005, April
27-28, 2005). Rules for Geospatial Semantic
Web Applications. Paper presented at the W3C
Workshop on Rule Languages for Interoperability,
Washington D. C., USA.
186
Geospatial Semantic Web
Chen, H., Finin, T., & Joshi, A. (2003, October
12-15, 2003). An Intelligent Broker for Context-
Aware Systems. Paper presented at the Ubicomp
2003, Seattle, Washington, USA.

Cohn, A., Bennett, B., Gooday, J., & Gotts, N.
(1997). Qualitative Spatial Representation and
Reasoning with the Region Connection Calculus.
GeoInformatica, 1(3), 275-316.
Di, L., Zhao, P., Yang, W., Yu, G., & Yue, P.
(2005). Intelligent Geospatial Web Services.
Paper presented at the IEEE 25th Anniversary
International Geoscience and Remote Sensing
Symposium.
Dublin. (2006). Retrieved March 20, 2006, from

Egenhofer, M. (2002, November 8-9, 2002).
Toward the Semantic Geospatial Web. Paper
presented at the The Tenth ACM International
Symposium on Advances in Geographic Informa-
tion Systems, McLean, VA.
Fonseca, F., & Sheth, A. (2002). The Geospa-
tial Semantic Web. Retrieved March 20, 2006,
from />2002researchPDF/shortterm/e_geoseman-
tic_web.pdf
Gruber, T. (1993). A Translation Approach to
Portable Ontologies. Knowledge Acquisition,
5(2), 199-220.
Hobbs, J. (2003). An Ontology of Spatial Rela-
tions for the Semantic Web. Paper presented at
the Workshop on the Analysis of Geographic
References.
Horrocks, I., Patel-Schneider, P., Boley, H., Tabet,
S., Grosof, B., & Dean, M. (2004). SWRL: A
Semantic Web Rule Language Combining OWL

and RuleML. Retrieved March 27th, 2006, from
/>Islam, A. S., Bermudez, L., Beran, B., Fellah, S.,
& Piasecki, M. (2004). Ontology for Geographic
Information - Metadata (ISO 19115:2003). Re-
trieved March 31, 2006, from xel.
edu/~wbs/ontology/iso-19115.htm
Lieberman, J., Pehle, T., & Dean, M. (2005, June
9-10, 2005). Semantic Evolution of Geospatial Web
Services. Paper presented at the W3C Workshop
on Frameworks for Semantic in Web Services,
Innsbruck, Austria.
Lin, K., Ludaescher, B., Broadaric, B., Seber, D.,
Baru, C., & Sinha, K. (2004). Semantic Mediation
Services in Geologic Data Integration: A Case
Study from the GEON Grid. Paper presented at
the GSA Annual Meeting.
O’Dea, D., Geoghegan, S., & Ekins, C. (2005).
Dealing with Geospatial Information in the Se-
mantic Web. Paper presented at the Australasian
Ontology Workshop, Sydney, Australia.
Ohlbach, H. J., & Lorenz, B. (2005). Geospatial
Reasoning: Basic Concepts and Theory. Retrieved
27th March, 2006, 2006
OpenCyc. (2006). Retrieved March 20, 2006,
from
Raskin, R. (2006). Guide to SWEET Ontologies.
Retrieved March 22, 2006, from .
nasa.gov/sweet
Sheth, A. (1999). Changing Focus on Interoper-
ability in Information Systems: From System,

Syntax, Structure to Semantics. In M. Goodchild,
M. Egenhofer, R. Fegeas & C. Kottman (Eds.),
Interoperating Geographic Information Systems
(pp. 5-30). New York: Kluwer.
W3C. (2006). Semantic Web. Retrieved March
25, 2006, from />Zhao, P. (2004). Geospatial Ontology. Retrieved
March 27, 2006, from .
edu/ontology
187
Geospatial Semantic Web
key t er Ms
Interoperability: Capability to communicate,
execute programs, or transfer data among various
functional units in a manner that requires the
user to have little or no knowledge of the unique
characteristics of those units [ISO 2382-1].
Geospatial Ontology: A formal vocabulary
that sufciently captures the semantic details
of geospatial concepts, categories, relations and
processes as well as their interrelations at dif-
ferent levels.
Geospatial Reasoning: Using logic to infer
implicit spatial relationships and knowledge from
given geospatial facts.
Geospatial Semantic Web: A domain-spe-
cic version of the Semantic Web which creates
and manages geospatial ontologies to exploit the
logical structure of geospatial world and allow
applications to take advantage of “intelligent”
geospatial reasoning capabilities.

Geospatial Web Service: A modular appli-
cation designed to enable the discovery, access,
and process of geospatial information across the
Web.
Ontolog: A specication of conceptualization.
In computer science domain, ontology provides
a commonly agreed understanding of domain
knowledge in a generic way for sharing across
applications and groups.
Semantic Web: A common interoperable
framework in which information is given well-
dened meaning such that the data and applica-
tions can be used by machine for more effective
discovery, automation, integration, and reuse
across various applications, enterprises and com-
munity boundaries.
Section V
Distributed Geoprocessing
189
Chapter XXIV
Geospatial Web Service
Chaining
Carlos Granell
Universitat Jaume I, Spain
Michael Gould
Universitat Jaume I, Spain
Miguel Ángel Esbrí
Universitat Jaume I, Spain
Copyright © 2009, IGI Global, distributing in print or electronic forms without written permission of IGI Global is prohibited.
Abstr Act

In the context of Geographic Information System’s evolution from monolithic systems to personal desk-
top GIS and then to collections of remote Internet services, we discuss the combination or chaining of
distributed geospatial web services. The adoption of web services technology provides remote access
to a diverse and wide array of geospatial datasets and allows developers to create web applications
(web browser-based or GIS client-based), hiding the underlying server functionalities from their public
interfaces. A major challenge in working with these remote services, as opposed to a single desktop
application, is to properly integrate ad hoc services to build a coherent service chain; this is especially
tricky in real-time scenarios where web applications need to be built on-the-y. This chapter discusses
strategies for geospatial web service chaining and poses some challenging issues, many related to se-
mantics, to be resolved for geospatial web service chaining to become a commonplace activity.
190
Geospatial Web Service Chaining
Int roduct Ion
The development of Geographic Information
Systems (GIS) has been highly inuenced by
the overall progress of Information Technology
(IT). These systems evolved from monolithic
systems to become personal desktop GIS with
all or most data held locally and then again to the
Internet GIS paradigm in the form of Web ser-
vices (Peng & Tsou, 2001). The highly distributed
web services model is such that geospatial data
are loosely coupled with the underlying systems
used to create and handle them, and geo-process-
ing functionalities are made available as remote,
interchangable, interoperable, and specialized
geospatial services.
In recent years the software industry has moved
away from complex architectures such as CORBA
(Common Object Request Broker Architecture)

(Vinoski, 1997) toward more universal and easily
dened architecture based on already-implement-
ed Internet protocols (Kaye, 2003). The success
factor of Web services technology has been to
promote service integration and interoperability
among heterogeneous distributed information
sources, without leaving the well-known and ac-
cepted Internet (Web) architecture. This has led
to de facto standards for delivery of services such
as Web Service Description Language (WSDL)
to describe the functionality of a service, Simple
Object Access Protocol (SOAP) to encapsulate
web service messages, and Universal Description,
Discovery, and Integration (UDDI) to register and
provide access to service offerings. Alternative
service architectures such as Representational
State Transfer, or REST, (Fielding, 2000) exist as
well; however the former de facto standards have
dominated. Perhaps the most benecial character-
istic of Web services technology is to provide not
only access to individual Web services but also
integrate several services assembling a new value-
added service chain. Adoption of Web services
technology as an option to xed, monolithic GIS
is an emerging trend, due in part to the diversity
and complexity of geospatial data, especially in
real-time scenarios such as emergency response,
where information systems often need to be built
on-the-y (Lemmens et al., 2006).
ch AInIng geosp At IAL ser vIce s

Interoperability, or the ability of software com-
ponents to interact with minimal knowledge of
the underlying structure of other components,
has become a basic requirement for distributed
information systems (Sheth, 1999), and so it is also
critical to GIS and to geospatial web services. The
Open Geospatial Consortium (OGC) has formed
working groups within the GIS community to
foster interoperability between geodata and
geospatial services in order to dene well-estab-
lished interfaces to a wider range of geospatial
web services (Whiteside, 2005). Table 1 lists a
sample of key geospatial web services interfaces
as dened by OGC.
The notion of chaining of geospatial web
services (Alameh, 2003) emerged as a mecha-
nism for assembling or combining individual
geospatial web services to create customized web
applications. A simple chain of the above listed
geospatial web services may be that constructed
to produce a coverage portrayal service (CPS)
that assembles an image retrieved from various
web coverage services (WCS) and portrays it to
a web coordinate transformation service (WCTS)
for the transformation of the composite image into
another coordinates reference system for proper
alignment with other geodata (Alameh, 2003).
The OGC and ISO Technical Committee 211
(ISO/TC211) have jointly developed international
standards for geospatial service architecture

and have dened interoperable geospatial web
service interfaces (ISO 19119, 2005; Percivall
2002). Transparent, translucent and opaque ser-
vice-chaining approaches have been dened by
these organizations according to the degree of
transparency of the web service chain complex-
ity to the client:
191
Geospatial Web Service Chaining
• Transparent or user-dened chaining:
the user or client application denes and
controls completely the structure and order
of execution of the individual services. Ob-
viously, the user has prior knowledge about
the geospatial web services to be used in the
chain.
•
Opaq
ue or aggregate service chaining:
a single aggregate service comprises and
coordinates all individual geospatial web
services in the chain. Opaque chaining pres-
ents the aggregate service to the user having
no awareness of the individual services.
•
Tra
nslucent or workow-managed chain-
ing: As the name implies, translucent
chaining takes place between transparent
and opaque chaining. The user invokes a

workow management service or mediated
service that acts as a broker and constructs
and manages chains of geospatial web
services. Translucent chaining promises to
reduce the transparency or exposure of geo-
spatial service chaining to the user because
mediated services handle most of the work
required to assemble, manage, and execute
the geospatial service chains. However, these
require a certain degree of intelligence, for
mediated services to be able to liberate us-
ers to the task of constructing the chain of
geospatial web services, encapsulating the
details and complexity. For this reason, they
could be inherently complex and challenging
to design and implement (Alameh, 2003).
This ‘intelligence’ comes in the form of se-
mantic discovery and interpretation of each
service’s capabilities and data consumption
needs.
ch ALLenges In ser vIce
ch AInIng
Several challenges remain before robust, reliable
and easily developed geospatial web service
chains become commonplace. Some of these
challenges stem from the use of web services in
general (Kaye, 2003) and others are specic of
geospatial needs.
Many of the important issues in web service
chaining that remain to be solved are semantic

in nature: creating, publishing and consuming
descriptions of how web services can be discov-
Table 1. Examples of geospatial web services
Service name Service description
Web Map Service (WMS) Dynamically produces spatially referenced maps of client-specied criteria from one or
more geographic datasets, returning pre-dened maps in an image or graphics format
(png, jpeg, gif)
Web Feature Service (WFS) Allows clients to lter and retrieves vector representation of geospatial features and
feature collections
Web Coverage Service
(WCS)
Retrieves client-specied coverage or image dataset
Catalog Service (CSW) Retrieves object metadata stored that meets specied query criteria
Gazetteer Service Retrieves location geometries for specied geographic names
Web Terrain Service (WTS) Dynamically produces perspective views from geographic feature and/or coverage data,
returning pictorial renderings of data in an image or graphics format
Web Coordinate Transforma-
tion Service (WCTS)
Transforms the coordinates of feature or coverage data from one coordinate reference
system (CRS) to another, including for example, conversions and rectication
Coverage Portrayal Service
(CPS)
Dynamically produces pictorial renderings in an image or graphics format of a coverage
subset dynamically retrieved from a Web Coverage Service (WCS)
192
Geospatial Web Service Chaining
ered and composed on-the-y according to user
needs. The research area of semantic web services
(McIlraith et al., 2001) proposes the markup of
web services description by means of semantic

web languages such as OWL-S (Semantic OWL-
based Semantic Web Service Ontology) and
WSMO (Web Service Modeling Ontology) (Lara
et al., 2004). Another important issue is related to
security, in the sense of setting policies for access
control and authentication. For example, consider
two or more services that need to interact properly
to complete a service chain in a critical scenario.
It is necessary to ensure that communications and
transactions among these services are conducted
in a secure environment and that messages are
reliably delivered to the correct destinations.
Also, the adoption of e-commerce features such
as service costs, billing and payment mechanisms
in web services offers a great opportunity for e-
commerce to take a central role in the growing
online economy (Medjahed et al., 2003). Table
2 summarizes some of these challenges for web
service chaining.
In addition to the challenges for web services
listed in Table 2, other challenges apply more
directly to geospatial web services (Tu et al.,
2004). In practice, chaining geographic services
is nontrivial (Lemmens et al., 2006), in part
because geographic data are quite varied and,
thus, different from other types of data in that
they may include multiple versions of the same
phenomena, and may be massive data sets. Geo-
spatial service monitoring, how to verify that what
is described is what is delivered, and geospatial

service compliance with standards, how to ensure
geospatial web service chains are compliant to
particular specications such those provided by
the OGC. In addition, common GIS scenarios or
use cases such as hydrologic analysis or environ-
mental walk-through simulations, require huge
amounts of geospatial data for browsing large
vector datasets (especially when represented in
XML such as the case for Geography Markup
Language, GML) and high-resolution satellite
imagery. Some effective methods are emerging
for integrating heterogeneous data sets from dif-
ferent application domains and visualizing them
to the user (Hobona et al., 2006). Due to the large
volumes of geospatial data, especially in XML
(GML) format, there is a critical requirement on
the performance on the data transfer. Table 3 sum-
marizes some of these challenges for geospatial
web service chaining.
conc Lus Ion
The future scenario for geospatial web service
chaining may not ever reach a wholly automated,
spontaneous service chaining for a set of self-de-
scribing geospatial web services, however in the
near term semi-automated solutions are emerging
to help users solve geographical problems with
remote services. Geospatial web services listed
in Table 1 mainly deal with the delivery of data
instead of advanced processing performed online.
More complex geospatial services have to be

specied in order to distribute over the Internet
all the functionalities (computation, analysis,
etc.) common in our desktop GIS. The rst steps
towards advanced geoprocessing services online
are outlined by the recently published OpenGIS
Table 2. Challenges in web service chaining
• Semantics: dynamic discovery, composition
• Security: access control, authentication
• Transactional integrity
• E-commerce features: billing, accounting, paying mechanisms
• Quality of Service (QoS): reliability, robustness
193
Geospatial Web Service Chaining
Web Processing Service discussion paper (Schut
and Whiteside, 2005), which provides interface
specications to enable geospatial web services to
support a limited range of geo-processing tasks,
by creating accessible libraries of geo-processing
algorithms under the appearance of geospatial
web service chains.
r eferences
Alameh, N. (2003). Chaining Geographic Infor-
mation Web Services. IEEE Internet Computing,
7(5), 22-29.
Fielding, R.T. (2000). Architectural Styles and the
Design of Network-based Software Architectures.
PhD thesis, University of California, Irvine.
Hobona, G., James, P., & Fairbain, D. (2006).
Web-Based Visualization of 3D Geospatial Data
Using Java3D. IEEE Computer Graphics and Ap-

plications, 26(4), 28-33.
ISO 19119:2005, Geographic Information – Ser-
vices. ISO/TC 211 Geographic information/Geo-
matics.
Kaye, D. (2003). Loosely coupled: the missing
pieces of Web services, Marin County, California:
RDS Press.
Lara, R., Roman, D., Polleres, A., & Fensel, D.
(2004). A conceptual comparison of WSMO and
OWL-S. Proc. of European Conference on Web
Services 2004, Springer LNCS, 254-269.
Lemmens, R., Wytzisk, A., de By, R., Granell, C.,
Gould, M., & van Oosterom, P. (2006). Integrating
Semantic and Syntactic Descriptions to Chain
Geographic Services. IEEE Internet Computing,
10(5), 42-52.
McIlraith, S.A., Son, T.C., & Zeng, H. (2001).
Semantic web services. IEEE Intelligent Systems,
16(2), 46-53.
Medjahed, B., Benatallah, B., Bouguettaya, A.,
Ngu, A.H.H., & Elmagarmid, A.K. (2003). Busi-
ness-to-business interactions: issues and enabling
technologies. Intl. Journal on Very Large Data
Bases, 12(1), 59-85.
Peng, Z R. & Tsou, M H. (2001). Internet GIS:
distributed geographic information services for
the internet and wireless networks, Hoboken,
New Jersey: John Wiley & Sons, Inc.
Percivall, G. (Ed.) (2002). The OpenGIS Abstract
Specication Topic 12: OpenGIS Service Archi-

tecture, Open Geospatial Consortium. Document
number 02-112
Schut, P. & Whiteside, A. (Eds.) (2005). Open-
GIS Web Processing Service. Open Geospatial
Consortium, Discussion paper. Document number
05-007.
Sheth, A. P. (1999). Changing Focus on Interop-
erability in Information Systems from System,
Syntax, Structure to Semantics. In M. F. Good-
child, M.J. Egenhofer, R. Fegeas & C.A. Kottman
(Eds.), Interoperating Geographic Information
Systems, Kluwer (pp. 5-30).
• Management: monitoring, compliance with standards
• Duplicity: multiple versions of the same geographical phenomena will exist
• Performance: GML vs. binary transmission formats
• Simulation: Hydrologic analysis, pollution analysis, trafc simulation, walk-through simula-
tions, terrain ight simulations, emergency response exercises
• Semantics: geo-ontologies will help to clarify meanings during geo-web service chaining.
Table 3. Challenges in geospatial web service chaining
194
Geospatial Web Service Chaining
Tu, S., Flanagin, M., Wu, Y., Abdelguer, M.,
Normand, E., Mahadevan, V., Ratcliff, J. & Shaw,
K. (2004). Design strategies to Improve Perfor-
mance of GIS Web Services. Proc. of International
Conference on Information Technology (ITCC)
2004 Volume II, Las Vegas, NV.
Vinoski, S. (1997). CORBA: Integrating diverse
applications within distributed heterogeneous
environments. IEEE Communications Magazine,

45(2), 46-55.
Whiteside, A. (Ed.) (2005). OpenGIS web ser-
vices architecture description. Open Geospatial
Consortium, Discussion paper.
key t er Ms
GML: The Geography Markup Language
is a XML grammar dened by OGC to express
geographical features. To help users and devel-
opers to structure and facilitate the creation of
GML-based application, GML provides GML
proles that are XML schemas that extend the
very GML specication in a modular fashion.
A GML prole is a GML subset for a concrete
context or application but without the need for the
full GML grammar, simplifying thus the adoption
of GML and facilitating its rapid usage. Some
common examples of GML proles that have been
published are Point Prole, for applications with
point geometric data, and GML Simple Features
prole, supporting vector feature requests and
responses as the case of WFS.
ISO/TC211: ISO Technical Committee 211 in
Geographic information/Geomatics is in charge
of establishing a set of standards for digital geo-
graphic information concerning objects or phe-
nomena that are directly or indirectly associated
with a location relative to the Earth.
OGC: The Open Geospatial Consortium is an
international industry consortium participating in
a consensus process to develop publicly available

interface specications. OGC members include
government agencies, commercial companies,
and university research groups.
Service Chain: Sequence of services where,
for each adjacent pair of services, occurrence of
the rst service is necessary for the occurrence
for the second service [paraphrased from ISO
19119].
Service: Functionality provided by a service
provider through interfaces [paraphrased from
ISO 19119].
Service Consumer: The role of service con-
sumer requires certain requirements and needs
that are fullled by one or more web services
available over the Internet.
Service Metadata: Metadata describing the
operations and geographic information available
at a particular instance of a service [paraphrased
from ISO 19119].
Service Provider: The role of service provider
provides software applications as web services,
creating functional descriptions and making them
available in public registries.
SOAP: A protocol for exchanging XML-based
messages between services over a computer
network, usually the Internet. A SOAP message
may think of as an envelope that wraps mainly
two elements: a header, with useful information
to interpret the data, and a body, which actu-
ally contains the exchanged data among web

services.
UDDI: A specication/protocol that allows
service providers to publish service descriptions
in a service registry and service consumers to
discover services in a service registry according
to their service descriptions, usually described in
WSDL. The main element of UDDI is the busi-
ness registry, a service registry based on XML
that contains three kind of information for each
195
Geospatial Web Service Chaining
web service published: white pages or contact
information, yellow pages include business
categorization, and green pages that comprise
technical information of the web service along
with a link to its WSDL description.
WSDL: A XML-based specication that al-
lows service providers to describe syntactically
service interfaces. Basically, a WSDL description
allows service providers to describe a web ser-
vice’s function and its input and output parameters
in order to be discovered and invoked by client
applications and other web services.
196
Chapter XXV
Multi-Agent Systems for
Distributed Geospatial
Modeling, Simulation and
Computing
Genong Yu

George Mason University, USA
Liping Di
George Mason University, USA
Wenli Yang
George Mason University, USA
Peisheng Zhao
George Mason University, USA
Peng Yue
George Mason University, USA
Copyright © 2009, IGI Global, distributing in print or electronic forms without written permission of IGI Global is prohibited.
Abstr Act
Multi-agent system is specialized in studying the collective effects of multiple intelligent agents. An
intelligent agent is a computer system with autonomous action in an environment. This technology is
especially suitable for studying geospatial phenomena since they are complex in nature and call for
intertwined actions from different forces. This chapter describes multi-agent systems and their applica-
tion in geospatial modeling, simulation and computing. Geospatial data integration and mining are
discussed.