KNOWLEDGE MOBILIZATION TO SUPPORT

ENVIRONMENTAL IMPACT ASSESSMENT

A Model and an Application

Juli

an Garrido

, Juan G

omez-Romero

, Miguel Delgado

and Ignacio Requena

Department of Computer Science and Artiﬁcial Intelligence, University of Granada, Granada, Spain

Applied Artiﬁcial Intelligence Group, University Carlos III of Madrid, Madrid, Spain

Keywords:

Knowledge mobilization, Environmental impact assessment, Ontologies, Context-aware systems.

Abstract:

EIA is a complex problem due to the wide range of different human impacts, the amount of different envi-

ronmental indicators to measure the effect of an impact, and the correlation among them. In this paper we

propose an application to support knowledge mobilization in Environmental Impact Assessment (EIA) based

on a formal model to represent relevance between context descriptions and domain-knowledge subsets. The

CDS (Context-Domain Signiﬁcance) pattern allows building a tool to assist in the data collection phase to

establish the relevant indicators in a particular scenario.

1 INTRODUCTION

Knowledge Mobilization comprises all the efforts

aimed to take advantage of the features provided by

mobile networks and devices in order to improve

Knowledge Management procedures. The mobiliza-

tion of knowledge consists of making “knowledge

available for real-time use in a form which is adapted

to the context of use and to the needs and cognitive

proﬁle of the user” (Keen and Mackintosh, 2001). In

other words, Knowledge Mobilization should furnish

users with suitable knowledge to support decision-

making wherever they are located.

From a pragmatical point of view, Knowledge

Mobilization addresses the challenge of developing

Knowledge Mobilization Systems, which are ubiqui-

tous, proactive, declarative, context-aware, integra-

tive, and concise (G

omez-Romero, 2008). Several

theories and technologies, most of them from Intel-

ligent Systems and Soft Computing areas, have been

proposed to accomplish mobilization. Among them,

Semantic Web and Ontologies play a crucial role in

building knowledge (Perttunen et al., 2009).

Knowledge management is especially relevant in

application domains affected by information over-

load. Information overload is described as a situation

in which a user is provided with more data than he can

digest, either because sifting through the information

received would take too much time or simply because

interesting facts cannot be separated from irrelevant

data (Eppler and Mengis, 2004). Knowledge Mobi-

lization systems must rely on suitable representations

that take into account what is signiﬁcant or relevant to

the user, which depends on certain factors other than

the query to be solved: environment, preferences, pre-

vious actions, etc. All these pieces of information,

used to characterize the situation of the entities, can

be regarded as context (Dey and Abowd, 2000), and

the results of the query must be adapted to it.

Accordingly, a Knowledge Mobilization system

must include two kinds of knowledge in its support-

ing ontologies: (i) knowledge speciﬁc of the domain

of the application; (ii) knowledge describing contex-

tual situations. The signiﬁcance of a piece of domain-

speciﬁc information in a given context is represented

by a relation between an ontological description of the

situation and an ontological deﬁnition of the domain

knowledge.

There are some approaches in the literature aimed

at the creation of ontologies representing this notion

of context-dependent signiﬁcance. This is the case of

the Context-Domain Signiﬁcance (CDS) pattern (Bo-

billo et al., 2008). This ontology design pattern de-

ﬁnes a set of rules to build a new ontology compliant

to the OWL standard (McGuiness and van Harme-

len, 2004) where context descriptions and domain

expressions are connected through constrained rela-

tions. The main reasoning task within a CDS ontol-

193

Garrido J., Gómez-Romero J., Delgado M. and Requena I..

KNOWLEDGE MOBILIZATION TO SUPPORT ENVIRONMENTAL IMPACT ASSESSMENT - A Model and an Application.

DOI: 10.5220/0003070101930199

In Proceedings of the International Conference on Knowledge Engineering and Ontology Development (KEOD-2010), pages 193-199

ISBN: 978-989-8425-29-4

 2010 SCITEPRESS (Science and Technology Publications, Lda.)

ogy is to retrieve the pieces of domain-speciﬁc knowl-

edge that are signiﬁcant in a given context.

In this paper, we study how the CDS model can

be applied in an Environmental Impact Assessment

(EIA) application to overcome mobilization issues.

EIA is a systematic process to assess the actual or

potential effects of human policies, objectives, pro-

grams, plans, or activities on the local or global en-

vironment. Thus, an EIA study analyzes a complex

system with several factors difﬁcult to understand

due to its large volume and the relationships among

themselves. For example, determining how a road

route can affect an ecosystem, or an industrial activ-

ity might affect the birds or an indigenous specie of

worm. The CDS model allows the summarization of

the number of factors that should be considered by

the auditor, according to the available extra informa-

tion about the environment analyzed and the purpose

of the study. We present a Knowledge Mobilization

system prototype for EIA based on the CDS that facil-

itates the labour of environmental experts by simpli-

fying data collection procedures and adding detailed

annotations to this information.

The paper is structured as follows. In Sect. 2,

we discuss some research works on the development

of knowledge-based solutions to EIA. In Sect. 3, we

describe the use of the CDS ontology-design pattern

to develop the knowledge base of our application, as

well as the functioning of the algorithm to retrieve

signiﬁcant knowledge from context descriptions. Sec-

tion 4 describes the implementation and a use case of

the Knowledge Mobilization system. The paper ends

with some conclusions and plans for future work.

2 RELATED WORK

Environmental applications require the participation

of a large amount of information sources, which

makes convenient the use of proper representation

formalisms. Different approaches to knowledge-

based environmental tools and models have been de-

veloped in the last years. These contributions incor-

porate formal knowledge representations in their data

model, and, speciﬁcally, ontologies.

OntoWEDSS is a decision-support system for

wastewater management. OntoWEDSS extends clas-

sic rule-based reasoning and case-based reasoning

with a domain ontology, which results in more ﬂex-

ible management capabilities. The ontology models

the treatment process, provides a shared and common

vocabulary, and improves the communication among

different elements and agents of the system (Cecca-

roni et al., 2004).

The SEEK project

aims at the development of

an infrastructure to support the whole process of

ecological and biodiversity data management, from

the acquisition stage to the synthesis and integration

stage. SEEK implements a semantic mediation mod-

ule, which is an advanced reasoning system that can

determine whether relevant data should be automat-

ically transformed for use with a selected workﬂow.

The Extensive Observation Ontology (OBOE) frame-

work has been developed in the context of the SEEK

project (Madin et al., 2007). OBOE includes an on-

tology for describing and synthesizing ecological ob-

servational data. The framework allows capturing the

process of ecological ﬁeld observation and measure-

ment, facilitating logic-based reasoning and making it

possible data discovery, summarization, and integra-

tion processes. A previous environmental ontology

is Ecolingua, which aims at representing ecological

quantitative data (Brilhante, 2004).

In (Oprea, 2005), the authors present a

knowledge-based system that addresses the problem

of air pollution control decision support. In (Batzias

and Siontorou, 2006), a decission support system

for bio-monitoring is described. This system uses

a bio-indicator ontology of environmental exposure

that links pollution to morphological response and

biochemical alterations. In (Chau, 2007), the authors

develop an ontology-based system for assisting

engineers in the management of knowledge about

ﬂow and water quality. The system aims at simulating

the behavior of human experts in problem solving

by using descriptive, procedural, and reasoning

knowledge. The problem of model-based water

management is also faced in (Scholten et al., 2007).

These authors have implemented an ontology-based

tool that supports the simultaneous treatment of

two or more domains as an integrated domain

(multidomain approach). The SOLERES project is

a spatial-temporal information system for environ-

mental management and automatic generation of

ecological maps from satellite images. SOLERES is

based on a knowledge representation module used to

model the environmental information (Padilla et al.,

2008).

3 DEVELOPMENT OF THE

CONTEXT-DEPENDENT

ONTOLOGY FOR EIA

In this section, we will explain how the CDS pat-

tern is applied to build the supporting ontology of the

http://seek.ecoinformatics.org

KEOD 2010 - International Conference on Knowledge Engineering and Ontology Development

194

knowledge mobilization system for EIA-assistance.

Essentially, the CDS pattern deﬁnes how to build a

new OWL ontology where concepts representing con-

textual situations and concepts representing domain

knowledge are connected through relations. We will

exemplify how to create context descriptions, domain

expressions, and relations in the EIA domain.

The new ontology is used to retrieve all the con-

cepts in the domain ontology which ought to be con-

sidered in a given context. The CDS pattern provides

a suitable algorithm to perform this operation based

on the execution of various basic DL inference tasks.

We will show the functioning of the algorithm with an

example in the EIA application.

For a comprehensive description of using and rea-

soning with the CDS pattern, we refer the reader to

(Bobillo et al., 2008). We refer the reader unfamil-

iar with Description Logics formalisms and the OWL

language to the book edited by (Baader et al., 2003).

3.1 Knowledge Base Development

A CDS ontology

is built from two basic sub-

ontologies, one representing domain-speciﬁc knowl-

edge and another deﬁning a vocabulary to describe

context situations. The domain-speciﬁc ontology

) contains the knowledge required to solve the

concrete problem that the system is facing. The con-

text ontology (K

) contains the knowledge required

to describe situations that determine which informa-

tion of the domain is relevant. The signiﬁcance on-

tology (K

) is a new ontology where complex con-

texts, complex domains, and links between them are

deﬁned.

Domain Ontology. In EIA, the domain ontology

abstractly represents the information that the audi-

tor has to collect, i.e., the indicators required by an

We note an ontology as a triple K = hT , R , Ai, where

T (the TBox) and R (the RBox) contain, respectively,

axioms about concepts and roles (terminological axioms),

and A (the ABox) contains axioms about individuals (as-

serts). The symbols used in K are its signature or vo-

cabulary. Formally, the signature is the disjoint union

S = C ] R ] I, where C =

{

}

is the set of atomic con-

cepts (or classes); R =

{

}

the set of atomic roles (or

properties); and I =

{

a, b, . . .

}

the set of individuals (or in-

stances). From the atomic elements in S, new complex con-

cepts Con(S) = {C

(i)

, D

(i)

, . . .}, roles Rol(S) = {R

(i)

}, and

axioms Ax(S) = {O

(i)

} can be composed (subscripts are not

used when disambiguation is not needed). By extension, the

signature S(O) of an axiom (respectively for roles and con-

cepts) is the set of atomic elements of S which are included

in O (respectively R and C). The signature of an ontology

S(K ) is the union of all the signatures S(O) of the axioms

in K .

environmental assessment methodology. An indica-

tor is a simple measurement of environmental fac-

tors or biological species that are representative of the

characteristics of a biophysical system. That is, in-

dicators provide interesting measures about the state

of the whole ecosystem or a speciﬁc part. Indica-

tors, represented as instances of the subclasses of the

concept Indicator, have been collected from spe-

cialized literature (Garmendia et al., 2005; Barettino

et al., 2005). A comprehensive description of the in-

dicators used is out of the scope of this paper –the on-

tology domain contains currently seventy six different

indicators.

Companies, represented as instances of the con-

cept Company, may have related several environmen-

tal assessments developed as a result of their ac-

tivities. Environmental assessments are stored as

instances of the concept EIA. The object prop-

erty hasEIA relates companies and EIAs. An

EIA instance represents a set of indicator measures

in a given period of time. EIA instances and

Indicator instances are related with the object prop-

erty hasMeasurement.

From the indicators ontology, complex domain ex-

pressions can be built. These new concept deﬁni-

tions built from elements of the domain-speciﬁc on-

tology and ontology concept constructors are called

complex domains (noted as D

). For example,

the complex domain D

≡ GroundwaterQuality t

BiodiversityIndex represents a set including indi-

cators about water quality and biodiversity.

Context Ontology. The context knowledge, in turn,

correspond to the possible activities that cause the de-

velopment of the EIA process. Depending on the type

of activity, the auditor focuses on different indicators.

Thus, the context ontology establishes a vocabulary

to describe activities from human actions, places, and

installations. For example, concepts of the context

ontology are Agriculture and WaterExtraction,

which are hazardous activities.

New concept deﬁnitions built from elements

of the context ontology and ontology concept

constructors are called complex contexts (noted

as C

). For example, an agrarian cultivation

which uses irrigation, fertilizers, and treatments

for insects is represented with the complex con-

text C

≡ Agriculture uFertiliserTreatment u

Irrigation u PesticideTreatment.

It is important to notice that D

and C

are not part

of the domain and the context ontology, respectively.

As mentioned, these ontologies provide a vocabulary

to build more complex concepts, which may not be

deﬁned in these ontologies.

KNOWLEDGE MOBILIZATION TO SUPPORT ENVIRONMENTAL IMPACT ASSESSMENT - A Model and an

Application

195

Table 1: Example of CDS knowledge model.

When agricultural activities involving pruning processes are going to be developed, it is neces-

sary to check the index of biodiversity, soil humidity and vegetal coverage.

≡ Agriculture u Pruning

≡ BiodiversityIndex t SoilHumidity t VegetalCoverage

1,1

≡ ∃hasAction.C

u ∃hasIndicator.D

When agricultural activities involving pruning processes and burning of its remainders are going

to be developed, it is necessary to check the index of biodiversity, soil humidity and vegetal

coverage, and the affected area by smells.

≡ Agriculture u Pruning u Burning

≡ BiodiversityIndextSoilHumidity tVegetalCoveragetAffectedAreaBySmells

2,2

≡ ∃hasAction.C

u ∃hasIndicator.D

When agricultural activities involving groundwater extraction and sloped land divisions into ter-

races are going to be developed, it is necessary to check the index of biodiversity, soil humidity,

vegetal coverage, soil loss and the alteration of the phreatic level.

≡ Agriculture u WaterExtration u TerraceControl

≡ BiodiversityIndex t SoilHumidity t VegetalCoverage t SoilLossRates t

PhreaticLevelAlteration

3,3

≡ ∃hasAction.C

u ∃hasIndicator.D

When agricultural activities involving groundwater extraction, fertilizer use, insects and weeds

control are going to be developed, it is necessary to check the affectation of groundwater quality,

biodiversity, soil humidity, vegetal coverage, alteration of the phreatic level, and soil contents in

metals and salts.

≡ Agriculture u WaterExtration u Fertilizer uInsectsControl u WeedsControl

≡ BiodiversityIndex t SoilHumidity t VegetalCoverage t

PhreaticLevelAlteration t GroundwaterQuality t ContentInMetal t

ContentInSalts

4,4

≡ ∃hasAction.C

u ∃hasIndicator.D

Signiﬁcance Ontology. The signiﬁcance ontology

is the model where complex contexts C

, complex do-

mains D

, and links between them are deﬁned. These

links, called σ-connections (connections representing

signiﬁcance), state that the domain-speciﬁc knowl-

edge D

should be considered in situation C

. In our

case, a σ-connection establishes that a set of environ-

ment indicators must be measured or calculated when

the impact of a set of human actions is evaluated.

A σ-connection concept is a new concept repre-

senting a σ-connection. It is deﬁned with existential

restrictions on the complex context and the complex

domain that it links (via properties R

and R

). When

there is no possibility of confusion, we use the term

σ-connection instead of σ-connection concept.

ontology is developed as follows:

Deﬁnition 1 . Let K

and K

be, respectively, the

domain and context ontologies, C

a complex con-

text, and D

a complex domain. The signiﬁcance or

CDS ontology that relates the set of pairs (C

, D

) is a

consistent ontology K

= hT

, R

, A

i such that T

(non-exclusively) includes deﬁnitions for the concepts

, C

, D

, C

, D

, P

i, j

, which satisfy:

1. P

, C

, D

are the superclasses “σ-connection

concePt’, “complex Context” and “complex Do-

main”:

• P

i, j

v P

, C

v C

, D

v D

2. R

is the bridge property linking σ-connections

and complex contexts:

• P

v ∀R

3. R

is the bridge property linking σ-connections

and complex domains:

• P

v ∀R

4. P

i, j

is the σ-connection linking the complex con-

text C

and the complex domain D

• P

i, j

≡ ∃R

u ∃R

KEOD 2010 - International Conference on Knowledge Engineering and Ontology Development

196

In its deﬁnition, the CDS pattern assumes that the

domain and the context ontologies are disjoint, since

context usually encompasses external to the interest

area. This is not our case and, without loss of gen-

erality, (non-complex) context and domain terms are

included in the same ontology.

Table 1 shows a brief example of the CDS ontol-

ogy for the EIA application. The bridge properties are

≡ hasAction and R

≡ hasIndicator. It can be

observed that C

v C

, D

v D

and, for the concept

≡ Agriculture t WaterExtraction, C

v C

and C

v C

3.2 Reasoning

The main reasoning task involving a signiﬁcance on-

tology consists in ﬁnding all the concepts in the do-

main ontology which ought to be considered in a

given context. This is the signiﬁcant domain for a

context, and is formally deﬁned as follows:

Deﬁnition 2 . Let K

be a domain-speciﬁc ontol-

ogy, K

a context ontology, and K

a CDS ontology,

with their respective signatures S(K

), S(K

), and

S(K

). Let Con(K

) be the set of composite con-

cepts that can be built from the primitive concepts of

. Let scenario E be a concept E ∈ S(K

The domain knowledge in K

that is signif-

icant in the scenario E w.r.t. K

, denoted as

D(E, K

), is a set which includes the concepts I such

that D(E, K

) = {I | I ∈ Con(K

) ∧ K

|= {E v

, P

n,m

v P

, I v D

}}

Concepts in I can be retrieved by performing sev-

eral subsumption tasks. Formally:

Algorithm 1: D(E, K

) can be computed in practice

as follows.

1. {C

} = {C

v C

| E v C

}

2. {P

k,l

} = {P

k,l

v P

| (P

k,l

v ∃R

)∧ (C

≡ C

)}

3. {D

} = {D

v D

| (P

k,l

v ∃R

) ∧ (D

≡

)}

4. D(E, K

) = {I ∈ Con(K

) | I v D

}

As a matter of example, assuming the CDS model

depicted in Table 1, let us suppose that we want

to retrieve the indicators that are relevant in a sce-

nario where the following actions are to be per-

formed: Agriculture, Pruning, Burning, and

WaterExtraction. This is accomplished by us-

ing Algorithm 1 to calculate the restricted domain

of the complex context concept E ≡ Agriculture u

Prunning u Burning u WaterExtraction.

1. C

= {C

, C

}

2. P

k,l

= {P

1,1

, P

2,2

}

3. D

= {D

, D

}

4. I = {BiodiversityIndex,

SoilHumidity,

VegetalCoverage,

AffectedAreaBySmells}

Accordingly, the signiﬁcant indicators in this case

would be those corresponding to the concepts in-

cluded in I.

4 EIA-ASSISTANCE

APPLICATION

The IASEIA application (Intelligent ASsistant for En-

vironmental Impact Assessment) is a prototype of a

Knowledge Mobilization system to support auditors

to carry out EIA process. This application uses a CDS

model developed according to the speciﬁcations and

the examples presented in the previous section.

To create the model, we have used the CDS plugin

for Prot

e and the CDS API (Bobillo et al., 2008).

The CDS plugin is a tool that allows knowledge engi-

neers to create, edit, test and reason with a CDS ontol-

ogy. The plugin adds a new tab to the Prot

e-OWL

environment where a user-friendly view of the CDS

ontology is displayed and queries can be introduced.

The CDS API is a Java library to programmatically

manage models created with the CDS pattern. It in-

cludes methods to carry out the most common pro-

cesses involving a CDS model: deﬁnition of complex

contexts and domains, creation of proﬁles, retrieval of

signiﬁcant domain knowledge, etc. A beta version of

these tools can be found in the web

The architecture of IASEIA is depicted in Fig-

ure 1. It is divided into three parts: the client agent,

the server agent, and the knowledge-data layer. The

client agent allows the user to access to the applica-

tion by using a standard web browser with an internet

connection. Since reasoning and complex operations

are accomplished in the server side, the device con-

sumes little resources and the application works suc-

cessfully with any common browser.

4.1 Architecture and Development

The server agent is responsible for the communica-

tion with the client side by providing forms based on

HTML, JSP (Java Server Pages) and AJAX (Asyn-

chronous JavaScript And XML) technologies to show

http://decsai.ugr.es/ jgomez/thesis/

KNOWLEDGE MOBILIZATION TO SUPPORT ENVIRONMENTAL IMPACT ASSESSMENT - A Model and an

Application

197

IASEIA

Server Agent

HTTP

OWL

IASEIA

Client Agent

Hazardous

Activities

CDS

Ontology

CDS API

Query resolution

JSP

AJAX

EIA

Indicators

Pellet

MySQL

Database

Web

Interface

Web

Browser

Figure 1: Architecture of the EIA support application.

and retrieve information. This module has also con-

trol over the query resolution and the CDS API mod-

ules (the knowledge-data layer), whose main task is

the extraction of the relevant knowledge for the sce-

nario considered by a speciﬁc context.

The knowledge-data layer consists of the con-

text ontology model, the domain ontology model, the

CDS ontology model, and a MySQL database. As it

is explained in Sect. 3, the three models are needed

for the query resolution and extraction of the relevant

domain and they are allocated into an ontology server.

The database is used to store the collected information

during the environmental assessment.

4.2 Use

The functioning of the IASEIA system is the follow-

ing. Let us suppose an auditor that needs to perform

an environmental assessment for a company, installa-

tion, or activity. The auditor can move to the place

to be evaluated bringing a smartphone or a similar

device with navigation capabilities. The auditor in-

spects the place and accesses to the query form in the

web application (see ﬁgure 2). At this point, the au-

ditor can choose the set of actions best suited to the

real scenario from the structured list retrieved of the

context knowledge base. Once the current scenario is

selected, the auditor sends back the form to the IA-

SEIA application.

Figure 2: Query form for IASEIA.

The server agent receives the description of the

impacting actions that will be performed and uses

them as the input for the CDS API implementation

of the CDS reasoning algorithm. The algorithm ﬁnds

out the signiﬁcant domain for the context description

–i.e., the relevant indicators–, which is used to ﬁll the

form that is sent to the auditor (see ﬁgure 3).

Figure 3: Data form for IASEIA.

The indicators form offers a list of indicators and a

corresponding text ﬁeld for each of them. The auditor

must ﬁll the form with the measured values, besides

some additional data: the date of the assessment, the

company, etc. Finally, the collected data are sent to

the server, which acknowledges the values by display-

ing them after checking that the transaction is right.

5 CONCLUSIONS

This paper presents an ontology-based system to as-

sist environment data acquisition for environmental

impact assessment, a problem affected by informa-

tion overload. The IASEIA application facilitates the

selection of the most appropriate environmental in-

KEOD 2010 - International Conference on Knowledge Engineering and Ontology Development

198

dicators to be evaluated for a given set of impacting

actions.

The knowledge base of IASEIA has been devel-

oped by relying on the CDS pattern, an ontology de-

sign pattern to create ontologies that allows the char-

acterization of the signiﬁcant domain information of

a given context. The CDS pattern promotes modu-

larity and reusing of the knowledge models. Thus, it

is possible to enhance the models by adding or cre-

ating more speciﬁc ontologies covering more detailed

aspects of the EIA process. The context and the do-

main ontologies can be easily extended with addi-

tional knowledge to describe additional actions and

indicators. Building a comprehensive actions ontol-

ogy is a difﬁcult task, and therefore it may be neces-

sary to incorporate new terms to describe actions that

were not predicted in the development of the context

ontology. Similarly, indicators may change if differ-

ent assessment methodologies are used.

Actually, our ongoing and future work is focused

on the enhancement of the knowledge models under

the supervision of environmental experts. The devel-

opment of more accurate models will serve as a basis

for the future real use of the system, both in indoors

and outdoors scenarios.

ACKNOWLEDGEMENTS

This work has been supported by the projects P07-

TIC-02913, P08-RNM-03584, TIN06-15041-C04-

01, and CAM CONTEXTS (S2009/TIC-1485).

REFERENCES

Baader, F., Calvanese, D., McGuinness, D., Nardi, D.,

and Patel-Schneider, P. F. (2003). The Description

Logic Handbook: Theory, Implementation, and Ap-

plications. Cambridge University Press.

Barettino, D., Loredo, J., and Pend

as, F., editors (2005).

Acidiﬁcation of soils and waters: problems and solu-

tions (In spanish). Instituto Geol

ogico y Minero de

Espa

na.

Batzias, F. A. and Siontorou, C. G. (2006). A knowledge-

based approach to environmental biomonitoring. En-

vironmental Monitoring and Assessment, 123(1-

3):167–197.

Bobillo, F., Delgado, M., and G

omez-Romero, J. (2008).

Representation of context-dependant knowledge in

ontologies: a model and an application. Expert Sys-

tems with Applications, 35(4):1899–1908.

Brilhante, V. (2004). An ontology for quantities in ecology.

Lecture Notes in Computer Science, 3171:144–153.

Ceccaroni, L., Cort

es, U., and S

anchez-Marr

eb, M. (2004).

OntoWEDSS: augmenting environmental decision-

support systems with ontologies. Environmental Mod-

elling and Software, 19(9):785–797.

Chau, K.-W. (2007). An ontology-based knowledge man-

agement system for ﬂow and water quality modeling.

Advances in Engineering Software, 38(3):172–181.

Dey, A. and Abowd, G. (2000). Towards a better under-

standing of context and context-awareness. In Pro-

ceedings of the Workshop on the What, Who, Where,

When, and How of Context-Awareness (CHI 2000),

The Hague, Netherlands.

Eppler, M. J. and Mengis, J. (2004). The Concept of In-

formation Overload: A Review of Literature from

Organization Science, Accounting, Marketing, MIS,

and Related Disciplines. The Information Society,

20(5):325–344.

Garmendia, A., Salvador, A., Crespo, C., and Garmen-

dia, L. (2005). Environmental Impact Assessment (In

spanish). Pearson Educaci

on.

omez-Romero, J. (2008). Knowledge Mobilization: Ar-

chitectures, Models and Applications. PhD thesis,

University of Granada.

Keen, P. G. and Mackintosh, R. (2001). The freedom econ-

omy. McGraw-Hill Professional.

Madin, J., Bowers, S., Schildhauer, M., Krivov, S., Pen-

nington, D., and Villa, F. (2007). An ontology for de-

scribing and synthesizing ecological observation data.

Ecological Informatics, 2(3):279–296.

McGuiness, D. and van Harmelen, F. (2004). OWL Web

Ontology Language Overview. Online. W3C Recom-

mendation.

Oprea, M. (2005). A case study of knowledge modelling

in an air pollution control decision support system. AI

Communications, 18(4):293–303.

Padilla, N., Iribarne, L., Asensio, J. A., Mu

noz, F. J.,

and Ayala, R. (2008). Modelling an Environmen-

tal Knowledge-Representation System. In WSKS

’08: Proceedings of the 1st world summit on The

Knowledge Society, pages 70–78, Berlin, Heidelberg.

Springer-Verlag.

Perttunen, M., Riekki, J., and Lassila, O. (2009). Context

Representation and Reasoning in Pervasive Comput-

ing: a Review. International Journal of Multimedia

and Ubiquitous Engineering, 4(4):1–28.

Scholten, H., Kassahun, A., Refsgaard, J. C., Kargas,

T., Gavardinas, C., and Beulens, A. J. M. (2007).

A methodology to support multidisciplinary model-

based water management. Environmental Modelling

and Software, 22(5):743–759.

KNOWLEDGE MOBILIZATION TO SUPPORT ENVIRONMENTAL IMPACT ASSESSMENT - A Model and an

Application

199