79

A publication of

CHEMICAL ENGINEERING TRANSACTIONS

The Italian Association
VOL. 31, 2013 of Chemical Engineering
Online at: www.aidic.it/cet
Guest Editors: Eddy De Rademaeker, Bruno Fabiano, Simberto Senni Buratti
Copyright © 2013, AIDIC Servizi S.r.l.,
ISBN 978-88-95608-22-8; ISSN 1974-9791

Territorial Risk Analysis and Mapping

Ales Bernatik*; Pavel Senovsky; Michail Senovsky, David Rehak
VSB-Technical University of Ostrava, Faculty of Safety Engineering, Lumirova 13, Ostrava -Vyskovice, 700 30,
Czech Republic
[email protected]

The aim of this contribution is to present the principles of risk analysis for territories that may have different
sizes, e.g. municipalities, towns, districts and regions. The article summarises approaches to the issues of
major risk assessment for stationary and mobile risk sources and natural disasters having effects on
human health and life, property and the environment.
Vulnerability and risk mapping is usually carried out using the HVA (Hazard Vulnerability Analysis) method.
For territorial assessment this method was promoted by the project SIPROCI (International Response to
Natural and Man-made Catastrophes SIPROCI), which in the years 2004-2007 asked participating
countries to predict, prevent and respond to natural and man-made disasters. One of the project outputs
was the development of a uniform method for risk mapping.
The SIPROCI method of risk mapping works with the calculation of individual values of hazard,
vulnerability and subsequently risk and their integration into the form of an index that can be simply
graphically represented on maps. The SIPROCI method is relatively general – creates merely a certain
methodical framework that is to be adapted for use in specific countries. During index selection, both
national specifics concerning the organization of individual rescue services of risk sources and data
available for the use of the method should be respected.
The application of the SIPROCI method in conditions of the Czech Republic and the pilot application of the
method were carried out by the Fire and Rescue Service of the Moravian-Silesian Region. The article
mainly deals with the presentation of a proposal for the method of optimized risk mapping, the results of
which can be transferred through target indexes to a GIS map layer.

1. Introduction
Territorial risk analysis and mapping represent a problem that is, in connection with considerations about
the protection of territorial critical infrastructure, becoming increasingly important. Traditionally, risks
following only from a certain considered scenario of an accident, e.g. in the chemical industry (loss of
control over hazardous substances) and impacts of natural disasters, such as floods in the territory, are
understood merely separately (Vallée and Duval, 2012; Renni et al., 2010).
With time, we have found that some independent events have a potential for significant increasing
negative consequences (synergic effects), whereas others can result in a cascade of failures (domino
effect). Thus, if we want to assess risks in a certain territory, we have to do this comprehensively for all
risks simultaneously.
st
This approach began to gain ground at the beginning of the 21 century. In this context, we can mention
the Swiss KATARISK (BZS, 2003) and perhaps the HVA method, sometimes designated as the HVE
method - H – hazard, V – vulnerability, E – value of the elements at risk (SIPROCI, 2007). For territorial
assessment the SIPROCI project (International Response to Natural and Man-made Catastrophes
SIPROCI) promoted this method; in the years 2004 – 2007, the project was aimed at helping participating
countries predict, prevent and respond to natural and man-made disasters. One of the project outputs was
the development of a uniform method for risk mapping.
 80

2. HVA Method in the Czech Republic

The SIPROCI method for risk mapping works with the calculation of individual values of hazard,
vulnerability and subsequently risk and their integration into the form of an index that can be simply
graphically represented on maps. The method calculates the value of risk as a function of hazard,
vulnerability and value of the element at risk, see Eq. 1 and 2.

R = f ( H ,V , E ) (1)
Rs = f ( H , V )
(2)

where: R risk
Rs simplified risk
H hazard
V vulnerability
E value of the element at risk.

The HVA is relatively general; it creates merely a certain methodical framework that is to be adapted for
use in specific countries. In the course of application, national specifics of the organisation of individual
rescue services of risk sources and also data available for processing are to be respected.
The HVA method was applied in conditions of the Czech Republic by Krömer et al. (2010) together with a
pilot study of the Moravian-Silesian Region. This application of HVA method will be henceforth referred to
as HVA CZ. From the point of view of concept of risk, the method works with a simplified risk – thus it
ignores the value of the element at risk. For particular types of hazards, Krömer et al. determine the risk
according to Eq. 3. In contrast to the HVA method itself, it however also deals with the cumulation of
individual risks for the given part of the territory (see Eq. 4).

Ri = MRi ⋅ Z
(3)
n
Rkum = ∑ Ri
i =1 (4)

where: Ri risk associated with the given type of hazard

Rkum cumulative risk for the assessed part of territory
MR risk level
Z territorial vulnerability.

Unlike the HVA method, Krömer et al. (2010) also propose the introduction of indexes that are easy to
visualise using GIS tools and that describe basic characteristics of the territory under evaluation. It is
above all a case of vulnerability index describing the proneness of the territory to damage, preparedness
index describing the ability of rescue services to respond to emergencies and minimise thus their
consequences. In the figure given below (Figure 1), a final map of cumulated risk is given as an example.

Figure 1: Map of cumulated risk

 81

3. Other Methods of Territorial Risk Analysis

3.1 Comparison of KATARISK and HVA CZ Methods

The KATARISK takes into account altogether 22 hazard types, whereas the HVA CZ considers merely 15
hazard types. In spite of this, differences are not too great and are caused rather by regional differences.
For instance, in the Czech Republic earthquakes do not threaten and meteorite impacts and some hazards
included in the social area, such as migration problems are not considered.
Much larger differences between both the methods consist in measurement of the extent of possible
damage. The KATARISK takes into account five indicators, namely the number of affected people, the
number of evacuees, the number of people who are a result of an extraordinary event in need, the extent
of the affected environment and damage. For the considered scenarios, these indicators are quantified,
subsequently converted into money, and the consequence is classified, according to the severity, extent
and incidence, into one of five classes.
The HVA CZ proceeds philosophically in a different way; it uses six indicators, namely the hazard to the
population, the affected area, the extent of the affected environment, the hazard to buildings and estates,
the hazard to animal husbandry, the interruption of transport. In contrast to the KATARISK, the dimension
of the class of events is missing here. This is given by the orientation of the whole method to the use of
GIS means – consequences are simply put as needed into map bases. The second difference is
intentional avoidance of the necessity of a financial expression of consequences of the considered
scenarios, because such an expression requires the long-term collection and evaluation of statistical data
on extraordinary events and their consequences.
3.2 General Methods
Another significant tool is the method of Spatial Development Safety Assessment that was developed in
the framework of a habilitation thesis (Rehak, 2011). The method is based on the principle of semi-
quantitative assessment of potential negative aspects of spatial development and areas of their possible
impact. The aim of the method is to assess realistically a potential risk following from spatial development,
and to contribute thus to ensuring social, environmental and technical safety. The method calculates the
level of risk as a function of probability of occurrence and severity of impact of an undesirable event due to
the influence of negatively acting spatial development and the sum of adopted safety measures, see Eq. 5.

f ( RS , RE , RT ) f [ f ( PS , RI S ), f ( PE , RI E ), f ( PT , RI T )]
R= n
= n

∑S ∑S
(5)
i i
i =1 i =1

where: RS(E,T) level of the potential risk of breach of social (environmental, technical) safety due to
negatively acting spatial development,
PS(E,T) probability of interaction between negatively acting spatial development and population
(the environment, technical infrastructure),
RIS(E,T) severity of the impact of undesirable interaction on population (the environment,
technical infrastructure),
Si degree of specific safety measure.
The final level of the risk of negative acting of spatial development on population (the environment,
technical infrastructure) is illustrated in a matrix of the influence of spatial development using three
possible categories of risk level. Category A signals a low risk that is evaluated as acceptable, it means
that the intended spatial development project can be fully implemented. Category B signals an increased
risk that will have to be reduced, i.e. the intended spatial development project has to be analysed, re-
planned and newly evaluated. Category C signals a high risk that is evaluated as inadmissible, i.e. by the
intended implementation of spatial development project, the population, environment and engineering
infrastructure would be damaged.
3.3 Specific Methods
One of significant specific areas of territorial risk analysis is flood risk assessment. Objective procedures
for the assessment of flood hazard level, the expression of flood risk and the determination of the amount
of possible damage belong to highly topical problems of importance to the whole society. The Directive of
the European Parliament and of the Council on the assessment and management of flood risks (Directive,
2007) requires Member States to assess gradually flood hazards and risks in their territories, and to
process acquired information into the form of a relevant map expression. Final flood risk management
 82

plans are to be completed by the 22nd December 2015. For their preparation, a risk matrix method was
recommended in the Czech Republic (Beffa, 2000; Riha et al., 2005).
The risk matrix method is one of the simplest procedures for the assessment of potential hazard and risk in
floodplains. The method does not require any quantitative estimate of damage caused by the overflow of
rivers, but it expresses the flood risk by means of a four-level scale. In the first step, the result of the risk
matrix method is hazard maps that show by means of a colour scale hazard categories for areas in a
floodplain (Drbal et al., 2005). These categories enable the assessment of suitability of existing and future
functional land use and of recommendations for limiting potential activities in areas in a floodplain having
rather high hazard levels. This procedure can be utilised, e.g. in the spatial planning process, in the
proposal for flood protection measures, and others.
In the risk matrix, the hazard level is given by combination of flood event probability and suitable
expression of flood hazard and consequences. Those can be expressed by the above-mentioned flood
intensity or by verbal evaluation of the hazard in a pre-selected range. The final hazard RI is evaluated in
this case as the maximum value of particular partial hazards RIi corresponding to the i-th scenarios of
hazard (passage of the N-year peak discharge) according to Eq. 6:
n
RI ( x, y ) = max RI i ( x, y ) (6)
i =1

where: n stands for the number of evaluated (input) flood hazard scenarios.
The levels of hazard and risk are plotted on hazard and risk maps. With individual risk levels expressed in
colours or figures, recommendations and limits for the use of a relevant territory are associated.
In connection with the risk matrix method, a method for the determination of flood risks and potential
damage in a floodplain (Drbal et al., 2008) was developed in the framework of the project DIBAVOD in the
year 2008. The essence of this method is an estimate of risk on the basis of potential flood damage. For
this purpose, potential damage is determined according to the categories of property and includes damage
to buildings, roads, public utilities, bridges and water management infrastructure, agriculture, industries,
facilities of buildings (dwellings and amenities), areas for sports and to large economic entities. The
expression of the flood risk on the basis of potential damage is then based on the distribution of probability
of annual peak discharge and is calculated by means of the following Eq. 7:
Qext

R= ∫ D(Q ) ⋅ f (Q ) ⋅ dQ
Qn
(7)

where: R average annual economic flood risk in [CZK/year],

3
D(Q) damage [CZK] at discharge Q [m /s],
f(Q) density of annual peak discharge probability,
Qn minimum discharge at which damage occurs,
Qext extreme discharge at which damage probability is close to zero.
Procedures applied in the method focus largely on key flood factors, namely the flood hazard, which
results in flooding, territorial vulnerability, which manifests itself in the proneness of buildings and facilities
to damage owing to low resistance to extreme flood load and owing to so-called exposure, moreover the
already-mentioned flood risk, which can be expressed, among other matters, by the level of probability of
occurrence of the undesirable event, and finally, flood damage, evaluated implicitly as direct effects of the
flood event that adversely affect a certain territory.

4. Proposal for Improving the HVA CZ Method

The authors of the HVA CZ replace the financial evaluation of indicators by the introduction of
dimensionless coefficients of intensity level of the manifestation of the given scenario in the given
indicator. As a result the calculated risk is then necessarily dimensionless as well. A weakness of the
procedure is just the determination of a function of the behaviour of the coefficient and also the
significance of individual indicators, to which the method applies weight coefficients. The weight value itself
is another problematical point of the method, because in this case, the weight contains both the technical
aspect of the subject matter and the dimension of social acceptance of impacts.
The outputs of the method are then easy to compare from the point of view of individual indexes of
observed characteristics. On the other hand, the back identification of causes – what is the reason, is very
difficult with the HVA CZ.
 83

Another significant proposal for a modernisation of the HVA CZ method is the extension of it by the
evaluation of already applied safety measures. As a consequence, the method will be usable in the cyclic
process of risk management, because in the existing form it is applicable only on the project base, i.e.
once for all. On this basis, a new formula for the calculation of territorial cumulative risk is proposed, see
Eq.8.
n

∑ MR ⋅ Z i
Rkum = i =1
n
(8)

∑S
i =1
i

where: Rkum cumulative risk for the assessed part of territory,

MRi level of risk for the given type of hazard,
Z territorial vulnerability,
Si degree of specific safety measure.
Generally, it is the acquisition of statistical data on extraordinary events for the need of these methods of
risk analysis that is a problem. Although in the Czech Republic, detailed statistics are kept, they are kept
by various institutions, especially the Fire and Rescue Service of the Czech Republic and insurance
companies. For successful quantification, data from these statistics should be processed together.
However, this requires a broad consensus on the provision of these data and also the method of
evaluation of the data, and last but not least, time and financial means. In such a case, cooperation should
be ensured by the state as e.g. in Switzerland.
The second reason is more complicated and its cause can be found in understanding the position of the
state and its inhabitants. Inhabitants in the Czech Republic in principle expect the state to take care of
them. In such environment, information that the risk of life hazard cannot be fully eliminated and that
emphasis should be put during the implementation of protection measures also on their financial
effectiveness is then disclosed only with difficulty. In the Czech Republic, this leads to the top level of
safety (e.g. in the area of protection of building constructions from the effects of the fire, in technical safety
of large transport construction works, such as road tunnels) and, on the other hand, such an approach
does not motivate investors (above all the state) to invest effectively, i.e. what is generally taken as a
problem and not necessarily what is technically justified as a problem is solved.

5. Discussion
The above-mentioned methods have one element in common – they deal with the problems of risk in a
pure technical way. Specific quantities will be quantified, put into formulae, and a cumulative risk value at a
chosen geographic scale will be the result. The value of resulting cumulative risk will enable us to compare
individual territorial units for which the calculation was performed (to decide which territory is more or less
at risk).
Such a piece of information is undoubtedly useful as such. However, the acquisition of it is not easy if the
assessment is to correspond to reality (especially in the sense of derivation of individual coefficients for
calculations). For this reason, certain simplifying assumptions are adopted in the case of application of the
methods in practice, e.g. assumptions of the character of development (e.g. the number of storeys of
blocks of flats, the number of persons living in the territory being assessed, etc.) in the case of flood risk
assessment.
When applying the above-mentioned methods, input data is assumed to be burdened with the random
error that will be compensated in a sufficiently large territory. However, let us ask the question to
ourselves: what are we able to do with the risk? The general theory of risk says that we can accept the
risk, try to reduce it by protection measure implementation or to prepare ourselves for handling
extraordinary events (preparation of reserve funds, etc.). Nevertheless, the value of cumulative risk for the
selected territorial unit does not make it possible to reveal back (in the sense of simple revelation) the
essence of risk – e.g. is the major risk a flood with estimated consequences or the presence of a risky
chemical works that uses a large amount of hazardous materials?
Subsequently, for ascertaining the details on possibilities and effectiveness of measures of protection
against individual risks, either the use of alternative methods based on the economic aspects of solving
(e.g. Cost-Benefit Analysis) or the use of a method of mapping the preferences of the population
(Willigness to Pay or Willigness to Accept) is necessary so that the soft component of risk perception by
the population may be considered as well.
 84

6. Conclusion
The main advantage of risk mapping application in connection with other analytical methods is the
strengthening of making decisions on protection measures made by public and government authorities.
Risk mapping can help select the areas that have the highest cumulative risk levels and in which priorities
for investment will be established. Thus, we are able to answer the question where to invest. The detailed
analysis of the selected territories will then specify the allocation of financial means to individual bodies
managing the identified key risk sources (e.g. specific ministries, regional authorities, and others).

Acknowledgements
This contribution was prepared as part of the grant project of the Ministry of the Interior of the Czech
Republic under number VF20112015018, titled Safety of Citizens – Crisis Management (in Czech).

References
Beffa C., 2000, A Statistical Approach for Spatial Analysis of Flood Prone Areas. International Symposium
on Flood Defence, D-Kassel, 8 p. or <www.fluvial.ch/pub/spatial_analysis_of_flood_prone_areas.PDF>
accessed 20.05.2012
BZS, 2003, KATARISK - Catastrophes and Emergencies in the Switzerland
<www.bevoelkerungsschutz.admin.ch/internet/bs/de/home/themen/gefaehrdungen-
risiken/studien/katarisk.parsys.0004.downloadList.00041.DownloadFile.tmp/methodeprint.pdf>
accessed 20.05.2012 (in German)
EU, 2007, Directive 2007/60/EC of the European Parliament and of the Council of 23 October 2007 on the
Assessment and Management of Flood Risks. Official Journal of the European Union, 50, L288, 27-34
Drbal K. et al., 2005, Design of the Methodology for Determining Flood Risk and Damages in the Area
and its Verification in the Elbe Basin
. VÚV TGM, Brno , Czech Republic (in Czech)
Drbal K. et al., 2008, Methodology for Determining Flood Risk and Potential Damage in the Flood Zone.
VÚV TGM, Brno, Czech Republic (in Czech)
Krömer A., Musial P., Folwarczny L., 2010, Risk Mapping. SPBI, Ostrava, Czech Republic: SPBI, (in
Czech)
Rehak D., 2011, Population Protection against the Indirect Negative Impacts of Spatial Development in the
Context of Sustainable Development of Territory (Habilitation Thesis), VSB – Technical University of
Ostrava, Ostrava, Czech Republic (in Czech)
Renni, E., Basco, A., Busini, V., Cozzani, V., Krausmann, E., Rota, R., Salzano, E., 2010. Awareness and
mitigation of NaTech accidents: Toward a methodology for risk assessment. Chemical Engineering
Transactions 19 , 383-389, DOI: 10.3303/CET1019063
Riha J. et al., 2005, Risk Analysis of Flood Zones. FAST VUT Brno, Brno, Czech Republic (in Czech)
SIPROCI, 2007, Risk Mapping a Proposal for a Common European Methodology <www.kraj-
jihocesky.cz/file.php?par[id_r]=22641&par[view]=0> accessed 20.05.2012
Vallée, A., Duval, C., 2012. Flooding of industrial facilities-Vulnerability reduction in practice. Chemical
Engineering Transactions 26 , 111-116, DOI: 10.3303/CET1226019