DOI: 10.3303/CET2290047
Paper Received: 17 January 2022; Revised: 24 March 2022; Accepted: 17 April 2022
Please cite this article as: Salvi O., Caillard B., Bolvin C., Lheureux E., Benjelloun F., 2022, Benchmark of European practices for land-use
planning around Seveso establishments, Chemical Engineering Transactions, 90, 277-282 DOI:10.3303/CET2290047
CHEMICAL ENGINEERING TRANSACTIONS
VOL. 90, 2022
A publication of
The Italian Association
of Chemical Engineering
Online at www.cetjournal.it
Guest Editors: Aleš Bernatík, Bruno Fabiano
Copyright © 2022, AIDIC Servizi S.r.l.
ISBN 978-88-95608-88-4; ISSN 2283-9216
Benchmark of European Practices for Land-use Planning
around Seveso Establishments
Olivier Salvia*, Bastien Caillarda, Christophe Bolvinb, Emmanuel Lheureuxc, Fessel
Benjellounc
a INERIS Développement, Verneuil-en-Halatte, France
b Institut National de l’Environnement Industriel et des Risques, Verneuil-en-Halatte, France
c SPW Direction des Risques Industriels, Géologiques et Miniers (DRIGM) – RAM, Namur, Belgium
olivier.salvi@ineris-developpement.com
The authorities in charge of Seveso establishments in the Walloon region of Belgium has organized a
benchmark study to understand the practices for land-use planning around Seveso establishments in application
of the article 13 of the European Directive 2012/18/UE of 4 July 2012, called Seveso III, concerning the control
of major accident hazards involving dangerous substances.
The objectives of the study were to:
• Analyse the practical implementation of the land use planning (LUP) procedure according to the Seveso
III directive in several member states
• Identify the legal basis and the explicit method & criteria for LUP
• Identify the measures to control the urbanization around hazardous establishments
• Understand the condition of revision of LUP documents in case of modifications of the installations or
changes in the determination of the zones (thresholds, calculation software, databases for failure rates…).
The benchmark provides detailed information on the implementation of the land-use planning procedure, and it
reveal the commonalities and differences in the approach among European countries and regions.
1. Method
The benchmark study reviewed the practices in 10 European Countries and in the 3 regions of Belgium, namely:
Austria, Czech Republic, The Netherlands, France, Germany, Greece, Italy, Luxembourg, Slovenia, Spain, and
the Regions of Belgium: Wallonia, Brussels Capital, and Flanders.
Among the methodological elements to be considered, it appeared particularly important to deal in detail with
the mechanisms for the revision or updating of land use planning zones around hazardous sites.
• How is the transposition of hazard zones into urban planning zones organized?
• What are the origins and treatments of the modifications of the danger zones?
• Is there a criterion for a tolerable increase of the hazard zones without modification of the planning zones?
The analysis was conducted in 3 steps during:
• Identification & analyse of regulations, guides and reports describing the LUP procedure in the Members
States and the regions of Belgium,
• Interviews with selected experts to understand the practical implementation and worries / issues related to
LUP,
• Analyse and compare the practices.
The interviews enabled to collect information on practical implementation of the land-use planning approach.
And the summaries of the interviews have been checked by the interviewees before the content has been
exploited for the study. All together the study is the most recent overview of the current land-use planning
practices around Seveso plants since the work carried out by Michalis Christou and al. (Christou, Gyenes,
277
Struckl, 2011) in 2010 and 2011. The analysis and the comparison of the practices have been primarily based
on what the experts said during the interviews.
This approach made it possible to quickly grasp field practices in application of regulatory texts and other
reference documents, which are difficult to grasp in a short time.
The following issues have been addressed:
• General approach to urban planning
• Actors for the determination of the zones used as a basis for urban planning zones
• Decision to define the zoning around Seveso establishments
• Determination of urban planning zones
• Models, software and thresholds
• Updating of urban planning documents
• Management of urban planning areas
• Impact of software updates
• Public information and publication
2. Results
The following paragraphs present examples of comparison of the main practices of the countries and regions
selected. The comparison is based on the criteria that have been defined to allow a certain comparison, find
commonalities and point out some specificities as they were perceived based on the interviews.
2.1 General approach to urban planning
In most cases, the hazard zones estimated by the competent authorities in charge of Seveso establishments
are transmitted to the authority in charge of urban planning for the preparation of land-use plans. Often these
plans are prepared by municipalities or groups of municipalities (e.g. "Gemeinde" in Germany). Countries with
a federal system or a high degree of regional autonomy have regulations for the control of urban planning around
Seveso establishments that are specific to each region or province (Germany, Austria, Belgium, Spain, Italy).
The other countries have an approach with national regulations applied at local level.
It also appears that the vast majority of countries and regions have formalized a guide or a publicly available
doctrine document, which makes it possible to move towards a coherent general approach, including the choice
of accident scenarios and their examination, if necessary.
The approaches in Flanders, the Netherlands and France are particularly well documented. In Slovenia,
everything is stated in the law.
Figure 1: Comparison of the general approach to urban planning
2.2 Actors for the determination of the zones used as a basis for urban planning zones
In the majority of the countries and regions studied, planning areas are determined at the local level, with the
exception of Greece and Slovenia, which indicate that the dossiers are managed at the level of the ministries in
charge of the environment and land use planning. However, in several countries, regional authorities consult
with central authorities for important installations. In most countries, the accident scenarios considered for
planning control are determined and evaluated under the responsibility of the local competent authorities without
being predetermined or studied by the operator in the safety report. The local authorities can hire accredited
experts or ask the operator to have the studies carried out by accredited experts, which allows a certain
consistency in the choice and examination of scenarios.
#EPARS! Ge
rm
an
y
Au
str
ia
BE
-B
RU
BE
-Fl
an
de
rs
BE
-W
all
on
ia
Sp
ain
Fr
an
ce
Gr
ee
ce
Ita
ly
Lu
xe
m
bo
ur
g
Ne
th
er
lan
ds
Slo
ve
nia
Centralized at the national level with
national regulations
X X X X X X
Decentralized to the regional or
provincial level with local regulation
X X X X X X
Doctrine and method formalized in a
publicly available document
X X X X X X X X X X X
General approach to urban planning
278
In France, the process is formalized and supervised at the national level and implemented with
regional/prefectural competencies. In Luxembourg, the consultants hired by the operators are approved by the
competent authorities. In Spain, the calculations are made by consultants approved by the provincial authorities.
Figure 2: Comparison of the actors involved for the determination of the zones for urban planning
2.3 Decision process to define the zoning around Seveso establishments
The decision-making process for the determination of urban planning control zones is difficult to understand. In
practice, it appears that the decision is the result of consultation or exchanges between the Seveso authorities
and the authorities in charge of urban planning, although in most cases the Seveso authorities only provide a
recommendation to the authorities in charge of urban planning.
In Italy, the use of a Regional Technical Committee for the examination of dossiers with representatives of
several authorities and agencies is rather unusual. This committee concept does not appear in any other
country.
In the Netherlands, it appears that the provincial authority can deviate from the recommendations by justifying
its decision with additional calculations using other models or other hypotheses for the examination of scenarios.
In France, it is possible to call on a third-party expert (second opinion) to examine all or part of the hazard study.
This third-party expert assessment requested by the competent authority is at the operator's expense. The
Ministry of the Environment maintains a list of recognized third-party experts for this work. This approach is an
alternative to accrediting organizations or experts to carry out studies that serve as a basis for urban planning.
Figure 3: Comparison of the decision process to define the zoning around Seveso establishments
#EPARS! Ge
rm
an
y
Au
str
ia
BE
-B
RU
BE
-Fl
an
de
rs
BE
-W
all
on
ia
Sp
ain
Fr
an
ce
Gr
ee
ce
Ita
ly
Lu
xe
m
bo
ur
g
Ne
th
er
lan
ds
Slo
ve
nia
Pre-determined areas under the
responsibility of the operator
X X X X X
Zones determined under the
responsibility of the competent
authority (Seveso) at a national level
X X X
Areas determined under the
responsibility of the competent
authority (Seveso) at a regional or
provincial level
X X X X X X X X X
Organizations or experts accredited by
the authority
X X X X X X X X
Actors for the determination of the zones used as a basis for urban planning zones
#EPARS! Ge
rm
an
y
Au
str
ia
BE
-B
RU
BE
-Fl
an
de
rs
BE
-W
all
on
ia
Sp
ain
Fr
an
ce
Gr
ee
ce
Ita
ly
Lu
xe
m
bo
ur
g
Ne
th
er
lan
ds
Slo
ve
nia
Decision by the same authorities that
investigate the safety reports
X X X X
Decision by the authorities in charge of
urban planning (receive a
recommendation from the authorities
in charge of Seveso establishments)
X X X X X X X
Collective decision with several
agencies or authority representatives
X X
Possibility of having recourse to a third
party expert (second opinion)
X
Decision to define the zoning around Seveso establishments
279
2.4 Determination of urban planning zones
For the determination of the hazard zones that are used to define the land-use planning zones, for a majority of
countries and regions, the approach is based on the evaluation of the consequences of accident scenarios
(deterministic approach).
Some countries have an approach that combines a first estimation of the zones according to a deterministic
approach to define a preliminary zone of consultation or attention, then within this zone, the estimation of the
individual risk is taken as reference to establish the zones of urbanization control.
Three countries (Germany, Austria and Spain) show a mixed approach that combines lumped distances and a
more refined estimation of accident scenarios to define urbanization zones.
In Spain, there is a strong diversity between the autonomous regions.
It should be noted that in France, for Seveso high threshold establishments, the individual risk is calculated by
taking into account all the accident scenarios, their probability and their intensity (effect distances). The
SIGALEA® software is used to represent the contours that serve as a basis for the establishment of
Technological Risk Prevention Plans (PPRT) and the corresponding urban planning documents. This procedure
for drawing up PPRTs is accompanied by formal consultation with the Site Monitoring Commissions (CSS).
The Netherlands has introduced the concept of a deterministic attention zone in a regulation that will come into
force in January 2022, whereas the country was known for traditionally using only a probabilistic approach that
takes into account individual and societal risk. The definition of attention zones in a first step has a pedagogical
virtue in informing the populations potentially affected by major accidents, given that the Dutch expert indicated
that the populations have difficulty understanding the probabilistic approach.
Figure 4: Comparison of the practices for the determination of urban planning zones
2.5 Models, software and thresholds
There is a wide variety of approaches to determining the zones on which LUP zones are based. Countries that
do not centralize the examination of scenarios do not prescribe models or software. But in Spain, the
autonomous region of Catalonia prescribes EFFECTS, RISK CURVES and ALOHA in the instruction 14/2008.
The Netherlands, which centralizes the calculations, uses a specific version of the DNV SAFETI tool which sets
parameters to limit the variations due to the choice of certain assumptions. The SAFETI-NL tool is evaluated for
each new version and the Ministry in charge of the Environment and Spatial Planning must give its approval to
use the new version.
In France, there is no official evaluation of the software. However, INERIS can be asked to evaluate some
software at the request of its supervisory ministry (it has done so in the past for PHAST, but not only). There is
no prior approval for the use of software since the models or software are not prescribed. However, the use of
CFD in hazard studies is supervised, it can be done if the configuration of the scenario requires it by relying on
the good practice guide of atmospheric dispersion modeling with CFD.
#EPARS! Ge
rm
an
y
Au
str
ia
BE
-B
RU
BE
-Fl
an
de
rs
BE
-W
all
on
ia
Sp
ain
Fr
an
ce
Gr
ee
ce
Ita
ly
Lu
xe
m
bo
ur
g
Ne
th
er
lan
ds
Slo
ve
nia
Fixed (distances not based on accident
scenario modeling and probability
calculations)
X X
Based on the evaluation of the
consequences of accident scenarios
(consequence-based approach)
X X X X X X X X
Based on the evaluation of the
individual risk at each point of the
environment taking into account all or
the most probable scenarios
(probabilistic approach)
X X X X X X
Mixed approach (fixed distances and
scenario-based)
X X X
Determination of urban planning zones
280
In Germany, the UBA agency has developed and made available a specific tool, AUSTAL. A project in Baden-
Württemberg is currently underway to evaluate the impact of software and its models in the estimation of the
consequences of accident scenarios.
In Slovenia, a study was carried out two years ago to evaluate the variations between the calculations made by
several organizations that are entrusted with the studies by the ministry in charge of the environment.
In Luxembourg, a benchmark is also in progress with the 3 engineering offices approved by the government to
understand the variations that can exist between the German, Dutch and French approaches accepted in the
country.
These recent or ongoing studies show that concerns about methods and practices are strong and that states
are seeking to evolve their doctrine.
Figure 5: Comparison of the models, software and thresholds used for the determination of the LUP zones
2.6 Impact of software updates
The analysis and in particular the exchanges with the national experts show that most of the modifications due
to changes in software or thresholds are taken into account at the time of the five-yearly revision of the safety
report. If there are changes in the hazard areas, then they are considered at that time. No country or region has
indicated that safety reports or planning zones have to be updated immediately after a new software or threshold
is introduced. The expert from Austria states that he has dealt with threshold changes and does not identify any
particular problems.
Figure 4: Comparison of the impact of software updates
3. Conclusions
The working method was influenced by the short time frame of the study and the summer period, which forced
us to conduct interviews with the experts at the beginning of the study. Moreover, the subject is very vast, and
the material collected can be deepened, if necessary, to better understand certain practices and to draw
inspiration from them.
In terms of the collection of regulatory texts, it appears that the subject of LUP is complex and difficult to
understand solely from a regulatory and legal perspective. The study highlighted the importance of doctrinal
documents, good practice guides and their sharing with the stakeholders and the public.
#EPARS! Ge
rm
an
y
Au
str
ia
BE
-B
RU
BE
-Fl
an
de
rs
BE
-W
all
on
ia
Sp
ain
Fr
an
ce
Gr
ee
ce
Ita
ly
Lu
xe
m
bo
ur
g
Ne
th
er
lan
ds
Slo
ve
nia
Software or calculations centralized by
the authority
X X X X X X X
Pre-assessment and decision to use
or not use the software
X
No prior assessment
Software (strongly) recommended X X X X X
No prescription of tools X X X X X X
Models, software and thresholds
#EPARS! Ge
rm
an
y
Au
str
ia
BE
-B
RU
BE
-Fl
an
de
rs
BE
-W
all
on
ia
Sp
ain
Fr
an
ce
Gr
ee
ce
Ita
ly
Lu
xe
m
bo
ur
g
Ne
th
er
lan
ds
Slo
ve
nia
Impact of software updates
No immediate impact, five-yearly
update of the safety report
X X X X X X X X X
Verification of changes to distances
before approval of updates
X X X X
Immediate consideration of new zones
including software updates or
thresholds.
281
In terms of the organization of LUP explained through interviews with the experts, it emerged that practices are
very diverse and are the result of a regulatory, industrial and cultural heritage. That said, several countries have
begun to reform their practices, such the Netherlands, which introduced the notion of attention zones
independent of the probability of occurrence. This shows that there is no single, immutable answer, and that
given the diversity and contexts, the search for harmonization is not obvious, as the MAHB of the Joint Research
Centre has already noted in previous studies.
The key words of the approaches seem to be transparency, rational justification and traceability.
Regarding the update of LUP zones based on updates of software, it appears that few countries are dependent
on software changes, and none of the countries studied reported major problems with software updates or
thresholds. For the majority of countries that have encountered similar situations, the new safety distances have
been introduced in new versions of urban planning documents or measures have been taken to impose a
reduction of risks at source so that the distances do not change.
The benchmark provides new inspiration to review the current practices in each country for land-use planning
practices and might offer a background to reinforce the harmonization of the implementation of the Seveso
Directive in Europe.
Acknowledgments
This work has received funding from the Walloon Authorities, SPW Agriculture, Ressources naturelles et
Environnement – DEE – DRIGM.
References
Dechy N., Gaston D., Salvi O., AZF : les leçons d'une catastrophe industrielle. Responsabilité & Environnement
- Annales des mines, 2007, n° 45, pp. 10-17.
Guide UNECE. Document d’orientation sur l’aménagement du territoire, le choix des sites d’activités
dangereuses et les aspects de sécurité s’y rapportant. e-ISBN: 978-92-1-363051-8. https://unece.org/
Gyenes Zsuzsanna, Wood Maureen Heraty, Struckl Michael. Handbook of Scenarios for Assessing Major
Chemical Accident Risks, JRC Technical Reports. 2017. doi:10.2760/884152
Land Use Planning and Risk-Informed Decision Making (Aménagement du Territoire et Prise de Décision en
Maitrise de Risques), Proceedings of the 43rd ESReDA Seminar. Hosted by INSA, Rouen, France Edited
by Mohamed Eid, Michalis Christou (2014). Report EUR 26719 EN.
Laurent André, Pey Alexis, Gurtel Peter, Fabiano Bruno, A critical perspective on the implementation of the EU
Council Seveso Directives in France, Germany, Italy and Spain, Process Safety and Environmental
Protection, Volume 148, 2021, Pages 47-74, ISSN 0957-5820, https://doi.org/10.1016/j.psep.2020.09.064.
Merad M., Rodrigues N., Salvi O., Urbanization control around industrial SEVESO sites : the French context.
International Journal of Risk Assessment and Management, 2008, vol. 8, n° 1-2, pp. 158-167
Salvi O., Debray B., 2006, A global view on ARAMIS, a risk assessment methodology for industries in the
framework of the SEVESO II directive. Journal of Hazardous Materials, 2006, vol. 130, n° 3, pp. 187-199.
Salvi O., Gaston D., 2004, Risk assessment and risk decision-making process related to hazardous installation
in France, Journal of Risk Research, 2004, vol. 7, n°6, pp. 599-608
Wood Maureen Heraty, Fabbri Luciano, Challenges and opportunities for assessing global progress in reducing
chemical accident risks, Progress in Disaster Science, Volume 4, 2019, 100044, ISSN 2590-0617,
https://doi.org/10.1016/j.pdisas.2019.100044.
282
lp-2022-abstract-203.pdf
Benchmark of European practices for land-use planning around Seveso establishments