Elisabeth Krausmann

The Great East Japan earthquake and tsunami damaged or destroyed many industrial facilities housing or processing hazardous substances, such as refineries, petrochemical facilities and other types of chemical industry. This showed that also generally well prepared countries are at risk of suffering natural hazard triggered technological (Natech) accidents. An analysis of data collected from open sources and through interviews with authorities was performed to understand the main reasons for the industrial damage and downtime as well as the extent of hazardous-materials releases and the associated impact on society. The analysis of the data set confirmed the findings from other studies with respect to main damage and failure modes, as well as hazardous-materials release paths. In addition, gaps in Natech risk management were identified. Based on the data analysis and interviews lessons learned in support of a more far-reaching Natech risk management are presented.

In recent years, severe natural events raised concern about so-called NaTech accident scenarios: technological accidents caused by the impact of a natural event on an industrial facility or infrastructure. Lightning strikes are one of the most important triggers of NaTech scenarios. Moreover, previous studies showed that lightning strikes are among the main causes of loss of containment (LOC) of atmospheric storage vessels containing hazardous materials. Although the lightning hazard is well known, well accepted quantitative procedures to assess the contribution of accidents triggered by lightning to industrial risk are still lacking. In particular, the approaches to the assessment of lightning strike probability and to the damage caused by lightning strike are mainly qualitative or semi-quantitative and are mostly based on expert judgment. In the present study, a quantitative methodology for the assessment of the equipment damage probability due to lightning is presented. The lightning severity was quantified by means of probability distribution functions of two parameters: peak current intensity and lightning charge. Through the application of a Monte Carlo simulation the expected frequency of lightning strikes on the equipment and the equipment damage probability were determined. The results of the equipment damage model were validated by available experimental data on metal perforation in simulated lightning strikes. The results of the validated Monte Carlo simulations were fit to empirical functions obtaining a simplified model suitable for use in a quantitative risk assessment framework.

Recognizing the need to raise awareness of the risk of severe space-weather impacts on critical infrastructures, the European Commission’s Joint Research Centre (JRC), in collaboration with the Directorate-General Enterprise and Industry, organized the Space-Weather Awareness Dialogue, a high-level event held in Brussels, Belgium, on 25–26 October 2011. The dialogue sought to highlight the potential effects of extreme space weather on technological infrastructures both in space and on the ground; to identify related scientific, operational, and policy challenges for disaster prevention, preparedness, and response; and to recommend actions that will reduce the vulnerability of critical infrastructures.
The event brought together about 70 representatives from European industry, government agencies, EU institutions, and the space-weather community. Officials from NASA and NOAA actively supported the event. The dialogue consisted of five panel sessions and a concluding session, with more than 30 panelists contributing during the two-day program.

A study was performed on the status of Natech risk reduction in EU Member States by means of a questionnaire survey. The results show that natural hazards are increasingly recognised as a possibly important external risk source for chemical facilities. The management of Natech risk is mainly addressed through the Member States’ legal framework for chemical-accident prevention whose effectiveness appears, however, inconclusive. Guidance on Natech risk reduction to support legislation is scarce and, where existing, did not always prevent
Natech accidents from occurring. In fact, in over half the responding countries Natech accidents have resulted in the release of toxic substances, fires or explosions with sometimes fatalities and injuries. The natural events that triggered these accidents were not necessarily the ones believed to be of most
concern in the Natech context, indicating an incongruity between accident causes and risk perception. Gaps in Natech risk reduction were recognised and are mostly due to budget
constraints and a lack of adequate resources which lead to the prioritisation of tasks deemed more important, a lack of training, and insufficient knowledge of the dynamics of Natech accidents. This has resulted in a lack of specific Natech risk-assessment methodologies and tools. Consequently, industry in almost half
of the responding countries is believed to insufficiently consider Natech risk in their facility risk assessment. The development of guidance on Natech risk assessment was indicated as the highest-priority need for effective risk reduction. The study concludes with a number of priority areas for future work to improve the management of Natech risk. The results of a Natech questionnaire survey in OECD Member Countries which was performed in parallel show the same trend.

The threat of natural hazards impacting chemi-
cal facilities and infrastructures with the subsequent release
of hazardous substances has been recognised as an emerging
risk which is likely to be exacerbated by the ongoing climate
change. Within the European FP7 project iNTeg-Risk, efforts
are dedicated to address the problem of Natech accidents by
trying to understand their underlying causes and by develop-
ing methodologies and tools to assess Natech risk. Special
attention is thereby given to the risk of chemical accidents
triggered by earthquakes, floods and lightning. This work
outlines the ongoing efforts in the development of new con-
cepts and tools for Natech hazard and vulnerability ranking,
risk assessment, risk-based design, and emergency planning
and early warning.

Natural hazards and disasters can cause major accidents in chemical and process installations. These so-called Natech accidents can result in hazardous-materials releases due to damage to process and storage units, or pipes. In order to understand the dynamics of Natech events, accidents triggered by earthquakes, floods and lightning recorded in industrial accident databases were analysed. This allowed the identification of the most vulnerable equipment types, their modes of failure due to natural-event impact and the final accident scenarios. Moreover, lessons learned for future accident prevention and mitigation were derived. The analysis showed that pipes and storage tanks are the most vulnerable equipment for earthquakes, floods and lightning, calling for more research of equipment behaviour under natural-event loading. The damage modes and states are strongly dependent on the characteristics of the impacting natural event. Toxic dispersion,
fires and explosions were observed as a consequence of all three types of analysed natural events. In the case of floods, two additional scenarios were identified. These are water
contamination and the formation of toxic and/or flammable vapours upon reaction of the released chemicals with the floodwaters. The overall number of recorded Natech accidents
was found to range from 2 to 5% of all reported accidents in the analysed databases.

Natural disasters can cause major accidents in chemical facilities where they can lead to the release of hazardous materials which in turn can result in fires, explosions or toxic dispersion. Lightning strikes are the most frequent cause of major accidents triggered by natural events. In order to contribute
towards the development of a quantitative approach for assessing lightning risk at industrial facilities, lightning-triggered accident case histories were retrieved from the major industrial accident databases and analysed to extract information on types of vulnerable equipment, failure dynamics and damage states, as well as on the final consequences of the event. The most vulnerable category of equipment is storage tanks. Lightning damage is incurred by immediate ignition, electrical and electronic systems failure or structural damage with subsequent release. Toxic releases and tank fires tend to be the most
common scenarios associated with lightning strikes. Oil, diesel and gasoline are the substances most frequently released during lightning-triggered Natech accidents.

"This work analyses the potential consequences of two tsunami scenarios and
their impacts on an oil refinery located in Sicily. Two credible tsunamis originating in the
Tyrrhenian Sea were selected based on historical data. The potential for damage and
hazardous materials releases resulting from the tsunami impacts to a refinery was assessed.
The results obtained by the JRC tsunami propagation and inundation code HyFlux2
indicate that in both scenarios there would be eighteen storage tanks (of 43 located within
400 m from the shoreline) at the refinery subject to flooding. Water flow velocities were
found to be generally low, <1 m/s, except for a central section of the refinery near the
shoreline where the water flow velocities reach 3–4 m/s. These results indicate that any
damage would most likely occur due to buoyancy loads particularly in the western part of
the facility where inundation levels are higher and storage tanks are less protected.
Potential damage caused by impact of floating debris may be a problem in the central area
of the refinery near the shoreline due to high flow velocities (3–4 m/s) in both tsunami
scenarios. Small hazardous materials releases could occur due to breakage of connected
pipes and flanges caused by floating off of almost empty storage tanks or other equipment.
Salt water intrusion could affect electrical equipment, such as control panels, pumps, and
motors that are not raised above the inundation level. We conclude that in the two tsunami
scenarios analysed, the risk to nearby residents and neighbouring facilities from potential
hazardous materials releases, fires or explosions triggered by the tsunamis is likely to be
small. Nonetheless, recommendations are made on prevention measures to reduce the risk
of tsunami-triggered accidents and to mitigate their consequences if they do occur. The
results of this study are limited by the uncertainty in the input data and most importantly by
the accuracy of the elevation data and the model resolution."

This study describes the results of a field trip to the area affected by the 12 May, 2008, Wenchuan
earthquake to analyse its impact on industrial facilities. The damage severity correlates well with the age
of the plant, with older facilities having suffered more extensive and severe damage than those built
more recently according to the latest design codes. The main cause of worker death and injury was the
collapse of warehouses, office and manufacturing buildings. This concerned mostly concrete structures
with insufficient confinement or poor reinforcement. The falling debris resulted in equipment damage
and loss, as well as pipe severing and crushing. Pipes were also severed or bent when connected tanks
were displaced or buildings collapsed. Numerous hazardous-materials releases occurred with spills being
the dominant accident scenario. In some sites soileliquefaction induced damage was evident, highlighting
the need to consider potential site effects when selecting the location for a facility. The impact of
the Wenchuan earthquake on chemical facilities confirms the findings from other earthquakes in terms
of typical Natech damage and failure modes, as well as of hazardous-materials release potential and
mechanisms.

Industrial accidents triggered by natural events (NaTech accidents) are a significant category of industrial
accidents. Several specific elements that characterize NaTech events still need to be investigated. In
particular, the damage mode of equipment and the specific final scenarios that may take place in NaTech
accidents are key elements for the assessment of hazard and risk due to these events. In the present study,
data on 272 NaTech events triggered by floods were retrieved from some of the major industrial accident
databases. Data on final scenarios highlighted the presence of specific events, as those due to substances
reacting with water, and the importance of scenarios involving consequences for the environment. This
is mainly due to the contamination of floodwater with the hazardous substances released. The analysis
of process equipment damage modes allowed the identification of the expected release extents due to
different water impact types during floods. The results obtained were used to generate substance-specific
event trees for the quantitative assessment of the consequences of accidents triggered by floods.

Tunnels in the trans-European road network (TEN) facilitate the transport of persons and goods on European
roads. Following a series of major tunnel accidents European Union Directive 2004/54/EC was
adopted to support the achievement of uniform and high tunnel safety levels. With future accident prevention
and mitigation in mind and in support of the effective implementation of Article 15 on Reporting
of the Directive we outline a procedure for learning lessons and discuss every step in the process with
specific regard for its implications on Article 15. This includes accident investigation, reporting, data collection
and analysis, learning lessons and their implementation. The realization that validated information
on tunnel accidents is not easily available or accessible, or suffers from a lack of detail or accuracy
fed into the development of a data-collection template. By its very nature the template development also
guided the formulation of key recommendations for accident investigation and reporting as the main
information source. In addition, key recommendations on data analysis, learning lessons and implementation
were also made to assist the actors responsible for reporting or sharing information under the EU
Directive.

Hurricane Katrina and Rita triggered numerous hazardous-materials releases from industrial facilities
and storage terminals onshore, as well as from oil and gas production facilities offshore in the Gulf of
Mexico. In this paper, we identify and analyze over 600 hazardous-materials releases triggered by
Hurricanes Katrina and Rita from offshore platforms and pipelines. The results of the study may assist
offshore industry owners/operators, government officials and policy makers by providing lessons learned
and recommendations for better disaster planning for major storms and flood events.

Hurricanes Katrina and Rita hit the centre of the American petrochemical industry, shutting down eight
refineries, hundreds of oil-drilling and production platforms, and many other industrial facilities.
Furthermore, it triggered unprecedented numbers of hazardous-materials releases from industrial
facilities and storage terminals onshore, as well as from oil and gas production facilities offshore in the
Gulf of Mexico (GoM). In this paper, we analyse the damage caused by the two hurricanes on the
offshore oil and gas industry. Hurricanes Katrina and Rita caused the largest number of destroyed and
damaged platforms and pipelines, and the highest number of mobile offshore drilling units set adrift in
the history of GoM operations. Following the hurricanes, changes have been proposed to operating and
emergency procedures, maintenance requirements, and design practices including mooring practices
for mobile offshore drilling units.

There is increasing evidence that natural disasters can trigger technological
accidents and damage. These so-called Natech accidents can pose a significant risk to
regions that are unprepared for responding to them. The European Commission’s Joint
Research Centre has recognised the risk associated with Natech events and has started
systematic research into Natechs and their underlying dynamics. This work investigates the
risk associated with the flooding of industrial installations through an analysis of past case
histories and using expert judgement. The potential impact of three levels of flood severity
on selected industrial facilities storing and/or processing (eco-)toxic, flammable or
explosive materials is analysed qualitatively and a scale is developed that links the flood
intensity to the level of potential damage. Our analysis indicates that natural disasters have
the potential for triggering hazmat releases and other types of technological accidents.
Hence, natural disasters should be considered as separate accident-triggering events in the
planning, design and operating stages of industrial facilities that process or store hazardous
substances. Our work revealed a lack of detailed information on the occurrence of Natech
events which indicates not necessarily a scarcity of Natechs but rather a lack of standardised
reporting and record keeping.

The THENPHEBISP 2-year thematic network started in December 2001, and was concerned with OECD/CSNI International
Standard Problem 46, itself based on the Phebus FPT1 core degradation/source term experiment. The aim was to assess the
capability of computer codes to model in an integrated way the physical processes taking place during a severe accident in a
pressurised water reactor, from the initial stages of core degradation, the fission product transport through the primary circuit
and the behaviour of the released fission products in the containment. ISP-46, coordinated by IRSN/DRS Cadarache, attracted
33 participating organisations, from 23 countries and international bodies, who submitted 47 base case calculations and 21
best-estimate calculations, using 15 different codes.

The International Phebus Fission Product programme, initiated in 1988 and performed by the French “Institut de Radioprotection et de Surete
Nucleaire” (IRSN), investigates through a series of in-pile integral experiments, key phenomena involved in light water reactor (LWR) severe
accidents. The tests cover fuel rod degradation and the behaviour of fission products released via the primary coolant circuit into the containment
building.
The results of the first two tests, called FPT0 and FPT1, carried out under low pressure, in a steam rich atmosphere and using fresh fuel for FPT0
and fuel burned in a reactor at 23 GWdt−1 for FPT1, were immensely challenging, especially with regard to the iodine radiochemistry. Some of
the most important observed phenomena with regard to the chemistry of iodine were indeed neither predicted nor pre-calculated, which clearly
shows the interest and the need for carrying out integral experiments to study the complex phenomena governing fission product behaviour in
a PWR in accident conditions. The three most unexpected results in the iodine behaviour related to early detection during fuel degradation of a
weak but significant fraction of volatile iodine in the containment, the key role played by silver rapidly binding iodine to form insoluble AgI in the
containment sump and the importance of painted surfaces in the containment atmosphere for the formation of a large quantity of volatile organic
iodides.
To support the Phebus test interpretation small-scale analytical experiments and computer code analyses were carried out. The former, helping
towards a better understanding of overall iodine behaviour, were used to develop or improve models while the latter mainly aimed at identifying
relevant key phenomena and at modelling weaknesses. Specific efforts were devoted to exploring the potential origins of the early-detected volatile
iodine in the containment building. If a clear explanation has not yet been found, the non-equilibrium chemical processes favoured in the primary
coolant circuit and the early radiolytic oxidation of iodides in the condensed water films are at present the most likely explanations. Models that
were modified or developed and embodied in the computer codes for organic iodide formation/destruction in the gas phase and Ag–I reactions in the
sump lead, in agreement with the Phebus findings respectively to greatly enhanced organic iodide formation kinetics and long term concentration
in the containment atmosphere on one hand and, in the conditions of Phebus experiments, to significantly limited molecular iodine volatilisation
from the sump in so far as silver was in excess compared to iodine, on the other hand. Organic iodides then quickly gain in importance and become
the predominant volatile iodine species at long term.

ABSTRACT Radioiodine entering the containment from the postaccident primary circuit in vapor or gaseous form, as observed in the Phebus FPT0 and FPT1 tests, has a direct impact on the source term evaluation. State-of-the-art fission-product transport codes based on the assumption of thermochemical equilibrium failed to predict this phenomenon. In this work the standard approach of assuming the instantaneous establishment of thermochemical equilibrium is questioned and it will be argued that kinetic limitations may have existed under the severe-accident boundary conditions of the FPT0 and FPT1 tests. To this end a simple monodimensional transport model was developed in an attempt at introducing kinetic aspects within the primary circuit. A number of homogeneous gas-phase reactions between selected fission products and structural materials, complemented by condensation reactions, underlies the kinetic model. In the absence of experimental data, the kinetic constants were estimated using the transition-state theory or semi-empirical methods. The kinetic model was then applied to the analysis of Phebus FPT0 and FPT1 yielding a satisfactory agreement between experimental data and model predictions.

Metallic silver and iodine form insoluble AgI in the containment sump which has a major impact on iodine volatility and hence on the source term to the environment. Resistance-in-series models are developed and validated against separate-effects tests. The reaction between I2 and Ag is limited by mass transfer in the liquid. The rate does not follow a parabolic law for the duration of the experiments. The extent of oxidation of the silver sample seems to play a decisive role for reactions with I− which proceed via a two-step process. The initial, rapid step is controlled by the reaction with the surface silver oxide with a contribution of mass transfer in the liquid. The subsequent, slow step is limited by reaction between I− and Ag+ at the solid–liquid interface. The reaction is probably negligible for pH > 7 and in the absence of oxidising conditions.