Keywords

2.1 Introduction

The argument of this chapter is that climate change is one change among many which affect the operating landscape of safety-critical systems. Climate change, which could, perhaps, preferably be described as global warming, should therefore not be considered in isolation but in relation to other changes (e.g., globalization, digitalization). One task for safety research is to identify, to empirically study and to explore the implications of such changes. Following a short presentation of a case study, the proposition of Post Normal Accident (Post NA, Le Coze 2021a, 2022, 2023a, b) is briefly introduced. It provides analytical lenses to conceptualize change through the notion of global scales shaping a new causal regime in safety.

2.2 An Empirical Illustration

In a case study of a chemical plant of a small US multinational in France (> 3000 employees worldwide), it appeared that within a decade, from the late 1990s to the early 2010s, this site of the company went through an important number of changes which substantially modified its mode of operating (Le Coze 2021b; Le Coze and Dupré 2022). These transformations amplified the digital, network and global properties of the plant. Automating (more sensors on chemical reactors connected to computers to supervise reactions), computerizing production (more information systems through workflow to manage logistics and maintenance), externalizing several of its activities (waste treatment plant, maintenance, boiler) and restructuring its organizational design toward a higher level of centralization and standardization (through a matrix structure) by the multinational’s head office in order to improve worldwide cooperation and control, the plant illustrates indeed this change of mode of operating, its level of interconnectedness across frontiers, organizations and distances that many safety-critical systems have experienced, including in the chemical industry (Avenas 2015).

Within a decade, the plant moved from a more autonomous, a more geographically “isolated” and a more “independent” mode of operating to a very different configuration which modified but also increased its level of interconnectedness. Automating and computerizing its processes created new ways, in real time and far more intrusively, of interacting with corporate actors of the multinational in the USA. These actors could indeed be in touch with daily practices of the plants thanks to the potentialities offered by its information infrastructure in ways which were simply impossible before. Externalizing activities increased the number of people to interact with because it multiplied the number of interfaces outside the company through multiple subcontracted organizations.

Modifying the organization through a matrix structure also meant far more links with headquarters, corporate and sites across the world to share information, to meet other employees during trips abroad (e.g., to India, to USA) and to implement standards audited by experts inside and outside the company, experts from sites who travel around the world to perform these audits. Within ten years, there was a profound modification of scale within which the plant operated and of work experience for employees. It was far more networked, digital and global. These had concrete implications for employees’ practices, social interactions and identities, from operators to site managers. Of course, the effects of such processes were not of the same kind depending on employees’ hierarchical levels and tasks (Le Coze and Dupré 2022). This description in the chemical industry is found in other safety-critical contexts which have been following similar paths (Dupré and Le Coze 2021).

An example is provided by Kongsvik et al., in the maritime industry, “while ships were traditionally autonomous organisational systems that the seafarers on board could – and were expected to – master alone, ships are now increasingly part of large networks of ships, internal and external IT systems, shipping companies, yards, certification agencies and national and international regulations” (Kongsvik et al. 2020) … in other words chemical plants (and ships) are connected to broader chains of causation allowed by global processes in different ways than in the past.

This is a materialization of the increased flow of people, data, goods or capital through expanding multinationals and tighter control of operations exerted across national borders (through standardization, computerization/digitalization) while also expanding their networks’ configuration into global value chains (externalization) (Baldwin 2016). This is a social, economic, political and cultural phenomenon with great consequences (Martell 2017). Their mode of operating is associated with a range of actors, organizations and institutions which are remote to their geographies or nationalities. They are connected to a global scale (Fig. 2.1).

Fig. 2.1
A world map connects the 2 points of a corporate in the United States and a chemical plant in France via the increase of standardization, externalization, and digitalization.

More globalized plant. Source Author’s own work

Full size image

Interestingly, there was another dimension linked to these changes of the plant, another change which situates it at another scale too, at a “second level” or referring to another “dimension” of global scales, yet of a different kind. During the study, two indications of this other change of scale were identified. The first was when operators mentioned how cold or hot it was at times in the building where they worked. This building was the main production one, with all the chemical reactors.

It was largely open to the outside through very large gates whose doors remained open most of the time for forklift trucks to circulate. It was an uninsulated warehouse made of metal, filled with chemical installations (pipes, reactors, valves). Their response was to organize work to cope with these extreme temperatures, wearing warm clothes in winter, drinking a lot in summer (but without the possibility of taking off their protective equipment) and staying inside the air-conditioned control room when possible.

However, this option was not possible for all the workers, because some activities had to be performed outside of the control room. These periods were not long (a few days over a year) but sufficient for the operators of the plant to mention them during the study. The second indication was the issue of accessing water to cool down chemical reactions during production stages, but also in case of a loss of control of an exothermic reaction. It is one of the safety barriers in the defense in depth of the system.

Because the plant was pumping water out of a well shared with a nearby town, they had designed sensors to monitor groundwater levels. These sensors were reported on computers’ interfaces which allowed them to make sure that they always had enough water in the system. In case of a low level in groundwater, alarms would be triggered, and they had, potentially, to reorganize production. However, this situation was never experienced, and sensors were almost forgotten by operators.

These two examples illustrate another change of scale, this time at the planetary level, not only as a connection to the trends of economic, social and cultural globalization discussed above but as a connection with the natural environment, which leads to another dimension of global scale. Environmental change at the planetary level through human activity is described as the phenomenon of anthropocene (McNeill and Engelke 2015). As much as economic, technological and cultural globalization modified the plants’ configurations and operating mode toward a global scale, the anthropocene was currently modifying the risk profile of safety-critical systems.

Today, a decade after this study performed in 2011/2012, heatwaves are more likely to occur, drought might result, and catastrophic natural events (e.g., floods, storms, megafires) might increase in intensity. What the plant experienced during the study was only a glimpse of what is potentially coming. Indeed, Europe, the location of the plant described in this paper, is particularly exposed to extreme events (e.g., France, Germany), although not as much as other countries (e.g., Haiti, Pakistan, Bangladesh and Thailand), yet, more than many others (e.g., USA, Brazil) (Germanwatch 2021).

As an illustration, the 2022 summer was particularly alarming, with many fires across France, in places without histories of such events (e.g., fire pits in Southwest—Gironde, and West—Brittany, of France), a topic which is part of the consequences of the global warming scenario (Zask 2019). Beyond France, the heatwaves also correspond to very unusual temperatures for European geography which also suffered several fires (e.g., Germany, Italy, Spain and Portugal), low level of rivers (e.g., the Thames, the Loire, the Gironde and the Rhine) and of hydraulic reservoirs (Norway’s reduction of electricity production); storms (a particularly devastating one in Corsica in August 2022); glaciers melting then collapsing (in Italy for instance in July 2022); and a marine heatwave (linked to Corsica’s storm in August).

2.3 From Natech to Socio-Natech

Events (e.g., droughts, storms, floods, megafires, but also snow) which can trigger accidental scenarios in safety-critical systems (e.g., refineries, chemical plants, nuclear power plants) have been conceptualized as Natech (natural events triggering technological accidents). Risk assessments take them into account to anticipate potential effects on critical infrastructures and safety-critical systems (Mikellidou et al. 2018). From a socio-technical point of view, global warming is therefore a new condition, bringing new threats to safety-critical systems. Coping with heatwaves in terms of operational practices or dealing with a lack of water to cool down reactors is significant modifications.Footnote 1 The same can be said concerning the exposition to floods (or megafires). These new potentialities have concrete operational, managerial and regulatory translations:

  • Modifying working hours to accommodate temperatures (i.e., starting earlier in the morning when temperatures are lower) is one option—beyond insulation or air conditioning—which can require new legal and human resources arrangements and negotiations. It might also create new issues for organizing production across departments, services, organizations (e.g., subcontracting) or sectors in safety-critical systems, with cases for which such arrangements are difficult to implement in practice.

  • Designing and authorizing an alternative mode of operating (for which one important barrier in the defense-in-depth protecting from the risk of runaway reactions—the cooling down systems based on water—might be inoperative during some periods in the year) is a challenge involving the collaboration of engineers, managers and regulators. Additionally, low level of river might also limit negotiated threshold of flows of pollutants for chemical plants, and heatwaves could alter water treatment processes, as much as electronic components of systems.

  • Preparing for extreme events such as flooding requires a thorough establishment of contingency planning based on risk assessments which incorporate the consequences of rising levels of water, with a certain speed and intensity in sometimes complex processes, as shown in the USA with the Arkema chemical plant during the Hurricane Harvey in 2017 (while the same applies to an exposition to megafires).

In other words, Natech is also very much a Socio-Natech problem. Indeed, how will these prevention and mitigation strategies (e.g., modifying working hours, designing and authorizing an alternative mode of operating, preparing for extreme events) be implemented in digitalized, externalized and networked safety-critical systems? What to make of all these global scale transformations and their effects on the reliability, safety and performance of safety-critical systems which has been a topic for several decades? How to conceptualize these large-scale evolutions, from increase of flows through globalized processes to climate change and their likely influence of operations?

Overall, these are indeed radical transformations of safety-critical systems contexts which have occurred within a few decades. Our argument is that these global scales are a level of change which is worth reflecting upon in the safety field. In this respect, a combined conceptual and empirical proposition to do so, to tackle this “challenge of change” (Hale and Baram 1998) is Post Normal Accident (Post NA, Le Coze 2021a, 2022, 2023a, b), to which we briefly turn.

2.4 Post Normal Accident (Post NA): Global Scales (New Causal Regime)

The Post NA proposition comes back on the seminal book, Normal Accidents (Perrow 1984), to provide an update for our contemporary era. There is no need to introduce in detail Normal Accidents (NA) (Perrow 1984). As most readers know, NA’s contention is that coupling and complexity of high-risk systems create opportunities for catastrophes, and some exhibit such features (e.g., nuclear power plants). The book has an iconic status, within and beyond academia (Clearfield and Tilcsik 2018). One reason is that the book helped us reason about the new level of complexity induced by the advent of large technical systems in the 1960s and 1970s (e.g., aviation, oil and gas, nuclear industry, nuclear weapons, dams).

The picture is now of course quite different from the 1980s. Beyond the flows triggered by globalization, the ecological crisis including global warming affects high-risk systems through an increase of natural catastrophes (e.g., storms, floods, droughts, heatwaves, megafires) which threaten their mode of operating. The notion of Natech (or, as suggested in this paper, Socio-Natech) has conceptualized these relationships between nature, socio-technical systems and safety, one important example for safety being the tsunami flooding the Fukushima Daïchi nuclear power plant in 2011, in Japan (Pritchard 2012). In this respect, socio-technical systems have become (global) eco–socio-technical systems.

These global trends create a very different situation for socio-technical systems in comparison to the 1980s, and two categories have been proposed in the past twenty years to update our perspectives on risks in the twenty-first century: systemic and existential risks. Systemic risks are these threats associated with the increase of flows that globalization entails (Goldin and Mariathasan 2015; Goldin 2020). A problem somewhere in the world can affect remote or distant places through the diversity of flows shaping globalization, through rippling effects.

Existential risks are these threats with the potential to affect societies’ survival and perhaps even that of humanity (Ord 2020). By the scale of their potentialities, they also address, like systemic risks, a global level of analysis. One existential risk is the prospect of a drastic degradation of living conditions due to the anthropocene, in a more or less distant future, depending on geographies, and societies’ actions in the decades to come (Gemene and Rankovic 2019).

Existential risks such as the anthropocene combine a number of highly interdependent dimensions, themes and measured variables at the global scale associated with diverse types of impacts including global warming (i.e., carbon dioxide emission, rising water levels, ice melting, average temperature increase, acidification of oceans, health-related effects), biodiversity loss (i.e., eutrophication of oceans, forests devastation, invasive species, agriculture extension and fishing—ocean depletion) and pollution (i.e., plastics, wastes, pesticides, endocrine disruptors). Much as for the category of systemic risks, the relationships are complex, in which complex circular causalities dominate.

Furthermore, many safety-critical systems which were core to Perrow’s argument (e.g., aviation, nuclear, oil and gas industry, maritime transport, chemical industry) are now at the heart of globalization and the anthropocene (large technical systems constitute the infrastructures of globalized flows and of nature’s degradation, Le Coze 2023a, b). Consequently, Post NA argues that these contemporary unfolding realities constitute a new causal regime at the global scale for safety. There is a level of connection between many safety-critical systems and their environments which has moved safety research to another level of causation and understanding (a change of our cosmology in the anthropological sense, see Latour 2015), which is also a change of our understanding of their scope, scale and time frame …

2.5 Conclusion

Over the past three decades, the operating landscape of safety-critical systems has profoundly evolved, and the argument of this chapter is that climate change should be linked to these profound changes of the past decades. This reformulates our understanding of the conditions of reliability, performance and safety of such systems. They are indeed connected to broader chains of causation, regime of causality, created by global processes which differ from the past. Post Normal Accident (Post NA) is one proposition to conceptualize this new situation, describing a new causal regime, combining systemic and existential risks into a perspective of the contemporary situation of safety-critical systems and expanding coupling and complexity at the global scale.