Accidental pollution events

Monitoring of Metals Under a Simulated Accidental Pollution Event Assessment of the Temporal Variability... 

SIPs are intended to prevent air pollution emergency episodes. The phms are directed toward preventing excessive buildup of air pollutants tliat me known to be harmful to the population and the enviroiunent when concentrations exceed certain limits. The compounds affected under the implementation plans are sulfur dioxide, particulate matter, ctirbon monoxide, nitrogen dioxide, and ozone. A contingency plan, which will outline the steps to be taken in tlie event tliat a particular pollutant concentration reaches tlie level at wliich it can be considered to be hannful, must be included in each implementation plan. The implementation plans are solely based on tlie continuous emission of tlie previously stated air pollutants. They do not mandate any actions to be taken in tlie event of an accidental toxic release. 

Ever since the inception of the petroleum industry the level of fires, explosions and environmental pollution that have precipitated from it, has generally paralleled its growth. As the industry has grown so has the magnitude of its accidental events. Relatively recent events such as the Flixborough incident (1974), Occidental s Piper Alpha disaster (1988), and Exxon s Valdez oil spill (1989) have all amply demonstrated the extreme financial impact these accidents can produce. 

The possibility of accidental events such as fire and chemical spills revealed some possible short term undesirable effects. These were in the area of obnoxious fumes or short-time-interval pollution. Virtually all of the sulfur pavement materials were difficult to ignite and were self extinguishing. 

Most laboratory-based calibration experiments are conducted under controlled conditions, with constant concentrations of metals in clean water. This does not necessarily reflect field conditions, and there is uncertainty as to how laboratory estimates of calibration parameters will transfer to the field, where concentrations of metals can fluctuate over several orders of magnitude. Fluctuations are often a result of an accidental discharge or weather event, and they can be accompanied by changes in, for instance, levels of dissolved and particulate organic material, pH and redox conditions. It is difficult to reproduce these conditions in the laboratory, but the use of tank studies where conditions can be controlled, and defined patterns of changing pollutant concentration can be achieved could provide a solution, and provides a compromise between the use of routine laboratory-based calibration, and the use of a reference site. 

The World Health Organization (WHO, 2001) defines chemical incidents as accidental or intentional events that threaten to expose or do expose responders and/or members of the public to a chemical hazard. Incidents can be sudden and acute, when hazardous chemicals are overtly released into the environment. Incidents can also have an apparently slow onset, when there is a silent release. The result could be substantial pollution of the immediate neighborhood, which might have health effect similar to the long-term consequences of short-term, major accidents. 

The BDMP supporting a risk analysis of this system is given in Fig. 5. It models the different scenarios that lead to pollution of the environment (the top event). There are three t3q>es of possible scenarios attack scenarios, accidental scenarios or hybrid scenarios. The first type of scenarios is a successful attack initiated by a malicious person, the second type is based on mere accidental events like failures of the system s components and the third type is a combination of attacks and components failures. This latter type best characterizes the possible interactions between safety and security events. 

The Accidental Scenario In this case pollution is caused if the pipeline breaks accidentally then the protection system fails to react. The protection failure is realized in two cases no instructions given by the RTU or the on-demand failure of the equipments (valves and pumps) to react properly. The first case is reahzed if the RTU fails or if it doesn t react which implies that it receives no instruction from CC and it does not activate its reflex action. Safety leaves of the BDMP detail the accidental events leading to such scenarios. 

The first accidental scenario given in Tab. 8 appears with a probability of 1.98e-5 and consists of accidental break of the pipeline and failure of the sensors to communicate correct measures to RTUs. Redundancy among sensors and the elimination of single points of failure could be considered to prevent such accidental scenarios. Results demonstrate that the hybrid scenario is more probable than the accidental scenario. Security events accelerate very much the realization of the undesired event (pollution). 

How to analyse the causes of accidental events involving releases of pollution Chapter 5 gives an overview of accident models that have been developed to support analyses of this type. Chapter 6 presents one particular analysis framework that suits this purpose. Chapter 8 focuses... 

---