Natural and manmade disasters all have their own scary associations; for example, people who live in the south Asian nation of Bangladesh face the prospect of cyclonic storms each year that could potentially kill tens of thousands, if they don’t end up just destroying much of the year’s crops. Earthquakes such as the one that led to the 1906 fire in San Francisco are an example of a multi-phase disaster, in which one terrible event leads to another, no matter what interventions people attempt. Nuclear disasters are another example of this sort of multi-phase mayhem, as the environmental damage that results from the initial radiation contamination can cause major problems for decades. The horrors of Hiroshima and Nagasaki, in which nuclear detonations caused the deaths of tens of thousands, come to mind every time there is a nuclear accident; indeed, since the Three Mile Island nuclear accident in 1979, the United States has not permitted the construction of any new nuclear power plants (Behr). Gas plants can have similar effects on the environment when there is an accident, as the escaping fuel can cause irreversible damage to the outlying environment. With multi-phase disasters, human response is crucial for either ameliorating or worsening the initial situation, as the major spills and accidents in Bhopal (India), Chernobyl and Three Mile Island (Pennsylvania) have demonstrated over the past forty years (Gilinski). While the discovery of nuclear and gas-based power has helped us operate power generation facilities more efficiently and cleanly than in the days when coal gave us all of our electricity, it is crucial to manage accidents correctly when plant accidents take place. Nuclear power and gas send fewer pollutants into the air as an immediate by-product of the power creation process; however, if a spill does happen, human response is crucial. In each of the three spills mentioned above, key errors in human response led to unnecessary damage.
All three of the plants under consideration here (Three Mile Island, Chernobyl, and Bhopal) had significant issues at the time of construction that came into play when disaster came. When Three Mile Island came online, Unit 2 (the one that ended up having the spill) had difficulty getting started. The overseeing agency had just reduced the maintenance crew because of a desire to cut costs, when the unit opened, and it is likely that many of the issues that played a role in the accident might have been caught with a more thorough inspection (Perrow). Because the methodology of inspections is similar from plant to plant, it is likely that other plants that opened during that time period may have similar maintenance issues, which is quite a chilling thought if one considers how many nuclear power plants there are around the world (Seale).
The Bhopal chemical plant, in several ways, was destined for failure from the very beginning. First of all, its location was highly unsafe, as it was approximately 0.6 miles from the nearest train station and less than two miles from two large hospitals. This went against the Bhopal Development Plan, which had urged a distance of at least 15.5 miles between this type of plant and any urban areas (Varma and Varma). Many studies supported this sort of distance between plants using chemicals such as methyl isocyanate (MIC) and their surrounding communities (Reinhold; R. Varma).
Several safety factors also came into play that were known well before the Bhopal tragedy. For example, in plants that use MIC, one storage tank is usually left empty in case of an emergency; in Bhopal, all three of the tanks were in use, and the company stored much more MIC dormant in the tanks than is standard procedure. On the night before the accident, Tank 610 was almost 90 percent full: standard procedure is to keep these tanks less than half full, and the maximum recommended level for that tank was 60 percent (Diamond; Varadarajan). Finally, standard procedure is to store MIC below 52 degrees Fahrenheit; in Bhopal, it was kept at or above 68 degrees Fahrenheit, without refrigeration (Varma and Varma).
Another safety issue involved two monitoring devices that were not functioning properly at the time of the accident. The scrubber and the flare tower were both out of order; the scrubber is designed to curb a flow of up to 200 pounds per hour of MIC at 95 degrees Fahrenheit, with a pressure limit of 15 psi. The flare tower can burn small amounts of MIC as it escapes in a leak, and it was not working. Also, the plant had reduced its safety staff as a cost-cutting measure. When the plant had opened, there were 12 operators, 3 supervisors, 2 maintenance supervisors and a superintendent. However, at the time of the accident, this staff had been basically cut in half: 6 operators and 1 supervisor (Varma and Varma). The primary method of leak detection was finding symptoms in workers, such eye and throat irritation.
Union Carbide, the company that operated the Bhopal plant, had identified many of these safety issues in a report in May 1982 (Union Carbide). This report identified several steps that the company needed to take to make the plant safe once again. However, there was no evidence that any of those changes took place (Varma and Varma). Other signs that the plant was in trouble included a phosgene spill in December 1981, which killed one of the plant operators. It took three years for the paperwork from this accident to make it to the right people in the Indian Department of Labor, but the government recommendations were never implemented. When company workers protested the lack of action, two of them were fired (Ramaseshan). Needless to say, while Union Carbide ended up paying the Indian government a huge settlement as a result of this tragedy, the regulatory structure in India did nothing to help keep this from happening.
Of the three accidents, the one at Chernobyl seems like it would have been the easiest to prevent. After all, the events leading up to it included a test of the durability of safety equipment during a shutdown. As a part of the test, technicians had turned off safety protocols within the reactor so that it would not shut down automatically when certain levels were reached. Unfortunately, this type of reactor (the Soviet RBMK) loses stability at low power (Marples). Add this instability to a crucial operator error that led to a poorly timed power surge, and you have the cause of one of the worst environmental disasters in history.
The accidents at all three plants started as a result of the failure of what might seem like an extremely minor part. At Three Mile Island, there were two different cooling systems designed to keep the core from overheating and melting. The primary system ran water through the core where the primary nuclear reaction took place. This water would run into a steam generator, to release the heat from the primary water system. The secondary system would also develop steam, but this steam is what would spin the turbines and create power. The turbine blades contained a great deal of precision, which means that they had to be extremely clean. Any pollutants in the water, such as resins, needed to be polished off the turbine blades for the steam to turn them properly. In this case, the polisher developed a leak in one of its seals, and moisture went into the instrument air system (Perrow). The sensors in that system had the feedwater pumps stop, which stopped cold water from entering the system. This was a design feature that ended up causing problems, because if you remove the primary cooling impulse from a nuclear reactor, it is difficult to keep temperatures under control, as the operators would soon discover (Bupp and Derian). The valves that released steam from the system closed; heat built in the core, as did pressure. The Pilot Operated Relief Valve (PORV) is told to close, and it indicates that it has closed, but it is really stuck in the open position. At this point, the emergency coolant pumps turn on, but there is not enough pressure, because the PORV is open, letting the pressure out. The next emergency stopgap was the Hi Pressure Injection system (HPI), which sent cold water at a rate of about 1,000 gallons per minute in through jets to bring the temperature down. Unfortunately, an operator turned the HPI system down to keep the pressurizer from failing. This was done in accordance with procedure but ended up causing the core pressure and temperature to rise (Kemeny, et. al.).
After the investigation concluded, another human error had to do with valves that had been left closed in pipes that fed the feedwater pumps, which were trying to send cold water in before they were automatically turned off. Maintenance a couple of days before the accident had left the valves closed, and the operators on the day of the accident did not know that they were shut. The maintenance workers remembered leaving the valves open after their tests were done, but the valves were still closed. While there were indicators on the instrument panel that they were open, one was hidden by a repair tag (Perrow), and operators did not look for the other one, because these valves were simply always open except during maintenance testing; this was not an occurrence that they even thought could be a problem. After eight minutes, they did notice the closed valve; but it only took those eight minutes for the core to heat enough to lead to failure.
In the case of the Bhopal accident, the chemistry involved is much less complicated than the meltdown of a nuclear reactor core. Here’s what you need to know: if methyl isocyanate mixes with water, the reaction is so exothermic that boiling is virtually instantaneous. While the system was being cleaned on the night of December 2, 1984, some water got into MIC Tank 610 (Varma and Varma). In a matter of seconds, a significant amount of liquid MIC evaporated, causing a spike in pressure that burst the seals, sending MIC gas out the ventilator system (Diamond).
This leak took place and was instantly noted at 11:30 P.M. However, the authorities did not activate the warning system until 1:30 A.M. on December 3, two full hours later. The leak had actually stopped between 12:15 and 12:30 A.M., and people in the area of the plant were already reporting throat and eye irritation so severe that it had woken them up. However, this first leak was followed by a secondary leak around 1:00 A.M. Unfortunately, the police were not trained in what to say, so in their alert messages over loudspeakers, they told people to run because poison gas was spreading (Varma and Varma). This got people moving faster and breathing harder, and as a result inhaling more of the poison. Instead of beginning immediate evacuations shortly after 11:30, the authorities turned this into an even larger disaster. Most of the deaths from this accident were a result of pulmonary edema, which is what you get when you inhale MIC gas (Varma and Varma). By sunrise, though, the streets of Bhopal were filled with corpses (or soon-to-be corpses) of people, cattle, buffalo, and other animals, and the foliage had fallen from most of the trees in the area.
When the Chernobyl reactor had that unexpected power surge on April 26, 1986, the reactor unit’s roof shot up into the sky. The reactor simply sent the radioactive items (primarily cesium-137, strontium-90 and iodine-131) inside into the atmosphere, and it took twelve days to contain the leak by dumping sand and boron on the graphite-composed flame. From beneath, miners dug an area to keep the toxins above the soil, so that they would not enter the water systems. The square mileage of the contaminated area would fit, but just barely, inside Kentucky. About a fifth of Belarus was now deadly (Marples).
Accountability for the first two of these accidents, as one might expect, remains a mishmash. In the case of Three Mile Island, according to the findings of the Kenemy Commission, the LOCA (loss-of-coolant-accident) was the result of operator error. They should have seen that the valves were closed from the indicators on the gauge panel, and they should have been able to deduce that the PORV was stuck open because of the continued low pressure readings. Sending in a solid pressurizer to fix the problem ended up causing just as many problems as it solved. Studies by Babcock and Wilcox and the British Secretary of State for Energy also blamed the operators. If the operators had noted the abnormalities in the drain tank pressure or the high temperature of the drain tank, which was filled with hot coolant instead of cold water, the situation might have turned out differently. However, the protocols that were in place, and which they followed, ended up causing significant damage as well (Perrow).
In the case of the Bhopal tragedy, accountability was much easier to affix. Union Carbide had used shoddy maintenance and upkeep practices and had understaffed its safety and maintenance personnel. Some of this was due to the decline in sales of the pesticide Sevin that Union Carbide was producing in this plant (Varma and Varma). In the days after the spill, Union Carbide continued to try to avoid blame, claiming that MIC was far less toxic than what it would have to be to produce the widespread mortality that had taken place. Until further studies were done, Union Carbide had the authorities chasing down the likelihood of a phosgene leak, instead of blaming MIC. However, the Indian government, which had been so lax in enforcing maintenance and upkeep standards, finally stepped in and took over administering accountability in the wake of the disaster. The government paid out 10,000 rupees (about US$800, in 1984) for each fatality and 1,250 rupees (about US$100) for each person who had needed hospital care. Then they took US$470 million from Union Carbide in a settlement, but it took 15 years for the Supreme Court of India to order the government to use that money to compensate victims and their families in an appropriate way (Varma and Varma).
Accountability for the Chernobyl disaster was much easier to assign. Initially, the government evacuated approximately 135,000 people from the area around the reactor, basically a zone with a radius of almost 19 miles. The town of Chernobyl, the town of Pripyat and other settlements were emptied of all residents. When fallout was shown to have spread farther than that radius, in studies done in 1989, about a quarter of a million more people were evacuated to new settlements (Marples). The global impact was much larger, as fallout spread throughout the Northern Hemisphere, particularly over Europe and Asia (Morrey). Studies have linked the disaster with all kinds of cancers throughout the region, including a spike in thyroid cancers in children in Belarus and northern Italy in the years after the explosion. Because it takes about a decade between exposure and development of the actual cancer, the full toll of the spill took years to hit home (Chiesa, et. al.; Anspaugh, et. al.; Zieglowski and Hemprich).
If there is an overriding lesson to these three accidents, and the other similar stories that have happened in the centuries since industrialization, it is that while cutting corners in terms of safety often saves money in the short term, it is never worth the long-term risk. Chernobyl, Bhopal and Three Mile Island all could have been avoided with more careful attention to detail in the construction of operational protocols. While Three Mile Island was more the result of poorly written policy than outright incompetence, the shoddy maintenance and upkeep that led to all of the deaths in Bhopal and the poor logic that went into the creation of the safety experiment in Chernobyl stand as points in history that will continue to tell all of humanity that it is always worth the cost of safety improvements in the handling of power production and chemical processing, if human lives will be saved as a result.
Anspaugh, L., et. al. The global impact of the Chernobyl reactor
accident. Science Vol. 242 (4885): 1513-1519.
Babcock and Wilcox, Press Conference, 1 April 1979.
Behr, P. (2009). Three Mile Island still haunts U.S. nuclear
industry. New York Times 27 March 2009. Web. Retrieved 5
December 2011 from
Bupp, I. and Derian, J. (1978). Light water: How the nuclear
dream dissolved. New York: Basic Books.
Chiesa, F., et. al. Thyroid disease in northern Italian children
born around the time of the Chernobyl nuclear accident.
Annals of Oncology Vol. 15 (12): 1842-1846.
Diamond, S. (1985). The Bhopal disaster: How it happened. New
York Times 28 January 1985. Web. Retrieved 5 December 2011
Gilinsky, V. (1982). Full ahead for nuclear power? Technology
Review February/March 1982: 10.
Kenemy, J. (1979). Report of the President’s Commission on the
Accident at Three-Mile Island. Washington, D.C.: Government
Printing Office, 1979.
Marples, D. (2004). Perspectives on Modern World History. Web.
Retrieved 5 December 2011 from http://tg-
Morrey, M., et. al. (1988). A preliminary assessment of the
radiological impact of the Chernobyl reactor accident on the
population of the European Community. Luxembourg: Euratom.
Perrow, C. (1984). Normal accidents: Living with high-risk
technologies. Princeton: Princeton University Press.
Ramaseshan, R. (1984). Profit against safety. Economic and
Political Weekly Vol. 19 (51): 2147-2150.
Reinhold, R. (1985). Disaster in Bhopal: Where does blame lie?
New York Times 31 January 1985, A8.
Seale, R. (1974). Consequences of criticality in accidents, in
Nuclear Criticality Safety, ed. R. Douglas O’Dell. U.S.
Atomic Energy Commission, Technical Information Center, 16-
The Bhopal tragedy (Editorial).(1984). The Wall Street Journal 10
December 1984, 26.
Union Carbide. (1982). Operational safety survey: Co/MIC/Sevin
units. Danbury, CT.
Varadarajan, S. (1985). A scientific enquiry into the methyl
isocyanate leak in Bhopal. New Delhi: Council of Scientific
and Industrial Research.
Varma, R. and Varma, D. (2005). The Bhopal disaster of 1984.
Bulletin of Science, Technology and Society Vol. 25 (1):
37-45. Web. Retrieved 5 December 2011 from
Varma, R. (1986). The Bhopal disaster: Implications for
policymaking on hazardous technology in the third world.
Unpublished master’s thesis, Rensselaer Polytechnic
Institute, New York.
Zieglowski, V. and Hemprich, A. Facial cleft birth rate in former
East Germany before and after the reactor accident in
Chernobyl. Mund Kiefer Gesichtschir Vol. 3(4): 195-199.