Dear Straight Dope:
After watching Dawn of the Dead, I am left to wonder about one thing: If we were to suffer an apocalypse where most of the living became flesh-eating zombies, how long, assuming I survived, would I continue to receive hydroelectricity from my power company? Is it a mean-time-before-failure situation, or would the system automatically shut itself down after a few days? (I am assuming that most of the people who were supposed to be maintaining things at my hydro company would be out looking for brains, and that the surviving hydro employees would be busy digging shelters, etc.) Also, what's the outlook like for people whose chunk of the power grid is supplied by coal, nuclear, and other types of energy? Just wondering how many solar panels I should be putting on my roof!
Believe it or not, this is a question I’ve been asked before. Many people wonder how key parts of civilized society might continue after a post-apocalyptic Dawn of the Dead / Night of the Comet / Omega Man / Teletubbies Go to Paris scenario. Your question has two possible answers depending on which scenario of zombie conquest you envision.
In Dawn of the Dead, the zombification process doesn’t happen all at once. We can imagine a gradual scenario in which the infrastructure systems controllers plan ahead for shortages of personnel and try to keep the power going as long as possible. Alternatively, zombification could happen fairly quickly – say, over a few hours. I’ll address the second, more dire scenario in detail first, then the first, slightly less alarming one briefly.
How long the power supply would last in the most critical zombie situation depends on two key factors – first, how long a given power plant can operate without human intervention, and second, how long before enough power plants fail to bring down the entire transmission grid. I’ll ignore the side issues of whether the zombies would want to try to run the power plant themselves, or if they would be a union or non-union shop.
Power plants are incredibly complex facilities with an enormous number of controls, and consequently an enormous number of things that can go wrong. The level of complexity and reliability of the plants is a function of the type of power plant, the control systems installed, and the plant’s age and condition. In addition to the possibility of unplanned events causing shutdowns, there is also the problem of maintaining a fuel supply without human intervention. Given all these variables, coming up with hard and fast numbers is difficult. To address your question as well as I can, I’ll break down power plants by type (coal, nuclear, hydro, and natural gas) and discuss each one separately, focusing on the U.S. and Canada, since their electrical systems are closely tied. I’ll ignore oil-based plants because, contrary to popular belief, oil provides only a small fraction of total utility power generation in North America.
About 51% of U.S. and 16% of Canadian electrical generation comes from coal-fired plants. Coal power plants are generally the most problematic in terms of supplying enough fuel to remain in operation, and I could write (and have written) hundreds of pages about them. Mercifully, I’ll summarize. At most coal power plants the coal is stored in a huge outdoor pile, where it is typically pushed by bulldozers onto a conveyor and carried to large silos or bunkers at an upper level of the plant, from which it is fed to the burners. When the plant is operating at full output, these bunkers theoretically have a capacity ranging from 8 hours to more than 24 hours. As a practical matter, depending on the amount of coal in the bunkers and the way the plant distributes coal to the burners, the plant may start losing power in as little as 2-4 hours. Whether or not this initial reduction in coal flow shuts the plant down depends on the sophistication of the control systems and the ability of the plant to continue at partial power output without operator intervention.
Coal plants commonly require a lot of operator input to keep running. The controls at coal plants vary tremendously, from systems that are essentially unchanged since the 1950s to modern closed-loop neural network predictive models. In my experience from many months spent in control rooms of power plants around the world, coal plants on average require some sort of operator response for a “critical alarm” every 1-3 hours. Sometimes this is a relatively minor issue, such as a warning to flush the ash systems; sometimes it’s more serious, such as excessively high steam temperature or low coal supply. Whatever the case, if the control room were left unattended, I think it’s likely that a large number of coal power plants would “trip” (automatically shut down and disconnect from the electrical grid) within 12-18 hours.
About 20% of United States’ and 12% of Canadian electrical generation comes from nuclear power plants. Nuclear plants can operate a long time between refuelings – 500 days is a typical quoted figure, and some plants (Brunswick 1 and Pickering 7) are notable for having gone more than 700 days between refuelings. Nuclear plants tend to be more stable in operation than coal plants, and generally have more advanced control systems that can correct for minor problems or routine fluctuations. Two nuclear plant operators I asked about this wondered what I had been drinking, then said that a modern North American nuclear plant would likely run unattended for quite a bit longer than a coal power plant barring a mandated operator response – perhaps as long as a few days to a week. This could vary considerably depending on the plant.
Hydroelectric plants supply roughly 60% of the electricity in Canada and 7% in the United States. In addition, the northern U.S. imports a significant amount of Canadian hydro power on top of that 7%. Hydro plants for the most part are highly reliable and require relatively few controls. Since their “fuel” is the water contained behind the dam, their “fuel reserve” can often be measured in weeks or months. Barring sudden equipment failure or other unusual circumstances, most hydroelectric plants in good operating condition would last days or weeks unattended.
Natural gas is the last significant fuel source for power plants in the United States and Canada. Most natural gas power plants in North America use turbines, which resemble a stationary jet engine. (Boilers, the other major gas technology used for electricity generation, typically are used for emergency power or startup power at coal plants.) A turbine receives its gas supply from a pipeline; as long as the pipeline has sufficient pressure, the turbine will have fuel. How long a pipeline would keep its pressure during a Dawn of the Dead event is difficult to determine. Experts I asked thought that pipelines in most regions would maintain pressure for only 1-3 days without human intervention – maybe less, depending on the status of power to the controls and other electrically-powered equipment. In other words, failure of a few key power plants or transmission systems could result in a cascade failure of natural gas supply to large portions of the system.
Simple-cycle natural gas turbines are highly automated systems with relatively few moving parts. I have worked at a power plant with simple-cycle natural gas turbines that ran essentially unattended for three days at a time, with operator input limited to dropping the power output at night and ramping it back up in the morning. That particular plant operated so well and so safely with minimal attention that the operators tended to read a lot, tie flies for fishing lures, and engage in Greco-Roman wrestling when the urge hit them (don’t ask). Combined-cycle gas turbines, which include a steam generation component, have more controls and moving parts and require greater attention. Combined-cycle gas turbines would likely operate unattended for a shorter length of time – perhaps only a day or two, depending on the age of the plant and the degree of automation.
Focusing on individual plants doesn’t give us the whole story, though. The North American power grid is a classic illustration of a chain being only as strong as its weakest link. As we saw during the blackout of August 2003, a relatively minor event or series of events can, under the right circumstances, bring down large portions of the whole system. During the August blackout, despite massive non-zombified human intervention, enough parts of the system failed to result in the loss of more than 265 power plants and 508 generating units within a few hours. As bad as the blackout was, without human intervention to shut down plants safely, balance load, transfer power to different lines, and disconnect salvageable chunks of the system from those that had totally collapsed, it could have been much worse. Quick intervention allowed isolated "islands" of power to remain in service – one large island in western New York supplied nearly 6,000 megawatts and was used to restart the power grid days later. But without humans working to isolate it, that island would not have been formed in the first place.
Bottom line? My guess is that within 4-6 hours there would be scattered blackouts and brownouts in numerous areas, within 12 hours much of the system would be unstable, and within 24 hours most portions of the United States and Canada, aside from a rare island of service in a rural area near a hydroelectric source, would be without power. Some installations served by wind farms and solar might continue, but they would be very small. By the end of a week, I’d be surprised if more than a few abandoned sites were still supplying power.
Now, let’s address a scenario where the zombification process is gradual. If the operators and utilities had sufficient advance warning they could take measures to keep the power going for a while. The first thing would be to isolate key portions of the grid, reducing the interties and connections, and then cease power delivery altogether to areas of highest zombie density. After all, it’s not like the zombies need light to read or electricity to play Everquest. Whole blocks and zones would be purposely cut off to reduce the potential drains (and to cope with downed lines from zombies climbing poles or driving trucks into transformers). Operators would work to create islands of power plants wherever possible, so if a plant were overrun by zombies and went down it wouldn’t drag others down with it. In cooperation with regional reliability coordinators, the plant operators would improve plant reliability by disabling or eliminating non-critical alarm systems that might otherwise shut down a power plant, and ignoring many safety and emissions issues.
Fuel supply would eventually be a problem. Hydro plants would fare best, essentially having an unlimited fuel supply given normal rainfall, and could operate until some essential component failed or wore out. Nuclear plants could run for perhaps a year or more before they would need refueling. Refueling is a tricky operation requiring many specialized personnel, and it’s doubtful that a nuclear plant could effectively refuel if 90% of the nuclear technicians and engineers in the country were running around glassy-eyed in the parking lot. Coal power plants on average have maybe 45-60 days’ worth of coal on hand. If the power output of the plant were reduced, this could be stretched for six months or more, but eventually it would run out unless deliveries could be maintained. There are a few mine-mouth coal power plants in the U.S. that could conceivably run for years, provided enough miners and operators remained un-zombified. Natural gas plants might be the most vulnerable, since maintaining the gas wells, balancing the gas flow, and otherwise keeping the pipeline system intact requires considerable effort. In addition, most power plants have little or no gas storage available on-site, so a zombie situation could put natural gas plants in a real bind.
So there you have it. As to your final question, I can suggest a better tactic than relying on solar. Go to the abandoned hardware stores, load up a flatbed trailer with gasoline generators, and take them and a few dozen tanker trucks of gasoline to your house. You could have power for a long time, possibly years or more, until the zombies finally come for you.
SDSTAFF Karen inquires:
What about random zombie sabotage? For example, if some zombies got into the power plant and started randomly pushing buttons, pulling levers, and yanking cables, how much damage could they do?
SDSTAFF Una replies:
Outside the control room, most essential wires and cables are contained in armored cable trays, or else are tucked well out of the way. However, once you get into that control room … well, the ones at the power plants I have been to are amazingly fragile. Most coal plants have an incredible number of exposed controls that can trip the unit, and I have met engineers who had accidentally done just that during a site visit. That’s why I instruct all the engineers working under me on their first visit to the control room to not only not touch anything, but to leave a "magic foot," or one-foot barrier, between them and any and all controls, tables, chairs, etc.
Sometimes that doesn’t work. A co-worker was notorious for years for having bumped an empty ceramic coffee mug that fell onto a control panel, hit a control, and ended up tripping the unit. A $20,000 mistake. Thankfully, I’ve never done that.
Gas turbine plants are typically self-contained and the controls are out of the way. However, punching or clawing at a few panels would shut them down hard. My understanding is nuclear plants have more safeguards, but they’re not my area of expertise, and times being what they are, I’d just as soon not know.
Canadian Electricity Association Website – http://www.canelect.ca/english/electricity_in_canada_snapshot_Demand_1.html
US DOE Energy Information Administration Website – http://www.eia.doe.gov/cneaf/electricity/epm/table1_1.html
Nuclear Energy Institute Website – http://www.nei.org/index.asp?catnum =2&catid=47
U.S.-Canada Power System Outage Task Force, Final Report on the August 14th Blackout in the United States and Canada – https://reports.energy.gov.
Send questions to Cecil via firstname.lastname@example.org.
STAFF REPORTS ARE WRITTEN BY THE STRAIGHT DOPE SCIENCE ADVISORY BOARD, CECIL'S ONLINE AUXILIARY. THOUGH THE SDSAB DOES ITS BEST, THESE COLUMNS ARE EDITED BY ED ZOTTI, NOT CECIL, SO ACCURACYWISE YOU'D BETTER KEEP YOUR FINGERS CROSSED.