Amusement park fire still haunts after 30 years

At the NFPA Conference and Expo in Las Vegas, I attended the session entitled “Analysis of Potential Fire Sprinkler Performance in the Great Adventure Fire of May 11, 1984.” The presenters, Messrs. Jack Fairchild and Brad Casterline, reported their post-fire analysis of this famous fire.

Their presentation was also the subject of an article in the June 2014 NFPA Journal..(Haunted by Fire (Conference Preview), Kathleen Robinson, NFPA Journal, May-June 2014).

Fairchild and Casterline revisited the post-fire analysis prepared by Russell Fleming in 1985, using today’s state-of-the-art engineering tools. (Analysis of Potential Fire Sprinkler Performance in the Great Adventure Fire, May 11, 1984, Russell P. Fleming, P. E., National Fire Sprinkler Association, November 12, 1985).

The fire occurred on May 11, 1984, at the Haunted Castle in the Great Adventure Park in New Jersey. Eight teenagers, ages 15 to 19, perished in the fire. The Haunted Castle was a haunted house amusement in which patrons walked through a maze-like structure and experienced various scary scenes. The house had two identical mazes. Both sides were open whenever the park was very busy. One side was designated the red side, the other side was designated the blue side. At the time of the fire, only the red side was open.

According to the fire investigation report prepared by John Bouchard of NFPA (available at, the fire was determined to have initiated in the Strobe Room when a 14-year-old set fire to a foam plastic pad using a cigarette lighter. The fire was ultimately ruled accidental. The area where the fire occurred was very dark, and patrons frequently used lighters to illuminate their paths.

The fire subsequently spread to combustible plywood and plastic interior finish materials and quickly developed to flashover. The building had no fire alarm and sprinklers.

The fire was important in many ways. It led to significant changes in the model codes with provisions dealing specifically with Special Amusement Buildings. In addition, the events occurring during the post-fire criminal trial resulted in a storm of controversy involving the chair of the NFPA Board of Directors.

Great Adventures, Inc. (aka Great Adventures Amusement Park), Six Flags, and two executives were indicted in September 1984 by the Ocean County (NJ) Prosecutor’s Office. The executives entered a pretrial community service program to avoid trial for manslaughter; they did not admit any wrongdoing. The corporations were tried for aggravated manslaughter charges.

Part of the defense team’s strategy was to argue that sprinklers 1) were not required by prevailing codes and 2) that if provided, sprinklers would not have been effective in preventing the fire deaths.

These arguments were justified by painting a picture that, given the inherent delays built-in to the time to sprinkler operation and the time to water application, sprinklers would not have operated soon enough to prevent the deaths. The key assumptions were:

• It was likely that standard response sprinklers would have been used instead of fast response sprinklers.
• It was likely that the sprinkler system would have been a dry pipe system.
• An exhaust fan operating directly above the point of origin removed all heat from the fire during the first 90 seconds at which time it was assumed the fan would have failed.

It is quite possible that, though fast response sprinklers were available in 1979 (when the structure was built), it is likely they would not have been used in such a sprinkler system. The other two assumptions, however, are subject to dispute; and there is no doubt that the three assumptions favor the defense contention that sprinkler operation would not have occurred soon enough to save lives.

The defense team assembled a veritable dream team of fire safety experts to support these arguments. The team included Professor Edwin Smith, of Ohio State University (OSU); Rolf Jensen, of Rolf Jensen & Associates; and Chet Schirmer, of Schirmer Engineering Corporation.

Professor Smith was a noted expert on fire combustion and heat release rates. He developed the OSU calorimeter, one of the most widely used bench-scale calorimeters, and is often used to determine heat release rates of materials. It is the precursor to the more widely used Cone Calorimeter.

In the Analysis of Potential Fire Sprinkler Performance in the Great Adventure Fire session, he testified as to the likely stages of fire development and growth within the Haunted Castle. Included in his testimony, was the opinion that “the fire was fast-developing and that an exhaust fan located above the point of fire origin would have removed the heat of the fire in its early stages.”

The following is from his testimony (Note: Unless otherwise noted, all testimony referred to and quoted in this column is extracted from the appendix of the Fleming report):

(Smith) Well, the ventilation, of course, delayed the rate of propagation of the fire down the trailer. At the same time, a relatively large area was involved before it did take off. So, once that ventilation fan was overwhelmed, then the rate increased dramatically.

Schirmer presented testimony as to whether or not sprinklers would have been required by applicable building codes. Both Schirmer and Jensen presented testimony regarding sprinklers and the life loss.

The following is from Jensen’s testimony on the impact of sprinklers:

Q: What is your opinion concerning what affect this sprinkler system would have had?

A: (Jensen) It's my opinion the system would have been successful in controlling the fire probably in about 25 percent actual fire spread on the red side, or the side of origin.

And, later in his testimony:

(Jensen) Now, once they (the sprinklers) came on, at that point they would have controlled the fire. There isn't any doubt in my mind of that, because we have a very good water supply on the site. But, by that time the fire would have spread so far that the level of lethal combustion products throughout the Haunted Castle would have been well above the level where anybody could have survived.

The following very long and carefully worded question was asked of Schirmer:

Q: Now, I want you also to assume that, given that construction, that nature, even though you've told us that the code does not require sprinklers to have been installed in those castles, in that castle, I want you to assume that there was a system within the castle, and it was a system similar in design and construction to that which we have discussed earlier today on those two charts that you identified for us, and that it was a system that met the standard that was applicable at the time. All right?

Finally, what I want you to assume for me are the following facts: That within one minute after the mat in the back of the strobe room was ignited, the plywood on the ceiling above the mat was involved in flame; within one and a half minutes after the mat was ignited, the exhaust fan, or for one and half minutes after the mat was ignited, the floor vent that we talked about was actually removing the majority of heat and smoke and air from that area; within two and a half minutes after the mat itself it had ignited, flashover had occurred in the north end there, the back end of the strobe trailer; within one minute after that flashover had occurred, I want you to assume that the level of carbon monoxide in this area of the castle, roughly adjacent to what's been described as the Phantom area, exceeded 75,000 parts per million. I want you to also assume that within three minutes after the mat was ignited that flashover had occurred over that entire strobe room; within three and a half minutes of the time it was ignited, flashover had occurred in the adjoining trailer; and finally, the fire itself had spread throughout the entire red side within four and a half minutes from the time that the mat itself was ignited. All right?

Now, based on those assumptions, on the information you had related to us and that you were familiar with concerning the tests that had been performed, the Norwood test, the Charlotte test, the Los Angeles test, in your opinion would an automatic sprinkler system installed in accordance with that NFPA 13-1976 that you mentioned that was the standard, would an automatic sprinkler system installed in accordance with that standard have prevented the fatalities which took place in the Haunted Castle fire?

A: (Schirmer) No.

Q: Would you tell us why not, please? In this particular circumstance, in this particular case.

A: (Schirmer) As I've indicated in the description of these residential fire tests, if the sprinkler system does not operate quickly enough or if the water does not hit high enough on the walls, either of those instances, you cannot control the survivability of people in the space of origin. In other words, it is necessary that water hit high enough on the wall to prevent fire spread, and it is also necessary that the sprinkler operate fast enough to knock the fire down and maintain tenability limits in that space of origin.

Later in Schirmer’s testimony:

Q: You told us that in your opinion a sprinkler system would have not prevented the fatalities, correct?

A: That's correct.

The jury probably did not remember that two-page question with all the assumptions. The jury did likely remember this later very simple question and answer. In my view, the attorney succeeded in eliciting a simple and concise statement that sprinklers would not have saved any of the lives.

John Bouchard, assistant division director of the Engineering Service Division at NFPA, who investigated the fire for NFPA, testified:

The fire would have been detected (by sprinklers) earlier and perhaps controlled to a tenable situation. (Analysis of Potential Fire Sprinkler Performance in the Great Adventure Fire of May 11, 1984, Jack Fairchiled and Brad Casterline, presented at the NFPA Conference & Expo, Las Vegas, June 2014).

The jury ultimately found the defendants not guilty.

At the time, Schirmer was chairman of the NFPA Board of Directors and, more importantly, he was chairman of the NFPA 13 Technical Committee on Automatic Sprinklers. Jensen was also a long-time member of the 13 committee.

At a subsequent NFPA meeting (I think it was the November 1985 Technical Session. I was present at the NFPA session that day.), a motion was made to remove Schirmer as chair of the Board of Directors. The motion failed.

As Fairchild and Casterline were quick to point out, Schirmer and Jensen made tremendous positive contributions to fire protection engineering and use of sprinkler technology as a life safety tool.

In their presentation in Las Vegas, Messrs. Jack Fairchild and Brad Casterline looked at the case in which all the assumptions used by the defense were in effect. Their analysis used FDS to model the fire. Fire Dynamics Simulator (FDS) is a computational fluid dynamics (CFD) model of fire-driven fluid flow developed by the National Institute of Standards and Technology. Tyco’s SprinkFDT QTM software to model dry pipe sprinkler system water delivery time.

The behavior of building occupants was modeled using the evac simulation module contained in FDS. This evac portion of the software randomizes occupant behavior, and will yield varied results for individual trial runs. They ran a limited number of trials to establish a range of results that could be statistically analyzed.

Their presentation included preliminary results of their analysis. This information was not included in the Fire Journal article. The presentation can be purchased and downloaded at With no sprinklers present, their limited number of trial runs indicated an average life loss of nine persons. With sprinklers present, the average life loss was two persons. Fairchild said that no definitive conclusion could be drawn, but that probabilities could be presented. It is interesting to note that in their analysis there was no difference in the effect of sprinkler Response Time Index on the average number of deaths. RTI is a measure of sprinkler sensitivity.

The 1985 Fleming analysis is a hallmark of elegant use of models available at the time, and is a must-read for all engineers that think they can do forensic work.

His analysis used two fire development curves 1) a fast growing t-squared fire and 2) a faster developing fire using a t-fourth power growth rate. Both fire growth curves were modified to eliminate all the heat release developed in the first 90 seconds to model the operating exhaust fan assumption. He models sprinklers with RTI’s of 50 (modern era fast response sprinkler), 210 (early era fast response sprinkler) and 500 (standard response sprinkler). Sprinkler activation times were predicted using DETACT-QS. Dry pipe valve activation time was estimated; sprinkler water delivery time was calculated using a program developed by Factory Mutual.

Life safety judgments were made based on whether or not sprinkler operation was predicted to occur before flashover. Life safety was also analyzed based on calculation of toxicity levels produced by burning polyurethane and plywood.

To make a long story short, Fleming concluded:

Using information on expected rates of fire growth and sprinkler system design provided by expert witnesses who testified on behalf of Six Flags Great Adventure, it has been shown that a properly designed and installed sprinkler system would likely have saved the lives of the eight teenagers who died in the fire.

It was reported that the families of 7 of the 8 victims settled out of court, and one family received $2.5 million. The eighth family went to trial, and was ultimately awarded $750,000.

Nowadays, DNA is often used to prove or disprove the guilt of someone for crimes committed long before the technology was available.

In the case of the Haunted Castle Fire, it could be argued that the science of 1985 was confirmed by the state-of-the-art science available 30 years later.

Samuel S. Dannaway, PE, is a registered fire protection engineer and mechanical engineer with bachelor’s and master’s degrees from the University of Maryland Department of Fire Protection Engineering. He is past president and a Fellow of the Society of Fire Protection Engineers. He is president of S. S. Dannaway Associates Inc., a 15-person fire protection engineering firm with offices in Honolulu and Guam. He can be reached via email at

Content Type: