Understanding and Fighting Basement Fires: The ISFSI & UL FSRI 2017 Joint Research Project

Preview for Dan’s #FDIC2020 Presentation/YouTube

By Dan Madrzykowski

Firefighters have always been taught that basement fires are one of the most dangerous fires to fight. They have been trained to operate above the fire and make their way down the throat of the fire to attack the fire on its own level. We understand from previous research that the floor systems on which we are walking or crawling fail quickly and without warning. Exposed wood-flooring systems do not maintain their structural integrity for long under fire conditions unless they are protected by drywall or a sprinkler system.

Since the double line-of-duty deaths (LODDs) on Cherry Road in Washington, D.C., in 1999, we have also been aware of  the “flow path” and its impact on firefighting, particularly in basement fires. What we must realize is that flow path didn’t first come about in 1999; we just became more aware of it then and, through technology, were able to recreate it and distribute simulations from the National Institute of Standards and Technology (NIST) for the fire service to study. This article provides an overview of the International Society of Fire Service Instructors (ISFSI) “Understanding and Fighting Basement Fires Project,” which includes a brief history of previous research, LODD basement fires, the research questions, and the tactical considerations developed as part of this research. In 2016, the ISFSI was awarded its most recent Assistance to Firefighters Grant Program’s Fire Prevention and Safety Research Grant. The study was funded to research basement fire attack tools and tactics and determine which are the most effective.   

The Study

The ISFSI developed a technical panel to guide the research to ensure that the end results would be beneficial and useful to the fire service. The panel included representatives from several sections of the International Association of Fire Chiefs (IAFC), the North American Fire Training Directors, the National Institute for Occupational Safety and Health (NIOSH), and from several fire departments that had suffered LODDs because of basement fires. The  ISFSI partnered with the Underwriters Laboratories (UL) Firefighter Safety Research Institute and the Delaware County (PA) Emergency Services Training Center to conduct the research.


The next step was to examine the previous research and fireground experience to understand the gaps in understanding basement fires. During the past 10 years, several floor-collapse studies have been conducted. In 2008, the UL Fire Safety Research Institute published its research findings on the structural stability of engineered lumber exposed to fire conditions.1 The experiments were conducted on a floor furnace.  The research demonstrated that “modern” engineered wood-floor assemblies failed faster than wood-floor assemblies with “legacy” designs. This study also pointed out that modern tools like thermal imaging cameras (TICs) had limited use in determining the condition of the floor assembly or the fire conditions under the floor. Further, the study questioned the use of the time-honored practice of “sounding the floor” to determine if it was safe to operate on the floor.       

In 2012, the  UL FSRI and NIST released a study on the examination of four types of flooring systems in a townhouse type of arrangement with a 720 square foot (67 square meter [m]) floor area and a 20 foot (6.1 m) span. These experiments examined the time to collapse for residential floor systems constructed with dimensional lumber, wood I-joists, parallel chord wood trusses, and lightweight steel C-channel. The results proved that any of the unprotected floor assemblies could collapse within the operational time frame of the fire department. This report also provided data that showed that current fireground practices of entering on the floor above the fire and working down to fire the basement would not provide the firefighter with the appropriate information to make decisions to provide a safe operating environment.2     

As a follow-up to UL’s floor furnace experiments, NIST conducted experiments in two-level wood structures with a 16 foot (4.8 m) span that supported the findings of the UL study on the value of gypsum board to protect the floor assembly and the challenges for TICs. Three TICs, each with a different type of sensor, were used to view and record the thermal conditions of the top of the floor assembly from the open doorway in the upper compartment. The times to the collapse of each floor were also noted. Given the insulating effects of the oriented strand board (OSB) and the floor coverings, the temperature increases or thermal signatures viewed by the TICs were small, given the fact that the ceiling temperatures below the OSB were in excess of 1,110°F (600°C).

These experiments further demonstrated that TICs alone cannot be relied on to determine the structural integrity of a wood-floor system. Therefore, it is critical for the fire service to review its practice of size-up and other fireground tactics needed to enable the location of the fire prior to conducting fire operations inside a building. This study also highlighted the relationship of increased ventilation to the growth of the fire in the lower level, which led to failure of the floor assemblies.3

Chief Gary Morris, representing IAFC Safety Health Survival Section and the Pine-Strawberry Fire District shared a study he had conducted on basement fires for the IAFC.4  Morris reviewed 21 incidents involving basement fires that resulted in 30 LODDs. In addition, he reviewed 400 near-miss reports related to basement fires that had been submitted to the National Near Miss Reporting System (NNMRS).  

The NNMRS data showed that the largest category for the near-miss reports related to basement fires involved the collapse of the first floor or firefighters falling through holes in the floor. The second largest category of near-miss events involved disorientation, separation, or entrapment of a firefighter operating in a basement. This information aligns with data from Chief Don Abbott’s Project Mayday, which revealed the second most common cause of a Mayday was a firefighter falling into a basement.

The technical panel then reviewed basement fire LODD case studies. There are many, including Joe Samec, Burr Ridge, IL; Joyce Craig, Philadelphia, PA; Richard T. Sclafani, New York City; Robin Broxterman and Brian Schria, Colerain Twp., OH; and Patrick Wolterman, Hamilton, OH. We know from the NIOSH LODD reports that the two most common causes of these deaths were a collapse and firefighters being caught in the flow path. Using the knowledge from previous LODDs along with the previous research the ISFSI and UL FSRI technical panel used to develop its research questions for the 2017 research, the technical panel posed the following research questions for further understanding how to fight basement fires:

              • What if the interior stairway is the only firefighter access?

              • What if there are no basement windows to the exterior?

              • Can door control be applied to basement fires?

              • Are alternative fire suppression techniques effective?

              • What if the basement is larger than those in previous studies?

The ISFSI basement fires study took place in 2017 at the Delaware County (PA) Fire and Emergency Services Academy. A ranch-style house with a full basement was built for this project. The floor area of the upper level was approximately 1,150 sq. ft. (106 sq. m.) and the basement floor area was approximately 1,050 sq. ft. (97 sq. m.).  Floor plans of the basement and upper level are shown in Figures 1 and 2, respectively.

Figure 1. Floor Plan of the Upper Level of the Test House

Figure 2. Floor Plan of the Basement of the Test House

The house was instrumented with sensors to measure temperature, pressure, gas velocity, and gas concentrations. These sensors provided information on the hazard conditions in different sections of the house to assess what occupants or firefighters would be exposed to at any given time in the fire. In addition, the data allowed the researchers to see if a given tactic or tool was working to decrease the hazard from the fire.   

Three types of basement access were considered: (1) No exterior access and no exterior vents–in other words, the only access was by an interior stairway; (2) No exterior access, but there were exterior vents that could be used to apply water into the basement; and (3) Exterior access at the basement level, which allowed firefighter entry.

Several firefighting nozzles and appliances were used during the experiments. Basic nozzles such as a smooth bore and a combination nozzle were used along with a cellar nozzle, a Bresnan distributor, and piercing nozzles. The nozzles were applied in favorable and less-than-favorable circumstances to see how well they controlled or extinguished the fire conditions and to examine the thermal conditions along the flow path and throughout the house.

The fires were started in one of four sofas in the basement. In addition, five wood pallets were arranged as a “desk” adjacent to each sofa. The exposed floor assembly was composed of wood I-joists on 16-inch centers and an OSB sub-floor. A full description of the experiments and the results are contained in the research report developed by UL FSRI. (4)

In addition to the graphs of all the data recorded, pictographs were developed to show the impact of a given tactic on the fire conditions. An example of one of the pictographs is shown in Figure 3. In the example, a 150-gallon per minute (gpm) straight stream was flowed into the small basement window remote from the fire and unable to hit the seat of the fire directly because of shielding from the stairway and a utility room. Even so, peak temperatures in the basement were reduced from over 1,000°F to 320°F; temperatures along the flow path were also reduced, making entry to the basement easy. An added benefit was that oxygen levels improved in remote areas of the house where victims may be trapped.

Related Articles:

Mayday Monday: Basement Fires

Humpday Hangout: ISFSI/UL Basement Fire Program

Humpday Hangout: Examining Basement Fires

Throw Back to Basics: Basement Fires

The Unexpected Basement Fire


Figure 3. Pictograph from UL FSRI Study

The tactical recommendations developed out of the research are the following:

              • Recognize that basement fires are a challenge because of the high potential for floor collapse and positioning in the exhaust portion of the flow path.

              • As with all fires, size-up is key.

              • If the fire is in the basement, do the following:

                             * identify potential flow paths–interior stair, side windows, ground level door

                             * coordinate ventilation with suppression tactics

                             * cool–piercing nozzles, cellar nozzles, hose streams through holes and windows, at grade attack (fastest, effective water)

                             * extinguish–after cooling, assess and enter if appropriate

                             * rescue–survival profiles, closed door, open door, gas concentrations

                             * continue to study for making/understanding your tactical choices. Know why.

              • Support codes and standards that support firefighters in achieving their goal of saving lives from fire.

              The ISFSI offered 50 Assistance to Firefighters Grant program-funded, eight-hour “Understanding and Fighting Basement Fire” courses throughout the United States in 2018 and is producing a training video featuring a general overview of the course. For additional information or to check the possibility of hosting a course, visit www.ISFSI.org or contact the ISFSI at (800) 435-0005. This course is dedicated to the memory of Lieutenant Matt LeTourneau, Philadelphia (PA) Fire Department, who passed away on Saturday, January 6, 2018, in a tragic LODD after succumbing to his injuries suffered in a structural collapse while fighting a rowhouse fire. Matt was an essential part of the research team and a contributor to the tech panel of the ISFSI Basement Fire Research and Curriculum.


1. Izydorek, MS, Zeeveld, PA, Samuels, MD, Smyser, JP, Report on Structural Stability of Engineered Lumber in Fire Conditions. Underwriters Laboratories, Northbrook, Illinois, September 2008.

2. Kerber, S, Madrzykowski, D, Dalton, J, and Backstrom, R, Improving Fire Safety by Understanding the Fire Performance of Engineered Floor Systems and Providing the Fire Service with Information for Tactical Decision Making

Underwriters Laboratories, Northbrook, Illinois, March 2012.

3. Madrzykowski, D and Kent, J, Examination of the Thermal Conditions of a Wood Floor Assembly above a Compartment Fire,  National Institute of Standards and Technology, Gaithersburg, MD,  NISTTN 1709, July 2011.   

4. Madrzykowski, D and Weinschenk, C, Understanding and Fighting Basement Fires: Report of Experiments, UL FSRI, Columbia, MD, January 2018.


Dan Madrzykowski, PhD, holds a Master of Science in fire protection engineering from the University of Maryland and a PhD from the University of Canterbury in New Zealand. Throughout his career, he has performed research focused on fire suppression, large fire measurements, fire investigation, and firefighter safety. He has conducted fire investigation-related studies over a wide range of topics that include firefighter line-of-duty deaths and injuries; the post-earthquake fires in Kobe, Japan; oil field fires in Kuwait; the Station Night Club fire; and the Cook County Administration Building fire in Chicago. He is a dedicated fire service researcher and educator. He was the recipient of the 2009 Fire Engineering/ISFSI Instructor of the Year Award.


No posts to display