Fire Dynamics Simulation of 2011 Baltimore County LODD- 30 Dowling Circle
By Christopher J. Naum, SFPE on Mar 25, 2012 with Comments 0
The Baltimore County (MD) Fire Department published the Line of Duty Death Investgation Report of the 30 Dowling Circle Fire recently.
The report was written by a Line of Duty Death Investigation Team comprised of departmental members, including representatives of the local firefighters’ union and the Baltimore County Volunteer Firemen’s Association.
An overview and executive narrative of the final report (PDF) on the apartment fire where Volunteer Firefighter Mark Falkenhan sustained fatal injuries was posed on CommandSafety.com HERE.
-
CommandSafety.com: Baltimore County (MD) Firefighter Falkenhan Line of Duty Death Report Issued
-
The entire report can be downloaded HERE .
-
Baltimore County (MD) Fire Department web site HERE
On Wednesday, January 19, 2011, a fire occurred in an apartment building located in the Hillendale section of Baltimore County, Maryland. This fire resulted in the line of duty death (LODD) of volunteer firefighter Mark G. Falkenhan, who was operating as the acting lieutenant on Squad 303 (for purposes of this report, Mark will be referred to as FF Falkenhan).
Upon their arrival, FF Falkenhan and a second firefighter (FF # 2) from Squad 303 deployed to the upper floors of the apartment building to conduct search and rescue operations. Other fire department units were already involved with both firefighting operations and effecting rescues of trapped civilians.
During these operations, FF Falkenhan and FF # 2 became trapped in a third floor apartment by rapidly spreading fire and smoke conditions. FF # 2 was able to self-egress the building by diving headfirst down a ladder on the front (address side) of the building. FF Falkenhan declared a “MAYDAY” and implemented “MAYDAY” procedures, but was unable to escape or be rescued.
FF Falkenhan was located and removed via a balcony on the third floor in the rear of the building. Resuscitative efforts began immediately upon removal from the balcony, and continued en route to the hospital. FF Falkenhan succumbed to his injuries and was pronounced deceased at the hospital.
The investigating team examined any and all data available, including independent analysis of the self contained breathing apparatus (SCBA), turnout gear and autopsy report. The Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF) produced a fire model to assist with evaluating fire behavior. Multiple site inspections were conducted. Extensive interviews were conducted by the team which also attended those conducted by investigators from the National Institute for Occupational Safety and Health (NIOSH). Photographic and audio transcripts were also thoroughly analyzed. A comprehensive timeline of events was developed. All information used to make decisions regarding recommendations was corroborated by at least two sources.
- In fairness to those units involved in this incident, the investigating team had the advantage of examining this incident over the period of several months. Furthermore, given the size and nature of the event, and the fact that arriving crews were met with serious fire conditions and several residents trapped and in immediate danger, all personnel should be commended for their efforts for performing several rescues which prevented an even greater tragedy.
- The team did not identify a particular primary reason for FF Falkenhan’s death.
- What were identified were many secondary issues involving but not limited to crew integrity, incident command, strategy and tactics, and communications.
- These issues are identified and discussed, and recommendations are made in appropriate sections of the report, as well as in a consolidated format in the Report Appendix.
Some of the issues identified in this report may require some type of change to current practices, policies, procedures or equipment. Most, however, do not. Specifically, the analysis and recommendations regarding Incident Command and Strategy and Tactics show that if current policies and procedures are adhered to, the opportunity for catastrophic problems may be reduced.
- Mark Falkenhan was a well-respected and experienced firefighter.
- He died performing his duties during a very complex incident with severe fire conditions and unique fire behavior coupled with the immediate need to perform multiple rescues of victims in imminent danger.
- It would be easy if one particular failure of the system could be identified as the cause of this tragedy.
- We could fix it and move on. Unfortunately it is not that simple.
- No incident is “routine”. Mark’s death and this report reinforce that fact.
On Wednesday, January 19, 2011 at 1816 hours, a call was received at the Baltimore County 911 Center from a female occupant at 30 Dowling Circle in the Hillendale section of Baltimore County. The caller stated that her stove was on fire and the fire was spreading to the surrounding cabinets. Fire box 11-09 was dispatched by Baltimore County Fire Dispatch (Dispatch) at 1818 hours consisting of four engine companies, two truck companies, a floodlight unit, and a battalion chief. All units responded on Talkgroup 1-2.
The location, approximately one mile from the first dispatched engine company, is a three story garden-type apartment complex, with brick construction and a composite shingle, truss supported roof. The fire building contained a total of six apartments divided by a common enclosed stairway in the center with one apartment on the left and one to the right of the stairs.
Fire Dynamics Simulation of 2011 Baltimore County LODD- 30 Dowling
Download the Fire Dynamics Report HERE
INTRODUCTION:
Assistance from the Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF) Fire Research Laboratory (FRL) was requested for a fire at 30 Dowling Circle by the Baltimore County Fire Investigation Division (FID) through the ATF Baltimore Field Division on the night of January 19, 2011.
ATF Fire Protection Engineers were asked to utilize engineering analysis methods, including computer fire modeling, to assist with determining the route of fire spread and the events that led to the firefighter MAYDAY and subsequent Line of Duty Death.
BACKGROUND:
Working closely with the Post Incident Analysis Team, the ATF Fire Research Laboratory created a computer simulation of the garden apartment building using Fire Dynamics Simulator (FDS). FDS is a computational fluid dynamics (CFD) modeling program developed by the National Institute of Standards and Technology (NIST).
FDS utilizes mathematical calculations to predict the flow of heat, smoke and other products of fire. Smokeview, a post-processer computer program also produced by NIST, was then used to visualize the mathematical output from FDS. The most current available versions of both programs were used: FDS 5.5.3 and Smokeview 5.6. Below are photographs of the front and rear of the fire building next to an image of the same building constructed in FDS.
The garden apartment building at 30 Dowling Circle was attached to two similar garden apartment buildings, one on each side. The fire damage was isolated to 30 Dowling Circle, so the exposure buildings were not included in the computer fire model. The entire six unit garden apartment building was modeled in FDS, including the patio and balconies on the rear of the building. FDS works by dividing a space into cubical “grid cells” for calculation purposes. FDS then computes various CFD calculations for each grid cell to predict the movement of mass, energy, momentum and species throughout a three-dimensional space.
The Dowling Circle model consisted of 2,560,000 total grid cells that were each 3.9 inch (10 cm) cubes. The model was used to simulate a total elapsed real time of 27.5 minutes, beginning before the 911 call and ending just after flashover of the third floor and the firefighter MAYDAY.
The model was synchronized in real time with the fireground audio throughout the duration of the fire.
FDS has been validated to predict the movement of heat and smoke throughout a compartment, however the accuracy of fire modeling depends on it being used appropriately by a trained user that is aware of its limitations. Due to lack of knowledge about the exact material properties for the various furnishings and other available fuels, a user-specified fire progression was used for this application.
For flame and fire gas movement after consumption of the original burning fuel packages, the fire model calculated smoke and ventilation flow paths through the building and was used to gain a better understanding of the rapid fire growth leading to flashover of the stairwell and third floor.
- In addition, FDS was utilized to illustrate the complex route of fire spread through the building as verified by witness statements, firefighter interviews, photographs and burn patterns.
- Input data for the computer model included heat release rate data and video from previous testing conducted by the ATF FRL and NIST.
- Ambient weather data was also input into the model, including temperature, as well as wind direction and magnitude at the time of the fire. In addition, several alternative compartmentation scenarios were modeled to explore the possible effects of closed stairway apartment entrance doors on the spread of smoke and flames in the stairwell.
- The statements of each firefighter were reviewed and their individual actions (breaking windows, opening doors, etc.) and observations (fire size, smoke conditions, etc.) were recorded on floor diagrams.
The actions and observations of the firefighters were then associated with specific times in the fireground audio to generate an overall event timeline. All events in the model are based on this master timeline of events. In addition, all photographs were time stamped and synchronized with the model. The Post Incident Analysis Team was consulted throughout the development of the event timeline and the computer fire model to ensure accuracy.
MODELING ANALYSIS:
1. Analysis of Fire Development in the Terrace Level
The fire originated on the stovetop of an occupied apartment on the right (south) side of the terrace level (apartment T2). Flames from a grease fire ignited kitchen cabinets, eventually causing the kitchen to flashover into the attached living room. Upon fire department arrival, a fully developed fire existed in the living room and kitchen of apartment T2. Prior to exiting the apartment, the occupant opened both the rear sliding door and the apartment entrance door in an attempt to ventilate smoke from the apartment.
An analysis of the ventilation flow path through the apartment with FDS indicated that a significant unidirectional flow path existed up the stairs with an inlet at the rear terrace sliding door and outlet at the front apartment entrance door leading to the stairwell.
This unidirectional flow path up the stairs is difficult to combat and is often experienced during basement fires as crews attempt to descend interior stairs. The model indicates sustained air temperatures in the stairwell of approximately 600 Fahrenheit (315 Celsius) at velocities of approximately 6 mph (2.7 m/s) from floor to ceiling as crews attempted to descend the stairs. This is consistent with statements from firefighting crews, who experienced extremely high heat conditions and indicated periodically seeing flames in the smoke layer flowing up the stairs.
The elevated air velocity of the stairwell flow path resulted in a high rate of convective energy transfer to the structural firefighting gear and high perceived temperatures as the firefighters attempted to descend the stairs. Firefighting crews flowed a hoseline down the stairs to combat the high temperatures; however no significant cooling was noticed by firefighters because the hose stream could not reach the seat of the fully developed fire in the kitchen area.
The crews were simply cooling the ventilation flow path without cooling the source of the energy in the apartment. It was not until a hose stream was directed through an exterior window and a portion of the fire was extinguished that gas temperatures and velocities began to decrease, allowing firefighters to make entry to the terrace apartment via the stairs.
Front photo of unidirectional flow of smoke up stairwell from apartment T2. Note the high volume of smoke from floor to ceiling as the stairwell door serves as the flow path outlet. The ground ladder in the foreground was used to rescue an occupant on the third floor trapped by heavy smoke in the stairwell. (Refer to Figure 014)
The first arriving engine, E-11, was staffed with a Captain, Lieutenant, Driver/Operator, and a Firefighter. Upon arrival at 1820 hours, the Captain gave a brief initial report describing a three story garden apartment with smoke showing from side Alpha: “The Captain of E-11 will have Command and we are initiating an aggressive interior attack with a 1 ¾” hand line”. Command also instructed the second due engine to bring him a supply line from the hydrant.
A female resident (victim # 1) appeared in a third floor apartment window, Alpha/Bravo side (Apt. B-1), yelled for assistance, and threatened to jump. Smoke or fire was visible from any of the third floor windows. At 1823 hours, Command advised Dispatch that he had a rescue and that he was establishing Limited Command. Fire Dispatch was in the process of upgrading the response profile to an apartment fire with rescue when the responding Battalion Chief requested that the fire box be upgraded to a fire rescue box. While the Firefighter and Lieutenant prepared for entry into the building, the Captain and Driver/Operator extended a ladder to the 3rd floor apartment window and rescued the resident. The first attempt by the Firefighter and Lieutenant to make entry into the side Alpha entrance was unsuccessful due to the extreme heat and smoke conditions.
The second due engine, E-10, arrived at 1823 with staffing of a Captain, Lieutenant, Driver/Operator, and a Firefighter. At 1823, E-10’s crew brought a 4″ supply line to E-11 from the hydrant at Deanwood Rd. and Dowling Circle and assisted the first-in crew with fire attack.
-
The Captain from E-10 conferred with Command and was instructed to advance a second 1 ¾” hand line.
-
The window to the first floor right apartment (Apt. T-2) was removed, and the second 1 ¾” line was advanced to the building by the crew of E-10.
-
Fire attack was initiated through the removed window. At 1827, Command requested a second alarm.
At this time, heat and smoke conditions just inside the front door improved enough to allow the Firefighter and Lieutenant from E-11 to make entry through the front door and into the stairwell. There they encountered heavy, thick black smoke and high heat conditions coming up the stairs from the terrace level apartment. The Lieutenant reported that the doorway to the first floor apartment was orange with fire and he had to fight his way through heavy heat and smoke conditions to attack the fire in the first floor right apartment (Apt. T-2). Entry was made approximately 3 feet into the doorway when the Firefighter’s low air alarm began to sound, and he exited the building. A member from E-10’s crew replaced the Firefighter from E-11 on the hose line.
At the same time, the Captain from E-11 proceeded to the rear of the structure to complete his initial 360 degree size up. He noted that there was fire emanating from the open sliding doors on the first floor Charlie/Delta apartment (Apt. T-2), extending to the balcony above. E-1, staffed by a Captain, Driver/Operator, and two Firefighters arrived and completed the hookup of the supply line that had been laid to the hydrant by E-10. The rest of Engine 1’s crew grabbed tools and an extension ladder and reported to the Charlie side of the building.
The Photo above referenced as Figure 015 shows conditions from rear of flames in apartment T2 and extension to the balcony above. Note the relative minimal volume of smoke as the sliding door serves as the inlet for ventilation into the apartment. The smoke and heat is flowing in from the rear, through the apartment and up the stairs.
This unidirectional flow path up the stairs is difficult to combat and is often experienced during basement fires as crews attempt to descend interior stairs.
- The model indicates sustained air temperatures in the stairwell of approximately 600 Fahrenheit (315 Celsius) at velocities of approximately 6 mph (2.7 m/s) from floor to ceiling as crews attempted to descend the stairs.
- This is consistent with statements from firefighting crews, who experienced extremely high heat conditions and indicated periodically seeing flames in the smoke layer flowing up the stairs.
- The elevated air velocity of the stairwell flow path resulted in a high rate of convective energy transfer to the structural firefighting gear and high perceived temperatures as the firefighters attempted to descend the stairs.
Firefighting crews flowed a hoseline down the stairs to combat the high temperatures; however no significant cooling was noticed by firefighters because the hose stream could not reach the seat of the fully developed fire in the kitchen area.
The crews were simply cooling the ventilation flow path without cooling the source of the energy in the apartment.
It was not until a hose stream was directed through an exterior window and a portion of the fire was extinguished that gas temperatures and velocities began to decrease, allowing firefighters to make entry to the terrace apartment via the stairs.
Plan view of flow path and temperatures within the apartment. Note the location of the seat of the fire and the location of initial hose stream application down the stairs.
Photograph of hoselines being positioned at the stairwell entrance door and front window. Note the heavy smoke venting from all front openings in apartment T2. (Figure 017)
The middle level apartment (A2) entrance door was opened by a second search crew around the same time as the second couch ignited, creating a ventilation flow path from the second floor balcony, through the apartment, and upwards into the stairwell (third floor). This flow path follows the same general route through the apartment and into the stairwell as was seen in the terrace level apartment below. Squad 303’s crew arrived on scene after the bulk of the fire in the terrace level apartment had been suppressed and appeared to be under control. The crew entered the front stairwell, which had minimal smoke up to the second level and the crew began to systematically search the building.
Squad 303’s crew proceeded to search two apartments before entering the third floor right side apartment to conduct a search, leaving the entrance door open. It should also be noted that carpeting impacted the bottom of the door and prevented the apartment entrance doors on the second and third levels from closing automatically. The entry doors had to be actively pushed closed to overcome the friction of the carpet.
Photo depicting building smoke and fire conditions around the arrival of Squad 303.
Note the lack of heavy smoke or fire in the stairwell or terrace level.
There is also no indication of the growing fire in the second (middle) level apartment.
When Squad 303’s crew of two firefighters entered the third level apartment (B2), smoke was banked about halfway down the walls with moderate visibility. The crew could clearly see the floor of the apartment without the need to crawl below the smoke layer to search. Squad 303’s crew was unaware of the flames spreading across the two couches in the second floor apartment below them. The crew split in order to search the apartment faster, with one firefighter searching the front bedrooms and the officer searching the kitchen and living room.
As flames in the second level began to rollover into the apartment entranceway, the smoke layer in the third level quickly dropped to the floor with a rapid increase in temperature. With Squad 303’s crew searching above, flames began to extend into the stairwell, supplied by sufficient ventilation flowing through the apartment. This combination of fuel, heat and oxygen rich fresh air resulted in a rapid increase in heat release rate and flashover of the second level apartment followed by full room involvement.
The open entrance doors on the second and third levels created a ventilation flow path through the second floor apartment, into the sealed stairwell and up through the third floor apartment directly above. The flames followed this flow path and extended from the second floor, through the stairwell and into the living room area of the third floor apartment. Flashover of the third floor occurred approximately 30 seconds after the second floor experienced flashover.
Command sounded the building evacuation tones as flames extended into the hallway and up to the third level apartment.
Two couches just inside the entrance door on the third level ignited, blocking the primary means of egress for both firefighters from Squad 303. Upon hearing the evacuation horns from the trucks, the second firefighter from Squad 303 (searching the front bedrooms) attempted to exit the apartment via the apartment entrance door, however he was blocked by flames in the living room and stairwell.
Trapped in the bedroom, the firefighter bailed out headfirst down a ground ladder on the front side from the third floor. Squad 303 officer’s means of egress through the apartment entrance door was also blocked by the flames in the living room and stairwell. There were no windows located in the rear of the apartment.
The only means of escape was the balcony slider, however the entire balcony was engulfed in flames from the fully involved apartment below. With both escape routes blocked by flames and experiencing extremely high heat conditions, Squad 303’s officer requested assistance and declared a MAYDAY from the rear of the third floor apartment.
Firefighters re-entered the structure to combat the fire and locate the trapped firefighter. The downed firefighter was eventually located on the third level just inside the sliding glass door and was removed to the rear balcony. The firefighter was then extricated in a stokes rescue basket down the aerial ladder of a truck located in the rear, where he was subsequently transported to the hospital.
Effects of Compartmentation on Fire Spread
The Post Incident Analysis Team requested that alternate modeling scenarios be conducted to explore the effects of compartmentation on fire spread throughout the building.
The team specifically wanted to know how the ventilation flow paths through the stairwell would differ if the second or third level apartment entry doors were shut after entering/leaving the apartments. Two alternate computer fire modeling scenarios were conducted.
The first alternative modeling run featured the exact same fire scenario, except the second (middle) level apartment door was closed after the last victim was removed from that apartment. The apartment entry doors from the stairwell were fire-rated doors constructed of solid wood.
- As soon as the door is shut, the ventilation flow path through the apartment and up the stairwell is blocked.
Even with the third floor apartment door left open, the model indicates that the stairwell and third floor remain tenable for firefighters. Flames eventually extend from the third floor balcony into the apartment, however the escape routes through the stairwell and the front apartment windows are accessible.
The model indicates that closing the second level apartment door prevents the flow of smoke, heat and other products of combustion from entering the stairwell, thus preventing flashover of the stairwell and the third level. As long as the second floor entry door remains shut, the model indicated that the conditions within the stairwell and third floor remain tenable for firefighters, even with the third floor apartment door open.
A second alternative modeling scenario was conducted where the third level entrance door was closed after crews made entry to search the apartment.The same fire conditions from the actual model were used.When the door remained closed, the outlet of the ventilation flow path was blocked at the top of the stairs. Without a complete flow path, there wasn’t sufficient oxygen flowing through the second floor apartment to support extended burning in the stairwell.
Consequently after flashover of the second floor, the flames in the stairwell only exist momentarily before consuming all available oxygen and becoming ventilation limited.The fire model indicated that temperatures within the third floor apartment stayed tenable for firefighters, even with a fully developed fire on the second floor and flames in the stairwell.
Flames would eventually extend up the rear balcony to the third level, however they would not block egress through the living room and front windows of the apartment.By closing the apartment door on the third floor and blocking the outlet for fire gases emanating from the second floor apartment, the third floor apartment remains tenable for firefighting crews and the temperatures only briefly spike in the stairwell before the fire becomes ventilation limited.The ventilation flow through the apartments results in an increased burning rate within both the second and third levels, as well as the stairwell.
Results of each modeling scenario describing extent of flame spread
By closing apartment unit entrance doors and interior hollow core doors, one can slow or even block the ventilation flow path through the structure, thus significantly reducing the rate of fire spread. The photos below represent the post-fire damage in all six apartments within the fire building. Four of the six apartment entry doors were open for the majority of the fire and the relative difference in damage is clearly evident.
Using doors to compartmentalize and limit fire and smoke spread in a structure is not limited to fire-rated entrance doors. Interior hollow core doors also offer considerable protection for compartmentation purposes.
A search crew utilizing the Vent, Enter and Search (VES) technique through a front window used a hollow core bedroom door to isolate themselves from the developing fire in the living room of apartment A2.
As the crews removed the second victim from the living room to the bedroom, they shut the bedroom hollow core door behind them.
The living room soon experienced flashover followed by full room involvement, however the bedroom remained isolated from the heat and smoke for the duration of the fire. The photos below illustrate this effective use of compartmentation to protect firefighters during a search.
While no fire model will exactly replicate a fire, this model provided insight on the route of fire spread, the rapid fire growth leading to flashover of the second and third level, and the benefits of compartmentation on slowing fire and smoke spread.
-
The unidirectional flow path up the stairs from the terrace level apartment resulted in a high rate of convective heat transfer to the firefighters initially attempting to descend the stairs, making attacking the seat of the fire very difficult.
-
The model then supported the fact that the main stairwell acted as an open channel for fire and smoke spread between the second and third levels, resulting in flashover of the third level in approximately 30 seconds after the second level.
-
This rapid fire growth leading to flashover is supported by photographs, witness statements and fireground audio.
-
The model was then utilized to explore the effects of compartmentation using apartment entrance doors.
-
The FDS model supported the scene observations and indicated that shutting the entrance doors blocked the flow of buoyancy driven fire gases through the structure, ultimately preventing fire extension to the third floor apartment via the stairwell.
-
The FDS model was utilized as part of the overall engineering analysis of this tragic fire and allowed for a better understanding of the events that led to the firefighter MAYDAY and subsequent Line of Duty Death.
-
The model was also used as an educational tool providing insight on potential methods of preventing similar tragedies in the future.
-
The results of this engineering analysis are intended to be reviewed by the Post Incident Analysis Team to assist in the creation of recommendations to mitigate the danger associated with future fire incidents.
References:
-
CommandSafety.com: Fire Modeling http://commandsafety.com/2011/11/fire-modeling-software/
-
Filed Under: Adaptive Fireground Management • Featured • Fire Dynamics & Behavior • Predictive Occupancy Performance • Probability of Adverse Consequences