Linear Disturbance Project

Test Burns at Fort McMurray and Slave Lake in Mowed and Unmown Grass Plots

Background

FERIC’s Wildland Fire Operations Research Group was tasked with investigating techniques to reduce fire behaviour potential along linear disturbances in Alberta. FERIC is approaching this as a fine fuels problem (particularly grass) along these right-of-ways. Research has been conducted in grass fire behaviour, but little of this has concentrated on the differences in fire behaviour between areas that have been mowed to reduce the hazard and natural, standing grass. FERIC believed this is where this project should begin as industry currently utilises mowing to reduce fire behaviour along their right-of ways. Changing the fuel structure by mowing is thought to have three advantages to fire behaviour potential. First, it reduces fire spread rates; secondly, it increases how quickly green-up occurs in the spring; and, thirdly, mowed areas should have higher moisture contents and thus burn less readily.

Earlier this year, a request was sent to Alberta Sustainable Resource Development and Industry looking for locations where experimental fires could take place. From this request, two locations were found and to date four sets of experimental burns have taken place. Each set of burns included the burning of a mowed plot and a natural, standing plot of grass at the same time. This brief report will summarize the results of these four burns. Data was collected on headfire rate of spread, moisture contents, degree of curing, fuel loading and weather observations at the time of the burn. Video footage and still photographs of the burn were also taken for documentation purposes.

April 26, 2004 Fort McMurray

FERIC located and established two sets of plots in the Waterways Region within the city for burning. Two plots were mowed in early April by Gary Dakin and used as the mowed plots. Ideally, it would have been better to mow the previous fall and burn in the spring, but the project had not yet begun. Snow did fall on the mowed plots to partially compress the grass. The following is the data collected prior to, during and following the burn. SRD was responsible for igniting and controlling the fire.

Figure 1. Mowed plot.

Figure 2. Standing grass.

Plot One (burned at 1453)

Weather Data (1453 observations)

Temperature                 21.4 C

RH %                           25%

Wind                            5 G 8 kmh – variable directions

Sky                              light overcast

Plot Two (burned at 1544)

Weather Data (1544 observations)

Temperature                 21.6 C

RH%                            30.5

Wind                            5 kmh

Sky                              90% light overcast

Observations:

Flamelengths were higher in the control plots (figures 3 and 4). Ranging from 1.5 – 2.5 m in the control plots to 0.5 – 1.0 m in the mowed plots. When the wind increased (gusted) both plots exhibited rates of spread that were very similar. During both fires the direction of the wind changed, altering the direction of burn for periods of time.

Figure 3. Fire behaviour in standing grass. Flamelengths are approximately 1.5 m.

Figure 4. Fire behaviour in mowed grass. Flamelengths are approximately 0.5 m.

May 18, 2004 Slave Lake (ATCO Powerline)

ATCO assisted FERIC while mowing their powerlines by leaving two unmowed sections so that we could carry out another two sets of burns along linear disturbances. The plots were mowed on April 13 by ATCO and sat for one month. Again, snow fell on the sites to help compress the mowed grass. SRD was responsible for igniting and controlling the fire. Thanks to Laurie Lake Contracting Type II crew from Wabasca and the Slave Lake Hac crew for their assistance on the burns.

Figure 5. Mowed grass along ATCO powerline near Slave Lake.

Figure 6. Standing grass plot along ATCO powerline near Slave Lake.

Plot One (burned at 1352)

Weather Data (1347 observations)

Temperature                 24.9 C

RH%                            16.8

Wind                            W 12 G 16 kmh

Sky                              clear, distant cloud

Plot Two (burned at 1233)

Weather Data (1225 observations)

Temperature                 24.0 C

RH%                            23.6

Wind                            W 12 G 15 kmh

Sky                              clear, distant cloud, moving closer

The mowed plot in Plot One was wet. Water was just below the surface in this location. This acted to influence the rate of spread of this plot and as a result this fire creeped along. The control plot (standing grass) raced off at 22 m/min to the 25 m marker, but a wind shift dramatically slowed its rate of spread. In both plots the degree of curing was lower (that is - greener) in the mowed plots indicating that mowing may increase the speed at which these sites green up in the spring. This also acted to produce rates of spread that were slower than those measured in the standing grass.

Figure 7. Fire behaviour in mowed grass.

Figure 8. Fire behaviour in standing grass. Note: the flamelengths are very similar to those from a mowed plot (Figure 7).

FBP Comparison

Weather data, fuel loads and degree of curing values were input into the FBP program for comparison purposes. Fuel moisture values were also used to calculate the FFMC value for the program (the general ‘rule of thumb’ equation:

FFMC = 101 – MC

was used to estimate the FFMC).

For the Slave Lake burns the FBP model calculated the following output:

Mowed Grass (O-1a FBP Fuel Type)

Head Rate of Fire Spread:                    7.3 m/min (observed was 11.2 m/min)

Intensity (kW/m)                                   176.4 kW/m

Standing Grass (O-1b FBP Fuel Type)

Head Rate of Fire Spread:                    10.7 m/min (observed was 13.3 m/min)

Intensity (kW/m)                                   257.0 kW/m

FBP model output is relatively close to what was documented during the fires. Differences between the model and actual could result from a wind that calmed during the fire but in the model was constant throughout the simulation.

A potential added value of this project is the collection of fire behaviour data to verify the grass models within the FBP model, as this has been of low priority in their modelling program.

Summary

In three of the four sets of plots burned in this research, the mowed plots averaged rates of spread that were 82% of those in the natural, standing grass over the length of the plots (50 m). This was higher than expected and as stated earlier when wind gusts influenced the plots, observed rates of spread were almost identical. This is an important finding for those thinking rates of spread would be significantly slower in mowed areas. On the other hand, for suppression purposes, flamelengths were observed to be significantly higher in the standing grass making the potential control of these fires more difficult. Flamelengths were less than one metre in all the mowed plots; this would allow direct attack to be taken these fires.

Data from the Slave Lake burns suggest that mowing does have an influence on how quickly the area greens up. Differences of 7% were collected in the two sets of plots. This would act to shorten the spring fire season by about a week.

The limited number of moisture samples collected does not allow a statement to be made on whether there is a difference between standing grass and mowed grass. The few samples collected varied from the mowed grass been ‘wetter’ to the standing grass been ‘wetter’. Site characteristics would have an influence on moisture contents.

Figure 9. Laurie Lake Contracting Type II crew from Wabasca.