Masticated Fuels Research

Fire behavior and above-ground fire effects

Masticated wood is typically very fragmented with high surface area: volume ratio, which would tend to increase fire-line intensity and reduce the duration of combustion. On the other hand, masticated fuel beds are also compact, which would tend to reduce fire-line intensity and increase the duration of combustion. When we initiated the study, there were some concerns that fires might have long residence times, causing greater than desired soil heating and the potential for cambium and root damage to standing trees.

Among our objectives was to conduct prescribed burns and estimate fire behavior in order to calibrate existing fuel models or develop custom fuel models for predicting fire behavior and fire effects in masticated fuel beds. We also wanted to better understand mechanisms of tree mortality when masticated fuel around them was burned.

Prescribed burns were conducted at two sites (Challenge Experimental Forest – May/June 2005, Whitmore – June 2006) in four replicate one-acre units per site. Prior to the burns, fuels around a randomly selected subset of trees were removed from around the base in order to investigate if mortality was due mostly to bole charring, root damage, or crown damage. Ignition was accomplished using strip head fires or backing fire when fire-line intensity became too great. Flame length and rate of spread were estimated during the burns and fire damage to individual trees (crown scorch height, percentage of crown volume scorched, bark char height etc.) measured after the burns.

Key findings:

• Under prescribed fire conditions, flame lengths were low to moderate (approx. 1 ft for backing fire, 2-3 ft for head fire) and rate of spread slow (0.2 chains/hr for backing fire, 1.3-2.8 chains/hr for head fire).

• Rate of spread and flame length were well predicted with the BehavePlus fire model, using either the sb1 or sb2 (low load activity fuel and moderate load activity fuel) models. However, scorch height was substantial and under-predicted by BehavePlus. At the high fuel loading site (Challenge), scorch height was approximately 4 times model predictions, while at the low fuel loading site (Whitmore), scorch height was approximately 2 times model predictions. We therefore created custom fuel models (below) that produced outputs more in line with field observations of both fire behavior and effects.

  Custom fuel model inputs for three different levels of masticated loading based on behavior of prescribed burns at two sites. Use the moderate load activity fuel model (sb2) of Scott and Burgan (2005), with the following modifications:

  	Masticated- Low load*	Masticated- Moderate load	Masticated- High load
  1-hr (tons ac-1)	3.5	5.7	7.8
  10-hr (tons ac-1)	2.5	5.9	13.1
  100-hr (tons ac-1)	0.3	1.2	5.8
  S:V ratio (ft2 ft-3)	750	750	750
  Fuelbed depth (ft)	0.36	0.52	0.89

  *Low load is similar to the “Shasta” and “Whitmore” sites in the fuel loading photo series; the “Iron Mountain”, “Sierraville”, and “Whiskeytown” sites depict a moderate load, whereas the “Taylor Ridge”. “Challenge”, “Applegate”, “Stanislaus” and “Mad River” sites show a high load.

• Tree mortality from prescribed burns appeared to be primarily the result of crown scorch. High soil and duff moisture at the time of the burning limited heat penetration into the soil. This suggests that prescribed burning can be successfully used to reduce masticated fuels without killing residual trees if fire-line intensity is kept low using conservative firing techniques. Greater use of backing fire, narrowing of distance between strips with strip headfires, or burning when air temperature is low might be options, especially if the residual trees are relatively small. Burning when soil moisture is high may also help reduce damage to trees.

Below-ground fire effects: soil heating

Temperatures as high as 212-572°F have been measured in upper soil horizons under burning masticated fuel beds (Busse et al 2005), when only 140°F are enough to kill plant roots. In order to develop a planning tool for managers interested in reducing masticated fuels using prescribed fire without causing adverse soil impacts, we conducted a study of heat penetration at varying soil moisture levels and in different soil types.

Twelve inch wide and 6-inch-deep soil cores were collected by pushing a PVC mould into the ground. This enabled us to collect the intact soil, without cracking and other changes to soil structure. Water was added to these soil cores, and then allowed to dry so that the soils were at: field capacity, 2/3 field capacity, 1/3 field capacity, and summer dry (approximately 40%, 30%, 20%, and 10% volumetric soil moisture). Once at the appropriate moisture level, the soil core was wrapped in plastic to prevent additional drying. When all moisture levels were ready, cores were placed in pits in an open field. Thermocouples were installed at the soil surface, and at 1, 2, 4, and 6 inch depths in the soil.

The pit was filled with soil around the core, and a masticated fuel bed, approximating some of the highest fuel loadings witnessed in the field, was created over the top. A final thermocouple was placed on top of the fuel bed. Fuels were ignited in backing fires and thermocouples removed approximately 24 hours later. Four replicate burns were done for each fuel moisture and soil combination. Three different soil types were investigated: a clay, a loam, and a pumice-sand.

Results of the controlled laboratory burning experiment were validated in the field by burying thermocouples at the same depths in the soil immediately prior to the prescribed burns at the Challenge and Whitmore sites. A total of 62 locations were instrumented.

Key findings:

• Heat penetration into the soil is highly dependent on soil moisture, with the greatest heat penetration when soils were driest. Heat penetration was greatly reduced with soil moisture at or exceeding 20%.
• There was no significant effect of soil type. Heat penetration was very similar for all three soil types.
• Results suggest that most roots and soil organisms will be little affected by prescribed burning of masticated fuels when soils are moist. Soil heating may be considerable with summer wildfires or prescribed burns conducted when soil moisture is low, especially if the loading of masticated residues is high.

Fuel loading in masticated areas

While small trees and shrubs can be chipped and sold to generate energy, market conditions for biomass are volatile and removal of this biomass is often not cost effective. Masticated biomass is therefore usually left on the forest floor to protect the soil from erosion and to retain nutrients. The biomass contained within the shrubs, hardwoods, and small conifers can be considerable. Masticated wood is often highly fractured and fragmented, with a high surface area: volume ratio. Size and shape of the individual pieces depends on the specific machinery used and amount of mastication effort (i.e. the amount of time spent per unit area).

We sampled fuels at 10 masticated sites from SW Oregon to the central Sierra Nevada in California, using both a plot-based method and Brown's planar intercept method. Final fuel loading values for fine woody fuels, litter, and duff were from plot data and fuel loading values for large woody fuels were from transect data.

Key findings:

• The surface fuel load at some sites was considerable. Most of the masticated wood at all sites fell within the 10 hr (¼ -1 in. diameter) size category.
• Fuel bed is quite compact, which is expected to moderate fire behavior. Compactness is related to operator effort (time spent per unit area) and the mastication machinery used, with drum-type cutting heads yielding finer particle size and more compact fuel bed than rotary cutting heads. Because of the variability in particle size, fuel loading is not well predicted by fuel bed depth.
• We recommend using a hybrid methodology for evaluating loading in masticated fuelbeds, where 1-h and 10-h fuels are estimated with a plot-based method, and larger fuels estimated with the standard Brown's transect method. If a high level of accuracy is not necessary, fuel loading can also be roughly estimated using a fuel loading photo series (below).

Treatment longevity: shrub and understory response

Among questions that managers frequently have about mastication treatments are: 

1. How long will the fire hazard reduction benefits last?
2. What is the effect of masticated wood covering the forest floor on understory plant species?

In order to address both questions, we recorded shrub species and shrub cover at each of the ten masticated fuels sites, along with the fuel loading data. Understory species were evaluated in a larger scale sampling at one of the sites with replicated treatment units (Challenge).

Key findings:

• Regrowth of shrubs after mastication depends strongly upon the growth habits of the shrubs. Resprouting species such as snowbrush grow back rapidly, while seeding species such as white-leaved manzanita, take longer to recolonize an area.
• Masticated wood on the soil surface did not suppress native species. In fact, significantly more species were found in masticated plots than in unmasticated control plots. Removing the masticated wood with a prescribed burn resulted in even more native species. Mastication also led to an increase in the density of non-native weedy species, but numbers were low relative to native species.
• Mastication did not depress shrub seedlings – density of seedlings in masticated units was not significantly different from the unmasticated control. The density of shrub seedlings did increase significantly when mastication was followed with prescribed burning.