Skip to main content

Using Bark Char Codes to Predict Post-fire Cambium Mortality

Abstract

Cambium injury is an important factor in post-fire tree survival. Measurements that quantify the degree of bark charring on tree stems after fire are often used as surrogates for direct cambium injury because they are relatively easy to assign and are non-destructive. However, bark char codes based on these measurements have been inadequately tested to determine how well they relate to live or dead cambium. Methods for assessing cambium injury through direct sampling have also been questioned as a potential factor for increasing tree mortality. In this study we used data collected from 11 wildfires and 6 prescribed fires in California, Idaho, Montana, and Wyoming to develop a relationship between bark char codes and cambium status for 14 coniferous species. Burned trees were assessed at groundline for bark char severity on each bole quadrant and then sampled at the center of each quadrant to determine cambium status (live or dead). We found that the moderate and deep bark char codes were strongly associated with dead cambium for thin-bark species: lodgepole pine (Pinus contorta), whitebark pine (P. albicaulis), western white pine (P. monticola), western redcedar (Thuja plicata), Engelmann spruce (Picea engelmannii), western hemlock (Tsuga heterophylla), and subalpine fir (Abies lasiocarpa). However, bark char codes were somewhat inaccurate in predicting cambium status of the thicker-bark species of white fir (Abies concolor), incense cedar (Calocedrus decurrens), ponderosa pine (Pinus ponderosa), Jeffrey pine (P. jeffreyi), Douglas-fir (Pseudotsuga menziesii), western larch (Larix occidentalis), and sugar pine (P. lambertiana). We also evaluated the effect of direct cambium sampling on ponderosa pine tree mortality in eastern Montana. Mortality rates were equivalent for eastern Montana ponderosa pines with and without cambium sampling. Our results support using bark char codes as surrogates for cambium sampling in tree species with thin bark, but bark char codes for thick-bark species, especially the moderate char code, are often not accurate fire-injury variables, as they do not correlate well with cambium status.

Introduction

A tree’s ability to survive after fire largely depends on the level of injury sustained to the crown, bole, and roots (Ryan 1982, Filip et al. 2007). The level of fire-caused crown injury is relatively easy to determine by direct examination, while injuries to the bole and roots are much more difficult to quantify without destructive sampling (Ryan 1982). As a result, surrogate measures have been developed to assess the degree of post-fire cambium injury and, to a lesser extent, root injury. However, relationships between these easily-observed surrogates and actual injury remain speculative (Fowler and Sieg 2004).

A tree’s susceptibility to cambium injury from heating is primarily influenced by heat duration and bark thickness (Ryan and Frandsen 1991), and secondarily by tree species morphology, due to varying bark thermal properties (van Mantgem and Schwartz 2003). The rate of cambium death also varies by species and season (Dickinson and Johnson 2004). The time required to cause cambium death from heat is proportional to the square of bark thickness (Martin 1963, Reifsnyder et al. 1967, Peterson and Ryan 1986, Rego and Rigolot 1990).

Various methods have been developed to estimate cambium injury from heating because direct sampling is invasive and time consuming, and because injury is difficult to predict from models (see Fowler and Sieg 2004 for a review). Jones et al. (2004) recently developed a stem heating model to predict cambium death from surface fires. However, the model requires duration of the heat pulse as an input, which is usually unknown. In addition, this model is still in the testing and developmental stages for many tree species. Other methods include estimating the percentage of the tree bole that was charred (Peterson and Arbaugh 1986, Beverly and Martell 2003, Keyser et al. 2006, Sieg et al. 2006), assessing the degree of charring around the tree bole (Ryan 1982, Harrington and Hawksworth 1990, McHugh and Kolb 2003, Thies et al. 2006), and assessing height of stem bark char (Regelbrugge and Conard 1993). Others have directly sampled a portion of the cambium (Mann and Gunter 1960, Hare 1965, Ryan et al. 1988, Peterson and Arbaugh 1989, Ryan 1990, Ryan and Steele 1990, Finney 1999, Ryan and Amman 1996, Hood and Bentz 2007, Hood et al. 2007), but this method still does not measure the total amount of cambium injury.

Ryan (1982) first developed categories (codes) of bark char severity to indicate stem injury resulting from fire, but he cautioned that these codes should only be used in conjunction with direct cambium sampling. However, these codes and others have since been used as a surrogate for stem injury in post-fire tree mortality models (Peterson and Arbaugh 1986, Peterson and Arbaugh 1989, Harrington and Hawksworth 1990, Regelbrugge and Conard 1993, Beverly and Martell 2003, Hély et al. 2003, McHugh and Kolb 2003, Kobziar et al. 2006, Sieg et al. 2006, Thies et al. 2006). While these studies reported that various measures of bark char were statistically significant variables for predicting delayed tree mortality, it is unknown how well bark char actually relates to cambium injury for many species (Fowler and Sieg 2004).

Few studies to date have tested the relationship between the Ryan (1982) codes and cambium status (Breece 2006, Hood and Bentz 2007, Hood et al. 2007). All reported an increase in cambium injury with increasing char severity, but noted that moderate charring is not clearly associated with either live or dead cambium. Our study is the first to examine the relationship of bark char to cambium injury for several species over a range of diameters at breast height (dbh) and fire types.

Bark char has been used as a surrogate for cambium injury sampling because it is readily observed and non-destructive (Beverly and Martell 2003, Thies et al. 2006). However, there is no empirical evidence that cambium sampling increases the probability of tree mortality. Van Mantgem and Stephenson (2004) found no difference in mortality between cored and uncored trees 12 years after they had been cored to determine age and growth rates. More injurious sampling of trees, however, may affect mortality rates. Heyerdahl and McKay (2001), for example, removed tree sections that averaged 8 % of the cross-sectional area and 7 cm thick to assess fire history and observed that mortality among the sectioned trees was low (8 %), but significantly higher than mortality for unsectioned trees.

In this study we used data collected from multiple wild and prescribed fires in the western US to: 1) assess the relationship between the Ryan (1982) bark char codes and cambium status (live/dead) for 14 conifer species over a range of tree sizes and fire types (prescribed fire and wildfire) and 2) evaluate the effect of direct cambium sampling on tree mortality rate. Combining fires allowed for a rigorous assessment of the validity of using bark char codes as a surrogate measure of cambium injury after fire. Species tested were lodgepole pine (Pinus contorta), whitebark pine (P. albicaulis), western white pine (P. monticola), western redcedar (Thuja plicata), Engelmann spruce (Picea engelmannii), western hemlock (Tsuga heterophylla), subalpine fir (Abies lasiocarpa), white fir (A. concolor), incense cedar (Calocedrus decurrens), ponderosa pine (P. ponderosa), Jeffrey pine (P. jeffreyi), Douglas-fir (Pseudotsuga menziesii), western larch (Larix occidentalis), and sugar pine (P. lambertiana).

Methods

Fire Descriptions

We collected fire-injury data from mixed-severity burned areas in 11 wildfires and 6 prescribed fires in California, Idaho, Montana, and Wyoming within one year of the fires (Table 1). All fires except the Power, Oops, Canyon Creek, Slowey, Air Patrol, Brewer, and Early Bird fires (Table 2) have been reported elsewhere and will not be described here.

Table 1 Trees sampled for each fire included in the data analysis. Additional fire information is described in the noted references.
Table 2 General location and physical features of sampled trees in fires that have not been previously described in literature.

The Power Fire began approximately 44 km east of Pioneer, California, in Amador County, and burned areas of private timberlands and the Eldorado National Forest, including a small portion of the Mokelumne Wilderness. It was primarily a moderate surface fire with some pockets of individual tree torching. Daytime temperatures at the nearby Beaver weather station were 24 °C to 25 °C with 10 hr fuel moisture between 6 % to 8 %.

The Canyon Creek Fire was ignited by lightning in Montana near the end of an unusually hot and dry fire season. The study site was near the North Fork of the Blackfoot River in an area burned by intense surface fire, essentially consuming all duff and completely scorching and killing Douglas-fir and lodgepole pine in the stand. Daytime temperatures at the nearby Lincoln ranger station were 31 °C to 34 °C with 10 hr fuel moisture between 3 % to 4 %.

The Early Bird, Air Patrol, and Brewer Fires in eastern Montana were ignited by lightning. Active surface fires burned the stands in the three study areas. The Early Bird Fire was east of Lame Deer, Montana, on the Northern Cheyenne Indian Reservation. Daytime temperatures were around 38 °C. The Air Patrol Fire was also on the Northern Cheyenne Indian Reservation approximately 16 km southeast of Lame Deer, Montana. Daytime temperatures were 30 °C to 36 °C with fine fuel moisture between 3 % to 5 %. The Brewer Fire began in the Long Pines Range of the Custer National Forest between Ekalaka, Montana, and Camp Crook, South Dakota. Daytime temperatures were between 37 °C to 44 °C accompanied by single digit fine fuel moisture.

The Slowey Fire was a prescribed burn set to reduce fuel hazards and restore wildlife habitat in ponderosa pine on the Lolo National Forest, 12 km northwest of Superior, Montana. A helicopter ignited the unit by using plastic spherical ignition devices. Conditions at the time of the burn were sunny, temperatures were 10 °C to 18 °C, relative humidity was 34 % to 50 %, and surface litter and upper duff moisture contents were 14 % and 12 %, respectively. Lower duff and mineral soil were moist. The resultant fire was an actively spreading surface fire with a light depth of burn (Ryan 2002).

The Oops Fire was a small wildfire that escaped from a nearby site preparation fire and burned a mixed conifer forest on the Priest River Experimental Forest in the Selkirk Mountains on the Idaho Panhandle National Forest, Idaho. The site was a mixed conifer forest dominated by western hemlock, with a heavy coarse woody debris component of rust-killed western white pine. The fire burned around 1.5 ha at night in mid-October 1983. No significant rain had fallen for a month, and fuels were very dry for that time of year. The fire consumed all of the duff and resulted in extensive basal injury. The burn was classified as an active surface, moderate depth of burn fire (Ryan 2002).

Field Sampling

We sampled randomly selected trees to represent a wide range of fire injuries and diameters (Table 3). We selected sample locations that burned under mixed-severity conditions, with either no plans for post-fire salvage or where only dead trees (i.e., no green needles) would be cut. Trees sampled from non-California fires were clustered into randomly chosen plots. For the California fires, individual fire-injured trees were selected over a range of fire severity levels in an attempt to fill a matrix of injury levels, size classes, and species. At each tree, we measured dbh at 1.4 m above the ground on the uphill side of the tree and recorded tree species. No distinction was made between ponderosa pine and Jeffrey pine for the McNally and Cone Fires due to similar morphological characteristics, so we will only refer to ponderosa pine hereafter. We visually divided the base of each tree into quadrants and assessed bark char and cambium condition in each quadrant. Quadrants for most fires were oriented with the slope, one quadrant being on the uphill side, one on the downhill side, and two on the cross-slope. In flat areas and in California fire sites, quadrants were oriented in the cardinal directions. We assigned each quadrant as unburned, light, moderate, or deep bark char (Ryan 1982) (see Table 4 for descriptions) based on its average level of charring near the groundline (non-California fires) or within 0.3 m (1 ft) of groundline (California fires). In the center of each quadrant, we directly sampled cambium at groundline by removing a small portion of the bark to reveal the cambium. We visually assessed each direct sample for cambium status as described in Ryan (1982) for all fires except those in eastern Montana. These samples were treated with a vital stain (Ryan 1982) to determine cambium status. Live cambium is light in color, moist, and pliable. Dead cambium is darker in color and either resinous or hardened (Ryan 1982).

Table 3 Tree characteristics of species sampled in study and included in data analysis. Species are arranged in ascending order of bark thickness using bark thickness equations in FFE-FVS.
Table 4 Bark char codes and description of bark appearance (adapted from Ryan 1982).

Ponderosa pine trees from the Air Patrol, Brewer, and Early Bird Fires in eastern Montana were part of a larger post-fire tree mortality study. We randomly selected approximately 20 % of the study trees for direct sampling to determine cambium status (307 of 1748 trees) by removing a sample of cambium at groundline from each quadrant using an increment borer. We did not sample cambium on the other trees. We also assessed percent crown volume scorched for each tree. We revisited all trees in this study annually for four years post-fire to assess tree mortality.

Data Analysis

We used logistic regression to evaluate the relationship between direct measurement of cambium status and external bole char ratings for each species. This included all visually assessed and stained cambium samples. Because four samples were taken per tree, we used estimates from generalized estimating equations to account for within tree correlation (Liang and Zeger 1986). We only included trees that were alive at the time of initial assessment in the analysis because we assumed cambium was dead for dead trees. We included tree dbh in the model to evaluate whether the predictive accuracy of the bole char codes varied with tree size. We evaluated fire type (wild or prescribed) for model inclusion for those species where sufficient data existed: lodgepole pine, Engelmann spruce, subalpine fir, ponderosa pine, and Douglas-fir. We retained all variables with p-values <0.05 in the full model. If dbh was significant in the model, we developed a second model without dbh in order to compare the added value of using dbh to predict cambium status from bark char. We performed all analyses in SAS (PROC GENMOD, SAS Institute, v. 9.1) using the following model form:

$${P_{DeadCambium}} = 1/[1 + \exp ( - ({B_0} + {B_1}{X_1} + \ldots {B_k}{X_k}))]$$

where P DeadCambium is the probability of dead cambium, B 0 , B 1 , and B k are regression coefficients, and X 1 and X k are independent variables (bark char code, dbh, and fire type).

All whitebark pine, western white pine, and western hemlock quadrants with deep bark char had dead cambium. Logistic regression models cannot converge when 100 % of categorical variable samples have the same response (Hosmer and Lemeshow 2000). Therefore, we did not include quadrants with deep char when developing the equations for whitebark pine, western white pine, and western hemlock.

All tables and figures list species in ascending order of bark thickness based on the bark thickness equations in the Fire and Fuels Extension to the Forest Vegetation Simulator (FFE-FVS) (Reinhardt and Crookston 2003). In this model, western white pine and western redcedar have the same bark thickness, as do ponderosa pine, Douglas-fir, and western larch. We did not measure bark thickness directly because this variable would be impractical for managers to measure due to the time required. Also, sampling occurred post-fire when a portion of bark on many trees had burned away and pre-fire bark thickness could not be determined.

We tested differences in four-year post-fire mortality, crown volume scorched (%), and dbh between ponderosa pines with and without cambium sampling from the Air Patrol, Brewer, and Early Bird Fires using a general linear mixed model (Littell et al. 1996). We included fire name and plot number within fire as random effects.

Results

For all species, the percentage of quadrants with dead cambium increased with increasing bark char severity (Figure 1), and bark char codes were significantly correlated with cambium status (Table 5). Over 80 % of the quadrants with moderate and deep bark char had dead cambium for species with thinner bark: lodgepole pine, whitebark pine, western white pine, western redcedar, Engelmann spruce, western hemlock, and subalpine fir (Figure 1a–g). Even light bark char indicated dead cambium for the majority of these thin-barked species. In contrast, cambium was dead for only 17 % to 54 % of the samples with moderate charring on species with thicker bark: white fir, incense cedar, ponderosa pine, Douglas-fir, western larch, and sugar pine (Figure 1h–m). Except for western larch, deep char on thick-barked species usually indicated dead cambium as well. Regression models also predicted a high probability (>0.8) of dead cambium when quadrants had moderate or deep bark char for lodgepole pine, whitebark pine, western white pine, western redcedar, Engelmann spruce, western hemlock, and subalpine fir—all species with relatively thin bark (Table 6).

Figure 1
figure1

Percent of bark char quadrants with dead cambium by bark char code. Species are arranged in ascending order of bark thickness using bark thickness equations in FFE-FVS.

Table 5 Full regression model coefficients by species to predict probability of dead cambium from bark char codes. Species are arranged in ascending order of bark thickness using bark thickness equations in FFE-FVS. N.S. indicates regression coefficient is not significant; all other coefficients are significant (p-value < 0.05). Empirical standard error estimates are in parentheses.
Table 6 Predicted probability of dead cambium by bark char code and fire type when dbh is excluded from the model. Species are arranged in ascending order of bark thickness using bark thickness equations in FFE-FVS.

A small percentage of unburned quadrants had dead cambium for all species. These quadrants either were dead before the fire, or a small portion of the quadrant may have been charred and that was where the cambium happened to be sampled, or we sampled uncharred bark directly adjacent to charred bark. The bark char codes were assigned based on the average level of charring in each quadrant and the cambium was only sampled at the center of each quadrant. Therefore, the codes did not always match where the cambium was sampled.

Diameter was a statistically significant variable for predicting the probability of dead cambium for all species except whitebark pine, Engelmann spruce, subalpine fir, and incense cedar (Table 5). For species where dbh was significant, the predicted probability of dead cambium decreased as dbh increased except for ponderosa pine, where the relationship was reversed (Figures 2 and 3). The predicted probability of dead cambium changed less than 20 % across the range of diameters for all species except western white pine, western hemlock, and western larch. The decrease in the predicted probability of dead cambium was particularly sharp for western larch as dbh increased, with little difference between bark codes for trees greater than 55 cm dbh (Figure 2e).

Figure 2
figure2

Predicted probability of dead cambium by bark char code for species where dbh is significant in the model. All western white pine and western hemlock quadrants with deep bark char were dead and could not be included in the regression analysis. Lodgepole pine, ponderosa pine, and Douglas-fir are shown in Figure 3 because fire type is significant in the model. Species are arranged in ascending order of bark thickness using bark thickness equations in FFE-FVS.

Figure 3
figure3

Predicted probability of dead cambium by bark char code and fire type for species where dbh and fire type are significant in the model. Species are arranged in ascending order of bark thickness using bark thickness equations in FFE-FVS.

Fire type (wild vs. prescribed) was significant in predicting lodgepole pine, subalpine fir, ponderosa pine, and Douglas-fir dead cambium (Table 5). It was not significant for Engelmann spruce. Predicted probability of dead cambium by bark char code was lower for prescribed fires than wildfires (Table 6, Figure 3). The predicted probability of dead ponderosa pine cambium from prescribed fires was low for all char codes and tree sizes (Figure 3).

There was no difference in the amount of ponderosa pine mortality four years post-fire between trees with cambium sampling and those that were not sampled for cambium injury (40 % trees died with cambium sampling compared to 39 % without cambium sampling, DF = 1746, P = 0.7667). The two groups did not have significantly different crown scorch (54 % scorch for both groups, DF = 1743, P = 0.8078) or dbh (29.2 cm for cambium sampled trees and 28.2 cm for trees without cambium sampling, DF = 1743, P = 0.0646).

Discussion

Many studies have found various measures of bark char to significantly predict delayed tree mortality after fire (see Fowler and Sieg 2004). This study shows that bark char severity is related to cambium death. The probability of dead cambium increased with increasing bark char severity for all species tested, but moderate char was not clearly associated with either live or dead cambium for thicker-bark species (Figure 1). Breece (2006) found an ambiguous relationship between Ryan’s (1982) bark char codes, actual cambium injury, and ponderosa pine tree mortality and reported that only 42 % of cambium samples with moderate char were dead. Breece et al. (2008) concluded that the codes should not be used for post-fire ponderosa pine mortality models. Moderate char was the most commonly assigned category for most of the species included in this study. Therefore, the use of moderate bark char as a surrogate for cambium status on thick-bark species could lead to large errors when estimating post-fire tree injury, thus limiting the utility of bark char codes.

Given the same dbh and bark char code, wildfires resulted in higher rates of dead cambium compared to those in prescribed fires for lodgepole pine, subalpine fir, ponderosa pine, and Douglas-fir. However, the actual differences in probabilities between wildfires and prescribed fires were only large for ponderosa pine (Figure 3). The reason for the higher proportions of dead cambium in wildfires is likely due to higher levels of fuel and duff consumption caused by burning under drier, more extreme weather conditions than those of prescribed fires. In addition, because duration of heating is longer in areas of deep duff combustion, greater charring and cambium injury would be expected (Ryan and Frandsen 1991). The Slowey, Danskin, and Lubrecht Fires—the prescribed fires from which ponderosa pine data were collected—were early season, low intensity surface fires with modest amounts of duff at the tree bases (K.C. Ryan and S.M. Hood, personal observation). While the pre-wildfire duff depths are unknown, differences in basal duff depths may be another reason why predicted probability of dead cambium was lower for prescribed fires than wildfires. In contrast, Ryan and Frandsen (1991) found that even in a relatively moist summer, extensive basal injury occurred in a late August prescribed burn which consumed deep duff mounds around large diameter, old ponderosa pine in Glacier National Park. Prescribed fires in areas with deep, dry duff mounds will likely result in cambium injury similar to that observed in our wildfires.

Although dbh was a significant explanatory variable for cambium mortality in many species, predicted probability of dead cambium changed little across the range of diameters observed except for western hemlock and western larch. This suggests that, for management purposes, the models that exclude dbh are probably sufficient for all examined species except western larch and western hemlock (Table 6). Keyser et al. (2006) reported that bark char measurements greatly improved prediction of ponderosa pine tree mortality for trees smaller than 40 cm dbh, but had a reduced value for predicting mortality of trees larger than 40 cm dbh. Bark thickness generally increases with increasing dbh for a given species. Therefore, equivalent bark char ratings on smaller diameter, and thus thinner-barked, trees should be associated with higher levels of cambium injury than in larger diameter, thicker-barked trees.

The ponderosa pines in our study were much larger than those in Keyser et al. (2006), and ponderosa pine was the only species in which predicted mortality of cambium increased as dbh increased. Ryan and Frandsen (1991) found a similar relationship for ponderosa pines in a prescribed fire in northwestern Montana. Other studies have found higher post-fire mortality rates for larger and smaller diameter trees than mid-diameter trees (McHugh and Kolb 2003, Hood et al. 2007) and attributed the relationship to higher levels of cambium injury resulting from consumption of deep duff mounds, which are common at the base of larger diameter trees. Complete consumption of deep duff at the bases of large diameter trees may be the reason for increased cambium injury in these large trees (Ryan and Frandsen 1991, Swezy and Agee 1991, Kaufmann and Covington 2001).

Sample size varied greatly between species (Table 3), and it is difficult to control for these differences (Rudy King, USDA Forest Service, personal communication). Models developed from small sample sizes may have less precision than those developed from larger samples, and it may only be possible to detect strong associations between the independent and dependent variables with small samples (Whittemore 1981). The significant relationship of bark char codes to cambium status indicate that it is a strong association for all species tested here. The differences in significance for dbh and fire type among species may be partially explained by the variation in sample sizes and indicate that these variables have a weaker association with cambium status than bark char codes.

Conclusions

Bark char codes are a good indicator of cambium status for the thin-bark species included in this analysis and can be used to save sampling time for many species when determining post-fire tree injury (Table 7). However, moderate bark char was not strongly associated with dead cambium for thicker-bark species. For these species (white fir, incense cedar, ponderosa pine, Douglas-fir, and sugar pine), cambium should be sampled directly to determine injury when bark char is moderate. Western larch cambium is most likely alive regardless of the degree of bark charring.

Table 7 Recommended management guidelines for using Ryan (1982) bark char codes as a surrogate for direct cambium sampling after fire. Species/code combinations not listed are not clearly associated with either live or dead cambium and should be sampled directly to determine injury.

It may be possible to improve the accuracy of predicting cambium status by initially comparing bark char codes to sampled cambium and developing an understanding of the association between bark char and the underlying cambium condition. Splitting the moderate char rating into two codes may also improve accuracy. Future research to test these ideas for improving the accuracy of bark char codes is needed.

Results from this study show that tree injury from direct sampling of the cambium using an increment borer does not contribute to additional post-fire ponderosa pine tree mortality. We expect a similar response from other tree species or from sampling with a hatchet due to similar wound response mechanisms among species and wound sizes (Shigo and Marx 1977). This research indicates that direct cambium sampling is a better variable than bark char codes for use in post-fire tree injury assessments for species with thick bark and that direct sampling can be performed without causing additional tree mortality.

Literature Cited

  1. Beverly, J.L., and D.L. Martell. 2003. Modeling Pinus strobus mortality following prescribed fire in Quetico Provincial Park, northwestern Ontario. Canadian Journal of Forest Research 33: 740–751.

    Article  Google Scholar 

  2. Breece, C.R. 2006. Effects of prescribed fire on bark beetle activity and tree mortality in southwestern ponderosa pine forests. Thesis, Northern Arizona University, Flagstaff, USA.

    Google Scholar 

  3. Breece, C.R., T.E. Kolb, B.G. Dickson, J.D. McMillin, and K.M. Clancy. 2008. Prescribed fire effects on bark beetle activity and tree mortality in southwestern ponderosa pine forests. Forest Ecology and Management 255: 119–128.

    Article  Google Scholar 

  4. Dickinson, M.B., and E.A. Johnson. 2004. Temperature-dependent rate models of vascular cambium cell mortality. Canadian Journal of Forest Research 34: 546–559.

    Article  Google Scholar 

  5. Dumm, G. 2003. Fire effects on fine roots and ectomycorrhizae of ponderosa pine and Douglas-fir following a prescribed burn in a central Idaho forest. Thesis, University of Idaho, Moscow, USA.

    Google Scholar 

  6. Eyre, F.H. 1980. Forest cover types of the United States and Canada. Society of American Foresters, Washington DC, USA.

    Google Scholar 

  7. Filip, G.M., C.L. Schmitt, D.W. Scott, and S.A. Fitzgerald. 2007. Understanding and defining mortality in western conifer forests. Western Journal of Applied Forestry 22: 105–115.

    Google Scholar 

  8. Finney, M. 1999. Fire-related mortality of ponderosa pine in eastern Montana. Unpublished Report INT-93800-RJVA on file at USDA Forest Service, RMRS Fire Sciences Laboratory, Missoula, Montana, USA.

    Google Scholar 

  9. Fowler, J.F., and C.H. Sieg. 2004. Postfire mortality of ponderosa pine and Douglas-fir: a review of methods to predict tree death. USDA Forest Service General Technical Report RMRS-GTR-132.

  10. Gundale, M.J., T.H. DeLuca, C.E. Fiedler, P.W. Ramsey, M.G. Harrington, and J.E. Gannon. 2005. Restoration treatments in a Montana ponderosa pine forest: effects on soil physical, chemical and biological properties. Forest Ecology and Management 213: 25–38.

    Article  Google Scholar 

  11. Hardy, C., H.Y. Smith, and W. McCaughey. 2006. The use of silviculture and prescribed fire to manage stand structure and fuel profiles in a multi-aged lodgepole pine forest. Pages 451–464 in: P.L. Andrews, and B. Butler, editors. Fuels Management—How to Measure Success: Conference Proceedings. USDA Forest Service Proceedings RMRS-P-41.

  12. Hare, R.C. 1965. Chemical test for fire damage. Journal of Forestry 63: 939.

    Google Scholar 

  13. Harrington, M.G., and F.G. Hawksworth. 1990. Interactions of fire and dwarf mistletoe on mortality of southwestern ponderosa pine. Pages 234–240 in: J. S. Krammes, editor. Effects of fire management of southwestern natural resources: symposium proceedings. USDA Forest Service General Technical Report RM-191.

  14. Hély, C., M. Flannigan, and Y. Bergeron. 2003. Modeling tree mortality following wildfire in the southeastern Canadian mixed-wood boreal forest. Forest Science 49: 566–576.

    Google Scholar 

  15. Heyerdahl, E., and S.J. McKay. 2001. Condition of live fire-scarred ponderosa pine trees six years after removing partial cross sections. Tree-Ring Research 57: 131–139.

    Google Scholar 

  16. Hood, S.M., and B. Bentz. 2007. Predicting post-fire Douglas-fir beetle attacks and tree mortality in the Northern Rocky Mountains. Canadian Journal of Forest Research 37: 1058–1069.

    Article  Google Scholar 

  17. Hood, S.M., S.L. Smith, and D.R. Cluck. 2007. Delayed tree mortality following fire in northern California. Pages 261–283 in: R.F. Powers, editor. Restoring fire-adapted ecosystems: Proceedings of the 2005 National Silviculture Workshop. USDA Forest Service General Technical Report PSW-GTR-203.

  18. Hosmer, D.W., and S. Lemeshow. 2000. Applied logistic regression, 2nd edition. John Wiley and Sons, New York, New York, USA.

    Book  Google Scholar 

  19. Jones, J.L., B.W. Webb, D.M. Jimenez, J. Reardon, and B. Butler. 2004. Development of an advanced one-dimensional stem heating model for application in surface fires. Canadian Journal of Forest Research 34: 20–30.

    Article  Google Scholar 

  20. Kaufmann, G.A., and W.W. Covington. 2001. Effect of prescribed burning on mortality of presettlement ponderosa pines in Grand Canyon National Park. Pages 36–42 in: R.K. Vance, C.B. Edminster, W.W. Covington, and J.A. Blake, editors. Ponderosa pine ecosystems restoration and conservation: steps toward stewardship: conference proceedings. USDA Forest Service Proceedings RMRS-P-22.

  21. Keyser, T.L., F.W. Smith, L.B. Lentile, and W.D. Shepperd. 2006. Modeling postfire mortality of ponderosa pine following a mixed-severity wildfire in the Black Hills: the role of tree morphology and direct fire effects. Forest Science 52: 530–539.

    Google Scholar 

  22. Kobziar, L., J.J. Moghaddas, and S. Stephens. 2006. Tree mortality patterns following prescribed fire in a mixed conifer forest. Canadian Journal of Forest Research 36: 3222–3238.

    Article  Google Scholar 

  23. Liang, K.Y., and S.L. Zeger. 1986. Longitudinal data analysis using generalized linear models. Biometrika 73: 13–22.

    Article  Google Scholar 

  24. Littell, R.C., G.A. Milliken, W.W. Stroup, and R.D. Wolfinger. 1996. SAS system for mixed models. SAS Institute, Cary, North Carolina, USA.

    Google Scholar 

  25. Mann, W.F., and E.R. Gunter. 1960. Predicting the fate of fire-damaged pines. Forests and People 10: 26–43.

    Google Scholar 

  26. Martin, R.E. 1963. Thermal properties of bark. Forest Products Journal 3: 419–426.

    Google Scholar 

  27. McHugh, C., and T.E. Kolb. 2003. Ponderosa pine mortality following fire in northern Arizona. International Journal of Wildland Fire 12: 7–22.

    Article  Google Scholar 

  28. Peterson, D.L., and M.J. Arbaugh. 1986. Postfire survival in Douglas-fir and lodgepole pine: comparing the effects of crown and bole damage. Canadian Journal of Forest Research 19: 1175–1179.

    Article  Google Scholar 

  29. Peterson, D.L., and M.J. Arbaugh. 1989. Estimating postfire survival of Douglas-fir in the Cascade Range. Canadian Journal of Forest Research 19: 530–533.

    Article  Google Scholar 

  30. Peterson, D.L., and K.C. Ryan. 1986. Modeling postfire conifer mortality for long-range planning. Environmental Management. 10: 797–808.

    Article  Google Scholar 

  31. Regelbrugge, J.C., and S.G. Conard. 1993. Modeling tree mortality following wildfire in Pinus ponderosa forests in the central Sierra Nevada of California. International Journal of Wildland Fire 3: 139–143.

    Article  Google Scholar 

  32. Rego, F., and E. Rigolot. 1990. Heat transfer through bark—a simple predictive model. Pages 157–161 in: J.G. Goldammer, and M.J. Jenkins, editors. Fire in Ecosystem Dynamics. Proceedings Third International Symposium on Fire Ecology. May 16–20, 1989, Freiburg, Federal Republic of Germany.

  33. Reifsnyder, W.E., L.P. Herrington, and K.W. Spalt. 1967. Thermophysical properties of bark of shortleaf, longleaf, and red pine. Yale University School of Forestry Bulletin 70: 1–40.

    Google Scholar 

  34. Reinhardt, E., and N. Crookston. 2003. The fire and fuels extension to the forest vegetation simulator. USDA Forest Service, Rocky Mountain Research Station General Technical Report RMRS-GTR-116.

  35. Ryan, K.C. 1982. Techniques for assessing fire damage to trees. Pages 1–11 in: J. Lotan, editor. Proceedings of the symposium: fire, its field effects. Intermountain Fire Council.

  36. Ryan, K.C. 1990. Predicting prescribed fire effects on trees in the interior West. Pages 148–162 in: M.E. Alexander and G.F. Bisgrove, editors. The art and science of fire management: proceedings of the First Interior West Fire Council Annual Meeting and Workshop. Forestry Canada Northwest Region Northern Forestry Centre Information report (Northern Forestry Centre (Canada)) NOR-X-309.

    Google Scholar 

  37. Ryan, K.C. 2002. Dynamic interactions between forest structure and fire behavior in boreal ecosystems. Silva Fennica 36: 13–39.

    Article  Google Scholar 

  38. Ryan, K.C., and G.D. Amman. 1996. Bark beetle activity and delayed tree mortality in the Greater Yellowstone Area following the 1988 fires. Pages 151–158 in: J. Greenlee, editor. The ecological implications of fire in Greater Yellowstone: proceedings of the second biennial conference on the Greater Yellowstone Ecosystem. International Association of Wildland Fire.

  39. Ryan, K.C., and W.H. Frandsen. 1991. Basal injury from smoldering fires in mature Pinus ponderosa Laws. International Journal of Wildland Fire 1: 107–118.

    Article  Google Scholar 

  40. Ryan, K.C., D.L. Peterson, and E.D. Reinhardt. 1988. Modeling long-term fire-caused mortality of Douglas-fir. Forest Science. 34: 190–199.

    Google Scholar 

  41. Ryan, K.C., and B.M. Steele. 1990. Cambium mortality resulting from broadcast burning in mixed conifer shelterwoods. Pages 108–116 in: D.C. MacIver, H. Auld, and R. Whitewood, editors. Proceedings of the 10th Conference on Fire and Forest Meteorology. Forestry Canada.

  42. Shigo, A.L., and H.G. Marx. 1977. CODIT (Compartmentalization of Decay In Trees). USDA Forest Service Bulletin No. 405.

  43. Sieg, C.H., J.D. McMillin, J.F. Fowler, K.K. Allen, J.F. Negron, L.L. Wadleigh, J.A. Anhold, and K.E. Gibson. 2006. Best predictors for postfire mortality of ponderosa pine trees in the Intermountain West. Forest Science 52: 718–728.

    Google Scholar 

  44. Swezy, D.M., and J.K. Agee. 1991. Prescribed-fire effects on fine-root and tree mortality in old-growth ponderosa pine. Canadian Journal of Forest Research 21: 626–634.

    Article  Google Scholar 

  45. Thies, W.G., D.J. Westlind, M. Loewen, and G. Brenner. 2006. Prediction of delayed mortality of fire-damaged ponderosa pine following prescribed fires in eastern Oregon, USA. International Journal of Wildland Fire 15: 19–29.

    Article  Google Scholar 

  46. van Mantgem, P., and M. Schwartz. 2003. Bark heat resistance of small trees in Californian mixed conifer forests: testing some model assumptions. Forest Ecology and Management 178: 341–352.

    Article  Google Scholar 

  47. van Mantgem, P.J., and N.L. Stephenson. 2004. Does coring contribute to tree mortality? Canadian Journal of Forest Research 34: 2394–2398.

    Article  Google Scholar 

  48. Whittemore, A.S. 1981. Sample size for logistic regression with small response probability. Journal of the American Statistical Association 76: 27–32.

    Article  Google Scholar 

Download references

Acknowledgements

We acknowledge funding from the Joint Fire Science Program under Project JFSP 05-2-1-105. The USDA Forest Service, Region 5, and Forest Health Protection funded data collection on the California Fires, with cooperation from personnel on the Eldorado, Lassen, and Sequoia National Forests. JFSP funded the Danskin (01B-3-2-01) and Lubrecht projects (99-S-01). USDA Forest Service, Forest Health Protection Region 1, Beaverhead-Deerlodge National Forest (0102-01010), and the Special Technology Development Program (R4-2004-02) funded the research on the Moose, Mussigbrod, and Green Knoll Fires. L. Bradshaw provided weather data for the Canyon Creek, eastern Montana, and Power Fires. The comments of R. Keane, M. Harrington, C. McHugh, and two anonymous reviewers greatly improved this manuscript.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Sharon M. Hood.

Rights and permissions

Open Access  This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.

The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

To view a copy of this licence, visit https://creativecommons.org/licenses/by/4.0/.

Reprints and Permissions

About this article

Cite this article

Hood, S.M., Cluck, D.R., Smith, S.L. et al. Using Bark Char Codes to Predict Post-fire Cambium Mortality. fire ecol 4, 57–73 (2008). https://doi.org/10.4996/fireecology.0401057

Download citation

Keywords

  • bark char
  • basal injury
  • cambium
  • tree mortality