Relict Ponderosa Pine in the Bob Marshall Wilderness

WILDERNESS MANAGEMENT The Complexity of Managing Fire-dependent Ecosystems in Wilderness: Relict Ponderosa Pine in the Bob Marshall Wilderness

by Robert E. Keane, Stephen Arno and Laura J. Dickinson

n recent years, some wildland fires have with mixed-severity or stand-replacement Ibeen allowed to burn to restore ecolog- fire regimes (Arno and others 2000). ical disturbance processes in large wilder- During the 1980s, some ecologists ness areas and national parks in the observed the successional replacement of northern Rocky Mountains of the United ponderosa pine by inland Douglas-fir States (Arno and Fiedler 2005). Prior to (Pseudotsuga menziesii var. glauca) and the application of fire suppression strate- other conifers, and an unprecedented gies in the early twentieth century, most buildup of litter, duff, and ladder fuels ABSTRACT of these lands had a history of mixed- that were endangering these isolated pon- severity fires or stand replacement fires derosa pine communities (Wendel Hann, Isolated wilderness ecosystems with a history of frequent, low-severity fires have been occurring at intervals averaging between personal communication). altered due to many decades of fire exclusion 50 and 300 years (Arno 2000). Programs We began studying various ecological and, as a result, are difficult to restore for for restoring fire in wilderness areas and features of the South Fork ponderosa pine philosophical and logistical reasons. In this national parks have permitted some man- communities during the 1990s (Arno and paper, we describe the successional conditions agers to use lightning fires as a form of others 2000, Östlund and others 2005). of ponderosa pine (Pinus ponderosa) communities along the South Fork of the Flathead burning under restricted conditions The fires of 2003, originating from light- River in the Bob Marshall Wilderness follow- (Parsons and Landres 1998, Zimmerman ning, burned through many of these ing decades of fire suppression, and then sum- and Bunnell 2000). These fires have been stands for the first time in nearly a century marize the first-year effects of the 2003 fires carefully monitored with some localized or more. In doing so, it provided a unique on these communities. We found that at least suppression used to protect facilities and opportunity to examine effects of return- 34 percent of the large ponderosa pine trees were dead or dying as a result of the fires, with private property. ing fire to a fire-dependent forest type much of this mortality due to cambial girdling Forests that were historically depen- after long periods of fire exclusion. following the burning of duff and litter dent on frequent, low-intensity fires—the In this paper, we describe the succes- buildup around the base of the trees. We “understory fire regime”—represent a sional conditions in the South Fork pon- explore possible strategies for, and barriers to, major challenge for the restoration of his- derosa pine communities and the first- the restoration of deteriorating ecosystems in wilderness and other similarly managed nat- torical wilderness fire regimes. An exam- year effects of the 2003 fires. Even though ural areas that historically depended on fre- ple of this is found along the South Fork mortality from bark beetle attacks and quent, low-intensity fires. We also discuss the of the Flathead River in the Bob Marshall other secondary effects linked to fire take complexity of managing fire-dependent Wilderness of northwestern Montana place during a period of about five years ecosystems in wilderness. (Figure 1). Here, an understory fire regime after a burn, the first-year findings already Keywords: Bob Marshall Wilderness, wilder- perpetuated relict old-growth ponderosa reveal trends that have important impli- ness management, Pinus ponderosa, ecological pine (Pinus ponderosa) communities cations for these wilderness ecosystems. restoration, ponderosa pine despite being surrounded by forest types We interpret the effects of the 2003 fires

ECOLOGICAL RESTORATION 24:2 ■ JUNE 2006 71 this context means also fails to acknowledge the role that “not subject to aboriginal people, through their extensive human controls and burning practices, may have played in manipulations that influencing wilderness landscapes. De- hamper the free play spite the recognition of this history, it is of natural forces” questionable whether, under today’s con- (Hendee and others ditions, regulated lightning fires alone can 1978, p. 68). In the restore and maintain ecosystems that were four decades since shaped historically by frequent fires, let this concept of alone those landscapes where aboriginal “wilderness” was ignitions were important (Brown 1992). codified, improved scientific under- standing has raised some questions The Fire Dilemma about the founda- Exemplified tional premises of The fate of relict ponderosa pine stands in the Wilderness Act. the Bob Marshall Wilderness illustrates Ecologists have deter- the dilemma associated with maintaining mined that most an understory fire regime in a wilderness wildland vegetation area. The Bob Marshall Wilderness types in western Complex (including the adjacent Great North America were Bear and Scapegoat wildernesses) covers shaped over thou- 1.5 million acres (0.6 million ha) of sands of years by rugged terrain straddling the Continental characteristic pat- Divide. The majority of this area is char- terns of wildland acterized by forests that burned histori- fire (Agee 1993, cally at medium to long intervals (50 to Frost 1998). In 250 years) in mixed-severity or stand- some regions, Native replacement fires (Gabriel 1976, Keane Americans started and others 1994, Arno and others 2000). Figure 1. A map of the ponderosa pine ecosystem in the Bob many of these fires In contrast, some river terraces in the Marshall Wilderness Area. The study area is shown as the area within during a period of broad valley of the South Fork of the the black line. The perimeter of the 2003 fires is shown as the cross- many centuries (An- Flathead River support old-growth pon- hatched area. derson and Moratto derosa pine that developed under the 1996, Boyd 1999, influence of relatively frequent, low- Kay and Simmons severity surface fires (Figure 1). Until the in communities that had already under- 2002, Stewart 2002, Williams 2004). early 1900s, historical fires typically gone profound successional change as a Most forest ecosystems whose composi- occurred at 20- to 30-year intervals and result of fire exclusion. We also discuss tion and structure was molded by the burned mostly through herbaceous fuels, possible strategies for, and barriers to, the understory fire regime have now been consuming light litter accumulations restoration of deteriorating ecosystems in greatly altered as a result of nearly a cen- under the large pines while killing many wilderness and other similarly managed tury of fire exclusion (Arno 2000, Keane of the small trees of Douglas-fir and other natural areas that historically depended and others 2002). conifers that became established since the on frequent, low-intensity fires. Language in the Wilderness Act sug- previous burn (Arno and others 1997). gests that permitting natural forces to Ponderosa pines ranging from 75 to resume without human interference will 600 years of age (with a few younger indi- restore “natural” conditions. Ecological viduals that are mostly found in mixed Precepts of Fire in evidence, however, indicates that allow- stands with other conifers) occupy about Wilderness ing lightning-caused fires to return after 2,000 to 3,000 acres (800 to 1,200 ha) in The Wilderness Act of 1964 (Public Law nearly a century of suppression can trigger the driest part of the South Fork valley. 88-577) designates lands as wilderness unprecedented changes in forests that On one exceptionally dry terrace (White areas “for preservation and protection in were historically dependent on frequent, River Park), mature ponderosa pine trees their natural condition… [in a state] low-intensity fires (Keane and others form pure, open stands above an extensive untrammeled by man.” Untrammeled in 2002, Mutch 1995). The Wilderness Act grass layer (Figure 2). However, in most of

72 ECOLOGICAL RESTORATION 24:2 ■ JUNE 2006 Study Area Ponderosa pine communities are scattered along an 8-mile (13-km) section of the South Fork valley in the vicinities of Big Prairie and White River junction (Figure 1). They occupy elevated terraces, mostly on the east side of the river, that have well-drained, coarse alluvial soils. The ponderosa pine communities consist largely of mature trees, sometimes accom- panied by younger individuals in the 70- to 100-year range. In some stands, ponderosa pines appear as widely dispersed, old trees within a community now dominated by Douglas-fir, lodgepole pine, and other conifer species. In other stands, ponderosa pine is a major component. Nearby low-lying, frost-prone sites have few, if any, ponderosa pine and are dominated by lodgepole pine accompa- nied by subalpine fir (Abies lasiocarpa) and Figure 2. The ponderosa pine (Pinus ponderosa) savanna ecosystem of the Bob Marshall Engelmann spruce (Picea engelmannii). Wilderness Area. This photograph was taken at White River Park, which is located near the Isolated fescue (Festuca spp.) grasslands northern end of the study area shown in Figure 1. All photos by Robert Keane, United States occupy exceptionally dry microsites, often Forest Service bordered by ponderosa pine (Koterba and Habeck 1971). Surrounding upland sites without ponderosa pine are largely domi- the valley area, this species is accompa- When lightning fires swept through nated by Douglas-fir and lodgepole pine. nied by an abundance of sapling- and many of the South Fork ponderosa pine Small numbers of western larch (Larix pole-sized trees of Douglas-fir, lodgepole stands in 2003 after a significant absence, occidentalis) occur with the ponderosa pine (Pinus contorta), and other conifers land managers were concerned that the pine, although larch is more abundant that developed in the modern period old pines, especially those surrounded by north of these ponderosa pine communi- when fires were actively suppressed. heavy fuel accumulations and weakened ties and on the north-facing slopes above by excessive competition with Douglas-fir the ponderosa pine ecosystem. and lodgepole pine, would suffer high lev- The relict ponderosa pine communi- The Fires of 2003 els of mortality. Previous studies of fire ties occupy some of the warmest and dri- During July 2003, several lightning fires effects in similar forest communities est sites in the South Fork valley, situated gradually merged and formed the Little showed that mortality is greatest among in a climatic rain shadow on the lee side Salmon complex. They burned southward old trees that had scars from pre-1930 fires of 9,000-ft (2,743-m) peaks in the Swan and eastward until early September, (Harrington 2000, Gray and Blackwell in and Mission ranges immediately to the entering more than half of the area in the press). Open scars, surrounded by thick west. These sites receive less snow and South Fork valley that harbors ponderosa layers of litter and duff, provide a means rain than areas upstream or downstream pine. The Little Salmon complex fire was for fire to burn deeply into the trunk and (Figure 1). In major valleys farther to the not actively suppressed (measures were kill cambial tissue (Figure 3). The vulner- west, north and south, ponderosa pine taken to protect administrative cabins ability of scarred trees might also extend occupies only the warmest and driest sites. and pack bridges) because exceptionally to the old trees that have bark-peeling The species is absent to the east in valleys dry conditions across the region focused scars made when Native Americans har- subject to a harsher, continental climate most suppression activities on non-wilder- vested the cambium for food. In the South (Arno and Hammerly 1984). Part of the ness areas where wildfires threatened Fork valley, bark-peeling scars on living reason for ponderosa pine’s limited extent developed areas. This return of fire after a pines date back as far as 1665, and any in the South Fork drainage is that this prolonged period of fire exclusion offered accelerated mortality of these trees would rain shadow area occurs in a relatively an opportunity to assess the effects of represent an unprecedented loss of living high valley (near 4,600 ft or 1,402 m), these fires on the relict ponderosa pines artifacts of a former culture (Östlund and where the frost-free season appears to be (Arno and others 2000). others 2005). marginally short and probably hinders

ECOLOGICAL RESTORATION 24:2 ■ JUNE 2006 73 ponderosa pine cone production and sub- sequent regeneration (Lunan and Habeck 1973, Arno and Hammerly 1984, Jacob- sen 1986). Due to fire exclusion, the South Fork ponderosa pine communities experienced very few small fires between roughly 1930 and 2002, even though the Spotted Bear Ranger District of the Flathead National Forest has had an active wildland fire use program for more than 20 years (Arno and others 2000, Östlund and others 2005). Lack of fires during this period allowed young lodgepole pine and Douglas-fir to colonize or increase in for- merly open stands of mature ponderosa pine (Figure 3). Mature lodgepole pines associated with the ponderosa pine communities commonly exhibit one to four well-formed fire scars created prior to 1930. Old-growth ponderosa pines sometimes possess scars from five to seven fires. Fire-scar dating thus far indicates that for three centuries prior to about 1930, fires occurred at average intervals of about 20 Figure 3. Deep duff mounts surrounding the bole of large ponderosa pine (Pinus ponderosa) to 30 years in the ponderosa pine commu- with an Indian scar in the Big Prairie area of the Bob Marshall Wilderness Area. Also note the nities (Östlund and others 2005). Fires increase in Douglas-fir (Pseudotsuga menzeisii) in the understory where it is crowding the clearly allowed individual ponderosa large ponderosa pine and weakening these trees. pines to survive for several centuries, and the species to dominate in relatively open stands. Östlund and his colleagues (2005) mortality (fire, beetles, unknown), and scar type (Table 1) shows that about 16 aged more than one hundred Native percent crown scorched by fire. We also percent of all ponderosa pine had died American bark-peeling scars and found recorded diameter at breast height (DBH), within a year after the fires, many from the peel scars were made between 1665 presence of fire scars, and bark-peeling secondary effects—possibly bark beetle and 1938. This, and other archeological scars (Figure 4). Geographic location was attacks. Another 18 percent are dying evidence and information, suggests a sig- identified with a GPS receiver, using and will probably be dead within a year or nificant amount of human occupation of UTM coordinates for zone 12 and NAD two. Thus, we estimate that at least 34 these sites for several centuries and that 83 projection. During visits prior to the percent of the mature ponderosa pine many of the historical fires may have been 2003 fires, we noted many trees on the site trees will be dead within a few years of the started by these Native Americans that were weakened by competition and 2003 fires (Table 1). Trees with old fire (Flanagan 2001). mountain pine beetle. We also observed scars or bark-peeling scars comprised duff mounds that were 6 to 15 inches deep about 25 percent of the mature ponderosa (15 to 40 cm) surrounding the base of pines in our sample (Table 1). About 42 Tree Sampling many old-growth ponderosa pines in areas percent of all scarred trees were dead and We sampled all mature (20 inches [50 cm] outside the 2003 burns. dying as a result of the 2003 fires com- or greater in diameter, and thus probably pared to the 31 percent of unscarred trees at least 100 years old) ponderosa pines in (Table 1). The fire commonly burned into broad transects (more than 100 meters Results old scars inflicting heat damage to the wide) extending through portions of the We sampled 455 mature ponderosa pine cambium or consuming wood needed for study area that burned in 2003 (Figure 1). trees within the 2003 fires in the vicinities structural support, causing the tree to On these transects, measurements and of White River Park and Big Prairie eventually fall. We estimate that about evaluations taken for all sampled trees that (Figure 5). This sample represents an half of living trees with historic bark-peel- were alive prior to 2003 included: health appreciable proportion of all mature pon- ing scars will die within two years of the status (1-satisfactory, 2-weak, 3-near derosa pine trees in the study area. The 2003 fires (Table 1). death, 4-died recently), probable cause of distribution of trees by health status and

74 ECOLOGICAL RESTORATION 24:2 ■ JUNE 2006 We speculate that most of the observed mortality came from cambial girdling at the ground line probably linked to the heavy duff and litter accumulations that almost always were completely consumed. However, the heavy crown scorch in the recently dead trees also contributed to the fire-caused tree mortality (Figure 6). Some post-fire mortality, which we estimate at about 20 percent, was attributed to observed mountain pine beetle infesta- tions in the weakened trees, although the specific cause of death was difficult to determine. Beetle mortality has also expanded to those ponderosa pines not exposed to recent fires. Several conditions in the South Fork ponderosa pine stands that lie outside the boundaries of the 2003 fires suggest that unburned stands are also vulnerable to accelerated attrition with or without fire.

1. Duff layers around mature trees have accumulated for 75 to 100 years, three Figure 4. Measuring diameter at breast height on an Indian-scarred tree that was burned in to four times as long as pre-1930 inter- the 2003 fires in the Bob Marshall Wilderness Area. Note the lack of vegetation surrounding vals between fires. Thus, any fire will these trees because of severe soil heating from the thick duff and litter layer resulting from produce an unusually severe heating fire exclusion. effect on tree cambium and roots (Harrington and Sackett 1992). 2. Competing conifers have had 75 to 100 Discussion years to become established instead of the 20- to 30-year intervals between Tree Mortality fires that were common historically. The 2003 fires generally burned with low This has favored Douglas-fir, which is intensities through the South Fork pon- quite fire sensitive when young but derosa pine communities causing only begins to develop corky bark and a moderate crown scorch in many places higher, more fire-resistant canopy after (Figure 6). In this sense, the effects of the about 30 years. Douglas-fir is consider- fires may be the best that can be hoped ably more shade-tolerant than pon- for without management intervention derosa pine and is the pine’s most (that is, raking the duff away from the abundant and aggressive competitor. It base of the tree) considering the abun- also produces cone crops more fre- dance of duff, woody, and ladder fuels. quently than the pine. Canopy torching and passive crown fire 3. Crowded ponderosa pines have low occurred locally in small patches, up to vigor and are vulnerable to bark beetle about 5 acres (2 ha), mostly in dense attack. Thus, they are unlikely to retain lodgepole pine/Douglas-fir stands. dominance and survive several hun- The estimated ponderosa pine dred years, as was the case historically mortality after the fires of 2003 (34 per- (for example, see Arno and others 1995 cent) may well increase as insects and and Biondi 1996). disease invade weakened trees (Table 4. Ponderosa pine cone crops and seedling 1). This may be a high rate of loss for a survival are generally poor in such rela- Figure 5. Trees sampled during the summer of species that in the past survived most tively cold sites, and are made worse by 2004. Only ponderosa pines that were more than fires and benefited competitively as a excessive competition linked to fire 20 inches DBH were sampled. result of them (Arno and Fiedler 2005). exclusion.

ECOLOGICAL RESTORATION 24:2 ■ JUNE 2006 75 Table 1. Summary of sampled ponderosa pine trees greater than 20 inches DBH by nerable, it is difficult to conceive that scar status and health. today’s communities can be considered “untrammeled by man.” Number of ponderosa pine trees (% of grand total) The recent exclusion of fires presents Scar Type Healthy Sick Dying Dead Total a dilemma for wilderness land managers. No Scars 174 (38) 60 (13) 66 (15) 38 (8) 338 (74) For several decades before and after the Fire Scarred 21 (4) 13 (3) 6 (1) 16 (4) 56 (12) passage of the Wilderness Act, land man- Indian Scar 25 (6) 5 (2) 9 (2) 14 (4) 53 (12) agers have suppressed the vast majority of Fire/Indian Scars 3 (1) 2 (>1) 1 (>1) 2 (>1) 8 (2) natural (lightning) fires in wilderness Total 223 (48) 80 (18) 82 (18) 70 (16) 455 (100) areas (Parsons and Landres 1998). This constitutes a profound human manipulation of ecosystems (Keane and others 2002). In some of the largest wilderness 100 areas, including the Bob Marshall, an 90 increasing proportion of natural fires 80 have been allowed to burn since 1970. This may well help restore ecosystems in 70 mixed or stand replacement fire regimes 60 that historically burned at intermediate 50 or long intervals (Arno and Fiedler 40 2005). However, in most of the South 30 Fork ponderosa pine communities, suc- 20 cessional replacement by other conifers and an excessive accumulation of fuel Average percent crown scorch crown percent Average 10 now portend a continuing loss of pon- 0 Healthy Sick Dying Dead derosa pine communities given either continued fire suppression or the resump- Tree health tion of lightning fires. The impact of anthropogenic burn- Figure 6. The average percent crown scorch by health class of the 455 trees sampled within ing in the South Fork river valley pre- the 2003 burned area near Big Prairie, Bob Marshall Wilderness Area. sents yet another dilemma for the management of wilderness. Native Management Dilemma Native American burning elsewhere in Americans are now thought to have set It is difficult to confirm that lightning igni- western Montana suggest that aboriginal many fires in the South Fork valley for tions alone could have been responsible for burning was likely an important ignition several centuries prior to European settle- the relatively high frequency of pre-1930 component of the historical fire regime in ment (Boyd 1999, Kay and Simmons fires in the South Fork ponderosa pine the South Fork valley (Barrett and Arno 2002, Östlund and others 2005). Their stands. Ponderosa pine communities cover 1982, Gruell 1985, Arno and others 1997). contribution to the ponderosa pine fire less than 2 percent of the 1.5 million-acre Prolonged exclusion of fires, from regime is difficult to determine, but the Bob Marshall Wilderness Complex, and both the direct suppression of fire and the evidence supports their probable influ- they are surrounded by forest types that elimination of the Native American cul- ence on maintenance of the old-growth had considerably longer fire intervals ture of burning, has had compounding ponderosa pine ecosystem. Neither (Gabriel 1976, Keane and others 1994). negative consequences for ponderosa pine today’s landscapes nor pre-European set- Furthermore, these ponderosa pine forests stands of the Bob Marshall Wilderness tlement landscapes can be considered are located in valley bottom sites that are Area regardless of the origin of historical “untrammeled by man.” Humans have less frequently struck by lightning than fires. These include allowing other, more been manipulating wilderness ecology for forests in higher mountain topography fire-sensitive conifer species to increase a long time, yet the Wilderness Act (Barrows and others 1977, Renkin and competitive stress on ponderosa pine as requires that wilderness be “not subject to Despain 1992). Severe drought years prob- well as creating dense ladder fuels and human controls and manipulations that ably provide the only opportunity for light- deep duff and litter accumulations that hamper the free play of natural forces” ning fires to spread from higher elevations now allows fire to kill the competition- (Hendee and others 1978). downslope into the South Fork valley. weakened ponderosa pine. Considering To have any hope of conserving the Long-term use of the South Fork pon- that many of the ponderosa pines are 400- late-successional ponderosa pine commu- derosa pine communities by Native to 600-years-old and survived numerous nities in the South Fork, management Americans and evidence of significant pre-1930 fires, but now appear highly vul- intervention is necessary to restore a sem-

76 ECOLOGICAL RESTORATION 24:2 ■ JUNE 2006 blance of historical fire regimes, stand After initial restoration treatments, it is Arno, S.F. and R. Hammerly. 1984. Timberline: structure, and fuel conditions (Figure 2). possible that natural fires can be permit- Mountain and Arctic forest frontiers. Seattle: The Mountaineers—Books. Fuel treatment options might include rak- ted to return with outcomes similar to Arno, S.F., D.J. Parsons and R.E. Keane. 2000. ing duff away from the base of old pon- those of the historical fires that perpetu- Mixed-severity fire regimes in the northern derosa pines to minimize excessive ated these communities. It will also be Rocky Mountains: Consequences of fire heating of the stem and roots, cutting necessary to monitor stand conditions and exclusion and options for the future. Pages some of the competing conifers and per- ensure that fire continues to occur at fre- 225-232 in Wilderness science in a time of change conference, Volume 5: wilderness haps burning them in piles to reduce the quencies and severities that mimic histor- ecosystems, threat, and management. potential for crown fires, and/or conduct- ical conditions. Until restoration of Missoula, Montana, May 23-27, 1999. U.S. ing prescribed burns during cool or moist historical fuel and structural conditions Department of Agriculture Forest Service seasons (early spring or autumn) to reduce has progressed, both protection from fire Rocky Mountain Research Station, Fort fire intensities (Covington and others and focused human manipulation, and Collins, CO. Arno, S.F., J. Scott and M. Hartwell. 1995. 1997, Harrington 2000, Friederici 2003, future wildfires will contribute to the Age-class structure of old growth pon- Arno and Fiedler 2005). Such treatments downward spiral of forest health and the derosa pine/Douglas-fir stands and its rela- are commonly practiced to restore old- eventual loss of the successional pon- tionship to fire history. USDA Forest growth ponderosa pine communities out- derosa pine communities. Without fire Service, Intermountain Research Station, side wilderness. A notable example of this and limited fuel treatments, competing Research Paper 481, Ogden, UT. Arno, S.F., H. Smith and M. Krebs. 1997. Old kind of fire regime restoration is begin- conifers will rapidly replace these pon- growth ponderosa pine and western larch ning on 200,000 acres (80,000 ha) of derosa pine forests. stand structures: Influences of pre-1900 wilderness and non-wilderness in Idaho’s Contrasting philosophical questions fires and fire exclusion. USDA Forest Salmon River canyon (Nez Perce remain: Will society accept the loss of Service, Intermountain Research Station, National Forest 1999). Although such ponderosa pine communities in wilder- Research Paper 495, Ogden, UT. Barrett, S.W. and S. Arno. 1982. Indian fires as treatments in the South Fork ponderosa ness because human intervention is unde- an ecological influence in the Northern pine communities would involve human sirable? Or, will society tolerate fuel Rockies. Journal of Forestry 80(10):647-651. manipulation, they should be limited to treatments in wilderness because the loss Barrows, J.S., D.V. Sandberg and J.D. Hart. that necessary to reverse the degradation of ponderosa pine forests is unacceptable? 1977. Lightning fires in Northern Rocky caused by a history of indirect human Mountain forests. Final Report for Contract Grant 16-440-CA USDA Forest manipulation (fire suppression) that can- ACKNOWLEDGMENTS Service Intermountain Fire Sciences not be reversed without active interven- We thank Gordon Ash, Debra Muklow and Laboratory, on file, USDA Forest Service, tion (Brown 1985, Mutch 1995, Agee Steve Wirt, USDA Forest Service, Flathead Intermountain Fire Sciences Laboratory, 2002, Thomas 2002). The South Fork National Forest for project support and tech- P.O. Box 8089, Missoula, MT. ponderosa pine forests may be somewhat nical review. We also thank Dave Parsons and Biondi, F. 1996. Decadal-scale dynamics at the Gus Pearson Natural Area: evidence easy to restore because they represent a Carol Miller of the Aldo Leopold Wilderness Research Institute, Missoula, MT; and Mick for inverse (a)symmetric competition? small area and more than 50 percent of Harrington, USDA Forest Service Rocky Canadian Journal of Forestry Research this area was burned in 2003. Yet, even Mountain Research Station, Missoula, MT for 26:1397-1406. with human intervention, it is still ques- additional review comments. Boyd, R., editor. 1999. Indians, fire and the land tionable whether managed lightning igni- in the Pacific Northwest. Corvallis: Oregon State University Press. tions can create fire regimes that will REFERENCES Brown, J.K. 1985. The “unnatural fuel sustain the older ponderosa pine trees Agee, J.K. 1993. Fire ecology of Pacific Northwest buildup” issue. Pages 127-128 in J.E. Lotan, (Brown 1992). These isolated ecosystems forests. Washington, D.C.: Island Press. B.M. Kilgore, W.C. Fischer and R.W. will surely decline if nothing is done to __. 2002. The fallacy of passive management of Mutch (eds.), Symposium and workshop ensure the continued survival of these western forest reserves. Conservation Biology on wilderness fire. U.S. Department of Agriculture, Forest Service, Intermountain relic ponderosa pines. in Practice 3(1):18-25. Anderson, M.K. and M.J. Moratto. 1996. Forest and Range Experiment Station, Native American land-use practices and Missoula, MT. ecological impacts. Pages 187-206 in Sierra __. 1992. A case for management ignitions in Conclusions Nevada Ecosystem Project: Final report to wilderness. Fire Management Notes 53- The relict ponderosa pine communities in Congress, Volume II. Davis: University of 54:3-8. Covington, W.W., P.Z. Fulé, M.M. Moore, S.C. the Bob Marshall Wilderness are charac- California, Centers for Water and Wildland Resources. Hart, T.E. Kolb, J.N. Mast, S.S. Sackett and terized by ancient trees that are often liv- Arno, S.F. 2000. Fire regimes in western forest M.R. Wagner. 1997. Restoring ecosystem ing artifacts of Native American land use ecosystems. Pages 97-120 in USDA Forest health in ponderosa pine forests of the as witnessed by bark-peeling scars. To sus- Service, Rocky Mountain Research southwest. Journal of Forestry 95:23-29. tain these communities, land managers Station, General Technical Report RMRS- Flanagan, D. 2001. Indian trails of the Northern Rockies. Stevenville, MT: Stoneydale Press. would need to reduce fuel hazard and P-42, Volume 2, Fort Collins, CO. Arno, S.F. and C.E. Fiedler. 2005. Mimicking Friederici, P. 2003. Ecological restoration of excessive competition before applying nature’s fire: Restoring fire-prone forests in the southwestern ponderosa pine forests. prescribed fire for more beneficial effects. West. Washington, D.C.: Island Press. Washington, D.C.: Island Press.

ECOLOGICAL RESTORATION 24:2 ■ JUNE 2006 77 Frost, C.C. 1998. Presettlement fire frequency Inland Mountain West Symposium. USDA Natural Areas Journal 23:315-325. regimes of the United States: a first approx- Forest Service, General Technical Report Parsons, D.J. and P.B. Landres. 1998. Restoring imation. Tall Timbers Fire Ecology INT-203. Ogden, UT. natural fire to wilderness: How are we Conference 20:70-81. Kay, C.E. and R.T. Simmons. 2002. Wilderness doing? Tall Timbers Fire Ecology Conference Gabriel, H.W. 1976. Wilderness ecology: The and political ecology: Aboriginal influences and 20:366-374. Danaher Creek drainage, Bob Marshall the original state of nature. Salt Lake City: Renkin, R.A. and D.G. Despain. 1992. Fuel Wilderness, Montana. M.S. thesis. The University of Utah Press. moisture, forest type, and lightning-caused University of Montana, Missoula, MT. Keane, R.E., P. Morgan and J.P. Menakis. fire in Yellowstone National Park. Canadian Gray, R.W. and B. Blackwell. In press. The 1994. Landscape assessment of the decline Journal of Forest Research 22:37-45. maintenance of key biodiversity attributes of whitebark pine (Pinus albicaulis) in the Stewart, O.C. 2002. Forgotten fires: Native through ecosystem restoration operations. Bob Marshall Wilderness Complex, Americans and the transient wilderness. In Proceedings from the 2002 Fire Montana, USA. Northwest Science 68:213- Norman, OK: University of Oklahoma Press. Conference: Managing fire and fuels in the 229. Thomas, J.W. 2002. Dynamic vs. static man- remaining wildlands and open spaces of the Keane, R.E., T. Veblen, K.C. Ryan, J. Logan, agement in a fire-influenced landscape— southwestern United States. USDA Forest C. Allen and B. Hawkes. 2002. The cas- the Northwest Forest Plan. Text of Service, Pacific Southwestern Research cading effects of fire exclusion in the Rocky presentation at the Conference on “Fire in Station, Berkeley, CA. Mountains. Pages 133-153 in J. Baron (ed.), Oregon Forests,” Bend, Oregon, October Gruell, G.E. 1985. Indian fires in the interior Rocky Mountain futures: An ecological 23. Oregon Forest Resources Institute, west: a widespread influence. General perspective. Washington, D.C.: Island Portland, OR. Technical Report INT-182, USDA Forest Press. Williams, G.W. 2004. The bibliography of Service. Koterba, W.D. and J.R. Habeck. 1971. Indian use of fire. USDA Forest Service, Harrington, M.G. 2000. Fire applications in Grasslands of the North Fork Valley, Washington, D.C. Available at www. ecosystem management. Pages 21-22 in Glacier National Park, Montana. Canadian fs.fed.us/fire/fmt/Bibliography/index.html. USDA Forest Service, Rocky Mountain Journal of Botany 49:1627-1636. Zimmerman, G. T. and D. L. Bunnell. 2000. Research Station, Proceedings 17, Ogden, Lunan, J.S. and J.R. Habeck. 1973. The effects The federal wildland fire policy: UT. of fire exclusion on ponderosa pine com- Opportunities for wilderness fire manage- Harrington, M.G. and S.S. Sackett. 1992. Past munities in Glacier National Park, ment. Pages 288-297 in D. N. Cole, S. F. and present fire influences on southwestern Montana. Canadian Journal of Forestry McCool, W. T. Borrie and J. O’Loughlin ponderosa pine old growth. Pages 44-50 in Research 3:574-579. (compilers), Wilderness science in a time Proceedings of a workshop: Old-growth Mutch, R.W. 1995. Restoring forest health: Do of change conference. Vol. 5, Wilderness forests in the Southwest and Rocky we have the will to apply science findings? ecosystems, threats, and management. Mountain regions. USDA Forest Service, Pages 18-22 in Proceedings of the confer- USDA Forest Service, Rocky Mountain Rocky Mountain Forest and Range ence on forest health and fire danger in Research Station, RMRS-P-15-Vol. 5, Experiment Station, General Technical inland western forests. Washington, DC: Ogden, UT. Report RM-213. Fort Collins, CO. American Forests. Hendee, J.C., G.H. Stankey and R.C. Lucas. Nez Perce National Forest. 1999. Salmon 1978. Wilderness management. USDA River Canyon Project: Draft environmen- Robert E. Keane, Stephen Arno (retired), and Forest Service, Miscellaneous Publications tal impact statement. Grangeville, Idaho. Laura J. Dickinson, USDA Forest Service, Rocky 1365, Washington, D.C. Östlund, L., R.E. Keane, S.F. Arno and R. Mountain Research Station, Fire Sciences Jacobsen, G.L. 1986. Are you getting pon- Andersson. 2005. Culturally scarred trees Laboratory, P.O. Box 8089, Missoula, MT derosa pine and Douglas-fir cone crops at in the Bob Marshall Wilderness, Montana, 59807; 406/329-4846, Fax: 406/329-4877, high elevations? Pages 47-49 in USA: Interpreting Native American his- [email protected]. Proceedings of the conifer tree seed in the torical land use in a wilderness area.

78 ECOLOGICAL RESTORATION 24:2 ■ JUNE 2006