Insect and Roadless Forests

8/14/2019 Insect and Roadless Forests

1/40

A Scientific Review ofCauses, Consequences and

Management Alternatives


2/40


3/40

INSECTS ANDROADLESS FORESTS

A Scientific Review ofCauses, Consequences and

Management Alternatives

S.H. Black1

D. Kulakowski2

B.R. Noon3

D. DellaSala4

1 Xerces Society for Invertebrate Conservation, Portland, Ore.

2 School of Geography, Clark University, Worcester, Mass.

3 Department of Fish, Wildlife and Conservation Biology,

Colorado State University, Fort Collins, Colo.

4 National Center for Conservation Science & Policy, Ashland, Ore

Suggested citation: Black, S. H., D. Kulakowski, B.R. Noon, and

D. DellaSala. 2010. Insects and Roadless Forests: A Scientific

Review of Causes, Consequences and Management Alternatives.

National Center for Conservation Science & Policy, Ashland OR.

Cover and opposite: photo of lodgepole pine forest by

Adam Jones/Visuals Unlimited via Getty Images; photo of beetle-stressed

lodgepole pine forest Drake Fleege/powderhillphotography.com.


4/40

Insects and Roadless Forests

A Scientific Review of Causes, Consequences and Management Alternatives


5/40



TABLE OF CONTE NTS

Executive Summary

Introduction

Colorados Forest Ecosystems

Important Insects in Colorados Forests

Root Causes of Insect Infestations in Colorados Forests

Forest Insects and Fire: Essential Elements of Ecosystems

Does Tree-Cutting to Address Bark Beetle Infestations

Reduce Stand Susceptibility to Outbreaks?

Colorados Roadless Forests

Managing Fire Risk Near Communities versus Roadless Forests

Conclusion

Literature Cited

1

3

5

7

9

10

13

17

21

23

25


6/40




7/40



e ndings in this report regarding the causesand appropriate treatment of insect outbreaksgenerally apply to both roaded and roadless forestareas. However, this report specically addressesa recent proposal to exempt national forest road-less areas in Colorado from protections underthe 2001 Roadless Area Conservation Rule, inpart to address insect outbreaks and perceivedre risk in these forests.

Colorados roadless areas have recognized eco-logical and social importance.e 2001 RoadlessArea Conservation Rule provides a consistentand scientically based national standard thatsafeguards more than 4 million acres of Colora-dos inventoried roadless areas from development.e state has submitted a proposal under the

Administrative Procedure Act to change the2001 rule to allow road construction and tree-cutting in inventoried roadless areas in part toaddress recent outbreaks of bark beetles andrelated perceived re risks. In this paper we out-line key aspects of and review research related tobark beetle outbreaks, their relationship to rerisk and the ecacy of silvicultural practices tocontrol outbreaks and reduce the risk ofre. Wealso summarize the importance of roadless areas

IN RESPONSE TO RECENT BARK BEETLE EPIDEMICS, decision-makers are call-

ing for landscape-level mechanical treatments to prevent the spread of these native insects and to

reduce the perceived threat of increased re risk that is believed to be associated with insect-killed

trees. e best available science indicates that such treatments are not likely to reduce forest suscepti-

bility to outbreaks or reduce the risk ofres, especially the risk ofres to communities. Furthermore,

such silvicultural treatments could have substantial short- and signicant long-term ecological costs

when carried out in national forest roadless areas.

EXECUTIVE SUMMARY

for wildlife and water quality.e keyndingsof this paper are presented below.

FINDING 1Insect outbreaks andres have been part of theecology of these forests for millennia.

Large outbreaks of native forest insects and largeforest res have played an important role in thedevelopment and maintenance of many foresttypes. In Colorado, as elsewhere, lodgepole pineand spruce-r forests are characterized by large,infrequent, high-severityres and occasionallarge-scale bark beetle outbreaks.

FINDING 2

Ongoing outbreaks of insects are probably caused

primarily by climate.

Climate, specically drought and high tempera-ture, may be the most important factor behindthe current bark beetle epidemic in the westernUnited States. Because silvicultural treatmentscannot effectively alleviate the overridingeects of climate, their application in roadlessareas is likely to do little to mitigate ongoing orfuture outbreaks.


8/40



FINDING 3Insect outbreaks in roadless areas are not likely toheightenre risk in adjacent communities.

Although it is widely believed that insect outbreaksset the stage for severe forest res, the few scien-

ti

c studies that support this idea report a very smalleect, and other studies have found no relationshipbetween insect outbreaks and subsequent re activ-ity in lodgepole pine and spruce-r forests.e bestavailable scientic knowledge, derived from empiri-cal and modeling studies from numerous independentresearchers, indicates that the assumed link betweeninsect outbreaks and subsequent forest re is not wellsupported, especially for the forest types that are cur-rently most aected by outbreaks. Furthermore, therisk ofres in remote roadless areas is not likely toinuence the risk ofres adjacent to communities.Instead, the presence of fuels (both living and dead)

near homes and other structures is likely to be mostimportant in determining the risk ofre in these areas.

FINDING 4

Tree-cutting is not likely to control ongoing bark beetleoutbreaks or other insect species common to Colorado.

Although individual trees may be saved by sprayingwith insecticides, eorts aimed at stopping outbreaksare unlikely to be eective once bark beetles reachepidemic levels and cause extensive tree mortality.

FINDING 5

inning in roadless areas is not likely to alleviatefuture large-scale epidemics of bark beetle.

inning is often recommended to control outbreaksof bark beetles, but the evidence is mixed as to itseectiveness at the stand level, and it is unlikelyto be eective in controlling or alleviating large-scale outbreaks. Experimenting with thinning inroadless areas also can cause short- and long-termecological eects.

FINDING 6

Tree-cutting in roadless areas will not keep communi-ties safe from wildre.

If the goal is to protect communities from wildre, thenmanagement would be most eective if it focused inand around communities and involved creating defen-sible space immediately adjacent to homes. Fire-hazardreduction eorts around communities are more eective,much less expensive and less ecologically damaging thantrying to make wholesale modication of forest structure.

e limited funds available for community protectionwould be most eective if used to create defensible spacearound homes and communities, rather than to buildroads and implement fuel treatments in remote road-less areas.

FINDING 7Building the roads necessary to enter roadless areasaects their ecological values.

The construction of temporary or permanentroads in roadless areas can have substantial short-and long-term ecological costs. e presence anduse of roads has been linked to increased wildreignitions, increased wildlife mortality and frag-mentation of habitat, changes in the physical andchemical environment, diminished water quality,introduction of invasive species and increased likeli-hood of landslides.

FINDING 8

Green and familiar forests will eventually return fol-lowing insect outbreaks in most locations.

Forests have continued to develop following pastinsect outbreaks. Although the current outbreaks are

very large and may even be unprecedented in extentand severity in recent history, there is no evidence thataected forests cannot regenerate following these dis-turbances.e forests that are now losing many treesto insect attack will not look the same in our lifetimes,but healthy trees and familiar forest structures will

eventually return in most locations. Although beetle-aected forests may look dierent to the human eye,they are still functioning ecosystems that provide foodand shelter for animals and water for sh and people.

FINDING 9

e 2001 Roadless Area Conservation Rule allows suf-cientexibility to manage Colorados roadless areas.

Anticipating the need for limited active management,the 2001 Roadless Area Conservation Rule allows suf-cient exibility locally to address public health andsafety,re and undesirable insects, while maintainingthe qualities and character of national forest roadlessareas. Under the states proposal, Colorados nationalforest roadless areas would be subjected to numerousexceptions to the protections that are provided underthe national rule, thereby degrading roadless qualitiesand providing fewer protections to these areas thanany state in the nation.


9/40



INTRODUCTION

Due to the ecological and social importance ofroadless areas, the 2001 Roadless Rule appliesa consistent and scientically based nationalstandard that safeguards the numerous ecolog-ical benets that roadless areas provide (USFS,2000). e state of Colorado has proposeda change to the 2001 rule and has proposedexemptions that would remove at least 246,000acres of roadless area from the national inven-tory or by allowing logging, road construction,oil and gas development, additional coal min-ing and exploration, and expansion of skiareas (scientists letter to Governor Bill Ritter

Jr., 2009).

One key reason given for the proposed exemp-tions is to address recent outbreaks of barkbeetles that have killed millions of trees across

THERE IS STRONG SCIENTIFIC EVIDENCE that roadless areas provide high-quality

habitat for threatened species; contain important concentrations of mature and old-growth forests,

aquatic strongholds and other sensitive and ecologically important habitat; and act as a buer against

invasive species (Strittholt and DellaSala, 2001; DeVelice and Martin, 2001; Loucks et al., 2003;

Peterson, 2005). Because more than half the nations roadless areas are at elevations above 7,000 feet

(U.S. Forest Service, 2000), they are vital for wildlife seeking cooler, moister conditions in the face

of a warming climate. Colorados roadless areas also provide recreational opportunities of all kinds,

including hunting and shing (Peterson, 2005) and world-class backcountry skiing and trekking.

e old trees in roadless areas also sequester and store carbon for centuries, playing a pivotal role

in absorbing greenhouse gas pollutants, and many of the nations roadless areas are at the source of

downstream drinking water supplies (USFS, 2000).

Colorado and other Rocky Mountain states. estates 2009 proposal would allow the follow-ing activities:

Tree-cutting anywhere in a roadless areawhere the regional forester determined it wasneeded to prevent or suppress an insect or dis-ease epidemic.

Tree-cutting, with accompanying roadconstruction, forre and insect managementwithin a 1.5-mile radius of any at-risk commu-nity (a term not dened in the states proposal).

Modifying the Roadless Rule in Colorado isintended to help managers mitigate the spreadof ongoing insect outbreaks, reduce suscepti-bility to future outbreaks and reduce the risk of

Johnny N. Dell/b


10/40



forest res that is believed to increase due to insectinfestations.

e widespread concern about recent outbreaks isunderstandable, because they may be unprecedentedin recent history. Furthermore, it is reasonable to be

concerned about the health of a

ected forests andthe potential for the elevated risk of forest res dueto outbreaks of bark beetles. However, those concernsneed to be informed by the best available science to

ensure that our responses do not have unintendedecological consequences with undesirable eects nowand in the future.

In this paper we outline key aspects of bark beetleoutbreaks as well as their relationship to re risk.

We also discuss the e

ects that the 2009 Coloradoproposal to modify the 2001 Roadless Rule wouldprobably have on bark beetle outbreaks,re risk andthe ecological values of roadless areas.


11/40



COLORADOS FOREST ECOSYSTEMS

Mature and old-growth forest stands are likelyto be more prevalent within roadless areas than

in areas where higher levels of logging and otherhuman activities have occurred (USFS, 2008).ese forests, in particular, are highly valuableas part of Colorados wildlife heritage, provid-ing habitat for countless species dependent onolder forests.ey also play an important role incarbon storage and carbon cycling. Finally theyoer scenic value to local residents and tourists.Of particular importance to the people of Colo-rado is the unique role that roadless areas play insupplying clean water to facilities that treat anddistribute drinking water.

Insects and Colorados Forest EcosystemsA Co-Adapted System

Bark beetles are native to the vast conifer forestsof temperate North America and the tree spe-cies have co-adapted with these organisms. Infact, these small insects play a vital role in eco-

NATIONAL FORESTS IN COLORADO provide a diverse array of vegetation types, ranging

from warm, dry pinyon-juniper woodlands at lower elevations to cold, moist subalpine forests at

upper elevations. While large expanses of grasslands, shrublands, croplands, rangelands and other

vegetation types dominate Colorado, its roadless areas are predominantly coniferous forests occu-

pying mountainous terrain. Approximately 72 percent of the states roadless areas contain various

forest communities.e largest types of forests in Colorados roadless areas are spruce-r (24 percent),

followed by aspen (20 percent), lodgepole pine (13 percent), Douglas r (8 percent), ponderosa pine

(3 percent), pinyon juniper (2 percent) and other tree species (2 percent) (USFS, 2008).

system function, and their absence could have aprofound eect on the function of forest ecosys-

tems (Black, 2005).

ere are more than 6,000 species of bark bee-tles worldwide. Most species cause little or noeconomic damage, normally infesting branches,stumps and stems of standing dead or severely

weakened trees or downed woody material. A fewspecies, including members of the Dendroctonusgenus, (which includes the mountain pine bee-tle) normally exist as small endemic populationsthat feed mainly on trees that recently died, but

when conditions are right, the populations cangrow rapidly to epidemic levels, overwhelmingthe defenses of live trees and resulting in wide-spread mortality.

Native insects, including those that attack andsometimes kill large patches or stands of trees,have been part of Rocky Mountain forests formillennia and have played an important role in

Wildern


12/40



the development of forest ecosystems. For exam-ple, there is abundant scientic data that documentepidemics of spruce beetles (Dendroctonus rupen-nis) having killed trees over extensive areas of forest,even long before Colorado was a state (Veblen et al.,1991; Veblen et al., 1994; Eisenhart and Veblen, 2001;Kulakowski and Veblen, 2006).ese and other nativeinsects evolved with their host trees over thousandsof years; they function as nutrient recyclers, agentsof disturbance, members of food webs and regula-tors of forest productivity, diversity and tree density(Clancy, 1993). Low-severity outbreaks that kill select

susceptible trees may increase growth rates of sur-viving trees (Haack and Byler, 1993). High-severityoutbreaks that reach epidemic levels can contributeto the development of complex, multi-aged stands(Axelson et al., 2009) and may ultimately contributeto increased biodiversity and resilience to future dis-turbances (Alfaro et al., 1982).

Source of Food Web Dynamics

Bark beetles are important parts of many forest foodwebs. A salient feature of bark beetle communities isthe staggering number of organisms associated withthem (Dahlsten, 1982).ese insects are hosts forparasites and are prey for a variety of animals, includ-ing spiders, birds and other beetles.

Bark beetle outbreaks can help create habitat andresources for a variety of species. By feeding on deador dying trees, bark beetles provide food to insect-eat-ing birds such as woodpeckers (Koplin and Baldwin,1970) and create snags that may be used by woodpeck-ers, owls, hawks, wrens and warblers, as well as such

mammals as bats, squirrels, American marten, Paci

cshers and lynx. Epidemics of bark beetles increasethe availability of plant material for foraging, brows-ing and nesting for wildlife such as small mammalsand birds (Stone and Wolfe, 1996). In a study of pon-derosa pine forest on the Front Range of Colorado,herbaceous biomass was 50 to 100 times greater instands ve years after an infestation of mountain pinebeetles than in uninfested stands (Kovacic et al., 1985).Kovacic et al. (1985) estimated that it is possible thatlevels of wildlife habitat will remain elevated abovepre-infestation levels for 10 to 15 years followingbeetle infestation. Forest insects ultimately contrib-

ute to recruitment of large coarse woody debris intoriparian areas and stream systems, which is essen-tial for building pools that provide habitat for trout.Compared to historic conditions, these large, deeppools are lacking in many streams today (Williamsand Williams, 2004).

Maintaining Forest Heterogeneity and Diversity

In many forest types, low-severity outbreaks of barkbeetles and defoliators reduce the density of treesand cull weak trees, relieving the stress on the survi-

vors, increasing the diversity of stands and creating

multi-aged stands (Schowalter, 1994). In many cases,the prime beneciaries of low-severity outbreaks arethe surviving trees (Schowalter and Withgott, 2001).For instance, the Douglas-r beetle (Dendroctonuspseudotsugae) may help maintain ponderosa pine bycontributing to a shift in dominance from Douglas rto ponderosa pine (Schowalter and Withgott, 2001).At the plant community level, insect outbreaks canalso increase the diversity of tree species over thelong term (Schowalter, 1994), which can promotefunctional stability and regeneration of forest ecosys-tems following subsequent disturbances. Trees thatare killed by insects and remain on site decomposeand ultimately contribute to improved soil fertility(Schowalter, 1994).

Bark beetles feed on the inner bark of trees, cutting off theflow of nutrients from the leaves to other parts of the tree.

EdwardS.Ross


13/40



IMPORTANT INSECTSIN COLORADOS FORESTS

Mountain pine beetles typically exist as small,endemic populations in many stands of lodgepolepine.ese populations feed on the innermostbark layer of trees (the phloem that transportssoluble organic material made during photosyn-

thesis) that have been weakened by disease orinjury. At low population levels, mountain pinebeetles do not usually successfully attack healthytrees. However, populations of mountain pinebeetle can sometimes erupt to epidemic levelsif stand structure and climatic conditions areappropriate. During outbreaks, large numbersof beetles launch pheromone-mediated attacks(beetles produce a chemical signal that attractsother beetles to the area) on healthy trees andcan overcome the defenses of even healthy trees,sometimes leading to widespread mortality ofhost species. Such outbreaks of mountain pinebeetle are part of a normal boom-and-bust cycle(Amman, 1977).ese cycles are likely to haveoccurred in lodgepole pine forests for thou-sands of years. Epidemics lasting ve to 20 yearsoccur at irregular intervals, aecting large areasby sometimes killing more than 80 percent oftrees that are 10 centimeters in diameter (about4 inches) or greater (Safranyik, 1989).

MOUNTAIN PINE BEETLE IN LODGEPOLE PINE STANDS

Lodgepole pines of the inland West often form extensive, even-aged stands across vast landscapes.

Stands of lodgepole pine are the main hosts for mountain pine beetles (Dendroctonus ponderosae)

and are most aected by the ongoing outbreak.e mountain pine beetle is a native species that has

played an important role in aecting the structure and dynamics of lodgepole pine ecosystems for

millennia (Fuchs, 1999). It is considered one of the most important insects of western forests and can

cause substantial mortality to lodgepole and ponderosa pine as part of the process of forest succession.

ere is considerable variation in the degreeof mortality in lodgepole pine forests aectedeven by severe outbreaks; and even in stands in

which all trees appear to have been killed, liveseedlings, saplings or trees of lodgepole pine

or other species are often present (Rocca andRomme, 2009; Axelson et al., 2009). In densestands, these seedlings, saplings or surviving trees

will be released from competition for light, wateror other resources and will be able to grow rap-idly to re-establish the canopy. In areas whereseedlings or saplings are not present in standsseverely aected by beetles, forest developmentfollowing outbreak will be via the establishmentof new seedlings from seeds in the soil.

Generally speaking, outbreaks of beetles canfacilitate the development of a forest that isstructurally, genetically and compositionallymore diverse (Axelson et al., 2009) and thereforeperhaps less prone to subsequent beetle attack(Amman, 1977). us, despite causing mortal-ity of many individual trees, outbreaks can alsoplay a critical role in ecosystem processes (Ber-ryman, 1982).

Whitney Cranshaw/www

Native mountainpine beetle.


14/40



Insects in Non-Lodgepole Colorado Forest Types

Most western pines are susceptible to mountain pinebeetle, but most of the mortality associated with theongoing outbreak is in lodgepole pine and ponder-osa pine forests. Recently, mountain pine beetle has

caused mortality of whitebark pine at higher eleva-tions. Although ponderosa pine has been less aectedthan lodgepole pine, there is a potential for increasedmortality of ponderosa pine along Colorados FrontRange as the outbreak progresses (Negrn and Popp,2004). High levels of mountain pine beetle wereobserved along the Colorado Front Range in the mid-1970s (Schmid and Mata, 1992). In several regionsof the western United States, past logging and resuppression have led to overly dense and simpliedponderosa pine stands that may be more vulnerableto insect outbreaks (Hessburg et al., 1994; Ferrell,1996; Filip et al., 1996; Maloney and Rizzo, 2002).

e Douglas-r beetle is often a secondary agent thatattacks low-vigor or damaged Douglas rs. Outbreaksusually occur in areas of wind-thrown trees, at sitesdamaged byre or during periods of extreme drought(Furniss and Carolin, 1977). e beetle often attacksDouglas rs that have been infected with root diseaseor defoliated by western spruce budworm (Choristo-neura occidentalis) or Douglas r tussock moth (Orgyia

pseudotsugata). In some areas, the Douglas-r beetlemay help maintain dominance of ponderosa pine by

thinning out individual Douglas rs (Schowalter andWithgott, 2001).

e most important insect of mixed forests of Engel-

mann spruce and subalpine r is the spruce beetle.Usually these beetles are restricted to recently wind-thrown trees or trees weakened by root disease, butthey can reach epidemic levels if the right stand struc-ture and climatic conditions are present (Romme etal., 2006). As mentioned above, there is abundantscientic evidence that epidemics of spruce beetleshave killed trees over extensive areas of forest overthe past centuries (Veblen et al., 1991; Veblen et al.,1994; Eisenhart and Veblen, 2001; Kulakowski andVeblen, 2006).

D

rakeFleege/pow

derhillphotography.com

Bark beetle infestation of a standof lodgepole pines.


15/40



ROOT CAUSES OF INSECT INFESTATIONSIN COLORADOS FORESTS

A warming climate during the last 100 years, par-ticularly in the last few decades, appears to haveplayed a major role in recent insect outbreaks.Numerous scientic studies indicate that climateis an important factor for outbreaks of bark bee-tles of various species (Bentzet al., 1991; Loganet al., 2003; Carroll et al., 2004; Breshears et al.,2005). Dry and warm conditions can stress hosttrees and make them less able to defend againstthe beetles; these same climatic conditions canalso accelerate the growth of beetle populationsand reduce winter mortality (Carroll et al., 2004).

In addition to climate, stand structure is impor-tant in outbreaks of bark beetles. Bark beetlesattack trees by boring through their bark (Saf-ranyik and Carroll, 2006). ey then lay theireggs in the phloem (the inner bark), the larvae

kill the tree by girdling it and by the introduc-tion of blue stain fungus, which can block watertransport. Finally, new beetles emerge and repeatthe cycle. Beetles prefer larger trees because thethicker phloem is better suited to supportingbeetle larvae. During low- to moderate-severityoutbreaks, trees in old or dense stands may bemore susceptible to beetles than trees in youngor less-dense stands because of competition forlimited water and nutrients (Shore and Saf-ranyik, 1992). For example, spruce-r forests

THERE IS NO SINGLE, SIMPLE REASON that a population of bark beetles reaches epidemic

levels. However, warm and dry conditions have been shown to be important to outbreaks in the

Rocky Mountains (Logan and Powell, 2001). Such conditions favor the reproduction and survival

of beetles and also stress host trees. Appropriate stand structure, including forests of a susceptible

age, tree size and density conducive to supporting large numbers of the beetle, also are important

(Shore and Safranyik, 1992; Shore et al., 2000; Safranyik and Wilson, 2006; Bentz, 2008; Kaufmann

et al., 2008; Bentzet al., 2009).

that became established after severe forest resin the late 19th century were less susceptibleto a severe spruce beetle outbreak in the 1940s(Veblen et al., 1994; Kulakowski et al., 2003; Bebiet al., 2003). If this relationship between res andoutbreaks holds, it would imply that the foreststhat are now aected by severe res may be lesssusceptible to outbreaks in subsequent decades.

e current epidemic of mountain pine beetle ispossible, in part, because vast areas of lodgepolepine provide suitable habitat for the mountainpine beetle (Hicke and Jenkins, 2008; Raa et al.,2008).is expanse of mature lodgepole pineforests in the central Rocky Mountains is in largepart the result of widespread severe wildres dur-ing several years of extreme drought in the 19thcentury (Veblen et al., 1994; Kulakowski and

Veblen, 2002; Kulakowski et al., 2003; Veblenand Donnegan, 2006; Sibold et al., 2006; Siboldand Veblen, 2006). Numerous scientic stud-ies have determined that these res were largelyresponsible for current landscape structure andare typical of the res that have shaped lodgepolepine forests in the Rocky Mountains for centuries.us, the presence of suitable habitat for moun-tain pine beetle is a factor that is characteristicof lodgepole pine-dominated landscapes in theRocky Mountains.


16/40



0

FOREST INSECTS AND FIRE:ESSENTIAL ELEMENTS OF ECOSYSTEMS

Although it is widely believed that insect out-breaks set the stage for severe forest res, thefew scientic studies that support this idea

report only a small e

ect, while other studieshave found no increase in re following out-breaks of spruce beetle and mountain pine beetle(Kulakowski et al., 2003; Bebi et al., 2003; Kula-kowski and Veblen, 2007; Simard et al., 2008;

Jenkins et al., 2008; Bond et al., 2009; Tinker etal., 2009).eoretically, the eect of outbreakson subsequent res will vary with the time sincethe outbreak occurred (Romme et al., 2006). Forexample, it is reasonable to expect that foliarmoisture in trees killed by beetles will decreaseand canopy density will be reduced during andimmediately after an outbreak; in subsequent

years, canopy density may be further reducedas dead needles and small branches fall fromkilled trees, which may be associated with anincrease in volume of large fallen fuel; and nallyincreased growth of smaller trees may lead togreater structural heterogeneity and fuel lad-ders (Romme et al., 2006; Bentzet al., 2009).Although such a model is theoretically possible,studies on the inuence of outbreaks on subse-

FIRE INTERACTS WITH BARK BEETLES IN MANY WAYS. Occurrence and severity

ofre following an insect infestation will depend on the forest type, intensity of the outbreak and

time since the outbreak. Despite the long-standing belief that insect outbreaks lead to increased

risk ofre, this assumed link is not well supported by the best available science for most of the for-

ests in Colorado currently aected by outbreaks (Romme et al., 2006; Jenkins et al., 2008; Simard

et al., 2008). Rather, the best available science indicates that the occurrence of large, severe res in

lodgepole pine and spruce-r forests is primarily inuenced by climatic conditions rather than fuels.

quent stand-replacing res over a range of yearssince outbreak have found little or no increasein re occurrence, extent or severity.

Fire and Mountain Pine Beetle Outbreaks

In Lodgepole Pine Forests

Although outbreaks of mountain pine beetle canalter fuel structure (Page and Jenkins, 2007a; Tin-ker et al., 2009; Klutsch et al., 2009), the actualeects of these changes in fuels on subsequentre risk are complex and may be counterintui-tive. Lodgepole stands that experienced highmortality from beetles (more than 50 percent ofsusceptible trees) in the ve to 15 years precedingthe 1988 Yellowstone res had a higher incidence

of crown re than stands that did not have highbeetle mortality (Turner et al., 1999). Stands withlow to moderate beetle mortality had a lowerincidence of high-severity crown res than standswith no beetle mortality. However, because beetlemortality occurs preferentially in older stands, itis not clear whether the changes in re behav-ior were the result of outbreak or pre-outbreakstand structure (Simard et al., 2008)that is,

Nix

Southern Coloradohardwood fire.


17/40



that stands with higher beetle-caused mortality mayhave been generally older stands, which are inher-ently more likely to burn at high severity than are

younger stands because of dierent fuel structure evenin the absence of beetle activity (Renkin and Despain,1992). Beetle-killed lodgepole pine stands, which

were characterized by lower density, experienced sig-nicantly lower re severity compared to adjacentburned areas that had not been aected by beetlesin the 3,400-hectare (8,398-acre) Robinson Fire thatburned in Yellowstone National Park in 1994 (Polletand Omi,2002).A possible explanation is that beetlekill may actually decrease the hazard of high-severitycrown re by reducing the continuity of the canopy.

Lynch et al. (2006) also examined the inuence ofprevious beetle activity on the 1988 Yellowstone resby testing whether re was more likely where beetleshad killed trees than in areas unaected by the bee-

tles.ese researchers found that stands aected bybeetles in 1972-5 had a higher probability of burn-ing but that the increase was only about 11 percentcompared to areas unaected by beetles. In contrast,stands that were aected by beetles in 1980-3 didnot increase the likelihood ofre in comparison touninfested stands (Lynch et al., 2006).

It has been hypothesized that the risk ofre mayincrease only during and immediately after outbreaksof bark beetles when the dry red needles are still onthe trees (Romme et al., 2006). However, Kulakowskiand Veblen (2007) found that ongoing outbreaks

of mountain pine beetle and spruce beetle did notaect the extent and severity ofre and suggestedthat changes in fuels brought about by outbreaksmay be overridden by climatic conditions. Tinkeret al. (2009) examined fuel conditions for 35 yearsfollowing outbreaks of mountain pine beetle in Yel-lowstone National Park.ey documented reducedcanopy moisture content after an outbreak, which wascoupled with reduced canopy bulk density. In simula-tion models ofre behavior, under intermediate windconditions (40 to 60 kilometers per hour, or 12.5 to37 miles per hour), the probability of active crown rein stands recently aected by beetles was signicantlylower than in stands not aected by beetles (Tinkeret al., 2009). If winds were below 40 kph, (12.5 mph)or above 60 kph (37 mph), stand structure had lit-tle eect on re behavior.us, although the canopy

was drier immediately after an outbreak, no increasein re risk was observed likely because of the moreimportant eect of reductions in canopy bulk density.Other independent modeling studies have also pre-dicted a reduced risk of active crown re ve to 60

years after outbreaks, due to decreased canopy bulkdensity (Page and Jenkins, 2007b; Jenkins et al., 2008).

e best available science indicates that outbreaksof bark beetles in lodgepole pine may have little orno eect on subsequent res and may in some cases

actually reduce the risk of

re. In contrast, there isstrong scientic evidence linking severe forest resin lodgepole pine to drought conditions (Bessie and

Johnson, 1995; Sibold and Veblen, 2006; Schoenna-gel et al., 2004).us, the occurrence of severe res inlodgepole pine forests is primarily inuenced by cli-matic conditions rather than changes in fuels causedby bark beetle outbreaks.

Fire and Spruce BeetleIn Subalpine Spruce-Fir Forests

ere is increasing evidence that spruce beetle out-breaks have little or no eect on the occurrence orseverity ofres capable of replacing stands of spruce-r forests (Simard et al., 2008). It is well establishedthat in spruce-r forests, extensive res are highlydependent on infrequent, severe droughts (Buechlingand Baker, 2004; Sibold and Veblen, 2006; Schoen-nagel et al., 2004). Under such extreme droughtconditions, increased dead fuels from bark beetleoutbreaks appear to play only a minor role, if any,in increasing re risk (Romme et al., 2006). After a1940s spruce beetle outbreak that resulted in dead-standing trees over thousands of acres of subalpineforests in the White River National Forest of west-ern Colorado, there was no increase in the numbersofres compared to unaected subalpine forests(Bebi et al., 2003). Likewise, beetle-aected standswere not more susceptible to a low-severityre that

Trees can sometimes fight off bark beetle attacks byblocking their way with flows of sap, called pitch tubes.

WhitneyCranshaw,ColoradoStateUniversity/bugwood.org


18/40



2

spread through adjacent forest several years after theoutbreak subsided (Kulakowski et al., 2003). Dur-ing the extreme drought of 2002, large res aectedextensive areas of Colorado, including some spruce-rstands that were previously aected by this outbreakof spruce beetle (Bigler et al., 2005). Despite the

expectation that these outbreaks would have led toan increased risk of severe res, the outbreak hadonly a minor inuence on re severity (Bigler et al.,2005). Likewise, ongoing outbreaks of spruce beetle(and mountain pine beetle) had no detectable eecton the extent or severity ofres in 2002 (Kulakowskiand Veblen, 2007). ese empirical ndings supportmodeling studies that predict likely reductions inthe probability of active crown re for one to twodecades after high-severity bark beetle outbreaks in

pure stands of Engelmann spruce (Derose and Long,2009). Other independent modeling studies have alsopredicted a reduced risk of active crown re ve to 60

years after outbreaks due to decreased canopy bulkdensity (Page and Jenkins, 2007b; Jenkins et al., 2008).

As with lodgepole pine forests, empirical and model-ing studies from numerous independent researchersindicate that increased risk of wildre is not an inevi-table consequence of bark beetle outbreaks. Instead,climatic conditions appear to have an overriding eecton re regimes in spruce-r forestsso much so thatchanges in fuels brought about by outbreaks of sprucebeetle have little or no eect on re occurrence, extentor severity.


19/40



DOES TREE-CUTTING TO ADDRESS

BARK BEETLE INFESTATIONS REDUCE STANDSUSCEPTIBILITY TO OUTBREAKS?

ere is very little reliable empirical evidence tosuggest that silvicultural treatments can eec-tively stop outbreaks once a large-scale insectinfestation has started. Despite nearly 100 yearsof active forest management to control the moun-tain pine beetle, evidence for the ecacy of thisapproach is scant and contradictory (Wood et

al., 1985). Citing multiple sources, Hughes andDrever (2001) found that most control eortshave had little eect on the nal size of out-breaks, although they may have slowed beetleprogress in some cases and prolonged outbreaksin others.ey also suggest that managementinterventions have never controlled a large-scaleoutbreak. Although control of such outbreaks istheoretically possible, it would require treatmentof almost all of the infected trees (Hughes andDrever, 2001), which may be possible only for asmall infestation.

In some situations, removing infested trees priorto the emergence of broods is recommended toprotect remaining trees. However, the overalleectiveness of this strategy over a large area isunproved (Wilson and Celaya, 1998). Further,in most situations, it is probably not logisticallyfeasible to locate and remove all trees beforethe emergence of adult beetles (Wilson andCelaya, 1998).

TWO IMPORTANT QUESTIONS need to be considered when making forest management

decisions in response to bark beetle outbreaks:

Considering specic goals, what are the ecacies of various management strategies?

What ecological and economic costs do these management strategies impose?

Amman and Logan (1998) point to failedattempts to use direct control measures, suchas pesticides and logging, after an infestationstarts.ey suggest that by the early 1970s, it wasapparent that controlling the extensive mountainpine beetle outbreaks that were occurring in thenorthern Rockies by directly killing the beetles

was not working.

Wickman (1990) detailed the eort to con-trol the mountain pine beetle at Crater LakeNational Park in Oregon from 1925 to 1934.More than 48,000 trees were cut down and thenburned in the last three years of the outbreak.e lesson learned was that once a mountainpine beetle population had erupted over a largearea of susceptible forest, and as long as environ-mental conditions remained favorable, there wasno eective way to stop the beetles until almostall the susceptible trees were either killed orremoved by logging or until climatic conditionsbecame unfavorable for sustaining an outbreak(Wickman, 1990).

e Crater Lake experience is not an isolatedone, as control eorts have been standard prac-tice across the West. Klein (1978) traced severalmountain pine beetle epidemics from beginningto end and detailed the control eorts. More

William M. Cies

Adult pine beetlein egg galleryunder bark.


20/40



4

than 30,000 infested ponderosa pine trees and 20,000infested lodgepole pine trees were treated in 1910and 1911 in the Wallowa-Whitman National For-est in Oregon.e treatments included felling andpeeling, felling and scoring the top, and felling andburning. Chemical methods were employed in the

1940s and 50s. DDT and other pesticides, such aslindane, were sprayed on thousands of acres across theintermountain West. In Operation Pushover, morethan 1,800 acres of lodgepole pine in the WasatchNational Forest in Utah were mowed down by heavytractors linked together, and the surrounding stands

were sprayed with pesticides. In spite of these controlattempts, mountain pine beetle outbreaks contin-ued (Klein, 1978). Klein (1978) ultimately suggeststhat letting infestations run their course may be a

viable option.

Pine beetle suppression projects often fail because

the basic underlying causes (e.g., stand structure, ageof trees, drought) of the outbreak have not changed(DeMars and Roettgering, 1982). Wood et al. (1985)point out that once bark beetles reach epidemic levelsand cause extensive tree mortality, treatments aimedat stopping the outbreak are futile because it is logis-tically impossible to eliminate all suitable habitat orto mitigate the overriding eect of climate.

Large-scale eorts to control beetles are also expen-sive and ecologically harmful.e uncertain benetsof control eorts should be weighed carefully againstcosts (Hughes and Drever, 2001). In fact, much of

the logging in stands infested with bark beetles hasbeen to log merchantable timber. In 1994, then-U.S.Forest Service Chief Jack Wardomas, in testimonybefore the Senate Subcommittee on AgriculturalResearch, Conservation, Forestry and General Leg-islation, acknowledged that the Forest Service logsin insect-infested stands not to protect the ecology ofthe area, but to remove trees before their timber com-modity value is reduced by the insects.ere is nodoubt that timber extraction is a viable and legitimateuse of national forest lands. However, it is importantthat ecosystem management be driven by clear andexplicit goals (Christiansen et al., 1996).

The Case for Thinning: Mixed Reviews

Because stressed and unhealthy trees may be moresusceptible to bark beetles, another managementapproach is to modify stand structure by thinningforests before an outbreak starts. Some thinning stud-ies show success in ameliorating mountain pine beetle

infestations in lodgepole and ponderosa pine forests(Amman and Logan, 1998). But the overall evidenceof the eectiveness of thinning is mixed.

e evidence for thinning

Most evidence supporting thinning as a control forbark beetles is based on tree vigor, not on directlymeasured insect activity in the stand. inning mayincrease tree vigor, which in turn may make trees lesssusceptible to insect infestation. e premise is thatif the trees are healthy and highly vigorous, they maybe able to pitch out the attacking beetles, essentiallyooding the entrance site with resin that can pushout or drown the beetle.

Some studies suggest that thinning forest stands toreduce competition for light and water may increasetree vigor, leaving what appear to be the best trees

and resulting in less successful bark beetle attacks(Sartwell, 1971; Schmid and Mata, 2005; Fettig et al.,2006). Larsson et al. (1983) examined the relationshipbetween tree vigor and susceptibility to mountain pinebeetle in ponderosa pine in central Oregon. Overall,low-vigor trees were more often attacked by beetlesthan high-vigor trees in early stages of outbreaks.

Perhaps the studies by Negrn most conclusively showthat beetle activity is associated with high densities ofstocking (Negrn, 1997; Negrn, 1998; Negrn et al.,2000; Negrn et al., 2001; Negrn and Popp, 2004).ese studies show a positive correlation between

attacked trees and poor growth. Research in Arizona,Utah and New Mexico showed roundheaded pinebeetles (Dendroctonus adjunctus) prefer stands andtrees exhibiting poor growth, and poor growth rates

were positively correlated to dense stands (Negrn,1997; Negrn et al., 2000). Similarly, research inColorados Front Range showed Douglas-r beetlesattacked stands containing a high percentage of basalarea represented by Douglas-r, high tree densitiesand poor growth during the ve years prior to attack(Negrn, 1998; Negrn et al., 2001). Negrn and Popp(2004) reported that ponderosa pine plots in Colo-rados Front Range infested by mountain pine beetlehad signicantly higher tree basal area and density.

Several studies in areas across the west have shownthat thinning reduces the amount of mortality causedby mountain pine beetle in ponderosa pine stands(McCambridge and Stevens, 1982; Fiddler et al., 1989,Schmid and Mata, 2005) and lodgepole pine (Coleet al., 1983; Whitehead, 2005), and some scientists


21/40



and managers recommend thinning as a viable man-agement strategy for managing bark beetles in theseforest ecosystems (Fettig et al., 2007).

Evidence against thinning

Other research has found bark beetles do not prefer-entially infest trees with declining growth (Santoroet al., 2001). Snchez-Martnez and Wagner (2002)studied bark beetles in ponderosa pine forests ofnorthern Arizona to see if dierences in speciesassemblages and relative abundance were apparentfor managed and unmanaged stands. ey found noevidence to support the hypothesis that trees growingin dense stands are more colonized by bark beetles.

Some scientists have suggested caution in using thin-ning to control bark beetles because geographic andclimatic variables may alter the eect (Hindmarch

and Reid, 2001). Hindmarch and Reid (2001) foundthat pine engravers had longer egg galleries, more eggsper gallery and higher egg densities in thinned stands.

Warmer temperatures in thinned stands also contrib-uted to a higher reproduction rate (Hindmarch andReid, 2001). However, pine engravers in Arizona hadthe opposite reaction to a similar thinning experiment(Villa-Castillo and Wagner, 1996).

ere is also evidence to suggest that thinning canexacerbate pest problems. Outbreaks of pine engravershave been shown to be initiated by stand manage-ment activities such as thinning (Goyer et al., 1998).

e process of thinning can wound remaining treesand injure roots, providing entry points for patho-gens and ultimately reducing the trees resistance toother organisms (Paine and Baker, 1993). Hagle andSchmitz (1993) suggest that thinning can be eectivein maintaining adequate growing space and resourcesto disrupt the spread of bark beetles; but note thatthere is accumulating evidence to suggest that physi-cal injury, soil compaction and temporary stress due tochanged environmental conditions caused by thinningmay increase susceptibility of trees to bark beetlesand pathogens.

Even if thinning does alleviate tree stress at the standlevel it is unlikely to be eective against large-scaleinfestations (Safranyik and Carroll, 2006). Preislerand Mitchell (1993) used autologistic regression mod-els to analyze mountain pine beetle colonization inthinned and unthinned lodgepole pine in Oregon.inned plots were initially reported to be unattractiveto beetles; but when large numbers of attacks occurred,

colonization rates were similar to those in unthinnedplots (Preisler and Mitchell, 1993). Similarly, Ammanet al. (1988) studied the eects of spacing and diam-eter of trees and concluded that tree mortality wasreduced as basal area was lowered. However, if thestand was in the path of an ongoing mountain pine

beetle epidemic, spacing and density of trees had littleeect (Amman et al., 1988).

Although thinning may be eective in certain circum-stances, it must signicantly reduce water stress to beeective, which is unlikely during severe droughtsassociated with many outbreaks.us, forest manage-ment, either in the form of searching for and removinginfested trees or thinning forest stands before out-breaks, is unlikely to prevent major outbreaks dueto the inherent diculties of manipulating standstructure over large enough areas of Colorado andthe overriding inuence of climatic stress in driv-

ing outbreaks.

In conclusion, if a bark beetle infestation is relativelysmall and concentrated in a limited area, it may befeasible to reduce the population growth by remov-ing infested trees from a forest stand or by thinninga stand to reduce stress on trees competing for lim-ited nutrients, sunlight and moisture. For example,if a small stand of spruce is blown down by a wind-storm and populations of bark beetles begin growingin fallen logs, then it may be feasible to remove allfallen, infested trees over a small area. However, giventhe climatic requirements for beetle population lev-

els to reach epidemic levels, it is not known whethersuch a situation would lead to an outbreak. In other

words, a small population of beetles is not sucientfor an outbreak to occur. Conversely, under climaticconditions favorable for an outbreak, such as those ofthe past decade, outbreaks of bark beetles can eruptsimultaneously in numerous dispersed stands acrossthe landscape. Unfortunately, even if a growing pop-ulation of beetles is successfully removed from onestand, under outbreak conditions beetles from otherstands are likely to spread over a landscape. Giventhat climate typically favors beetle populations andstresses trees over very large areas, it is unlikely thatmanagement could successfully identify and removeall populations of beetles over an extensive region.inning and associated roads can also have a negativeimpact on wildlife and water quality. Experiment-ing with thinning for bark beetle control should bedone only on a limited scale in areas that are alreadyroadednot in ecologically sensitive roadless areas.


22/40



6

After the Fact:

Post-Disturbance Logging After Outbreaks

Timber production may be an appropriate activity inthe right places at the right times and with the rightmethods, and there may be economic benets from

utilizing some of the dead trees that are now abundanton the landscape, but from an ecological standpointthere is little or no need to remove trees killed byinsects (Romme et al., 2006), and tree removal maycause ecological harm and exacerbate insect outbreaks.Standing snags and fallen logs contribute to a numberof ecological values in forests, including maintenanceof natural forest structures and processes, protectionof soils and water quality and preservation of speciesat risk from the eects of roads, exotic species andhabitat alteration (Romme et al., 2006).

Depending on how it is done, logging after a natural

disturbance (so-called salvage or post-disturbancelogging) can also inadvertently lead to heightenedinsect activity. Specically, logging after insect out-breaks can reduce parasites and insect predators byeectively eliminating their habitat of standing anddowned trees (Nebeker, 1989).erefore, outbreakscould be prolonged because of a reduction in the

eectiveness of the beetles natural enemies (Nebeker,1989). Standing dead trees are important for severalbirds that feed on mountain pine beetles (Steeger etal., 1998), and the widespread removal of dead anddying trees eliminates the habitat required by insec-tivorous birds and other species with the result that

outbreaks of pests may increase in size or frequency(Torgerson et al., 1990). Post-disturbance logging dif-fers from natural disturbance as it tends to decreasehabitat complexity and diversity by removing largelegacies (e.g., standing dead and downed logs), whichcan lead to an increase in insect activity (Hughes andDrever, 2001).

Furthermore, logging following insect outbreakscan seriously damage soil and roots by compactingthem (see Lindenmayer et al., 2008, for synthesis),leading to greater water stress. Soil damage result-ing from logging with heavy equipment can increase

the susceptibility of future forests to insects and dis-ease (Hagle and Schmitz, 1993; Hughes and Drever,2001), reduce conifer regeneration by increasing sap-ling mortality (Donato et al., 2006) and, in general,cause more damage to forests than that caused bynatural disturbance events (DellaSala et al., 2006).


23/40



COLORADOS ROADLESS FORESTS

A broad scale program to treat stands of pon-derosa and lodgepole pine and spruce-r foreststhat have been aected by the bark beetle willrequire an extensive road system. e eects ofroads on ecological systems have been thoroughly

studied and summarized in several recent publi-cations (Trombulak and Frissell, 2000; Forman etal., 2003; Con, 2007). Signicant unintendedconsequences can result from individual roadsand from the cumulative eects of extensive roadnetworks whether they are temporary, long-termtemporary (as in the case of the Colorado pro-posal) or permanent. Notably, temporary roadsmay cause even more short-term damage than

ENVIRONMENTAL CONSEQUENCES OF

MANAGEMENT IN ROADLESS AREAS

Forman (2000) estimated that 20 percent of the land surface in the United States is directly aectedby roads. In addition, in a recent analysis of road impacts, Ritters and Wickham (2003) report that

60 percent of the total land area in the United States is within 382 meters (about 1,250 feet) of the

nearest road. Closer to home, national forest lands within the southern Rocky Mountain ecosystem

have about 45,000 kilometers (28,000 miles) of roads for an average density of about 0.5 km/km2

(Baker and Knight, 2000).is density is relevant to deer and elk populations since large vertebrates

may show a negative threshold response to road density if it is approximately 0.6 km/km2 or more

(Forman et al., 2003). Over the next 20 years, the projected increase in population and road density

for private lands along the Front Range in Colorado is pronounced (eobald, 2005). As private land

areas are developed and transformed, they will no longer provide the environmental benets that they

did prior to development. One consequence is that the value of roadless areas, and public lands in

general, for providing key ecological services increases as private lands are developed for other uses.

permanent roads as they are seldom engineeredto permanent road standards.

In general, researchers have found roads to havemostly adverse eects on ecological and physical

processes and on sh and wildlife populations(Forman et al., 2003).ese eects can be eitherdirect or indirect and chronic or acute as sum-marized below.

Eects on Aquatic Habitats and Fishes

By creating generally impermeable surfaces, roadsdisrupt the natural inltration of water into the

Aron Ralston/Wilder

There are 14.5 macres of Nationawithin Colorado.


24/40



8

soil and increase runoto streams. Roads eectivelyincrease the area of the drainage resulting in increasedpeakows, erosion of channel banks and increaseddeposition of debris (Forman and Alexander, 1998).ese eects are particularly pronounced in moun-tainous regions, especially on high gradient streamsand headwaters (Ziegler et al., 2001). Increased sedi-ment input to streams can result in changes to channelmorphology and channel substrate, and the creationof shallow pools (Beschta, 1978).ese changes tostream structure, an indirect eect of road construc-tion, often adversely aect native sh habitat. Logging

of trees within the riparian zone, coupled with sed-iment deposition and increased light penetration,often result in increases in stream temperature that

will adversely aect cold-water game shes (Myrick,2002). In Colorado, trout that occupy headwaterstreams are particularly vulnerable to logging (Peter-son, 1995). Any road network constructed to thin orharvest insect-infested stands will have to be care-fully engineered to prevent increased sedimentationrates or alteration of hill slope processes (Reid andDunne, 1984; Beschta, 1978). While proper engineer-ing can help mitigate some negative eects, it doesnot mitigate the overall impact of roads on hydrologicprocesses, water ow and fragmentation of habitat.

e eects of roads on watersheds and sh popula-tions are often greater than the eects of wildres(Neville et al., 2009). In general, disturbances such aswildres are known as pulse disturbances, becausetheir eects amount to short-lived bursts, while roads,with their more permanent footprint, contribute toongoing and cumulative press disturbances. e

type of press disturbance from roads occurs becauseroads are permanent sources of sediment and alteredwater ow, whereas re eects may be relatively short-lived (Burton 2005). In fact, some wildre events maypositively aect the natural function of stream ecosys-tems over the long-term (Bisson et al., 2003; Minshall

2003). In contrast, roads and stream culverts in par-ticular often act as barriers to sh passage and mayreduce the genetic variability of native trout popula-tions (Neville et al., 2009).

Importantly, there is high value to leaving dead anddying trees on steep slopes and near streams andrivers. For example, the presence of such trees foreventual recruitment into Colorados stream chan-nels may actually improve stream habitat for shesover the long-term (Riley and Fausch, 1995; Rich-mond and Fausch, 1995). Trees in stream channelsprovide a source of large wood to create complex

habitat structures (e.g., deep, still pools for safe spawn-ing) benecial to many aquatic organisms (Gregoryet al., 1991).

Eects on Terrestrial Landscapes

e major physical results of roads on the terrestrialenvironment are increases in forest fragmentationand disruption of the movement of organisms andow of ecological processes across the landscape (Lin-denmayer and Fisher, 2006). For the most part, roadsare permanent transformations of the landscape; assuch, their eects accumulate over time and space.

As fragmentation increases, so will edge eects withaccompanying decreases in habitat quality (Fahrig,1997). In addition to the direct loss of habitat fromthe imposition of roads, changes to landscape connec-tivity arise because roads act as barriers (or facilitators)to the transport of matter, nutrients and organisms.For example, invasive, exotic plants are often foundalong the edges of roads because seed dispersal isfacilitated by vehicle or human transport (Formanet al., 2003).

e impact of roads at a landscape scale requires aconsideration of cumulative eects. Even if indi-

vidual road segments are constructed to minimizelocal environmental eects, they can still have sig-nicant adverse eects when considered collectivelyand as part of a larger network (or matrix) of roadsand related disturbances (Forman et al., 2003). Inthe context of cumulative eects, of most concernare nonlinear changes to abiotic and biotic processessuch as nutrient transport and wildlife dispersal thatmay show threshold eectsthat is, sudden changes

RussellSnitzer,ColoradoTroutUnlimited

Ecosystems benefit from healthy forests.


25/40



to ecological processes as a result of small changes inthe environment (Con, 2007; Frair et al., 2008).

Direct and Indirect Eects on Wildlife

Eects on wildlife can be categorized as direct or indi-

rect consequences, and population responses may benumerical or behavioral. Direct eects leading to anumerical response by wildlife populations are thoserelated to the immediate loss of habitat due to roadcreation and any wildlife mortality that may resultfrom subsequent vehicle collisions or by increases inharvest. Direct mortality following road constructioncan be signicant. For example, data for white-taileddeer and mule deer suggest that an estimated 720,000animals are killed on U.S. roads each year (Conoveret al., 1995). However, direct eects beyond habi-tat loss may be relatively minor for forest roads thatreceive little trac.

In the context of possible extensive silvicultural treat-ment of insect-infected forests in Colorado, however,it may be that indirect eects are most relevant. Forexample, it is important to recognize that the totalarea aected by a road may be considerably largerthan its immediate footprint. For example, animalsmay avoid roads because they associate them withhuman disturbance or with increased mortality risk.As a result, habitat quality is eectively reduced oversome boundary zone that may extend a considerabledistance from the road (Rost and Bailey, 1979; Row-land et al., 2000; Trombulak and Frissell, 2000; Fahrig

and Rytwinski, 2009). A threshold eect is possiblebecause the overlapping eects of neighboring roadscause a nonlinear accumulation of road eects in thelandscape (Frair et al., 2008).ese eects can leadto reduced and isolated wildlife populations that arevulnerable to continuing decline because of smallpopulation size, loss of landscape connectivity andreduced recolonization of vacant habitat patches.

In a recent meta-analysis, Fahrig and Rytwinski(2009) summarized the eects of roads on the abun-dance of wildlife. Based on 79 studies, with results for131 species, they found that the number of negativeeects of roads on animal abundance outnumberedpositive eects by a factor ofve. Particularly vul-nerable were species such as deer, elk and bear thatmove over large distances and are thus more likely toencounter roads. Highly mobile animals with largebody size often occur at lower densities and havelower reproductive rates and, as a result, are less ableto rebound from small population size. For example,Rost and Bailey (1979) documented a negative eect

of roads on mule deer and elk in Colorado. Morerecent studies on mule deer response to roads in areasof oil and gas development in Colorado also havereported avoidance and reductions in habitat quality(Sawyer et al., 2006). Studies conducted to date in thewestern United States suggest that ungulate popula-

tions may be at risk from increasing road densities.

In summary, there are at least three population-levelconsequences of road construction on wildlife (Bis-sonette, 2002). First, behavior and movement ofanimals are often altered signicantly by the presenceof roads. Second, roads lead to forest fragmentation,which in turn leads to smaller and more isolated pop-ulations vulnerable to further loss.ird, declines inlandscape connectivity cause isolated populations tolose genetic variation and thus the ability to adapt tochanging environmental conditions (Reh and Seiz,1990; Frankham, 2006).

Benefits of Maintaining Roadless Areas

Roadless areas often serve as refuges for wildlife, pro-viding a source of dispersing animals and enablingthem to recolonize depleted populations. In Colorado,wilderness areas and national parks also serve this pur-pose, but most are at high elevations with extensiveareas of rock and ice. As a result, mid- and low-ele-

vation forest communities are poorly represented inprotected areas except for designated roadless areas.In the context of climate change, persistence of nativeplant and animal communities may require the unim-

peded opportunity to migrate to higher elevationsthat unroaded landscapes uniquely provide. ere-fore, protecting native plant communities along entireelevational gradients may be required for successfuladaptation.

Roadless areas may have their greatest value in termsof protecting watersheds that can maintain high waterquality and predictable ows throughout the year. Asan example of the increasing value of water in thewestern United States, consider that a gallon of waterin the grocery store may cost more than a gallon of gas.As water becomes increasingly scarce in the south-ern Rocky Mountains, the hydrological signicanceof roadless areas for providing clean water to down-stream users and oering other ecosystem servicesto local communities will increase. Roadless areasalso serve as refuges for aquatic populations and forreplenishing source populations of degraded biotain downstream aquatic ecosystems (Peterson, 2005).

In addition to the role of roadless areas for watershed


26/40



0

protection and for maintaining landscape connec-tivity, roadless areas are also important as ecologicalbenchmarks for the management and restoration ofthe roaded landscape.ese areas provide referenceconditionsthe condition in the absence of humandisturbance which is used to describe the standard,

or benchmark, against which the current condition iscompared (Snchez-Montoya et al., 2009). Referenceconditions from roadless areas are needed to assessthe eects of management in the roaded landscape.

In combination with wilderness areas and parks,roadless areas act as natural laboratories for study-

ing eects of various environmental stressorsforexample, global climate change, invasive species, airpollution impacts and rein areas not confoundedby management or direct human impacts.

In sum, Colorados 4.3 million acres of inventoried

roadless areas provide unique bene

ts to Colorado-ans. Managing these areas consistent with the 2001Roadless Area Conservation Rule is the highest andbest use of these natural resources and the ecosystemservices they provide in an increasingly human-dom-inated landscape. A rapidly changing global andregional climate makes this even more imperative.


27/40



MANAGING FIRE RISK NEAR COMMUNITIES

VERSUS ROADLESS FORESTS

ough outbreaks do not necessarily increasethe risk ofres in Colorados forests, droughtconditions can lead to a high risk ofres withor without outbreaks. Although a major goalofre hazard mitigation is to protect commu-nities, a recent analysis of the location of fuel

treatments found that only 11 percent of fed-eral fuel treatments in the western United Stateshave been within 2.5 km (1.5 mi) of the wild-land-urban interface, where they would be mosteective at reducing re risk to homes and settle-ments (Schoennagel et al., 2009). Schoennagelet al. (2009), as well as other scientists (Cohen,2000), conclude that greater priority should begiven to treating fuels in and near the wildland-urban interface.

Defensible SpaceFirst Line of Defense

Perhaps the most important component of thewildland-urban interface is the space imme-diately surrounding homesalso known asdefensible space. Building roads and cuttingdown trees far from communities is not likelyto reduce re risk to homes and neighborhoods.Forest Service re expert Jack Cohen states cat-egorically that logging away from communities

LARGE FIRES AND EXTENSIVE GROWTH of residential communities adjacent to re-

prone forests has raised public awareness of the overall risks of forest res. As discussed previously,

res, even severe and extensive ones, are essential and normal components of Colorados forests. Fur-

thermore, rigorous scientic research has found that the inuence of outbreaks on subsequent re

risk in Colorados spruce-r and lodgepole pine forests is minimal (Romme et al., 2006).

does not eectively change home ignitability(Cohen, 2000). Cohen points to a 40-meterzone (about 130 feet) around the home thatdetermines a homes ignitability (Cohen, 1999).Reducing ammable material from the immedi-ate vicinity of structures and replacing ammable

building materials such as wooden decks withnonammable alternatives has been shown toeectively protect structures against re dam-age (Cohen, 1999).

Materials explaining how to protect commu-nities are available from several sources, mostnotably from the national Firewise Communi-ties, a multi-agency eort designed to involvehomeowners, community leaders and planners toprotect people and property. Its Web site (www.rewise.org) oers brochures, videos and articles(Firewise, 2009).

As discussed above, silvicultural treatments (oftenmechanical thinning of merchantable trees) arenot a viable option for controlling bark beetleepidemics. Similarly, conventional timber har-

vests will do little to reduce re risk at any scaleif it removes primarily large trees, because smallertrees, brush and dead fuels often are the major

NFPA Firewise Comm

Houses among tand vegetation.


28/40



2

carriers of a spreading re (Romme et al., 2006).us,to be eective at reducing re hazard to communities,tree-cutting must be executed in a way that removesall ammable material (not just economically valu-able timber) and must be located in the immediate

vicinity of homes and settlements.

Overall, it is going to be much less expensive, moreeective and less damaging to focus re-hazard reduc-tion eorts around communities and homes than it

would be to try to make a wholesale modication offorest structure over large landscapes.


29/40



ere is substantial evidence that logging inroadless areas will not stop bark beetle outbreaksand would be ecologically damaging because

such mechanical treatments would necessitatean expansive and potentially damaging roadsnetwork.e best available science indicates thatonce a large infestation has started, it is not pos-sible to stop the outbreak.

Even forest thinning, which is widely promotedas a solution by reducing tree susceptibility tooutbreaks, has had mixed results and is unlikelyto stem bark beetle epidemics on a large land-scape scale, especially during drought cycles.Further, this type of thinning would not be aone-time treatment, but would require regular

thinning of all treated stands every decade or sobecause thinning tends to promote rapid growthof understory vegetation, making it a potentialfuel ladder. Moreover, too much thinning canmoderate stand climates, which may be favor-able to some beetles, and increase wind speedsadding to crown re spread.

CLIMATE CHANGE and other factors are leading to unprecedented changes in Colorados forest

ecosystems. One likely consequence of a changed climate is increased bark beetle activity leading to

tree mortality over large areas of the state, including its roadless areas. Although ongoing outbreaks

have led to widespread public concern about increased re risk, the best available science indicates that

outbreaks of mountain pine beetle and spruce beetle do not lead to increased risk of re. Neverthe-

less, forests of lodgepole pine and spruce-r are naturally prone to high-severityres during drought

conditions, regardless of the inuence of bark beetle outbreaks.us, there is a need to take eective

steps to protect homes and communities from re risk that is associated with climatic conditions.

e 2001 Roadless Area Conservation Ruleprovides a consistent and scientically basednational standard that safeguards roadless areas

from development. Colorados petition to addressrecent outbreaks of bark beetles by proposinglogging and road building in inventoried road-less areas is inconsistent with the best availablescience and would degrade roadless area values.Colorados 2009 proposed rule would potentiallyput hundreds of thousands of acres of roadlessareas at risk. Furthermore, it is unlikely to beeective in mitigating the risk of insect outbreaksor severe re. If there are cases in which it is nec-essary to remove trees in a roadless area, the 2001rule allows sucient exibility while maintainingneeded rigorous protections for roadless qualities.

Scientists, land managers and residents of Col-orado are concerned about how wildre mightaect our forests and communities. If the goalis to protect communities,re-mitigation eortsshould be focused around those communitiesand homes, not in remote and ecologically valu-

CONCLUSION

Johnny N. Dell/b


30/40



4

able roadless areas. Prioritizing re-hazard reductionaround communities is much less expensive, moreeective and less damaging to roadless values thantrying to make a wholesale modication of foreststructure across the landscape.

Although the current insect outbreaks are very largeand their scope and size may even be unprecedentedin recent history, there is strong evidence that aectedforests will regenerate in time.e forests that are

losing many trees to insect attack will not look thesame in our lifetimes, but current patterns of regrowthindicate that green forests will eventually return inmost locations (Kaufmann et al., 2008; Rocca andRomme, 2009; Axelson et al., 2009).ese forests maylook dierent to us, but beetle-aected forests are still

functioning ecosystems that provide food and shelterfor animals, cool clear water for sh and humans, andirreplaceable refuges for wildlife from the eects oflogging, road building and climate change.

ACKNOWLEDGMENTS

For helpful comments on earlier drafts of this report, we thank J. Hicke, D. Jarvis, W.H. Romme, T. Schoen-nagel, M. Simard, R. Spivak, T. Veblen and C. D. Williams.e authors of this report take sole responsibilityfor its content.


31/40



L I T E R A T U R E C I T E D

Alfaro, R.I., G.A. Van Sickle, A.J.omson and E. Wegwitz. 1982. Tree mortality and radial growth losses

caused by the western spruce budworm in a Douglas-r stand in British Columbia. Canadian Journal of For-est Research 12:78087.

Amman, G.D. 1977.e role of the mountain pine beetle in lodgepole pine ecosystems: Impact of succes-sion. Ine Role of Arthropods in Forest Ecosystems: Proceedings in the Life Sciences, W.J. Mattson, ed. Pp. 318.New York: SpringerVerlag.

Amman, G.D., M.D. McGregor, R.F. Schmitz, R.D. Oakes. 1988. Susceptibility of lodgepole pine to infesta-tion by mountain pine beetles following partial cutting of stands. Canadian Journal of Forest Research 18:68895.

Amman, G.D., and J.A. Logan. 1998. Silvicultural control of mountain pine beetle: Prescriptions and theinuence of microclimate.American Entomologist44(3):16677.

Axelson, J. N., R.I. Alfaro et al. 2009. Inuence ofre and mountain pine beetle on the dynamics of lodge-pole pine stands in British Columbia, Canada. Forest Ecology and Management257(9):187482.

Baker, W.L., and R.L. Knight. 2000. Roads and forest fragmentation in the southern Rocky Mountains. Pp.97122. In Forest Fragmentation in the Southern Rocky Mountains, R.L. Knight, F.W. Smith, S.W. Buskirk, W.H.Romme and W.L. Baker, eds. University Press of Colorado.

Bebi, P., D. Kulakowski and T.T. Veblen. 2003. Interactions between re and spruce beetles in a subalpineRocky Mountain forest landscape.Ecology 84(2):36271.

Bentz, B.J., J.A. Logan, J.C. Vandygri, J.C. 2001. Latitudinal variation in Dendroctonus ponderosae(Coleop-tera: Scolytidae) development time and adult size. Canadian Entomologist133:37587.

Bentz, B. 2008. Western U.S. Bark Beetles and Climate Change. U.S. Department of Agriculture, U.S. ForestService, Climate Change Resource Center. www.fs.fed.us/ccrc/topics/bark-beetles.shtml.

Bentz, B, C.D. Allen, M. Ayres, E. Berg, A. Carroll, M. Hansen, J. Hicke, L. Joyce, J. Logan, W. MacFarlane, J.MacMahon, S. Munson, J. Negrn, T. Paine, J. Powell, K. Raa, J. Rgnire, M. Reid, W. Romme, S. Seybold,D. Six, D. Tomback, J. Vandygri, T. Veblen, M. White, J. Witcosky and D. Wood. 2009. Bark Beetle Outbreaksin Western North America: Causes and Consequences. University of Utah Press, 42 pp.

Berryman, A.A. 1982. Population dynamics of bark beetles. In Bark Beetles in North American Conifers, J.B.Mitton and K.B. Sturdgeon, eds. Pp. 264314. Austin: University of Texas Press.

Beschta, R.L. 1978. Long-term patterns of sediment production following road construction and logging in

the Oregon Coast Range. Water Resources Research 14:10116.

Bessie, W.C., and E.A. Johnson. 1995. e relative importance of fuels and weather on re behavior in sub-alpine forests.Ecology 76:74762.

Bigler, C., D. Kulakowski and T.T. Veblen. 2005. Multiple disturbance interactions and drought inuence reseverity in Rocky Mountain subalpine forests.Ecology 86:301829.


32/40



6

Bisson, P.A., B.E. Rieman, C. Luce, P.F. Hessburg, D.C. Lee, J.L. Kershner, G.H. Reeves and R.E. Gresswell.2003. Fire and aquatic ecosystems of the western USA: Current knowledge and key questions. Forest Ecologyand Management178:21329.

Bissonette, J.A. 2002. Scaling roads and wildlife: the Cinderella principle. Zeitschrift fr Jagdwissenschaft48:20814.

Black, S.H. 2005. Logging to Control Insects:e Science and Myths Behind Managing Forest Insect Pests.XercesSociety for Invertebrate Conservation. Portland, Ore.

Bond, M.L., D.E. Lee, C.M. Bradley and C.T. Hanson. 2009. Inuence of Pre-Fire Tree Mortality on FireSeverity in Conifer Forests of the San Bernardino Mountains, California.Open Forest Science Journal2:417.

Breshears, D.D., N.S. Cobb, P.M. Rich, K.P. Price, C.D. Allen, R.G. Balice, W.H. Romme, J.H. Kastens, M.L.Floyd, J. Belnap, J.J. Anderson, O.B. Myers and C.W. Meyer. 2005. Regional vegetation die-oin response toglobal-change-type drought. Proceedings of the National Academy of Sciences 102(42):151448.

Buechling, A., and W.L. Baker. 2004. A Fire History From Tree Rings in a High Elevation Forest of RockyMountain National Park. Canadian Journal of Forest Research 34:125973.

Burton, T.A. 2005. Fish and stream habitat risks from uncharacteristic wildlife: Observations from 17 yearsofre-related disturbances on the Boise National Forest, Idaho. Forest Ecology and Management211:1409.

Carroll, A.L., S.W. Taylor, J. Regniere and L. Safranyik. 2004. Eects of climate change on range expansionby the mountain pine beetle in British Columbia. T. Shore, J.E. Brooks and J.E. Stone, eds.Mountain PineBeetle Symposium: Challenges and Solutions. Natural Resources Canada, Canadian Forest Service, Pacic For-estry Centre, Kelowna, B.C. Pgs. 22332.

Christensen, N.L., A.M. Bartuska, J.H. Brown, S. Carpenter, C. DAntonio, R. Francis, J.F. Franklin, J.A.MacMahon, R.F. Noss, D.J. Parsons, C.H. Peterson, M.G. Turner, R.G. Woodmansee. 1996. Report of theEcological Society of America, Committee on the Scientic Basis for Ecosystem Management.Ecological

Applications 6:66591.

Clancy, K.M. 1993. Research approaches to understanding the roles of insect defoliators in forest ecosystems.In Sustainable Ecological Systems: Implementing an Ecological Approach to Land Management. Pp. 2117. USDAGeneral Technical Report RM-247.

Con, A.W. 2007. From roadkill to road ecology: A review of the ecological eects of roads.Journal of Trans-port Geography 15:396406.

Cohen, J.D. 1999. Reducing the Wildland Fire reat to Homes: Where and How Much? USDA, USFS,General Technical Report PSW-GTR-173.

Cohen, J.D. 2000. Preventing Disaster: Home Ignitability in the Wildland-Urban Interface.Journal of For-estry. 98:1521.

Cole, W.W., D.B. Cahill, G.D. Lessard, 1983. Harvesting Strategies for Management of Mountain Pine Bee-tle Infestations in Lodgepole Pine: Preliminary Evaluation, East Long Creek Demonstration Area, ShoshoneNational Forest, Wyoming. RP-INT-318. USDA, USFS, Intermountain Research Station, Ogden, Utah. 11 pp.

Conover, M.R., W.C. Pitt, K.K. Kessler, T.J. DuBow and W.A. Sanborn. 1995. Review of human injuries, ill-ness, and economic losses caused by wildlife in the United States. Wildlife Society Bulletin 23:40714.


33/40



Dahlsten, D.L. 1982. Relationships between bark beetles and their natural enemies. In Bark Beetles in NorthAmerican Conifers, J.B. Mitton and K.B. Sturgeon, eds. Pp. 14082. University of Texas Press, Austin.

DellaSala, D.A., J.R. Karr, T. Schoennagel, D. Perry, R.F. Noss, D. Lindenmayer, R. Beschta, R.L. Hutto, M.E.Swanson and J. Evans. 2006. Post-re logging debate ignores many issues. Science314:512.

Donato, D.C., J.B. Fontaine, J.L. Campbell, W.D. Robinson, J.B. Kau

man, B.E. Law. 2006. Post-Wild

reLogging Hinders Regeneration and Increases Fire Risk. Science311:352.

DeMars, C.J. Jr., and B.H. Roettgering. 1982. Forest Insect and Disease Leaet 1: Western Pine Beetle. USFS,Pacic Southwest Region. www.na.fs.fed.us/spfo/pubs/dls/we_pine_beetle/wpb.htm.

DeRose, R.J., and J.N. Long. 2009. Wildre and spruce beetle outbreak: Simulation of interacting disturbancesin the central Rocky Mountains.Ecoscience16:2838.

DeVelice, R.L., and J.R. Martin. 2001. Assessing the extent to which roadless areas complement the conser-vation of biological diversity.Ecological Applications 11(4):100818.

Eisenhart, K., and T.T. Veblen. 2000. Dendrochronological identication of spruce bark beetle outbreaks in

northwestern Colorado. Canadian Journal of Forest Research 30:178898.

Fahrig, L. 1997. Relative eects of habitat loss and fragmentation on species extinction.Journal of WildlifeManagement61:60310.

Fahrig, L., and T. Rytwinski. 2009. Eects of roads on animal abundance: an empirical review and synthesis.Ecology and Society 14:21.

Ferrel, G.T. 1996.e Inuence of Insect Pests and Pathogens on Sierra Forests. In Sierra Nevada EcosystemProject: Final Report to Congress, Vol. II. Davis, Calif. Pp. 117792.

Fiddler, G.O., D.R. Hart, T.A. Fiddler, P.M. McDonald. 1989. inning Decreases Mortality and IncreasesGrowth of Ponderosa Pine in Northeastern California. RP-PSW-194. USDA, USFS, Pacic Southwest

Research Station, Berkeley, Calif. 11 pp.

Filip, G.M., T.R. Torgersen, C.A. Parks, R.R. Mason and B.E. Wickman. 1996. Insects and disease factorsin the Blue Mountains. In Search for a Solution: Sustaining the Land, People and Economy of the Blue Mountains,R.G. Jaindl and T.M. Quigley, eds. Pp. 169202. American Forests, Washington.

Fettig, C.J., J.D. McMillin, J.A. Anhold, S.M. Hamud, R.R. Borys, C.P. Dabney, S.J. Seybold, 2006.e eectsof mechanical fuel reduction treatments on the activity of bark beetles (Coleoptera: Scolytidae) infesting pon-derosa pine. Forest Ecology and Management230:5568.

Firewise (2009). www.rewise.org.

Forman, R.T.T. 2000. Estimate of the area aected ecologically by the road system of the United States.Conservation Biology 14:315.

Forman, R.T.T., and L.E. Alexander. 1998. Roads and their major ecological eects.Annual Review of Ecol-ogy and Systematics 20732.

Forman, R.T.T., D. Sperling, J.A. Bissonette, A.P. Clevenger, C.D. Cutshall, V.H. Dale, L. Fahrig, R. France,C.R. Goldman, K. Heanue, J.A. Jones, F.J. Swanson, T. Turrentine and T.C. Winter. 2003. Road Ecology:Science and Solutions. Washington, D.C.: Island Press.


34/40



8

Frair, J.L., E.H. Merrill, H.L. Beyer and J.M. Morales. 2008. resholds in landscape connectivity and mor-tality risks in response to growing road networks. Journal of Applied Ecology 45:150413.

Frankham, R. 2006. Genetics and landscape connectivity. K.R. Crooks and M. Sanjayan, eds. In ConnectivityConservation, Cambridge University Press. Pp. 7296.

Fuchs, M. 1999. e Ecological Role of the Mountain Pine Beetle(Dendroctonus ponderosae):A Description ofResearch From the Literature. Prepared by Foxtree Ecological Consulting for British Columbia Parks Service,Victoria, B.C.

Furniss, R.L., and V.M. Carolin. 1977. Western forest insects. Misc. Publication No. 1339. USDA, USFS, Wash-ington, D.C. Pp. 35361.

Goyer, R.A., M.R. Wagner and T.D. Schowalter. 1998. Current and proposed technologies for bark beetlemanagement.Journal of Forestry 96 (12):2933.

Gregory, S.V., F.L. Swanson, W.A. McKee and K.W. Cummins. 1991. An ecosystem perspective of riparianzones. Bioscience41:54051.

Haack, R.A., and J.W. Byler. 1993. Insects and pathogens, regulators of forest ecosystems.Journal of Forestry91(9):327.

Hagle, S., and R. Schmitz. 1993. Managing root disease and bark beetles. In Beetle-Pathogen Interactions inConifer Forests,T.D. Schowalter and G.M. Filip, eds. Pp. 20928. New York: Academic Press.

Hessburg, P.F., R.G. Mitchell and G.M. Filip. 1994. Historical and Current Roles of Insects and Pathogens inEastern Oregon and Washington Forested Landscapes. USDA, USFS, General Technical Report PNW-GTR-327.

Hicke, J.A., and J.C. Jenkins. 2008. Mapping lodgepole pine stand structure susceptibility to mountain pinebeetle attack across the western United States. Forest Ecology and Management225:153647.

Hindmarch, T.D., and M.L. Reid. 2001. Forest thinning aects reproduction in pine engravers (Coleoptera:

Scolytidae) breeding in felled lodgepole pine trees.Env

Insect and Roadless Forests

Documents

Transcript of Insect and Roadless Forests