simplified forest management to achieve watershed …coastrange.org/documents/forestreport.pdf ·...

a report of the

National Wildlife Federation

sponsored by the Bullitt Foundation

The Scientific Panel on

Ecosystem Based

Forest Management:

Jerry Franklin

David Perry

Reed Noss

David Montgomery

Christopher Frissell

S I M P L I F I E D

F O R E S T

M A N A G E M E N T

T O A C H I E V E

W A T E R S H E D A N D

F O R E S T H E A L T H :

A C R I T I Q U E

The Scientific Panel on Ecosystem

Based Forest Management:

Jerry Franklin

David Perry

Reed Noss

David Montgomery

Christopher Frissell

S I M P L I F I E D

F O R E S T

M A N A G E M E N T

T O A C H I E V E

W A T E R S H E D A N D

F O R E S T H E A L T H :

A C R I T I Q U E

National Wildlife Federation418 First Avenue WestSeattle, Washington 98119phone: 206/285-8707email: [email protected]: www.nwf.org

S I M P L I F I E D F O R E S T M A N A G E M E N T T O A C H I E V EW A T E R S H E D A N D F O R E S T H E A L T H : A C R I T I Q U E

N AT I O N A L W I L D L I F E F E D E R A T I O N

Simplified Forest Management to Achieve Watershedand Forest Health: A Critique

The National Wildlife Federation has worked for decades to protect habitats that wildlife andpeople depend upon. As our understanding of habitat function evolves, so too must our pre-scriptions for maintaining their health. This report is a step in that evolution.

Today, the decline of salmon and other natural systems, combined with the growing humanpopulation along the west coast, is forcing us to explore new ways of managing our watershedsand forests. But while the concepts of watershed or ecosystem management are broadly appeal-ing, it is often unclear what they really mean. Thus, NWF asked a multi-disciplinary group ofscientists to explore what kind of management will keep our watersheds and forests functioningproperly. Through their findings, reported here, we seek to prompt an overdue conversationthat can culminate in a stronger scientific footing on which to base management decisions.

A prescription for better management is just one part of a bigger picture. Through this reportand its other ongoing efforts, NWF seeks to build public awareness of the problems facing ourforested watersheds and the strategies for solving them. This will entail working for better laws,helping states and localities to safeguard their own watersheds, promoting efficient use of natu-ral resources, and, most importantly, getting individuals involved. Only in these ways can weprogress in making all our waters fishable, swimmable and drinkable.

The National Wildlife Federation rejects the outdated myth that people must choose betweeneconomic prosperity and healthy natural resources; our communities can only thrive when wehave both. We value the lives and livelihoods of those who work in our forests, we want tobring stability to rural communities, and we want to conserve our renewable resources too. Iurge you to join us in working toward these goals using the common sense conservationapproaches exemplified by the contents of this report.

Thank you to all of those who have helped to make the report possible. And thank you to thosewho will use its findings to help us shape a better future for people and wildlife alike. Together,we can and will make a difference.

Mark Van PuttenPresident and CEO

i



I. Introduction – “Maintaining the Integrity of Forest Ecosystems” ................................. 1What is Old-Growth? .............................................................................................................3

II. “The Importance of Considering the complexity of Natural Forests” ............................. 5Simplified Structure-Based Management ................................................................................5

1. The complexity of natural forests. .................................................................. 6Multiple Structures in Addition to Live Trees .........................................................................6Spatial Pattern as an Important Element of Stand Structure ...................................................7Biological Legacies ................................................................................................................10Richness and Duration of Development Processes ................................................................11

2. The resistance of older forests to catastrophic disturbance. .......................... 16Fire Susceptibility .................................................................................................................17Does Logging in Older Forest Stands Reduce Fire Hazard? ..................................................20Insects and Pathogens ...........................................................................................................20Can Logging in Older Forests Reduce Risk of Serious Infestations? ......................................22Landslides and Debris Flows ................................................................................................23

3. Habitat requirements of species of concern. ................................................. 25Species of Concern ...............................................................................................................25Habitats of Concern .............................................................................................................26

4. Understanding the impacts of roads. ........................................................... 27

5. The importance of regional context. ............................................................ 31

6. The role of reserves in ecological management. ............................................ 33Questions to consider: Natural forest stands are structurally and functionally complex .........36

III. Considerations for Ecosystem-Based Management Approaches ................................... 36Questions to consider:Older forests are less susceptible to catastrophic disturbances than younger forests ...............36Not all species and habitats are of equal concern ..................................................................37Roads dramatically alter forest ecosystems ............................................................................37Regional context is critical to ecologically-based forest management .....................................37Forest reserves are an important component of ecological management ................................37General planning considerations ..........................................................................................38

Scientific Panel on Ecosystem Based Forest Management — Bios ...................................... 39

Literature Cited ................................................................................................................. 41

Contents

ii



Much has been written over the last de-cade about strategies to restore and maintainwhat is most succinctly called the integrity offorest ecosystems. In this context, ecologicalintegrity has been described best as “a system’swholeness, including presence of all appropri-ate elements and occurrence of all processes atappropriate rates” (Angermeier and Karr1994). A primary challenge to managing for-ests for ecological integrity—particularly onpublic lands—has been to find levels of tim-ber and other resource extraction consistentwith a broad set of social goals. These goalsinclude maintaining environmental services(e.g., watershed protection and carbon seques-tration), protecting biological diversity,providing goods (e.g., wood and forage),nurturing aesthetic and spiritual values, andassuring the sustained long-term health ofentire forest ecosystems and landscapes. Sug-gested approaches for achieving a balancebetween extractive and non-extractive manage-ment of forests cover a spectrum ranging fromtotal preservation with no logging at one ex-treme, to active logging of entire landscapes atthe opposite extreme. This critique focuses ona set of widely accepted management ap-proaches that lie at the extractive end of thatspectrum.

The concept that all forests must be silvicul-turally manipulated (logged) and eventuallyreplaced in order to provide desired goods andservices, including the continued health of for-est landscapes, is an old and honored tradition

I. Introduction – “Maintaining theIntegrity of Forest Ecosystems”

among many forestry profes-sionals. The “fully regulated”forest landscape with its “bal-anced” distribution of forestage classes, or developmentalstates, has been a goal and iconof forest management for overa century. Another traditionalview is that forests must be ac-tively replaced, because without humanintervention their ability to provide goods andservices will decline and fire, storm, insects ordisease will eventually destroy them.

Proposals for widespread logging as the mecha-nism to create and provide for all forest valuesare therefore not surprising. These approachescontinue to be advanced by advocates oftimber harvesting under such rubrics as “Struc-ture-Based Management” and “High QualityForestry.” Indeed, such approaches have beenproposed as optimal to achieve forestry goalsin the United States (Oliver et al. 1997). Theseproposals assert that managed forests, includ-ing plantations, are not highly susceptible todestruction by natural disturbances whereasunmanaged forests are. Therefore, proponentsargue that forest reserves are a poor conserva-tion strategy since they will inevitably bedestroyed, and that active logging is thus nec-essary to maintain desired forest values. Aparallel — although often unstated — premiseis that foresters know how to grow new foreststhat will provide desired goods and services.

1

“Intensively managed

forests cannot fully

duplicate the

role of natural

forests.”


N A T I O N A L W I L D L I F E F E D E R A T I O N

The proposition that forest values are protectedwith more, rather than less logging, and thatforest reserves are not only unnecessary, but

undesirable, has great appeal tomany with a vested interest inmaximizing timber harvest.These ideas are particularly at-tractive to institutions andindividuals whose incomes de-pend upon a forest land base.On the other hand, approachesthat involve reserving of a por-tion of the land base, or harvestpractices that leave commer-cially valuable trees uncut toachieve ecological goals, areoften considered much lessdesirable as they reducetraditional sources of timber in-come.

Our interpretation of the sci-entific literature, combinedwith our professional experi-ence, leads us to some verydifferent conclusions aboutappropriate approaches. Scien-tifically based strategies for theconservation of forest ecosys-

tems, with a sound theoretical basis inconservation biology — including biodiversityand critical ecological services —have inevi-tably incorporated reserves along withecologically sensitive management of unre-served areas (e.g., FEMAT 1993). At a regionalor large landscape level, this may be partiallyaccomplished by balancing different prioritiesamong multiple landowners. For example, inthe Pacific Northwest the responsibility for

maintaining terrestrial biological diversityhas been placed largely on federal lands(Tuchmann et al. 1996).

Our objective in this critique is to address thevalidity of the concepts and assumptions sup-porting forest policies and plans that call foractive management of essentially the entire for-est estate, and which specifically reject theconsideration of biological reserves and non-traditional harvest techniques, such asstructural retention. Its advocates have labeledthis approach “structure-based management”or “landscape management” (e.g., Oliver 1981,Oliver and Lippke 1994). (Much of forestrypractice and forest management is determinedand informed by the structural characteristicsof forest stands; we do not wish to see this termappropriated for limited and specific manage-ment proposals). Consequently, we specificallycriticize the “simplified structure-based man-agement” approaches derived from simplestructural models and traditional silviculturalsystems such as clearcutting.

In our view, the assumptions underpinningsimplified structure-based management(SSBM) are not supported by the publishedscientific literature on structural developmentof natural forests, disturbance ecology, land-scape ecology and conservation biology, or bythe relationships between ecosystem structuresand processes. In this report, we review scien-tific findings associated with each of these areaswith particular attention to the over-simpli-fied structural models associated with SSBMand the importance and viability of forestreserves to achieve various ecological goals.

I. Introduction – “Maintaining the Integrity of Forest Ecosystems”

2

“In this report, we

review scientific

findings associated with

the development of

natural forests,

landscape ecology and

conservation biology

with particular

attention to the over-

simplified structural

models associated with

SSBM and the

importance of forest

reserves to achieve

various ecological goals.



We agree that many issues raised by theproponents of SSBM about ecosystem man-agement are valid and require attention. Forexample:

• Reserve strategies must account for habitatloss through natural disturbances. Inrecognizing this point, we also note thatmanagement implications will be differentfor areas with different disturbance regimes.

• Fire exclusion, high-grade logging and ac-tive management to create fully-stocked haveincreased the susceptibility of many dry for-est types in the West to catastrophicdisturbance, a situation that puts importanthabitat at risk. Well-planned, carefully ex-ecuted thinning of smaller trees may helpreduce the risk of stand-replacement fire.

Thinning densely stocked, young to mid-agedstands can increase tree vigor, encourage un-derstory growth, and enhance structuralcomplexity. Management regimes such as thoseinvolving variable density thinning and the cre-ation of coarse woody debris and decadencein living trees (e.g., cavities), may be very ef-fective in creating structural complexity inmanaged stands (see, e.g., Carey et al. 1996).

Hence, we support the notion thatsilvicultural activities — structure-based man-agement in the broadest sense — can assist inthe maintenance of biological diversity andother ecological services in managed stands.We do not believe, however, that scientific lit-erature or forestry experience supports thenotions that intensively managed forests can

duplicate the role of natural forests, or thatsufficient knowledge and ability exist to cre-ate even an approximation of a naturalold-growth forest stand.

What is Old-Growth?

Much of this report, and in-deed many of the issuesfacing ecosystem-based man-agement in general, have todo with the functional as-pects of old-growth forestsand the degree to which old-growth functions can beduplicated in forests man-aged for commodities. In

discussions of these issues, itis imperative to defineclearly what is meant by “oldgrowth.” This is especiallyimportant because two quitedifferent definitions appearin the literature.

Because ecological character-istics of old growth vary fromone forest type to another, no single specificdefinition is appropriate. However, US For-est Service definitions for all major types inthe Pacific Northwest adopt five criteria:number of large, old trees; variation in treediameter; degree of tree decadence; amountof large, dead wood; and the degree ofvertical heterogeneity and/or horizontalpatchiness in the canopy. (See NRC 2000for a more thorough review). These are thecharacteristics used in this report.

continued


3

“We do not believe that

the scientific literature

or forestry experience

support the notions that

intensively managed

forests can fully

duplicate the role of

natural forests, or that

sufficient knowledge

and ability exist to

create even an

approximation of a

natural old-growth

forest stand.”



Oliver and Larson (1996) adopt a different,much more restrictive definition. Accord-ing to them, forests don’t enter a “true oldgrowth stage” until the trees that becameestablished immediately after the last stand-initiating disturbance have died and beenreplaced by trees (virtually always shade tol-erant species) that established under thecanopy of the original invaders.

The difference in these definitions cannotbe overstated, especially in forests dominatedby long-lived, fire resistant conifers such asDouglas-fir, ponderosa pine, longleaf pine,and redwood. In these forest types, treesestablished after a given disturbance char-acteristically live many hundreds of years,and the stands typically exhibit the set ofcharacteristics most commonly associatedwith old growth by their 200th year (NRC2000), remaining in an old-growth phasefor centuries thereafter and, in some cases(e.g. ponderosa pine), in perpetuity barringa change in climate or disturbance regime.Using Oliver and Larson’s definition, how-ever, the old forests of Douglas-fir andponderosa pine that dominated much of thewest when EuroAmericans arrived wouldnot have been “old growth,” and in fact verylittle old growth would have existed.

Names such as “old growth” are a conve-nient and necessary shorthand for discussingwhat often are complex and variable sets ofconditions. Whatever we choose to call aparticular stage of forest development, thebasic question remains the same: what arethe relationships among its structure, itsfunctions, and the objectives we are tryingto achieve.

In setting forth the following critique andmanagement considerations, we have, bynecessity, summarized a great deal of infor-mation. Whole texts have been written onsubjects restricted to short summary analy-ses in this report. We urge readers to consultthe extensive bibliography when preparingor critiquing forest management plans orprescriptions. In addition, our examplesdraw largely from the forest ecosystems ofthe Pacific Northwest with which we aremost familiar. Our generalized conclusionsand considerations, however, will apply tomany additional forest types throughoutthe U.S.


4



The following critique is organized aroundsix key considerations that we believe areessential to ecologically based forest manage-ment, but which the simplified structure-basedapproaches address either inadequately or notat all. These are:

(1) The complexity of natural forests;(2) The resistance of older forests to

catastrophic disturbance;(3) Habitat requirements of species of

concern;(4) The importance of understanding

the impacts of roads;(5) The importance of regional context;

and(6) The role of reserves in ecological

management.

The literature describing and promotingSSBM is somewhat inconsistent, but review-ing it we found some common themes thatrepresent what we consider serious errors ofcommission or omission. Perhaps more impor-tantly, these six considerations respond to thethemes land managers and policy makers havemost often taken from the literature of SSBMand applied in management plans or silvicul-tural practice. We offer examples from severalregions but our discussion is most pertinentto the forests of the Pacific Northwest, both“westside” and “eastside” (i.e., both west andeast of the Cascade-Sierra cordillera).

II. “The Importance of Consideringthe complexity of Natural Forests”

Simplified Structure-BasedManagement

Simplified Structure-Based Management(SSBM) is not a single theory or practice. Itis a set of loosely associated forestry conceptsdrawn principally from traditional silvicul-tural science, and applied to landscape-levelforest management. SSBM relies on tradi-tional silvicultural techniques—harvest,thinning, chemical application (herbicideand pesticide), and pruning—to create sal-able timber and other forest “products,”including wildlife habitat. SSBM is, in fact,simply an updated version of the traditionalforesters’ “regulated forest” with a few standstructural types substituted for age classes.

Management plans relying on SSBM gen-erally share the following characteristics:

• Recognition of only four or five foreststand structures: Savanna, StandInitiation, Stem Exclusion, UnderstoryRe-initiation, and Old Growth.

• Use of intensive silvicultural managementtechniques in lieu of natural forest man-agement approaches.

• Reliance on intensive management tomove the forest through each of the standstructures.

• Intensive management of stands acrossthe forested landscape to provide a con-stant flow of wood products and wildlifehabitat diversity.

5



1. The complexity of naturalforests.

Simplified structural models(e.g., Oliver 1981) gloss overmany complexities and painta dangerously abstracted pic-ture of forests, potentiallyleading to poor forest manage-ment decisions. One of thegreatest revelations in forestecosystem science over the pastseveral decades has been therecognition of structural com-plexity and its importance inproviding habitat and regulat-

ing natural processes at a variety of scales. Withrare exceptions, natural forests are diverse interms of individual structures and are highlyheterogeneous in their spatial arrangement. In-deed, more than three decades of ecologicalresearch has shown that natural forests are farmore than a collection of living trees distrib-uted uniformly over a landscape. Managingfor ecological functions requires recognizingthe complex and varied structures found in avariety of spatial patterns and scales in naturalforests, and understanding the implications ofdifferent structural patterns for habitat andecological processes. Those numerous pro-cesses include gene flow, migration, hydrology,nutrient cycling, the spread of disturbances,herbivory, predation, and system recovery fromdisturbance.

The following outline of four key componentsof structural complexity, should be addressedin the development of ecosystem-based

management plans — and are notably absentin simplified structure-based approaches.

Multiple Structures in Addition toLive Trees

Traditional forestry models of stand develop-ment recognize living trees as the only structurein a forest landscape. Yet, managing for eco-logical functions requires a recognition andunderstanding of the complex and varied struc-tures found at a range of scales in naturalforests. Natural forests, or forests managed forall their ecological functions, can provide en-vironmental services ranging from habitat forconservation of biological diversity andconsistent yield of high-quality water, to

non-traditional forest commodities (e.g.,mushrooms and pharmaceuticals), recreation,and high-quality timber.

Even live trees can and should be differenti-ated by such variables as size, canopy position,vigor, species, and levels of decadence. Otherstructural elements important to ecosystemmanagement include standing dead trees(snags), boles on the forest floor, and otherlarge pieces of wood (coarse woody debris)(Franklin, Shugart and Harmon 1987,Harmon et al. 1986, Maser et al. 1988). Suchstructures have great importance as habitat forforest-dwelling species, including rare species,and for other ecological processes. Both stand-ing riparian forests and dead and downed bolesand branches are critical to shaping the habi-tat structure and dynamics of forest streamsand rivers (Harmon et al. 1986, Abbe and

II. “The Importance of Considering the complexity of Natural Forests”

6

“Simplified stand

structural models

present an extremely

limited view of forest

structural development

with respect to time

and ecological process.”



Montgomery 1996, Bilby and Bisson 1997).Recruitment of coarse wood plays a crucial rolein the recovery of streams and rivers fromnatural disturbances such as wildfire (Minshall1997, Gresswell 1999) and volcanism (Sedellet al. and Dahm 1984).

The fact that dead trees and logs are as impor-tant to ecosystem function as living treeschallenges traditional forestry models that treatsuch materials as waste, fire hazards, and me-chanical impediments. To move away fromecologically simplistic models, new forest man-agement regimes must addressquestions such as: How muchcoarse woody debris is needed?and: How many snags invarious stages of decay arerequired? to fulfill important eco-logical functions (see e.g., Maseret al. 1988, Franklin 1992).

Other important individualstructural elements include un-derstory plant communities(layers of shrubs, herbs, andmosses), and surface organic mat-ter or litter layers. The extent ofsuch structural elements mayalso be critical to decomposer or-ganisms, wildlife species, andprocesses that support them. Forexample, the development ofthick litter layers in old-growthforest stands has been identifiedas a key factor in production oftruffle-like fungi which are, in


turn, critical food sources for small mammalssuch as flying squirrels (North et al. 1999).

Spatial Pattern as an ImportantElement of Stand Structure

Traditional models of forest structuraldevelopment, including the simplified struc-ture-based approaches, portray forest stands as

Dead and downed trees and logs are an

important component of forest ecosystems

that goes unacknowledged under most

renderings of SSBM.

7



simple, homogenous structural units (e.g.,Oliver and Larson 1996). They do not addressspatial patterning in structure as an important

element of forest development.Yet this spatial patterningshould be fundamental to anystrategy that purports toaddress a broad range of eco-logical objectives.

Natural forests are inherentlyspatially heterogeneous. For

example, trees of particular sizes or species of-ten are aggregated, rather than distributedeither randomly or uniformly throughout for-est stands. Two important and widely knownspatial patterns are heterogeneity in the hori-zontal and vertical dimensions. Canopy gapsare an important form of horizontal diversitytypically present in mature and old forests (al-though they also may occur in earlier stages ofstand development). Such gaps typically de-velop as individual and small groups of treesdie and function as an important habitat ele-ment. For instance, richness and density of birdspecies are often higher in gaps than in undis-turbed forest; some species are gap specialists(Blake and Hoppes 1986, Martin and Karr1986, Noss 1991, Noss and Cooperrider1994). An important counterpoint to gaps —sometimes called “anti-gaps”— may occur if adense cohort of tree saplings develops in re-sponse to a gap or other disturbed portions ofa stand. The low light levels in these anti-gapstypically results in loss of most, or all, of thevascular plant understory within that portionof the stand.

In many older forests, foliage is distributed es-sentially from ground level to the top of thecrown. This is consistently the case in old-growth forests of the Pacific Northwest.However, depending on forest type and dis-turbance history, trees of different sizes andages may be relatively evenly distributedthroughout the forest, occur in discretepatches, or exhibit some combination of thesetwo patterns (e.g., Parker 1997). Old-growthtrees are usually relatively widely spaced, andtherefore maintain crowns extending down-ward for one-half to three-quarters of theirtotal height. Consequently, especially whenold-growth forests are viewed as spatial mosa-ics, the distribution of foliage is heavilyweighted toward the lower part of the canopy.Therefore, most of the upper crown in thesestands is well lighted. In contrast, dense youngstands typically have all of their foliage in asingle, high, layer — resulting in dense shadethroughout the interior of the stand. Such dif-ferences in canopy structure have profoundecological implications. For example, multiplecanopy layers can increase biological diversityby dramatically expanding the number of avail-able habitats. Vertical heterogeneity also resultsin diverse thermal environments, producing amultitude of microclimates. Furthermore,older forests contain a more diverse andabundant understory of plants which supportmore diverse invertebrate, vertebrate, andfungal communities, including epiphyticcommunities.

Ultimately, most older forest stands developvery complex spatial patterns of small, con-trasting structural patches. This is true of


8

“Ecological research

has shown that

clearcuts are quite

unlike any natural

disturbance.”




almost all old forests in temperate regions, in-cluding the Pacific Northwest coastal forestsof Douglas-fir, western hemlock, and westernred cedar, although the dense vegetation inthese forests can obscure the pattern. Horizon-tal diversity is much more evident in forestssubjected to frequent disturbances of low-to-moderate intensity, such as the pine andmixed-conifer forests of the Sierra Nevada andColumbia Basin, which are disturbed by wild-fire (Franklin and Fites-Kaufmann 1996).

Natural forest development ultimately pro-duces an old-growth state in which stands arecomposed of many small but distinctive struc-tural units. Traditionally, foresters haverecognized these structural units as “stands”but functionally, the old-growth stand is theentire forest patchwork. In effect, to under-

Biological legacies—such as this down and

decaying Jeffrey pine—provide continuity

from one forest stand to the next.

stand the habitat functions, reproductivepotential, and ecological services characteris-tic of older natural forests, one must look atthe entire mosaic of structural units orpatches — from the functional viewpoint, themosaic is the stand. Foresters have tradition-ally recognized and proposed managing eachstructural unit as a stand or, alternatively, tohomogenize them through even-aged manage-ment. This difference in perspective betweentraditional silviculture and more ecologicallybased approaches to forest management, is oneof reasons for the intense arguments over thenature of old forests in places such as the

9



interior Columbia Basin and Sierra NevadaRange (Sierra Nevada Ecosystem Project1996-1997).

Biological Legacies

Many of the structures and organisms foundin natural forest systems originate in preced-ing stands. Such “biological legacies” includesnags, living trees, soil fungi, and a host of otherorganisms that survive stand-regeneratingdisturbances. Biological legacies provide con-tinuity from one forest stand to another andacross disturbance episodes, greatly enrichingthe structural complexity of young regenerat-ing stands. Simplified stand-structure models(e.g., Oliver 1981) generally omit the conceptof legacies.

One of the myths of traditional timber-focusedforest management is that clearcutting mim-ics fire. Ecological research, however, hasshown that clearcuts are quite unlike any natu-ral disturbance. In contrast to traditionalclearcuts that retain essentially none of theharvested stand, natural disturbances rarelyclear away or even kill all elements of the pre-ceding stand. Wildfire converts many treesfrom living to standing-dead and downedwoody debris with varying — but usuallysmall — amounts of organic material con-sumed by fire. Conditions following fires oftenfavor the establishment of shade-intolerantspecies, such as Douglas-fir. However, many,if not most, natural Douglas-fir stands estab-lished following wildfire incorporate survivinglarge old trees — legacies from the previousstands.

Catastrophic windthrow typically convertsoverstory trees to logs and woody debris onthe forest floor. Quite unlike a clearcut, noneof the organic matter in windthrow events is

Unlike clearcuts, wildfire disturbances often

leave dead trees standing and unconsumed

wood debris on the forest floor.


10



consumed or removed by the disturbance. Ifthe stand has an understory of trees that haveadvanced far enough in their regeneration,then this cohort of trees will respond toaltered light conditions provided by the dis-turbance (these are likely to be at leastmoderately shade-tolerant species). There are

many examples of such stands in the PacificNorthwest, including the dense westernhemlock-dominated stands that developed fol-lowing the 1921 windstorm on the OlympicPeninsula (Henderson et al. 1989).

Richness and Duration ofDevelopment Processes

Finally, and perhaps most importantly, sim-plified stand structural models (e.g., Oliver


Windthrow disturbances leave all

organic matter in the forest system.

11



An example of the “stages” of forest

development derived from a simplified

structure-based model (Oliver et al. 1997)


SAVANNA

Changes in stand structures following growths (dashed arrows) and disturbances (solid arrows).

OPEN DENSE UNDERSTORY COMPLEX

1981) present an extremely limited view of for-est structural development with respect to timeand ecological process. Only half of the pro-cesses and developmental stages we considerimportant are represented in the widely citedmodel (Oliver 1981, Oliver et al. 1997). Thismodel describes all of the post-maturationprocesses of vertical and horizontal diversifi-cation — critical in creating the diverse habitatsand processes of the late-successional forest —simply as an “Old-Growth” or “StructurallyComplex” stage. This provides no insight intothe important developmental and structuralattributes of late-successional forests or howtheir functions might be conserved in man-aged forests.

In places like the Pacific Northwest, many pro-cesses are associated with the structuraldevelopment of natural stands (Spies andFranklin 1991). Each of these processes takes

place over an extended time, rather than indistinct and mutually exclusive stages as sim-plified structural models suggest (e.g., Oliverand Larson 1996). As discussed in the previ-ous section, natural disturbances vary in type,intensity, size, frequency, and homogeneity.Such variation in disturbances results in widelycontrasting starting points for stand develop-ment (in terms of structural legacies), and hasprofound implications for the speed, compo-sition, and density of tree regeneration. Theconsequence of varied time-frames for theseprocesses and successional stages means thatforest management aimed at reproducing (asmuch as possible) the structural complexity ofnatural forests, cannot be designed and sched-uled based on simple time-staged actions.Instead of the generically programmedmanagement steps that the simplified struc-ture-based models call for, forest managementshould aim toward the development of standstructure and successional processes actuallyobserved. The following six critical points out-line the complex structural stages. For furtherdetailed discussions see Franklin et al. andCarey (1999).

12



Cohort Establishment. The time required forthis developmental stage – which produces agroup of individual trees of the same age,established concurrently – varies greatly, de-pending upon seed source, environmentalconditions, and competing vegetation. Regen-eration may be almost immediate and dense(as in the case following many windthrowevents), or may take decades. In some standsregeneration may be insufficient to produce a“fully stocked” forest, and the “self-thinningstage” may never occur.

Canopy Closure. Canopy closure is probablythe most dramatic process in stand develop-ment in terms of the degree and rate of changethat occurs in the ecosystem. Assuming regen-eration of a fully stocked stand, as the overstorytree canopy closes, environmental conditionsat the forest floor undergo major changes inlight, temperature, humidity, and wind speed.Canopy closure mediates important microcli-matic and hydrologic changes that influencethe rate of runoff, and subsequent nutrient loss,soil and channel erosion (see e.g., Berris andHarr 1987).

Competitive Exclusion or Self-Thinning.This can be an extended period of dominanceby the new tree cohort if the new growth isdense. Intra- or inter-specific competitionamong the trees — which results in density-dependent tree mortality — is a major processduring this phase. Overstory dominance andcompetition is most intense early in the self-thinning stage, but eases gradually as the standmatures. Forest ecosystem structure andcomposition is usually simple early in the

self-thinning stage. At this stage, species di-versity typically is low, due to limitedavailability of resources in the understory, andlimited spatial variety in the overstory (Harris1984). Re-establishment of understory plantsand tree regeneration comes late in the self-thinning stage, as mortality and growth ofdominant trees creates increased light levels inthe understory. Many natural stands actuallybypass this stage of development due to lowinitial stocking levels.

Maturation. The maturation stage of standdevelopment has many distinctive structuralfeatures and processes: maturation of thepioneer species cohort, development of shade-tolerant trees in the understory, developmentof minimal masses of coarse woody debris, anda shift from density-dependent to density-in-dependent tree mortality processes. Thedominant tree cohort reaches maturity at thisstage, attaining maximum height and crownspread. It is important to note that this devel-opment requires about 200 to 250 years forDouglas-fir on productive sites in the PacificNorthwest. Douglas-fir typically only have 60-65% of their eventual height at 100 years. Themass of coarse woody debris is typically at itslowest level at this stage in development (Ma-ser et al. 1988) because the inherited legacy oflarge snags and boles has undergone substan-tial decay, and significant inputs from thecurrent stand are just beginning. Developmentof height and mass can be very important increating the potential for other processes, suchas massive uproots.


13



Vertical Diversification. The major processduring this stage is the growth of shade-tolerant associates into intermediate and co-dominant positions in the canopy anddevelopment of decadence in the overstory inDouglas-fir forests. The growth of the shade-tolerant associates gradually results in thedevelopment of a continuous canopy from theground level to the top of the crown. Dou-glas-fir trees redevelop branch systems on themiddle and lower bole from epicormic (de-fine) branches, a process initiated during thematuration stage. The vertical diversificationstage may develop very slowly. Observationsin natural, mature stands of Douglas-fir indi-cate that re-establishment of shade-toleranttree associates and their movement into mid-and upper canopy levels may take several cen-turies (W. Keeton, personal communication,T. Spies, personal communication). Decadencedevelops in earnest through death and break-age in the tops of the dominant Douglas-fir,development of decay via a variety of entrypoints, and damage to residual trees from boleand butt rots.

Horizontal Diversification. This stage is char-acterized by increased horizontal diversificationin environmental and structural conditionswithin a stand. Such diversification resultsfrom many processes including creation ofcanopy gaps by wind, insects, and disease, aswell as establishment of dense patches of shade-tolerant associates. Clearly, gaps are not simplyareas of greater light, but also places wherecoarse woody debris and other resources maybe located. Indeed, the spatial complexity in-troduced by gaps is far greater than would

appear at first glance. Some resources — suchas increased moisture, nutrients and woody de-bris — largely coincide with the gaps, whileother resources —such as increased light andheat — are displaced to the north of the gap(Van Pelt and Franklin 1999).

The horizontal diversification stage shifts astand from homogeneity, and a relatively uni-form distribution of structural features andenvironmental conditions, to a highly hetero-geneous condition, with high levels of nichediversity and therefore, biological diversity. Thecomplexity resulting from both vertical andhorizontal diversification clearly is one of themajor factors producing the combinations ofhabitat conditions needed by species which fa-vor, or require, late-successional forestconditions (Ruggiero et al. 1991, Noss andCooperrider 1994).

Most of the six processes described take placethroughout stand development. In some cases,stages are skipped and multiple pathways offorest development are possible. For example,the development of distinct structural unitswithin a stand (horizontal diversification) maybegin with a stand-regenerating disturbanceand the establishment of the new tree cohort,although it is often identified with the devel-opment of gaps during advanced stages ofstand development. Nevertheless, specific pro-cesses typically dominate at particular pointsand can, therefore, be used to recognizeprocess-based stages in stand development.


14



2. The resistance of older foreststo catastrophic disturbance.

Two implicit assumptions run throughsimplified structure-based managementapproaches (e.g., Oliver 1981) and are madeexplicit in the 1997 Report on Forest Healthof the United States (Oliver et al. 1997): (1)older forests are highly vulnerable to naturalcataclysms, particularly stand-replacementwildfires and pest epidemics and (2) foresthealth can be restored and maintained throughactive management. In fact, quite the oppo-site is often the case. While there are someinstances in which these assumptions hold, ingeneral they are supported neither bydata or experience. Older forests aremore resistant to catastrophic distur-bance than younger stands, andmanagement often has increased,rather than decreased, the severity offire and pests (Perry 1988a, 1998,Franklin et al. 1989).

The questions surrounding vulner-ability and appropriate managementpractices, however, are not a simpleeither-or situation. Management —done with the aim of protecting allvalues — can, in some cases, reducefire hazard and susceptibility to somepests (although that remains largelya hypothesis to be tested). Instancesin which careful management mayconfer health benefits are most com-mon in the dry forest types of theinterior West. But the current vulner-ability of those forests to disturbance

is the result of many factors including fire ex-clusion, high-grade logging, livestock grazingand active conversion of stands from low tohigh levels of tree stocking rather than naturalconditions (Agee 1993, Belsky and Blumenthal1997).

To analyze the relationship between age andvulnerability, and provide managers with anappropriate framework to evaluate susceptibil-ity to disturbance, we focus on several commonnatural disturbances in North American for-ests: fire, insects and pathogens, and landslides.Under each category, we briefly evaluate the


16



effectiveness of management in mitigating orexacerbating the effects of disturbance.

Fire Susceptibility

With the exception of lodgepole pine, andsome other high-elevation species in fire-proneenvironments, old-growth conifer forests in theWest (and old-growth longleaf pine in theSouth) are relatively resistant to stand-destroy-ing wildfires. Depending on the forest type,their resistance can be explained by three pri-mary factors: (1) the ability of older trees tosurvive ground fires, (2) the fact that frequentground fires kept fuels from building to levelsthat would carry fire into crowns (in areas offrequent fire) and (3) the extraordinarily cool,

moist microclimates of the westside forests(i.e., west of the Cascade-Sierra Nevada cor-dillera).

Old-growth forests of ponderosa pine, sugarpine, Jeffrey pine, longleaf pine, and dry-siteDouglas-fir, have a history of frequent groundfires that kept fuels (dead biomass and flam-mable understory trees) from building to levelsthat would support large, stand-destroying fires(Agee 1993, Hermann 1993). Because oldertrees of most of these species have thick bark,they are more likely to survive ground firesthan young trees. And because fuels rarely ac-cumulated to levels that would carry fire intothe tall crowns of the older trees, theseold-growth forests were quite resistant to stand-destroying wildfire (Agee 1993, Perry 1994).

Drawing upon historical surveys, (e.g., Cowlinet al. 1942), Henjum et al. (1994) calculated

that nearly 90% of low and mid-elevation pon-derosa pine forests in eastern Oregon andWashington had been old growth prior to thecommencement of logging. This is similar toconclusions reached about the pre-settlementextent of Sierran mixed-conifer forest (Franklinand Fites-Kaufmann 1996). The only way sucha proportion could be maintained within a re-gime of frequent fire is if older forests werefire resistant. Given the measured age-class dis-tributions, it can be roughly calculated that,on average, stand-replacement disturbanceevents (fire or insects) during the 600-800 yearsprior to 1936, occurred on less than one-tenthof one percent of the landscape per year. (Thatdoes not include small-scale patch regenera-tion, the norm in those forest types.) Evidenceof frequent ground fires in dry forest types isconfirmed by studies that have dated the scarsleft on living trees by fire (e.g., Bork 1985). Incontrast to eastern Oregon and Washington,two general types of fire regime have beendocumented for the ponderosa pine forests inthe Black Hills of South Dakota — frequent,low-intensity and infrequent, catastrophic —depending on location and topography. BlackHills ponderosa pine forests apparentlyexperienced a greater proportion of stand-destroying wildfire than in Oregon andWashington (Shinneman and Baker 1997).

Historically, fires in the west side moist, tem-perate conifer forests—the giant Douglas-firand redwood forests of the Pacific North-west—were less frequent but on average, moresevere than those in dry forest types. Therewere also similarities, most notably the rela-tive fire resistance of old-growth trees.


17



Furthermore, the cool, moist microclimaticconditions found within old-growth stands arenot favorable to the ignition and spread of fire.

Extreme conditions were nec-essary for them to burn and,in fact, major fires that affectedwestside old-growth forests(e.g., the Tillamook and,Yacholt Burns and their mul-tiple reburns) were initiated byhuman activities outside of theold-growth forest and driveninto them by strong winds un-der severe burning conditions.Old-growth westside forestsare very resistant to fire igni-tion and spread. As we discuss

in more detail later, experiencein various Pacific Northwestforest types shows young plan-tations to be more susceptibleto intense crown fires than oldgrowth (Andrews and Cowlin1940, Cowlin et al. 1942, U.S.Forest Service 1988).

Mid- to high-elevation conifer forests includeforest types that are among the mostsusceptible of North American forests to stand-destroying wildfires. Lodgepole pine, alongwith its close cousins jack pine and sand pine,are classic examples of forest types with a highprobability of stand-replacement fires (andbark beetles) before trees reach 200 years ofage. Because of its frequent serotinous conehabit, lodgepole pine in the Rocky Mountainsoften regenerates abundantly following fire,and that species tends to occupy areas withhigh lightning frequency. Such a dynamic

cannot be generalized to other high-elevationconifers, however. Stand-replacement fires re-turn, on average, every 800 years in subalpineforests in Oregon and Washington, and every140-340 years in subalpine forests of Mon-tana (an area with greater frequency oflightening) (Agee 1993). White fir-grand firand mixed conifer forests experienced relativelyfrequent low-intensity fires that reduced fu-els, while sparing many older trees, a dynamicmuch like the dry forests, but with a longerinterval between fires (Agee 1993). Althoughlonger fire-free intervals, and consequentgreater build-up of fuels, suggest a greater prob-ability of stand-replacement fires in true fir andmixed conifers than in ponderosa pine, age-class distributions in the former types suggest

the older forests were not highly susceptibleto crown fires. The 1936 forest survey of east-ern Oregon and Washington classified 96%of white fir stands as “large,” with mostvolume in trees greater than 12 inches diam-eter-at-breast-height (DBH) (Cowlin et al.1942). Seventy-one percent of upper slopemixed conifers was classed as “large.” At mid-and high-elevations in the Interior West it maytake from 50 to over 200 years for a tree toreach 12 inches in diameter (Perry and Huang1998). Large blocks of old-growth forests –rather than large contiguous blocks of younggrowth or highly simplified forests – are thebest scenario for reducing catastrophic wild-fire. Once west-side forests reach 350-400 yearsof age, they tend to be resistant to both fireignition and initial spread, with forests of1,000 years or older even more resistant thanyounger forests (J.F. Franklin, personalobservation).


18

“The current

vulnerability of dry

forests of the interior

West to disturbance is

the result of many

factors including fire

exclusion, high-grade

logging, livestock

grazing and conversion

of stands form low to

high levels of tree

stocking, rather than

natural conditions.”



There is considerable evidence that, in con-trast to old growth, young stands are highlysusceptible to crown fires. Early foresters of-ten commented on the flammability of youngstands. For example, Andrews and Cowlin(1940) documented and commented on firein young second-growth stands in the Dou-glas-fir region. Early foresters in the ponderosapine region (eastern Oregon and Washington)documented a similar pattern. Between1924 and 1932, 32% of the total areaof saw-log stands that experienced fireburned at stand-replacement intensity,whereas 73% of the area of second-growth stands (less than saw-log size)that experienced fire burned at stand-replacement intensity (Cowlin et al.1942). In other words, once a fire wasin a stand, the probability the standwould be destroyed was 2.3 timesgreater in small second growth (lessthan saw-log size) than in large secondgrowth or old growth. That statistic ap-plied to all commercial forest types, notjust ponderosa pine.

More recent fires tell the same story.For example, in the 1987 Silver fire insouthwestern Oregon, 45% of old-growth and mature stands (averageDBH > 21 inches) and only 20% ofsmall saw-timber stands (average DBH12-21 inches) escaped with less than10% mortality. Stands with averageDBH less than 12 inches were mostheavily affected, 65% having “less thanadequate stocking” after the fires (U.S.Forest Service 1988, Perry 1994). In

these and other examples, burned plantationsoften were destroyed.

Various factors may explain the susceptibilityof young stands. Typically, a high level of con-tinuity or contact exists between adjacent treecrowns. Combined with a high volume ofcrowns close to the ground, this makes youngstands highly flammable (Andrews and Cowlin


19



1940). Large amounts of debris, either largeor small in size, left as logging slash, or as lega-cies from earlier fires, can greatly increase firehazards (see, e.g., Huff et al. 1995).

Does Logging in Older Forest StandsReduce Fire Hazard?

The variety of forest types, environmental con-ditions and approaches to logging precludes asimple yes-or-no answer to the question ofwhether or not logging older stands reducesfire hazard. In the Douglas-fir region, and up-per-slope true fir stands of the Interior West,logging is much more likely to increase ratherthan decrease fire hazard for two reasons. First,old-growth stages are generally the most fire-resistant structural stages a forest can attain.Any activity that replaces fire-resistant old treeswith fire-susceptible young trees, or that frag-ments the old-growth forest, increases hazardto the entire stand and surrounding landscape.Second, logging generates slash that increasesfire hazard, and methods to reduce slash carryrisk in themselves. The same arguments holdfor the dry forest types of the West, but theincursion of young trees following fire exclu-sion and livestock grazing introduce acomplication not found in more mesic old-growth forests (or at least not to the samedegree). In forests with a dense growth of un-derstory trees, the probability of crown fireshas almost certainly increased over what it wasunder natural conditions. Using logging to re-duce this hazard must be carefully assessed,because logging may exacerbate other environ-mental problems or create new ones.

Any logging that reduces average tree size, ateither the stand or landscape scale — includ-ing clearcutting, shelterwoods, seed tree cuts,selective cutting of larger trees, or thinning thatlowers average stand diameter —will increasethe risk of stand-replacement fires rather thandecrease it. Thinning only small and interme-diate trees less than 100 years old coulddecrease fire risk, depending on how much newrisk is introduced by logging slash (or its dis-posal). Under-thinning done carefully can bea useful tool to reduce fire risk in dry foresttypes. Logging that compacts soils, createsroads, or depletes nutrient stocks simply tradesone kind of problem for others. The challengeis to alleviate one problem without exacerbat-ing others or creating new ones (Perry 1995).Therefore, each project requires carefulthought and analysis.

Insects and Pathogens

Proposals for active management to maintainforest health (e.g., Oliver et al. 1997) are of-ten predicated on the flawed assumption(similar to that of fire), that older forests aremore susceptible to insects and pathogens thanyoung forests. Susceptibility, however, dependson the particular insect or pathogen, the treespecies, and the environmental context. Insome cases, older trees are more susceptible,in other cases, younger trees are. In most cases,factors other than tree age are the primarydeterminants of susceptibility. The dynamicsof host-pest relations emerge from complexinteractions among climate, efficacy of thenatural enemy complex, uniformity of hostspecies across landscapes, and vigor of


20



individual trees (Perry 1994). In many in-stances, even the assumption that healthy,vigorous trees will be the least susceptible toinsects or disease proves incorrect, for example,vigorous Pacific silver fir and subalpine fir aremore susceptible to balsam woolly aphid thanslow growing specimens.

Among insects, defoliators often preferentiallykill suppressed and intermediate trees ratherthan larger ones in a stand — a phenomenonmore related to tree vigor than age per se (War-ing and Schlesinger 1985), whereas barkbeetles often kill older, senescent trees. Forwestern pine beetle in ponderosa pine, tree-kill is more of a plucking out of scatteredindividuals than a widespread killing of trees.What foresters called a “severe epidemic” ofbark beetles in ponderosa pine during the1930s, resulted in loss of all trees “on areas upto 10 acres” and “losses of 15% of the standover large areas” (Cowlin et al. 1942) — a levelof tree-kill that created holes within a basicallystable landscape. On the other hand, moun-tain pine beetles periodically erupt to kill treesover wide areas in lodgepole pine, a patternassociated with that species’ tendency to growin even-aged stands. Lodgepole is well adaptedto and benefits from such episodic outbreaks.

Among fungal pathogens, obligate parasitessuch as rusts prefer vigorous young trees, while(facultative parasites define) such as some heartrots — which never cause epidemics — preferweakened or slow-growing trees (which mayor may not include older trees). Root rots —one of the disease complexes exacerbated bymodern forestry practices — attack both young

and old trees (Manion 1981). Oregon StateUniversity pathologist Greg Filip (in personalcommunication) attributes an ongoingepidemic of Swiss needle cast in the OregonCoast Range to widespread conversion of oldgrowth to Douglas-fir plantations.

Even in cases where low vigor reduces the pro-duction of chemical defenses by older trees,compensatory factors often come into play atthe ecosystem level. The diversity and com-plex structure associated with older forestsreduces the continuity of hosts, and increasesdiversity and abundance of the natural enemycomplex, both of which act to dampen pestoutbreaks (e.g., Schowalter 1989, 1995). Inboth the northwestern and southeastern U.S.,old growth supported a more complex com-munity of insects that prey on foliage-feedinginsects than that found in younger forests. Inthe Pacific Northwest, old growth supportedthe greatest diversity, and more than threetimes greater biomass of predatory insects perkilogram of foliage than young stands, whereasthe ratio of folivores to predators was approxi-mately 1:1 in old growth, and 7:1 in youngstands (Schowalter 1989). Another studyfound that genetic diversity of foliar endo-phytes, symbiotic fungi that help defend treesagainst insects and pathogens, was highest inolder Douglas-fir trees, intermediate in youngDouglas-fir near old trees, and lowest inyoung trees within a superior tree orchard(McCutcheon and Carroll 1993). One of thebiological legacies that appears to be passedfrom old trees to nearby young ones is a diver-sity of foliar endophytes, which should increasethe ability of young trees to resist pathogens


21



and defoliating insects. Preserving legacies suchas endophytes, which depend on the presenceof old, living trees, is one of the primary

objectives of green-treeretention as an alternative toclearcutting.

Ironically, factors such as heartrots in living trees and snags,large downed logs and canopygaps resulting from tree death,diversify forests and createhabitats for cavity nesting birdsand ants that are importantcomponents of the natural en-emy complex (Torgersen et al.1990). In short, risk to pestand pathogen epidemics is acomplex function of ecosystem

complexity, landscape patterns, and individualtrees. Theories of pest and pathogen manage-ment based solely on tree age are overlysimplistic and, more often than not, in error.

Can Logging in Older Forests ReduceRisk of Serious Infestations?

In limited cases — as in lodgepole pine sus-ceptibility to mountain pine beetle — loggingold-growth stands in order to reduce the riskof serious infestations may be effective. For themajority of situations, however, the experiencein North America (and elsewhere) has beenquite the opposite. Logging, and other activi-ties associated with forestry, (especially roadbuilding, fire exclusion, and stand simplifica-tion) have increased the spread of a numberof insects and fungal pathogens, particularly

the latter (Perry 1998). According to Manion(1981), some of the more common anddestructive root rots (Armillariella mellea,Fomes annosus, and Phellinus wierii) “…areimportant diseases only because of modern for-estry practice…” Indeed, a root diseaseepidemic in Port-Orford-cedar, in southwestOregon and northern California, was spreadprimarily through the construction and use oflogging roads that serve as vectors for sporedispersal (Zobel et al. 1985).

The most striking effect of land-use on defoli-ating insects is associated with the eastern andwestern spruce budworms (Anderson et al.1987, Wickman et al. 1993). During the 20thcentury, depending on location, budworm in-festations have become longer, more extensivein area and have resulted in greater tree death(Perry 1988b, Wickman et al. 1993). In andthe southern boreal zone of eastern Canada,fire exclusion and logging of the valuable old-growth spruce, hemlock and hardwoods,allowed the smaller and shorter-lived balsamfir to spread, effectively setting the lunch tablefor the eastern spruce budworm. In westernNorth America, logging old-growth ponderosapine, along with fire exclusion, allowed grandand white fir — prime hosts of western sprucebudworm (and Armillaria root rot) — tospread, with the same results.

In instances, the right kind of logging can re-duce susceptibility to pests. The best exampleis thinning tightly stocked stands to increaseindividual tree resistance to bark beetles (War-ing and Pitman 1985). Such instances virtuallyalways involve thinning young to mature,


22

“The diversity and

complex structure of

older forests reduces

the continuity of hosts

and increases diversity

and abundance of

natural enemies, both

of which act to dampen

pest outbreaks”



Older forests are not only less susceptible todamage from landslides and debris flows, theyalso mediate or limit the frequency of suchdisturbances to aquatic andriparian environments. To thisend, older forests providestructural soil stability via treeroots, buffer watershed hydrol-ogy, and supply large logs thatanchor natural logjams andlimit the propagation of debris-flow disturbances.

Tree roots reinforce the shearstrength of soils. Mature treeshave substantially greater rootstrength than young trees (Burroughs and Tho-mas 1977, Schmidt 1999). For example, theroot strength of mature Douglas-fir providessufficient cohesion to stabilize even cohesion-less soils on steep slopes (Montgomery et al.2000). Prior to decay, even Douglas-fir stumpsprovide sufficient root strength to hold thinsoils on topographic noses and side slopeswhere soils are relatively well drained. In con-trast, steep unchanneled valleys, or topographichollows, define areas particularly susceptibleto debris-flow initiation following forest cut-ting. Consequently, timber harvest andsubsequent herbicide application that reducethe apparent cohesion of the soil would be ex-pected to accelerate the frequency of shallowlandslides in hollows and steep side slopeswhere landslides typically occur after timberharvest.

Another hydrologic effect of forests has to dowith the forest canopy. The rainfall intercepted

even-aged stands. Removing true firs fromareas they have invaded since fire exclusionwould probably help control spruce budwormoutbreaks (the spruce budworm is actually per-forming the thinning chore itself ). Neither ofthese prescriptions, however, involves cuttingold-growth trees. What is true for the previ-ous explanation regarding managing for fire,also applies to managing for infestations. Anyproposal to improve forest health throughlogging must account for the fact that loggingand associated activities can create problemsthemselves.

Landslides and Debris Flows

Although natural and management relatedlandslides in steep terrain are produced by avariety of factors, it is well established that olderforests are relatively resistant to landslidingwhen compared to younger forests (Sidle etal. 1985). Moreover, the relationship betweenclearcutting in steep terrain and increased fre-quency of debris-flow has been widelydocumented (e.g., Anderson 1954, Bishop andStevens 1964, Swanston 1969, Gray 1970,Fredriksen 1970, Brown and Krygier 1971,Mersereau and Dryness 1972, O’Loughlin1974, Brown and Sheu 1975, Swanson andDyrness 1975, O’Loughlin and Pearce 1976,Swanston and Swanson 1976, Burroughs andThomas 1977, Gresswell et al. 1979, Wu etal. 1979, Wu and Swanston 1980, Sidle et al.1985, Montgomery et al. 2000). The result-ing increased sediment delivery to downslopechannels can trigger both short and longtermchannel response, and impact downstreamaquatic resources.


23

“Any proposal to

improve forest health

through logging must

account for the fact that

logging and associated

activities can create

problems themselves.”

S I M P L I F I E D F O R E S T M A N A G E M E N T T O A C H I E V EW AT E R S H E D A N D F O R E S T H E A L T H : A C R I T I Q U E


by the forest canopy significantly affects de-velopment of pore-water pressure in the soil.Removing the trees and canopy thereby

increases the chance of a land-slide on steep slopes. Olderforests have far greater leaf areain their canopies than recentclearcuts and very youngforests. Hence, a significantfraction of the rain falling onan older forest is interceptedand evaporated, never reachingthe ground (Leonard 1961,Rothacher 1963, Rogersonand Byrnes 1968, Pearceand Rowe 1979, Rowe 1979,Teklehaimanot et al. 1991).The amount of canopy inter-ception varies with both tree

spacing (Teklehaimanot et al. 1991) andrainfall intensity (Rothacher 1963). In aDouglas-fir forest, rainfall interception variesfrom about 20 percent for storms with greaterthan 5 cm (2”) total rainfall, to 100 percentfor very small storms (Rothacher 1963). A 20percent increase in rainfall intensity would re-sult in more frequent debris-flows in clearedforest than would occur under preclearanceconditions.

A forest canopy also may affect the timing ofmoisture delivery to the ground, further re-ducing the intensity of the shortterm rainfall.For example, the canopy architecture of olderforests can buffer the intensity of rain on snowflood events, thus affecting the amount andtiming of storm runoff (Berris and Harr 1987).

In addition, the erosive effects of flood eventscan be dampened by mature timber that pro-vides cohesion to stream banks.

Debris-flows scour steep headwater channelsand deliver both sediment and wood debris todownstream channels, often creating logjamswhere they deposit. The disturbance associ-ated with scour and passage of a debris-flow isgenerally considered to affect aquatic ecosys-tems adversely. Nevertheless, in certain cases,debris-flow deposition also can create fishhabitat (Reeves et al. 1995). In particular,logjams formed either by deposition of debris-flows, or by the direct recruitment of large,“keymember” logs from streamside forests, canstore large amounts of sediment and thereby

expand the extent of alluvial valley bottoms(Keller and Swanson 1979, Abbe and Mont-gomery 1996, Montgomery et al. 1996).Logjams can enhance fish habitat in moun-tain streams by increasing the frequency and/or depth of pools, (Montgomery et al. 1995,Abbe and Montgomery 1996) and in some lo-cations by converting bedrock channels toalluviumfloored channels (Montgomery et al.1996). The size and abundance of inchannelwood debris is central to both of these pro-cesses; debris derived from oldgrowth forestscan trigger profound effects on channel habi-tat, whereas debris derived from smaller treesin plantations provides less, and probablyshort-lived habitat benefits (Montgomery etal. 1995, Montgomery et al. 1996).

The types of logjams that occur in stream chan-nels reflect log size, channel size, and the natureof the processes that deliver wood to streams.


24

“Older forests are not

only less susceptible to

damage from landslides

and debris flows, they

also mediate or limit

the frequency of

such disturbances to

aquatic and riparian

environments.”



Most stable jams are founded on large,keymember logs that anchor natural logjamstructures (Abbe and Montgomery 1996).Application of a simple, modified version ofWigmosta’s (1983) model of the debris-flowimpact force required to break a log, predictsthat the reduction of log size in headwaterchannels will lead, on average, to longer de-bris runout pathways and therefore, greaterdisturbance to aquatic ecosystems. Unfortu-nately, few data are available at present to testthis expectation.

In summary, old-growthforests differ from youngerforests, in that old-growthforests reduce the likelihoodof debris-flows, and if flowsdo occur, they are more likelyto be beneficial because of theinclusion of large wood andlimited runout lengths

3. Habitatrequirementsof species ofconcern.

To provide for biodiversity,among other things, simpli-fied stand structural modelsseek to manage for a varietyof structures across the land-scape (Oliver 1992). At facevalue, this may seem areasonable approach. How-ever, it is predicated on the

questionable assumption that all species andhabitats are of equal management concern.They are not.

Large woody debris from older forests plays

a critical role in the creation of fish habitat

in mountain streams by creating pools

and storing sediment.


25



Species of Concern

Species adapted to early successional stages andedge habitat are often “weedy”(Terborgh 1976, Noss 1983).Many of these are opportunis-tic generalists that flourish in avariety of habitat conditions.Typically, their powers of dis-persal are great, which helpsthem locate in recently dis-turbed areas. Many of thesespecies thrive under physicallyharsh conditions that other,more specialized, forest speciesfind stressful. There aremany examples of disturbance-adapted plant and animal

species in western Oregon and Washington in-cluding red alder, bitter cherry, Scotch broom,brush rabbits, coyotes, red fox, raccoons, black-tailed deer, garter snakes, brown-headedcowbirds – a brood parasite – and songsparrows. Importantly, disturbed habitat con-ditions, and the species dependent on them,usually are abundant in human-dominatedlandscapes, including intensively managedindustrial forests, agricultural lands, and manyurban areas. While some early successionalspecies are of interest to forest managers (e.g.,elk, deer and bear) disturbance-adapted spe-cies are only of concern when they createproblems for sensitive species (e.g., as does thebrown-headed cowbird) (Brittingham andTemple 1983). Of course, many of the weedyopportunists on harvested sites are actuallyexotics that compete with and displacenative species.

As noted earlier, one of the most importantways in which simplified stand structural mod-els and traditional silvicultural practices differfrom natural forest dynamics, is by truncatingforest development before mature, let aloneold-growth conditions, are attained. Conse-quently, species sensitive to human activities,or dependent on old-growth forests, will farepoorly under simplified structure-based man-agement approaches, regardless of whatmanagement planning documents claim. Inthe Douglas-fir region, such sensitive speciesinclude the northern spotted owl, marbledmurrelet, Vaux’s swift, Myotis bats, cavity-nest-ing birds, the northern flying squirrel, andseveral salamander species (Carey 1989). For-est carnivores such as the fisher and Americanmarten also are at risk (Ruggiero et al. 1994).These are examples of species whose numbersand habitats have dwindled under traditionalforest management regimes (Noss andCooperrider 1994) and about which forestmanagers should be concerned.

Ironically, a number of the proponents of thesimplified stand structural models point to theCanadian lynx, wolves, bears, bighorn sheepand several other wide-ranging species asevidence for the need for additional early-suc-cessional habitat (“stand-initiation structures”)(Oliver and Lippke 1994). What these authorsfail to acknowledge is that the major factorlimiting most of these species across their rangeis the lack of large, secure, roadless areas freeof human disturbance (Noss et al. 1996,Weaver et al. 1996, Mladenoff et al. 1999) Theintensive management and high road densitydemanded by traditional silvicultural


26

“Large woody debris

from older forests plays

a critical role in the

creation of fish habitat

in mountain streams

by creating pools and

storing sediment.”



models — and recent plans based on simpli-fied stand structural models — would posesevere threats to these species.

Habitats of Concern

Some proponents of simplified forest standstructural models have warned that where allharvesting is curtailed, there will be a shortageof open habitat and therefore a reduction ofbiodiversity (e.g., Oliver 1992). Even if the ar-gument that open habitat will be scarce iscorrect, the case can be made that the type ofhabitat that is really at risk in this region isunsalvaged, legacy-rich, early-successionalhabitat. For instance, Wisdom et al. (in press),summarizing findings of the Interior Colum-bia Basin Ecosystem Management Project,note that “current early-seral communitieswere found to commonly be devoid of largetree emergents and snags, to have compara-tively high levels of disturbed soil, and containexotic weeds.”

Another assumption implicit in simplifiedstand structural approaches (e.g., Oliver 1992),is that forests managed for commercial timberproduction can also provide the entire suite ofhabitats found in a natural successional se-quence. We are aware of no evidence to supportthis assumption. As discussed earlier, asidefrom the fact that living trees are killed, inten-sive forest management has little similarity tonatural disturbances, which leave a plethoraof structural legacies, usually have a frequencydistribution characterized by many small anda few large events, and hence leave a relativelyvariable patch mosaic (Perry 1998). Moreover,

where economics is the motivation, forests willbe harvested long before reaching old-growth,the developmental stage withgreatest biological diversity.

Silviculture can, and in somecases has been adapted to moreclosely approximate naturaldisturbances, grow trees tolonger rotations, and with thehope of providing habitat forat least some species requiringcomplexly structured forests.These silvicultural approacheshold substantial promise astechniques that provide forhigher levels of native forestdiversity following timber har-vest, both initially and over thelong term. Extensive researchis currently underway andwork during the precedingdecade has already documented substantialdifferences between areas harvested usingtraditional clearcut prescriptions and thoseusing structural retention prescriptions withregards to varied organismal groups such asinvertebrates, lichens and birds. ( e.g., Asreplacement for clearcutting, the Canadiancorporation MacMillan-Bloedel has adoptedthis approach. The Weyerhaeuser Company,who recently purchased MacMillan Bloedel,has agreed to continue this practice.)However, much remains to be learned withregard to how many old-growth species andprocesses can actually be sustained using suchan approach.

“Species sensitive to

human activities, or

dependent on old-

growth forests, will

fare poorly under

simplified structure-

based management

approaches, regardless

of what management

planning documents

claim.”


27



4. Understanding the impactsof roads.

Logging roads have a profoundeffect on forest ecosystems— increasing erosion andstream sedimentation, servingas vectors for diseases andinvasive species, and fragment-ing habitat.

Silvicultural science has longsuffered from a myopic focuson the dynamics of regenera-tion, tree and stand growth,

with much less attention to the logging andtransportation systems necessary to implementits prescriptions. The forest stand structuralmodels scrutinized in this report are no

exception. Several of these approaches empha-size management of the entire forestedlandscape through silvicultural operations. Toaccess every stand across the landscape, exten-sive road systems would need to be built andmaintained. These roads, in turn, would in-troduce a broad suite of environmental impactsto aquatic and terrestrial ecosystems.

As described in the literature, and in the 1997Report on Forest Health in the United States

Traditional forestry practices, including

dispersed-patch clearcutting, produces

highly fragmented habitat conditions

and large amounts of biologically poor

early-successional habitat.


28

“Aside from the fact

that living trees are

killed, intensive forest

management has little

similarity to natural

disturbances.”



(Oliver et al. 1997), these simplified standstructural management approaches appear toassume and require a permanent, high-den-sity road network distributed across thelandscape. Yet discussion of the logging andtransportation systems necessary to implementthese management approaches is conspicu-ously absent in the literature presenting thosemodels (e.g., Oliver and Larson 1996). AsMatthews (1989) and others have pointed out,design, construction costs, maintenance, andthe potential environmental impacts of thenecessary transportation system are central tothe evaluation of any silvicultural system.

Logging roads are now generally recognizedas the most pervasive source of damage tostreams from forest management activities(Furniss et al. 1991, Noss and Cooperrider1994, Frissell and Trombulak 2000). Roadspermanently alter the hydrology of slopes (e.g.,Megahan 1972, Montgomery 1994, Wempleet al. 1996,), and thus contribute to manyforms of hillslope erosion and sediment con-tribution to streams (e.g., Reid and Dunne1984, Hagans et al. 1986, Hicks et al. 1991).Erosion from roads can chronically elevate sus-pended sediment, reducing the growth andsurvival of aquatic species (Newcombe andJensen 1996) and otherwise impair fish (Buck1956). Roads also threaten terrestrial speciesin a multitude of ways (see review in Frisselland Trombulak 2000). For example, forestroads can serve as vectors for tree diseases(Zobel et al. 1985) and non-native weeds(Tyser and Worley 1992). In addition,logging roads provide hunters and otherrecreationists with access to previously remote


Some silvicultural activities including

selective harvest, green tree retention,

and extended rotation, have shown some

promise in providing diverse forest habitats.

29



areas, increasing human harassment,exploitation and inadvertent road-kill of sen-sitive species (see reviews in Noss and

Cooperrider 1994 and Frisselland Trombulak 2000).

Many recent studies havedemonstrated that large forestland tracts that are roadless, orof low road density, supportsensitive fish and wildlife spe-cies (e.g., Brody and Pelton1989, Eaglin and Hubert1993, Thurber et al. 1994,Rieman et al. 1997, Baxter etal. 1999). As such, preserva-

tion or restoration of roadless areas is anessential component of management strategiesdesigned to protect biological diversity(FEMAT 1993, Noss and Cooperrider 1994,

Noss et al. 1999). The success of watershedand aquatic restoration programs in forestedlandscapes will depend upon a series of mea-sures related to roads, including improvedmaintenance and reconstruction of permanentroad obliteration of unnecessary road andavoidance of road construction in watershedsthat are currently road-free. (Weaver et al.1987, Harr and Nichols 1993, Frissell andBayles 1996). Moreover, naturally function-ing watersheds with low road density can serveas regional landscape refugia for sensitive spe-cies and ecosystems (Reeves and Sedell 1992,

Logging roads have a profound effect

on forest ecosystems— increasing erosion

and stream sedimentation, serving as

vectors for diseases and invasive species,

and fragmenting habitat.


30

“Logging roads

are now generally

recognized as the most

pervasive source of

damage to streams from

forest management

activities.”



Noss and Cooperrider 1994, Frissell andBayles 1996).

A silvicultural system that requires an exten-sive road network is inimical to regionalconservation needs for many sensitive and pro-tected species. The repeated harvest entriesdispersed extensively over the landscape of in-tensive silvicultural systems that numeroussimplified structure-based management forestplans require, appear almost certainly to pre-clude the persistence of any large-scaleecological refugia free from human disturbanceRoad systems and their impacts are an impor-tant element of any management strategy, onethat SSBM fails to address.

All of the aforementioned factors — amongothers — must be considered when prescrib-ing an intensive silvicultural managementregime. By failing to account for such factorsand some of the more severe impacts associ-ated with intensive silviculture, the simplifiedstand structure model now being applied tovarious management regimes (e.g., Oliver andLarson 1996) fails to address road systems andtheir impacts.

5. The importance of regionalcontext.

The simplified stand structural model is predi-cated on creating a “balance” of standstructures across the landscape, but does sowithout considering historical managementpractices. It also has been applied without re-gard for regional context (e.g., Oliver 1992,Oregon Department of Forestry 1999).

Prior to European settlement it has been esti-mated that before European settlement,60-70% of the commercial timberland in theDouglas-fir region was coveredby old growth. (see, e.g.,Franklin and Spies 1984 andWimberly et al. 2000). Bycomparison, Norse estimatedthat in 1990, only 13% of theregion’s old growth in the re-mained. Significantly more hasbeen lost since 1990. Similarreductions in old-growthponderosa pine forests have oc-curred in interior Oregon andWashington (Henjum et al.1994). This loss of old growth has serious

consequences for the species dependent on, orclosely associated with this stage, includingthose of Douglas-fir forests cited earlier (Carey1989). A recent study of disturbance historyin forests of the Coast Range (Wimberly et al2000) indicates that the present proportion ofold growth in the region is far below the natu-ral range of variability in the region over thelast 3000 years.

This historical perspective suggests that main-taining — indeed, re-growing and restoring— old growth should be a high priority forforest management on public lands, particu-larly where public and private lands areintermingled, such is the case in the OregonCoast Range, where state forest lands existprimarily as islands within a landscape ofindustrial forests managed intensively for woodproducts on short rotations. These and otherplaces where intensive forestry has been

“Preservation or

restoration of roadless

areas is an essential

component of

management strategies

designed to protect

biological diversity.”


31



practiced long enough to change the overallmatrix of the forest will almost certainly bedeficient in high quality habitats for popula-tions of locally sensitive or rare species. Forexample, old-growth forests and associated spe-cies populations have become highly isolatedin Oregon’s northern Coast Range (Noss 1993,Oregon Department of Forestry 1999).

It is beyond the scope of this report to cri-tique the Oregon Department of Forestry’s(ODF) management plan for their lands inthe northern Coast Range. (A thoroughscientific review has already been done; Hayes1998). It is instructive, however, to brieflymention some aspects of the ODF plan, as itincorporates positive examples of movementaway from traditional intensive management,

toward a more ecologically based silviculture.It also illustrates what many conservationscientists believe to be unwarranted optimismregarding the degree to which a completelymanaged, forested landscape can contribute toconservation. Dealing almost exclusively withsecond-growth stands originating afterwildfire earlier in the century, the ODF planfor the Tillamook State Forest includessignificantly longer rotations than those in sur-rounding industrial lands, with 20%-30% offorest lands targeted to achieve “older forest

Understanding regional context — including

past management on adjacent forested

parcels — is a key to planning appropriate

management activities.


32



structure.” The plan proposes regenerationharvest in a patchwork of different sizes andretains living trees to provide structure.

In terms of reestablishing and maintainingcomplex forest structure, the ODF plan defi-nitely moves in the right direction (thoughquestions remain about some details — suchas level of retention). On the other hand, theplan leaves no lands free from eventual regen-eration harvest. From the standpoint ofconservation biology, this is a significant weakpoint, especially in a region where ownershippatterns provide no other options for reserves(Noss 1993). ODF acknowledges the Depart-ment has a role in regrowing habitat forold-growth associates such as spotted owls andmarbled murrelets, but assumes a priori thatthis goal can be accomplished without reserves,an assumption with which most conservationbiologists would disagree (e.g., Hayes 1998),and that is at best an untested hypothesis.

Practices such as retention harvests and longrotations clearly produce more complexlystructured forests than intensive forest man-agement. But are they alone sufficient toproduce high quality habitat for all the manyold-growth associates? The majority of biolo-gists who reviewed the Tillamook plan weredoubtful (Hayes 1998). We agree that the planinadequately considers the regional and his-torical context of the planning area. Specieswhose populations or metapopulations oper-ate on spatial scales larger than the limitedplanning area, or which require large blocksof late-seral forest, are unlikely to farewell under the ODF plan. Consequently, a

prudent forest manager interested in assuringmaintenance of biological diversity would in-corporate a system of reserved areas. Someconservation biologists have characterized thedecision to fully manage the landscape as a lostopportunity for restoring refugia in the heavilydegraded northern Coast Range landscape. Onthe other hand, if the time ever comes whenreserves are shown (by rigorous and widelyaccepted peer-reviewed science) to be unnec-essary for conservation, a fully managedlandscape might become a viable option forachieving the conservation and timber harvestgoals for the Tillamook, rather than wishfulthinking on the part of forest planners.

6. The role of reserves inecological management.

Some of the simplified stand structural mod-els explicitly dismiss the need for ecologicalreserve areas as part of a healthy landscape. Inplace of reserves, some have called for a “land-scape approach” to ecosystem managementthat entails active management of all forest-lands. Oliver and Larson (1996) define the twoapproaches:

1) The ‘landscape’ approach is based on the‘dynamic’ theory of forest development...and advises active silvicultural manipula-tions to imitate, avoid, and/or recover fromnatural disturbances to maintain the fullrange of stand structures, landscape pat-terns, processes, and species across thelandscape.


33



2) The ‘reserves’ approach establishes largeareas where stands are expected to developtoward the old-growth development

stage over large areas by settingaside large forest areas wherehuman activities are restrictedor excluded...This approachrelies on the ‘steady state’ecological theory. (italics inoriginal)

Such characterizations of thetwo approaches are highly bi-ased and misleading. Oliverand Lippke (1994) claim thatthe steady-state ecologicaltheory underlies the reserve ap-proach and is “outdated.” Theyadvocate the dynamic theoryof forest development, whichthey claim supports the “moreaccepted” landscape approach.In fact, the strategy of creatingand managing protectedareas is the only strategy thathas been shown — at least oc-casionally — to work (Noss

and Cooperrider 1994, Meffe and Carroll1997). Interestingly, the landscapeapproach (Oliver and Lipkke 1994) is actu-ally a steady-state or equilibrium landscapeconcept. The idea of a balanced or targeteddistribution, or percentage, of each of 4 or 5structural types in each 2,000 to 10,000-acre“landscape” is nothing more – or less — thanan updated version of the traditional foresters’“fully-regulated” forest in which a perfect

distribution of age classes provides a steadyflow of timber. Furthermore, the reserve strat-egy does not depend at all on steady-state orequilibrium theory, nor are reserves usually en-visioned as “unmanaged” areas free fromhuman disturbances. The modern literature ofconservation biology is replete with case stud-ies of reserve management prescriptions andchallenges, with few champions of a “hands-off ” approach, and no hint of dependence onsteady-state theory. As pointed out by Nosset al. (1997):

Maintaining ecological processes at appropri-ate levels usually requires active managementand, in many cases, restoration. In most if notall conservation plans, we cannot count onnatural processes operating effectively if we es-tablish reserves and then leave them entirelyalone. This problem arises largely becausemany natural processes operate on spatial scalesmuch more vast than our reserve networks.

Perpetuation of natural disturbance regimes —not simply preservation of a particular seralstage — is often discussed as a design and man-agement goal for protected areas (Pickett andThompson 1978, Baker 1992, Noss andCooperrider 1994). On the other hand, aspointed out earlier, if examination of theregional landscape context shows that old-growth is depleted — which is the case in mostforest regions — then protection and restora-tion of old-growth forests in reserves is alegitimate management goal, and casts doubton analyses that dismiss reserves as part of an“outdated steady-state” paradigm.


“If a forest is managed

according to the

principles of the

simplified structural

models, with no stands

permitted to attain true

old-growth conditions,

then (JR) some species

dependent on, or closely

associated with, old

growth will have a high

probability of

disappearing from the

managed landscape.”

34



While the emerging simplified stand structureapproaches establish a false dichotomy betweenso-called “landscape approaches” and theestablishment of reserves, reserve establishmentis, in fact, a crucial component of landscapeor ecosystem management (Noss 1983).Although biologists differ in the emphasis theygive reserve design versus management of thelandscape matrix in conservation plans, fewwould deny that both are necessary. Reservesmake at least three important contributionsto a landscape conservation and managementstrategy.

First, reserves serve as habitat for species thatare unlikely to persist in the multiple-use land-scape. If, for example, a forest is managedaccording to the principles of the simplifiedstructural models (e.g., Oliver 1992), with nostands permitted to attain true old-growth con-ditions, then some species dependent on, orclosely associated with, old growth will have ahigh probability of disappearing from the man-aged landscape. In this case, a network ofold-growth reserves will provide the only refu-gia for such species. These reserves also willserve as sources from which disturbance-sen-sitive species can recolonize the broaderlandscape after disturbance. Many species alsoare sensitive to exploitation, persecution, orharassment by humans — or, in some cases,to the mere presence of humans. Large andmedium-sized carnivores (e.g., grizzly bear,wolf, wolverine, and lynx) are examples of such

species. Roadless reserves with restricted trailsystems can provide security to these sensitivespecies. The larger and less accessible thereserve, the greater the securityoffered (Noss and Cooperrider1994).

Reserves, especially when largeand roadless, serve as referencesites and control areas for man-agement experiments. Fewscientists would dispute theneed for control areas in anycredible approach to adaptivemanagement, yet the simpli-fied structural approaches (e.g.,Oliver and Lippke 1994) donot acknowledge the need for such controlareas. Proponents of these emerging manage-ment schemes appear to assume that forestersunderstand forest ecosystems well enough tomanipulate them for long-term commodityproduction and other uses, without losingbiodiversity and ecosystem function. To testthe validity of this assumption by comparingtreated areas to untreated or natural areas, re-serves are necessary (Frissell and Bayles 1996).When (adaptive) ecosystem management ex-periments are carried out on a landscape scale,control areas (reserves) that span entire water-sheds are required.


“Roadless reserves with

restricted trail systems

can provide security

to sensitive species.

The larger and less

accessible the

reserve, the greater

the security offered.”

35

S I M P L I F I E D F O R E S T M A N A G E M E N T T O A C H I E V EW A T E R S H E D A N D F O R E S T H E A LT H : A C R I T I Q U E


In this section, we provide managers with achecklist of important considerations thatshould be included in fashioning and/or evalu-ating an ecosystem-based management plan.This list is by no means exhaustive. Rather, itis a distillation of the issues raised in the pre-ceding text. It should be treated as a startingpoint for thinking about ecosystem manage-ment and reviewing forest plans that purportto be ecosystem-based. [See also Noss (1999)for a listing of “green lights” and “red flags”for evaluation of ecosystem managementplans.]

Questions to consider:Natural forest stands are structurallyand functionally complex

• How well does the management prescrip-tion account for multiple forest structuresand spatial heterogeneity in structure, in-cluding but not limited to living trees? Forexample, does it provide for coarse woodydebris and give a rationale for the amountand type of coarse woody debris to be main-tained? Does it provide for multiple canopylayers and variable stand densities, includ-ing gaps?

• If the plan purports to mimic natural dis-turbances, does it recognize the differencesbetween natural and anthropogenic distur-bances? For example, does it incorporatestructural legacies that provide continuityfrom one stand to the next? What is therationale for the quantity and type of lega-cies, if any, in the plan?

• How much does the plan simplify foreststand development? For example, does theplan account for multiple developmentpathways?

• Does the plan address spatial planning ofstructural units? Does it consider the pat-tern, juxtaposition, and connectivity ofhabitat types across the landscape? Does itevaluate the consequences of alternativepatterns in terms of the life-histories andpopulation viabilities of particular species,and the operation of natural processes?

Questions to consider:Older forests are less susceptible tocatastrophic disturbances thanyounger forests

1. Does the management prescription for oneproblem (e.g., fire risk, pathogens) inad-vertently create other problems? Forexample, do management plans account forthe soil compaction, erosion, and increasedfire risk from logging residues, and othereffects associated with intensive silviculturalactivities? Has the plan accounted for thefull range of relevant disturbances that willimpact the forest in preparing its plannedactivities?

2. Does the plan account for potentiallyunstable terrain?

3. Does the plan account for the effects of re-duced root strength on landslide frequency?Does the plan account for the effect of tim-ber harvesting on slope stability? Does theplan account for the effect of vegetation/tree/stand age on slope stability?

III.Considerations for Ecosystem-BasedManagement Approaches

36



4. Does the plan specifically account for his-torical patterns as well as recent and futurechanges in riparian and floodplain foreststructures and landscapes?

Questions to consider: Not all speciesand habitats are of equal concern

• Does the plan specify which wildlife spe-cies the prescriptions intend to benefit? Arethe species rare or threatened, or are theyweedy species that can get along in a vari-ety of habitat conditions? In the case of rareor threatened species, how does the planprovide specifically for their needs? Are spe-cies dependent on or closely associated withlate-seral forests accounted for in the plan?

• What habitats is the plan designed to cre-ate or protect? Are these common or rarein the region? Are protected habitats of ad-equate size and connectivity to meet theneeds of the most demanding native spe-cies?

Questions to consider: Roadsdramatically alter forest ecosystems

1. Does the management prescription accountfor the ecological effects of the road con-struction and maintenance activitiesassociated with carrying out such activities?

2. Have alternatives to road building beenconsidered? How does the plan attempt toaddress the effects of roads? Does the plancall for obliteration and revegetation ofroads no longer needed for management?

3. Does the plan identify and maintain (orcreate) roadless areas and low road-densitywatershed as refuges from human activity?

Questions to consider: Regionalcontext is critical to ecologically-based forest management

1. Does the plan look beyond the boundariesof the planning area and consider histori-cal and regional context? For example, doesthe plan consider the history of habitatchange in the broader region and strive toprotect or restore habitat types and specieswhich have declined most in the regionsince European settlement?

2. Has the plan used adequate spatial and tem-poral scales for meeting its internalobjectives? (For example, does it considerwhat is occurring on adjacent land as wellas what might occur on those lands in thefuture?)

3. Does the management plan consider broad,regional strategies for conservation, andwork to further those strategies? For ex-ample, does the plan seek to link protectedareas within the plan boundaries to othersuch areas established or proposed withinthe region or adjacent regions? Does theplan contribute to conservation strategiesfor wide-ranging species (e.g., forest carni-vores), whose conservation must beaccomplished on vast spatial scales?

4. Has the plan sought to minimize habitatfragmentation? In what ways?

Questions to consider: Forest reservesare an important component ofecological management

1. Does the management plan evaluate theneed for reserves by looking at the regionaland landscape context of lands covered inthe plan? Does the management plan call

III. Considerations for Ecosystem-Based Management Approaches

37



for the creation of reserve areas as part of itsstrategy to conserve forest and wildlife di-versity? If not, how has the plan establisheda margin of safety, should the managementprescriptions not meet their stated objec-tives? If yes, have the reserves been designedto meet specific conservation objectives?What is the control area for the grand ex-periment with nature?

III. Considerations for Ecosystem-Based Management Approaches

Questions to consider: Generalplanning considerations

1. After having articulated its objectives, doesthe plan follow through on accomplishingthose objectives?

2. Does the plan use the best availablescientific information from all relevant dis-ciplines, including wildlife biology, fisheriesscience, ecosystem ecology, conservationbiology, landscape ecology, and hydrology?Have independent scientists been consultedto review the plan, and have their recom-mendations been taken to heart?

3. Does the plan explicitly acknowledge un-certainty and risk, and account for thesefactors by incorporating appropriate mar-gins of error?

4. Does the plan incorporate adaptive man-agement that provides opportunities tolearn from, and makes real changes basedon, management experiences? Is the designfor adaptive management scientifically rig-orous?

5. Does the plan include a credible monitor-ing program that will provide usefulinformation for adaptive management? Isa mechanism provided to assure that themonitoring program is well funded and willcontinue in perpetuity?

6. Have sources and levels of risk beenidentified and fully disclosed? Haveoptions for reducing risk been consideredand disclosed? Has a framework fordecision-making or a rationale for copingwith risk been articulated?

38



Jerry Franklin is Professor of Ecosystem Analy-

sis in the College of Forest Resources at the Uni-

versity of Washington, Land Steward for the Rio

Condor Project, Chile, a director of the Wilder-

ness Society, and a fellow of the American Asso-

ciation for the Advancement of Science. Dr.

Franklin has served as President of the Ecologi-

cal Society of America, a member of the execu-

tive committee of the Forest Ecosystem Manage-

ment Assessment Team and on the Board of Gov-

ernors of the Nature Conservancy. Among Dr.

Franklin’s awards are the William B. Greeley

Award from the American Forests Association

(1996) and the Barrington Moore Award from

the Society of American Foresters (1986).

Reed Noss is President and Chief Scientist for

Conservation Science, Inc., an international con-

sultant on biodiversity issues, and President of

the Society for Conservation Biology. For several

years he edited the journal Conservation Biol-

ogy, and has worked for several government

agencies and conservation organizations. He is

the author of four books and more than 170

articles on ecological topics. Dr. Noss has adjunct

appointments in Fisheries and Wildlife and For-

est Science at Oregon State University. His cur-

rent work involves science-based conservation

planning at regional to continental scales.

David Montgomery is an Associate Professor

in the Department of Geological Sciences at the

University of Washington, where he studies ero-

sion and fluvial processes in mountain drainage

basins. Dr. Montgomery’s research focuses on

tectonic, hillslope, and fluvial geomorphology,

with particular interest in GIS applications, land-

form evolution, and relations to ecological sys-

tems. He is the author of more than 80 articles

on geomorphological topics.

David Perry is Professor Emeritus of Ecosystem

Studies and Ecosystem Management in the De-

partment of Forest Science at Oregon State Uni-

versity, and is the Resident Ecologist, Ka Ha o Ka

‘Aina, Kapa’au, Hawai’i. Dr. Perry is author of

the textbook Forest Ecosystems and lead

editor of Maintaining Long-term Productivity of

Pacific Northwest Ecosystems. He also is the

author or co-author of approximately 100 pa-

pers on various aspects of ecosystem structure

and dynamics.

Christopher Frissell is Research Associate Pro-

fessor at the Flathead Lake Biological Station and

the Division of Biological Sciences of the Univer-

sity of Montana. He formerly held research posi-

tions at Oregon State University, where he com-

pleted an MS and PhD in Fisheries Science. He

and his students study the effects of human man-

agement and natural landscape processes on

freshwater ecosystems, especially with regard

for the conservation of fishes and aquatic

biodiversity. His recent research concerns the role

of groundwater-surface water exchange in shap-

ing stream and floodplain habitats, the effect of

invasive freshwater species, the effects of wa-

tershed disturbance on stream habitats and fish

populations, regional strategies for salmon re-

covery, and the role of adaptive management in

conservation planning.

Project Manager: Jay Chamberlin. Project con-

sultant: Rick Brown, Conservation Science Sup-

port Center. Editorial assistance by Elizabeth

Grossman. Design and production by Vicki Moser

(CopperMoon Design). Photos by Jerry Franklin.

Special thanks to Sybil Ackerman and to NWF’s

Western Natural Resource Center. Special thanks

also to the Bullitt Foundation whose support

made this report possible.

Scientific Panel on Ecosystem BasedForest Management — Bios

39

N A T I O N A L W I L D L I F E F E D E R A T I O N40



Abbe, T.B., and D.R. Montgomery. 1996. Interac-tion of large woody debris, channel hydraulicsand habitat formation in large rivers. RegulatedRivers: Research & Management 12: 201-221.

Agee, J.K. 1993. Fire ecology of Pacific Northwestforests. Island Press. Washington, D.C. andCovelo, CA.

Anderson, H.W. 1954. Suspended sediment dis-charge as related to stream flow, topography,soil, and land use. EOS, Transactions of theAmerican Geophysical Union 35: 268-281.

Anderson, L., C.E. Carlson, and R.H. Wakimoto.1987. Forest fire frequency and western sprucebudworm outbreaks in western Montana. For-est Ecology and Management 22: 251-260.

Andrews, H.J. and R.W. Cowlin. 1940. Forest re-sources of the Douglas-fir region. USDAMiscellaneous Publication 389. Washington,DC.

Angermeier, P.L. and J.R. Karr. 1994. Biological in-tegrity versus biological diversity as policydirectives. BioScience 44: 690-697.

Baker, W.L. 1992. The landscape ecology of largedisturbances in the design and management ofnature reserves. Landscape Ecology 7: 181-194.

Baxter, C.V., C.A. Frissell, and F.R. Hauer. 1999. Geo-morphology, logging roads, and the distributionof bull trout (Salvelinus confluentus) spawningin a forested river basin: implications for man-agement and conservation. Transactions of theAmerican Fisheries Society.

Belsky, A.J. and D.M. Blumenthal. 1997. Effects oflivestock grazing on stand dynamics and soilsin upland forests of the Interior West. Conser-vation Biology 11 (2): 315-327.

Berris, S.N., and R.D. Harr. 1987. Comparative snowaccumulation and melt during rainfall in for-ested and clear-cut plots in the westernCascades of Oregon. Water Resources Research23: 135-142.

Bilby, R.E., and P.A. Bisson. 1997. Function and dis-tribution of large woody debris. Pages 324-346in R.J. Naiman and R.E. Bilby, editors. River Ecol-ogy and Management: Lessons from the PacificCoastal Ecoregion. Springer. New York, NY.

Bishop, D.M., and M.E. Stevens. 1964. Landslideson logged areas in southeast Alaska. USDA For-est Service Research Paper NOR-1.

Blake, J.G. and W.G. Hoppes. 1986. Influence ofresource abundance on use of treefall gaps bybirds. Auk 103: 328-430.

Bork, J. 1985. Fire history in three vegetation typeson the eastern side of the Oregon Cascades.PhD Thesis. Oregon State University Departmentof Forest Science.

Brittingham, M.C. and S.A. Temple. 1983. Havecowbirds caused forest songbirds to decline?BioScience 33: 31-35.

Brody, A.J., and M.R. Pelton. 1989. Effects of roadson black bear movements in western NorthCarolina. Wildlife Society Bulletin 17: 5-10.

Brown, G.W., and J.T. Krygier. 1971. Clear-cut log-ging and sediment production in the OregonCoast Range. Water Resources Research 7:1189-1198.

Brown, C.B. and M.S. Sheu. 1979. Effects of defor-estation on slopes. Journal of the GeotechnicalEngineering Division, American Society of CivilEngineers 101: 147-165.

Buck, D.H. 1956. Effects of turbidity on fish andfishing. Transactions of the North AmericanWildlife Conference 21: 249-261.

Burroughs, E.R. Jr., and B.R. Thomas. 1977. Declin-ing root strength in Douglas-fir after felling asa factor in slope stability. USDA Forest ServiceResearch Paper INT-190.

Carey, A.B. 1989. Wildlife associated with old-growth forests in the Pacific Northwest. NaturalAreas Journal 9: 151-162.

Carey, A.B., et al. 1996. Washington Forest Land-scape Management Project-a pragmatic,ecological approach to small landscape man-agement. Washington State Department ofNatural Resources Report No. 2. Olympia, WA.

Carey, A.B. 1998. Ecological foundations forbiodiversity: lessons from natural and managedforests in the Pacific Northwest. Northwest Sci-ence 72 (special issue): 127-133.

Cowlin, R.W., P.A. Briegleb, and F.L. Moravets. 1942.Forest resources of the ponderosa pine region.USDA Miscellaneous Publication 490.

Diamond, J. 1976. Island biogeography and con-servation: strategy and limitations. Science 193:1027-1029.

Eaglin, G.S., and W.A. Hubert. 1993. Effects of log-ging roads on substrate and trout in streams ofthe Medicine Bow National Forest, Wyoming.North American Journal of Fisheries Manage-ment 13: 844-846.

Literature Cited

41

S I M P L I F I E D F O R E S T M A N A G E M E N T T O A C H I E V EW A T E R S H E D A N D F O R E S T H E A LT H : A C R I T I Q U E


Forest Ecosystem Management Assessment Team(FEMAT). 1993. Forest ecosystem management:an ecological, economic, and social assessment.U.S. Department of Agriculture, U.S. Depart-ment of the Interior (and others). Portland, OR.

Franklin, J.F., K. Cromack Jr., W. Dennsion, A.McKee, C. Maser, J. Sedell, F. Swanson, and G.Juday. 1981. Ecological characteristics of old-growth Douglas-fir forests. USDA Forest ServiceGeneral Technical Report PNW-118. Portland,OR.

Franklin, J.F., and T.A. Spies. 1984. Characteristicsof old-growth Douglas-fir forests. Pages 328-334 in New forests for a changing world. Societyof American Foresters. Bethesda, MD.

Franklin, J.F., H.H. Shugart and M.E. Harmon. 1987.Tree death as an ecological process. BioScience37 (8): 550-556.

Franklin, J.F. 1992. Scientific basis for new perspec-tives in forests and streams. Pages 25-72 in: R.J.Naiman, editor. Watershed management: bal-ancing sustainability and environmental change.Springer-Verlag. NY.

Franklin, J.F., D.A. Perry, T.D. Schowalter, M.E.Harmon, A. McKee and T.A. Spies. 1989. Im-portance of ecological diversity in maintaininglong-term site productivity. Pages 82-97 in D.A.Perry, et al., Maintaining the long-term produc-tivity of Pacific Northwest forest ecosystems.Timber Press. Portland, OR.

Franklin, J.F., and J.A. Fites-Kaufmann. 1996. As-sessment of late-successional forests of theSierra Nevada. Pages 627-657 in Sierra NevadaEcosystem Project: final report to Congress, Vol.II. Assessments and scientific basis for manage-ment options. University of California Centerfor Water and Wildland Resources. Davis, CA.

Franklin, J.F., D. Rae Berg, D.A. Thornburgh, andJ.C. Tappeiner. 1997. Alternative silvicultural ap-proaches to timber harvesting. Pages 111-139in Kohm, K.A., and J.F. Franklin, editors. Creat-ing a forestry for the 21st Century. The scienceof ecosystem management. Island Press. Wash-ington, DC and Covelo, CA.

Franklin, J.F., T.A. Spies, R. Van Pelt, A.B. Carey, D.Thornburgh, W. Keeton, and J.A. Fites-Kaufmann. 1999. Structural development ofnatural forest ecosystems including the role ofdisturbances. Manuscript submitted to ForestEcology and Management.

Fredriksen, R. L. 1970. Erosion and sedimentationfollowing road construction and timber harveston unstable soils in three small western Oregonwatersheds. USDA Forest Service Research Pa-per PNW-104.

Frissell, C.A., and D. Bayles. 1996. Ecosystem man-agement and the conservation of aquaticbiodiversity and ecological integrity. Journal ofthe American Water Resources Association 32:229-240.

Trombulak, S.C. and C.A. Frissell. 2000. Reviewof ecological effects of roads on terrestrial andaquatic communities. Conservation Biology 14(1) 18-30.

Furniss, M.J., T.D. Roeloffs, and C.S. Yee. 1991.Road construction and maintenance. Pages 297-323 in W.R. Meehan, editor. Influences of forestand rangeland management on salmonid fishesand their habitats. Special Publication 19. Ameri-can Fisheries Society. Bethesda, MD.

Gray, D.H. 1970. Effects of forest clear-cutting onthe stability of natural slopes. Bulletin of theAssociation of Engineering Geologists 7: 45-66.

Gresswell, S., D. Heller, and D.N. Swanston. 1979.Mass movement response to forest manage-ment in the central Oregon Coast Ranges. USDAForest Service Resources Bulletin PNW-84.

Gresswell, R.E. 1999. Fire and aquatic ecosystemsin forested biomes of North America. Transac-tions of the American Fisheries Society 128:193-221.

Hagans, D.K., W.E. Weaver, and M.A. Madej. 1986.Long-term and off-site effects of logging anderosion in the Redwood Creek basin, northernCalifornia. In: Papers presented at the Ameri-can Geophysical Union meeting on cumulativeeffects. National Council for Air and Stream Im-provement Technical Bulletin 490: 38-66.

Harmon, M.E., J.F. Franklin, F.J. Swanson, P. Sollins,S.V. Gregory, J.D. Lattin, N.H. Anderson,S.P. Cline, N.G. Aumen, J.R. Sedell, G.W.Lienkaemper, K. Cromack, Jr., and K.W.Cummins. 1986. Ecology of coarse woody de-bris in temperate ecosystems. Advances inEcological Research 15: 133-302.

Harr, R.D., and R.A. Nichols. 1993. Stabilizing for-est roads to help restore fish habitat: anorthwest Washington example. Fisheries 18:18-22.

Harris, Larry D. 1984. The fragmented forest. Uni-versity of Chicago Press. Chicago, IL.

Hayes, J. 1998. An independent scientific reviewof Oregon Department of Forestry’s proposedwestern Oregon State Forests Habitat Conser-vation Plan. College of Forestry, Oregon StateUniversity. Corvallis, OR.

Henderson, J.A., D.H. Peter, R.D. Lesher, and D.R.Shaw. 1989. Forested plant associations of theOlympic National Forest. USDA Forest ServiceTechnical Paper R6 ECOL 001-88. Portland, OR.

42



Henjum, M.G., J.R. Karr, D.L. Bottom, D.A. Perry,J.C. Bednarz, S.G. Wright, S.A. Beckwitt, andE. Beckwitt. 1994. Interim protection for late-successional forests, fisheries, and wildlife. TheWildlife Society Technical Review 94-2

Hermann, S.M., editor. 1993. The longleaf pine eco-system: ecology, restoration, and management.Proceedings of the 18th Tall Timbers Fire Ecol-ogy Conference, May 30-June 2, 1991. TallTimbers. Tallahassee, FL.

Hicks, B.J. J.D. Hall, P.A. Bisson, and J.R. Sedell.1991. Response of salmonids to habitat change.Pages 483-518 in W.R. Meehan, editor. Influ-ences of forest and rangeland management onsalmonid fishes and their habitats. SpecialPublication 19. American Fisheries Society.Bethesda, MD.

Huff, M.H., R.D. Ottmar, E. Alvarado, R.E. Vihnanek,J.F. Lehmkuhl, P.F. Hessburg, and R.L. Everett.1995. Historical and current landscapes in east-ern Oregon and Washington. Part II: Linkingvegetation characteristics to potential fire be-havior and related smoke production. USDAForest Service Pacific NW Experimental StationGeneral Technical Report 355. Portland, OR.

Keller, E.A., and F.J. Swanson. 1979. Effects of largeorganic material on channel form and fluvialprocesses. Earth Surface Processes & Landforms4: 361-380.

Leonard, R.E. 1961. Interception of precipitation bynorthern hardwoods. USDA Forest ServiceNortheastern Forest Experimental Station. Sta-tion Paper 159.

Ludwig, D., R. Hilborn. and C. Walters. 1993.Uncertainty, resource exploitation, and conser-vation: lessons from history. Science 260: 17,36.

Manion, P.D. 1981. Tree disease concepts. PrenticeHall. Englewood Cliffs, NJ.

Martin, T.E. and J.R. Karr. 1986. Patch utilizationby migrating birds: resource oriented? OrnisScandinavica 17: 165-174.

Maser, C., R.F. Tarrant, J.M. Trappe, and J.F. Franklin,technical editors. 1988. From the forest to thesea: A story of fallen trees. USDA Forest ServiceGeneral Technical Report PNW-GTR-229.

Matthews, J.D. 1989. Silvicultural systems. OxfordUniversity Press. New York, NY.

McCutcheon, T.L. and G.C. Carroll. 1993. Geno-typic diversity in populations of a fungalendophyte from Douglas-fir. Mycologia 85: 180-186.

Meffe, G.K., and C.R. Carroll. 1997. Conservationreserves in heterogeneous landscapes. Pages305-343 in G.K. Meffe and C.R. Carroll, edi-tors. Principles of conservation biology. Secondedition. Sinaver. Sunderland, MA.

Megahan, W.E. 1972. Subsurface flow interceptionby a logging road in the mountains of centralIdaho. Pages 350-356 in Proceedings of a na-tional symposium on watersheds in transition.American Water Resources Association,Bethesda, MD.

Mersereau, R.C., and C.T. Dryness, 1972. Acceler-ated mass wasting after logging and slashburning in western Oregon. Journal of Soil andWater Conservation 27: 112-114.

Minshall, G.W., C.T. Robinson, and D.E. Lawrence.1997. Postfire responses of lotic ecosystems inYellowstone National Park, U.S.A. CanadianJournal of Fisheries and Aquatic Sciences 54:2509-2525.

Mladenoff, D.J., T.A. Sickley, and A.P. Wydeven.1999. Predicting gray wolf landscape recolon-ization: logistic regression models vs. new fielddata. Ecological Applications 9: 37-44.

Montgomery, D.R. 1994. Road surface drainage,channel initiation, and slope stability. Water Re-sources Management 30: 1925-1932.

Montgomery, D.R. 1999. Process domains and theriver continuum. Journal of the American Wa-ter Resources Association 35: 397-410.

Montgomery, D. R., Schmidt, K. M., Greenberg, H.,and Dietrich, W. E., 2000. Forest clearing andregional landsliding, Geology, v. 28, p. 311-314.

Montgomery, D.R., T.B. Abbe, N.P. Peterson, J.M.Buffington, K.M. Schmidt, and J.D. Stock. 1996.Distribution of bedrock and alluvial channels inforested mountain drainage basins. Nature 381:587-589.

Montgomery, D.R., J.M. Buffington, R. Smith, K.Schmidt, and G. Pess. 1995. Pool spacing in for-est channels. Water Resources Research 31:1097-1105.

Montgomery, D. R., Zabowski, D., Ugolini, F. C.,Halberg, R., and Spaltenstein, H., 2000. Soils,Watershed Processes, and Marine Sediments,in Earth System Science, edited by M. C.Jacobson, R. J. Charlson, H. Rodhe, and G. H.Orians, Academic Press, p.159-194.

Morrison, P.H. and S.J. Swanson. 1990. Fire historyand pattern in a Cascade Range landscape.USDA Forest Service General Technical ReportPNW-GTR-254. Portland, OR.

National Research Council. 2000. EnvironmentalIssues in Pacific Northwest Forest Management.National Academy Press. Washington, DC. p.259.

43



Newcombe, C.P., and J.O.T. Jensen. 1996. Chan-nel suspended sediment and fisheries: asynthesis for quantitative assessment of risk.North American Journal of Fisheries Manage-ment 16: 693-727.

Norse, E.A. 1990. Ancient forests of the PacificNorthwest. The Wilderness Society and IslandPress. Washington, D.C.

North, M.P., J.F. Franklin, A.B. Carey, E.D. Forsman,and T. Hamer. 1999. Forest stand structure ofthe northern spotted owl’s foraging habitat. For-est Science 45: 520-527.

Noss, R.F. 1983. A regional landscape approach tomaintain diversity. BioScience 33: 700-706.

Noss, R.F. 1991. Effects of edge and internal patchi-ness on avian habitat use in an old-growthFlorida hammock. Natural Areas Journal 11: 34-47.

Noss, R.F. 1993. A conservation plan for theOregon Coast Range: some preliminary sugges-tions. Natural Areas Journal 13: 276-290.

Noss, R.F., and A. Cooperrider. 1994. Saving nature’slegacy: Protecting and restoring biodiversity.Defenders of Wildlife and Island Press. Wash-ington, D.C.

Noss, R.F., M.A. O’Connell, and D.D. Murphy. 1997.The Science of Conservation Planning: Habitatconservation under the Endangered Species Act.World Wildlife Fund and Island Press. Washing-ton, D.C.

Noss, R.F., H.B. Quigley, M.G. Hornocker, T. Merrill,and P.C. Paquet. 1996. Conservation biologyand carnivore conservation in the Rocky Moun-tains. Conservation Biology 10: 949-963.

Noss, R.F. 1999. A citizen’s guide to ecosystem man-agement. Biodiversity Legal Foundation.Boulder, CO.

Noss, R.F., J.R. Strittholt, K. Vance-Borland, C.Carroll, and P. Frost. 1999. A conservation planfor the Klamath-Siskiyou ecoregion. NaturalAreas Journal 19: 392-411.

Oliver, C.D. 1981. Forest development in NorthAmerica following major disturbances. ForestEcology and Management 3: 153-168.

Oliver, C.D. 1992. A landscape approach: Achiev-ing and maintaining biodiversity and economicproductivity. Journal of Forestry 90: 20-25.

Oliver, C.D., and B.R. Lippke. 1994. Wood supplyand other values and ecosystem managementin the western interior forests. In R.L. Everettand D.M. Baumgartner, compilers. Ecosystemmanagement in western interior forests. Sym-posium proceedings. May 3-5, 1994, Spokane,WA. USDA Forest Service, Pacific Northwest Ex-periment Station. Portland, OR.

Oliver, C.D., and B.C. Larson. 1996. Forest standdynamics. Update edition. John Wiley & Sons.New York, NY.

Oliver, C.D., D. Adams, T. Bonnicksen, J. Bowyer, F.Cubbage, N. Sampson, S. Schlarbaum, R.Whaley, and H. Wiant. (Forest Health SciencePanel). 1997. Report on Forest Health of theUnited States. Chartered by Charles Taylor,Member, United States Congress, 11th District,North Carolina.

Oliver, C.D., A. Osawa, and A. Camp. 1998. Forestdynamics and resulting animal and plant popu-lation changes at the stand and landscape levels.Journal of Sustainable Forestry 6: 281-312.

O’Loughlin, C. 1974. The effect of timber removalon the stability of forest soils. Journal of Hy-drology (New Zealand) 13: 121-134.

O’Loughlin, C. L., and A. J. Pearce. 1976. Influenceof Cenozoic geology on mass movement andsediment yield response to forest removal, NorthWestland, New Zealand. Bulletin of the Inter-national Association of Engineering Geology 14:41-46.

Oregon Department of Forestry. 1999. WesternOregon State forests Habitat Conservation Plan,Salem, OR.

Parker, G. G. 1997. Canopy structure and light en-vironment of an old-growth Douglas-fir/westernhemlock forest. Northwest Science 71: 261-270.

Pearce, A.J., and L.K. Rowe. 1979. Forest manage-ment effects on interception, evaporation, andwater yield. Journal of Hydrology (New Zealand)18: 73-87.

Perry, D.A. 1988a. An overview of sustainable for-estry. Journal of Pesticide Reform 8: 8-12.

Perry, D.A. 1988b. Landscape patterns and forestpests. Northwest Environmental Journal 4: 213-228

Perry, D.A. 1994. Forest ecosystems. The JohnsHopkins University Press. Baltimore, MD.

Perry, D.A. 1995. Landscapes, humans, and othersystem-level considerations: a discourse on ec-stasy and laundry. Pages 177-191 in R.L. Everettand D.M. Baumgartner, editors. Ecosystemmanagement in Western Interior forests. Wash-ington State University Cooperative Extension.Pullman, WA.

Perry, D.A. 1998. The scientific basis of forestry. An-nual Review of Ecological Systems 29: 435-466.

Perry, D.A., and J. Huang. 1998. Structure of old-growth forests on the east slopes of theCascades in central Oregon. Unpublished reportto the Deschutes National Forest.

44



Pickett, S.T.A., and J.N. Thompson. 1978. Patch dy-namics and the design of nature reserves.Biological Conservation 13: 27-37.

Reeves, G.H., L.E. Benda, K.M. Burnett, P.A. Bisson,and J.R. Sedell. 1995. A disturbance-based eco-system approach to maintaining and restoringfreshwater habitats of evolutionarily significantunits of anadromous salmonids in the PacificNorthwest. In J. L. Nielsen, editor. Evolution andthe aquatic ecosystem: defining unique units inpopulation conservation. American Fisheries So-ciety Symposium 17: 334-349.

Reeves, G.H. and J.R. Sedell. 1992. An ecosystemapproach to the conservation and managementof freshwater habitat for anadromous salmo-nids in the Pacific Northwest. Proceedings ofthe 57th North American Wildlife and NaturalResources Conference: 408-415.

Reid, L.M., and T. Dunne. 1984. Sediment produc-tion from forest road surfaces. Water ResourcesResearch 20: 1753-1761.

Rieman, B.E., D.C. Lee, and R.F. Thurow. 1997. Dis-tribution, status, and likely future trends of bulltrout within the Columbia River and KlamathRiver basins. North American Journal of Fisher-ies Management 17: 1111-1125.

Rogerson, T.L., and W.R. Byrnes. 1968. Net rainfallunder hardwoods and red pine in central Penn-sylvania. Water Resources Research 4: 55-57.

Rothacher, J. 1963. Net precipitation under a Dou-glas-fir forest. Forest Science 9: 423-429.

Rowe, L.K. 1979. Rainfall interception by a beech-podocarp-hardwood forest near Reefton, NorthWestland, New Zealand. Journal of Hydrology(New Zealand) 18: 63-72.

Ruggiero, L.F., K.B. Aubry, A.B. Carey, and M.F. Huff,editors. 1991. Wildlife and vegetation inunmanaged Douglas-fir forests. USDA ForestService. Pacific Northwest Research Station Gen-eral Technical Report PNW-GRT-285. Portland,OR.

Ruggiero, L.F., K.B. Aubry, S.W. Buskirk, L.J. Lyon,and W.J. Zielinski, editors. 1994. The scientificbasis for conserving forest carnivores: Ameri-can marten, fisher, lynx, and wolverine. GeneralTechnical Report RM-254. USDA Forest Service,Rocky Mountain Forest and Range Experimen-tal Station. Fort Collins, CO.

Schmidt, K.M. 1999. Root strength, colluvial soildepth, and colluvial transport on landslide-pronehillslopes. PhD dissertation. Department of Geo-logical Sciences, University of Washington.

Schowalter, T.D. 1989. Canopy arthropod commu-nity structure and herbivory in old-growth andregenerating forests in western Oregon. Cana-dian Journal of Forest Resources 19: 318-322

Schowalter, T.D. 1995. Canopy arthropod commu-nities in relation to forest age and alternativeharvest practices in western Oregon. Forest Ecol-ogy and Management 78: 115-125.

Sedell, J.R., and C.N. Dahm. 1984. Catastrophicdisturbance to stream ecosystems: volcanismand clear-cut logging. Pages 531-539 in M.J.Klug and C.A. Reddy, editors. Current perspec-tives in microbial ecology. American Society forMicrobiology, Washington, DC.

Shinneman, D.J. and W.L. Baker. 1997. Noneq-uilibrium dynamics between catastrophicdisturbances and old-growth forests in Ponde-rosa pine landscapes of the Black Hills.Conservation Biology 11: 1276-1288.

Sidle, R.C., A.J. Pearce, and C.L. O’Loughlin. 1985.Hillslope stability and land use. American Geo-physical Union Water Resources Monograph 11.Washington, D.C.

Sierra Nevada Ecosystem Project. 1996-1997. Sta-tus of the Sierra Nevada. Sierra NevadaEcosystem Project Final Report to Congress (5volumes). Wildland Resources Center Reports.University of California. Davis, CA.

Spies, T.A. and J.F. Franklin. 1988. Old growth andforest dynamics in the Douglas-fir region ofwestern Oregon and Washington. Natural Ar-eas Journal 8: 190-201.

Spies, T.A., and J.F. Franklin. 1991. The structure ofnatural young, mature, and old-growth Dou-glas-fir forests in Washington and Oregon.Pages 91-111 in L. F. Ruggiero, K. B. Aubry, A.B. Carey, and M. H. Huff, editors. Wildlife andvegetation of unmanaged Douglas-fir forests.USDA Forest Service General Technical ReportPNW-GTR-285. Pacific Northwest Research Sta-tion. Portland, OR.

Swanson, F.J., and C.T. Dyrness. 1975. Impact ofclear-cutting and road construction of soil ero-sion by landslides in the western Cascade Range,Oregon. Geology 3: 393-396.

Swanson, F.J., T.K. Kratz, N. Caine, and R.G.Woodmansee. 1988. Landform effects on eco-system patterns and processes. BioScience 38:92-98.

Swanston, D.N., 1969. Mass wasting in coastalAlaska. USDA Forest Service Research PaperPNW-83.

Swanston, D.N., and F.J. Swanson. 1976. Timberharvesting, mass erosion and steepland forestgeomorphology in the Pacific Northwest. Pages199-221 in D. R. Coates, editor. Geomorphol-ogy and Engineering. Dowden, Hutchinson &Ross, Inc., Stroudsburg, PA.

45



Tappeiner, J.C., D. Huffman, D. Marshall, T.A. Spies,and J.D. Bailey. 1997. Density, ages, and growthrates in old-growth and young-growth Douglas-fir in coastal Oregon. Canadian Journal of ForestResources 27: 638-648.

Teklehaimanot, Z., P.B. Jarvis, and D.C. Ledger.1991. Rainfall interception and boundary layerconductance in relation to tree spacing. Jour-nal of Hydrology 123: 261-278.

Terborgh, J. 1976. Island biogeography and con-servation: strategy and limitations. Science 193:1029-1030.

Terwilliger, V.J. 1990. Effects of vegetation on soilslippage by pore pressure modification. EarthSurface Processes and Landforms 15: 553-370.

Thurber, J.M., R.O. Peterson, T.D. Drummer, andS.A. Thomasma. 1994. Gray wolf response torefuge boundaries and roads in Alaska. Wild-life Society Bulletin 22: 61-68.

Torgerson, T.R., R.R. Mason, and R.W. Campbell.1990. Predation by birds and ants on two for-est insect pests in the Pacific Northwest. Studiesin Avian Biology 13: 14-19.

Torres, R., W.E. Dietrich, D.R. Montgomery, S.P.Anderson, S.P., and K. Loague. 1998. Unsatur-ated zone processes and the hydrologicresponse of a steep, unchanneled catchment.Water Resources Research 34: 1865-1879.

Tuchmann, E.T., K.P. Connaughton, L.E. Freedman,and C.B. Moriwaki. 1996. The Northwest For-est Plan. A report to the President and Congress.USDA Office of Forestry and Economic Assis-tance. Portland, OR.

Tyser, R.W., and C.A. Worley. 1992. Alien flora ingrasslands adjacent to road and trail corridorsin Glacier National Park, Montana (USA). Con-servation Biology 6: 253-262.

U.S. Forest Service. 1988. Silver fire recovery plan.Draft Environmental Impact Statement. SiskiyouNational Forest. Grants Pass, OR.

Van Pelt, R. and J.F. Franklin. 1999. Response ofunderstory trees to experimental gaps inold-growth Douglas-fir forests. Ecological Ap-plications 9: 504-512.

Waring, R.H. and G.B. Pitman. 1985. Modifyinglodgepole pine stands to change susceptibilityto mountain pine beetle attack. Ecology 66:889-897.

Waring, R.H. and W.H. Schlesinger. 1985. Forestecosystems. Academic Press. Orlando, FL.

Weaver, J.L., P.C. Paquet, and L.F. Ruggiero. 1996.Resilience and conservation of large carnivoresin the Rocky Mountains. Conservation Biology10: 964-976.

Weaver, W.E., M.M. Hektner, D.K. Hagans, L.J.Reed, R.A. Sonneville, and G.J. Bundros. 1987.An evaluation of experimental rehabilitationwork, Redwood National Park. Technical Report19 Redwood National Park. Arcata, CA.

Wemple, B.C., J.A. Jones, and G.E. Grant. 1996.Channel network extension by logging roadsin two basins, western Cascades, Oregon. Wa-ter Resources Bulletin 32: 1195-1207.

Wickman, B.E., R.R. Mason, and T.W. Swetnam.1993. Searching for long-term patterns of for-est insect outbreaks. In: proceedings of aconference: Individuals, Populations, and Pat-terns. Norwich, England. 7-10 November 1992.

Wigmosta, M.S. 1983. Rheology and flow dynam-ics of the Toutle debris flows from Mt. St. Helens,unpublished M.S. Thesis, University of Wash-ington.

Wimberly, M.C., T.A. Spies, C.J. Long, and C.Whitlock. 2000. Simulating historical variabilityin the amount of old forests in the Oregon CoastRange. Conservation Biology 14:167-180.

Wisdom, M., R. Holthausen, B. Wales, D. Lee, C.Hargis, V. Saab, W. Hann, T. Rich, M. Rowland,W. Murphy and M. Eames. In press. Source habi-tats for terrestrial vertebrates of focus in theInterior Columbia Basin: Broad-scale trends andmanagement implications. USDA Forest ServiceGeneral Technical Report.

Wu, T.H., W.P. McKinnell III, D.N. Swanston. 1979.Strength of tree roots and landslides on Princeof Wales Island, Alaska. Canadian GeotechnicalJournal 16: 19-33.

Wu, T.H., and D.N. Swanston. 1980. Risk of land-slides in shallow soils and its relation to clearcutting in southeastern Alaska. Forest Science26: 495-510.

Ziemer, R.R., J. Lewis, R.M. Rice, and T.E. Lisle. 1991.Modeling the cumulative watershed effects offorest management strategies. Journal of Envi-ronmental Quality 20: 36-42.

Zobel, D.B., L.F. Roth, and G.M. Hawk. 1985.Ecology, pathology and management of Port-Orford-cedar (Chamaecyparis lawsoniana).USDA Forest Service. General Technical ReportPNW-184. Pacific Northwest Forest and RangeExperiment Station. Portland, OR.

46

The mission of the National Wildlife

Federation is to educate, inspire and

assist individuals and organizations

of diverse cultures to conserve wildlife

and other natural resources and to

protect the earth’s environment in

order to achieve a peaceful, equitable

and sustainable future.

National Wildlife Federation418 First Avenue WestSeattle, Washington 98119phone: 206/285-8707email: [email protected]: www.nwf.org

simplified forest management to achieve watershed …coastrange.org/documents/forestreport.pdf ·...

Documents