reprinted from forest ecology and management forest ...coweeta.uga.edu/publications/63.pdf ·...

Reprinted from

Forest Ecology and

Management

Forest Ecology and Management 114 (1999) 173-197

Ecosystem management decision support for federal

forests in the United States: A review1'2

H. Michael Rauscher*

USDA Forest Service, Southern Research Station, Bent Creek Research and Demonstration Forest, 1577 Brevard Rd.,AshevUle, NC 28805, USA

ELSEVIER

Forest Ecology

and Management

Aims and scope. Forest Ecology and Management publishes scientific articles concerned with forest management and conser-vation, and in particular the application of biological, ecological and social knowledge to the management of man-made andnatural forests. The scope of the journal includes all forest ecosystems of the world. A refereeing process ensures the qualityand international interest of the manuscripts accepted for publication. The journal aims to encourage communication betweenscientists in disparate fields who share a common interest in ecology and natural-resource management, and to bridge thegap between research workers and forest managers in the field to the benefit of both. The journal should be of interest toresearch workers, managers and policy makers in forestry, natural resources, ecological conservation and related fields.FOUNDING EDITOR ' ,Laurence L Roche, Murroe, IrelandEDITORS-IN-CHIEFFor the Americas, Australia, New Zealand and the Pacific:R.F. FisherDepartment of Forest ScienceTexas A&M UniversityCollege Station, TX 77843-2135, USA

BOOK REVIEWS EDITORMargaret R. GaleSchool of Forestry and Wood ProductsMichigan Technological University1400 Townsend DriveHoughton, Ml 49931. USAEDITORIAL ADVISORY BOARDG. Abrahamsen, Agricultural University of Norway, As,

NorwayR. Alfaro, Canadian Forestry Service, Victoria, B.C., CanadaF. Andersson, Swedish University of Agricultural Sciences,

Uppsala, SwedenP.M.S. Ashton,Yale University, New Haven, USAP. Attiwill, University of Melbourne, Parkville, Vic., AustraliaJ. Boyle, Oregon State University, Corvallis, USAS. Brown, US Environmental Protection Agency, Corvallis,

USAL Burke, Colorado State University, Fort Collins, CO, USAJ.C. Calvo, Institute of Technology, Cartago, Costa RicaR.-J. Chen, University of Hong Kong, Pokfulan Road,

Hong KongJ.D. Deans, Institute of Terrestrial Ecology, Penicuik,

Midlothian, UKRM DeGraaf, USDA Forest Service, University of

Massachusetts, Amherst, MA, USAS. Diamandis, Forest Research Institute, Thessaloniki,

GreeceD.P. Dykstra, CIFOR, Jakarta, IndonesiaEP. Farrell, University College Dublin, Dublin, IrelandP.M. Fearnside, Institute Nacional de Pesquisas da

Amazonia-INPA, Manaus-Amazonas, BrazilP.H. Freer-Smith, Forestry Commission, Farnham, UKO. Garcia, ENGREF, Nancy, FranceD. Gilmore, Canadian Forest Product Ltd., Alberta, CanadaR.A. Goyer, Louisiana State University, Baton Rouge, LA,

USAJ.B Hall, University College of North Wales, Bangor, UKF. Houllier, Campus International de Baillarguet, LaboratoireAssocie de Moderation des Plantes (AMAP), Montpellier, France

For the rest of the world:G.M.J. MohrenForest Production Ecology GroupDepartment of Vegetation EcologyDLO-lnsutute for Forestry and Nature ResearchP.O. Box 236700 AAWageningen.The Netherlands

B.M Kumar, Kerala Agricultural University, Kerala, IndiaJ.P. Lassoie, Cornell University, Ithaca, NY, USAJ.N. Long, Utah State University, Logan, UT, USAAJL Lugo, International Institute of Tropical Forestry, Rio

Piedras, PR, USAJ.A. Maghembe, SADCC/ICRAFAgroforestry Project

Zomba, MalawiF. Makeschin, Institut fur Bodenkunde und Standortslehre,

Tharandt, GermanyDC. Malcolm, University of Edinburgh, Edingburgh, UKE Malkonen, Rnnish Forest Research Institute, Vantaa,

Rnland :MAR. Nahuz, Institute de Pesquisas Tecnologicas,

Sao Paulo, SP, BrazilR. Paivinen, European Forestry Institute, Joensuu, RnlandS.G. Pallardy, University of Missouri, Columbia, MO, USAR.F. Powers, Pacific Southwest Research Station, Redding,

CA.USAT. Pukkala, University of Joensuu, Joensuu, RnlandF.E Putz, University of Rorida, Gainesville, FL, USAL. Rasmussen, RISO, Roskilde, DenmarkD.D. Reed, Michigan Technological University, Houghton,

Ml, USAR. Sands, University of Canterbury, Christchurch, NZJ.A. Stanturf, Stoneville, MS, USAQ Sziklai, University of British Columbia, Vancouver, B.C.,

CanadaJ.R. Toliver, USDA Forest Service, Washington, DC, USAK. von Weissenberg, University of Helsinki, Helsinki, RnlandQ Whitehead, Manaaki Whenua Landacre Research,

Lincoln, New Zealand

Publication Information: Forest Ecology and Management (ISSN 0378-1127). For 1999 volumes 111-122 are scheduled for publication. Subscrip-tion prices are available upon request from the Publisher. Subscriptions are accepted on a prepaid basis only and are entered on a calendaryear basis. Issues are sent by surface mail except to the following countries where air delivery via SAL mail is ensured: Argentina, Australia,Brazil, Canada, Hong Kong, India, Israel, Japan, Malaysia, Mexico, New Zealand, Pakistan, PR China, Singapore, South Africa, South Korea,Taiwan, Thailand, USA. For all other countries airmail rates are available on request Claims for missing issues should be made within sixmonths of our publication (mailing) date.Orders, claims, and product enquiries: please contact the Customer Support Department at the Regional Sales Office nearest you:New Yoric Bsevier Science. PO Box 945, New York, NY 10159-0945, USA; TeL (+1) (212) 633 3730, [toll free number for North American cus-tomers: 1-888-4ES-INFO (437-4636)]; fax (+1) (212) 633 3680; e-mail [email protected]: Elsevier Science, PO Box 211,1000 AE Amsterdam, The Netherlands; TeL: (+31) 20 485-3757; fax: (+31) 20 485 3432; [email protected]: Bsevier Science, 9-15, Higashi-Azabu 1-chome, Minato-ku, Tokyo 106, Japan; Tel. (+81) (3) 5561 5033; fax (+81) (3) 5561 5047; e-mail:[email protected]: Elsevier Science, No. 1 Temasek Avenue, #17-01 Millenia Tower, Singapore 039192; TeL (+65) 434 3727; fax: (+65) 337 2230; e-mail:[email protected] de Janeiro: Bsevier Science, Rua Sete de Setembro 111/16 Andar, 20050-002 Centra, Rio de Janeiro - RJ, Brazil; phone: (+55) (21) 5095340; fax: (+55) (21) 507 1991; e-mail: [email protected] [Note (Latin America): for orders, claims and help desk information, pleasecontact the Regional Sales Office in New York as listed above]

Forest Ecologyand

Management

ELSEVIER Forest Ecology and Management 114 (1999) 173-197 -

Ecosystem management decision support for federal

forests in the United States: A review1'2

H. Michael Rauscher*

USDA Forest Service, Southern Research Station, Bent Creek Research and Demonstration Forest, 1577 Brevard Rd.,AshevUle, NC 28805, USA

Abstract

Ecosystem management has been adopted as the philosophical paradigm guiding management on many federal forests in theUnited States. The strategic goal of ecosystem management is to find a sensible middle ground between ensuring long-termprotection of the environment while allowing an increasing population to use its natural resources for maintaining andimproving human life. Ecosystem management has all the characteristics of 'wicked' problems that are tricky, complex, andthorny. Ambiguities, conflicts, internal inconsistencies, unknown but large costs, lack of organized approaches, institutionalshock and confusion, lack of scientific understanding of management consequences, and turbulent, rapidly changing powercenters all contribute to the wickedness of the ecosystem management paradigm. Given that ecosystem management, likehuman survival and welfare, is a wicked problem, how can we proceed to tame it? Managers need to use the same tools thatpeople have always used for handling such problems - knowledge, organization, judicious simplification, and inspiredleadership. The generic theory of decision support system development and application is well developed. Numerous specificecosystem management decision support systems (EM-DSS) have been developed and are evolving in their capabilities. Thereis no doubt that given a set of ecosystem management processes to support and adequate time and resources, effective EM-DSS can be developed. On the other hand, there is considerable doubt that sufficiently detailed, explicitly described andwidely accepted processes for implementing ecosystem management can be crafted given the current institutional,educational, social and political climate. A socio-political climate hi which everyone wants to reap the benefits and no one

• wants to pay the costs, incapacitates the federal forest management decision making process. Developing a workableecosystem management process and the decision making tools to support it is probably one of the most complex and urgentchallenges facing us today. This paper offers a concise review of the state of the art of decision support systems related toimplementing ecosystem management A conceptual model of the context hi which ecosystem management is expected tofunction is'presented. Next, a candidate for an operational ecosystem management process is described and others arereferenced. Finally, a generic ecosystem management decision support system is presented and many existing systems brieflydescribed. © Published by Elsevier Science 1999.

Keywords: Ecosystem management; Decision analysis; Management systems; Management philosophy; Decision support system

"•Corresponding author. Tel: +1-828-667-5261; fax: +1-828-667-9097; e-mail: mrauscher/[email protected]'Presented at the Firest Biennial North American Forest Ecology Workshop, 23-26 June 1997, Raleigh, NC.2The opinions expressed by. the author do not mecessarfly represent the policy or position of the U.S. Department of Agriculture or the

Forest Serviece. .

0378-1127/997$ - see front matter © Published by Elsevier Science 1999.PII: 80378-1127(98)00350-8

174 H.M. Rauscher/Forest Ecology and Management 114 (1999) 173-197

1. Introduction

After almost 20 years of increasingly contentiouspublic unhappiness with the management of nationalforests, the USDA Forest Service officially adoptedecosystem management as a land management para-digm (Overbay, 1992). Other federal forest land man-agement agencies such as the USDI Bureau of LandManagement, the USDI National Park Service, theUSDI Fish and Wildlife Service, the USDC NOAA,and the Environmental Protection Agency have alsomade the commitment to adopt ecosystem manage-ment principles (FEMAT, 1993; GAO, 1994). Manyexcellent historical reviews trace the history ofenvironmental management on forest lands in theUnited States (Botkin, 1990; Kimmins, 1991; Ken-nedy and Quigley, 1993; Caldwell et al., 1994; Shands,1994).

Ecosystem management represents different thingsto different people. A recent report by the UnitedStates general accounting office states that ecosystemmanagement is a popular concept partly because"there is not enough agreement on the meaning ofthe concept to hinder its popularity" (GAO, 1994). Atthe heart of the ecosystem management paradigm liesa shift in emphasis away from sustaining yields ofproducts towards sustaining the ecosystems that pro-vide these products (Thomas, 1995). Overbay (1992)provided the definition of ecosystem management that

. the USDA forest service uses: 'Ecosystem manage-ment is the means we use to meet the goals specified inour programs and plans. Ecosystem management isthe means to an end. It is not an end itself. We do notmanage ecosystems just for the sake of managingthem or for some notion of intrinsic ecosystem values.

, We manage them for specific purposes such as produ-cing, restoring, or sustaining certain ecological con-ditions; desired resource uses and products; vitalenvironmental services; and aesthetic, cultural, orspiritual values".

In contrast, non-governmental scientists tend todefine ecosystem management hi terms of sustaining'intrinsic ecosystem values.' For example, Grumbine(1994) identified 10 dominant themes of ecosystem

. management which led to his formulating the follow-ing definition: "Ecosystem management integratesscientific knowledge of ecological relationshipswithin a complex sociopolitical and values framework

toward the general goal of protecting native ecosystemintegrity over the long term."

In this view, ecosystem management is specificallynot aimed at resource management, but rather onprotecting ecosystem integrity and the needs ofnon-human life for its own sake - both 'intrinsicecosystem values' (Grumbine, 1994). As the ecosys-tem management concept evolves, debates over defi-nitions, fundamental principles, and policyimplications will probably continue and shape thenew paradigm in ways not yet discernible. A strategicgoal for ecosystem management on federal forestsmight be to find a sensible middle ground betweenensuring the necessary long-term protection of theenvironment and protecting the right of an ever-grow-ing population to use its natural resources to maintainand improve human life.

The ecosystem management paradigm was adoptedquickly. No formal studies were conducted to identifythe consequences of the changes ushered in by thisnew approach nor was any well-documented, widelyaccepted organized methodology developed for itsimplementation (Thomas, 1997). Today, ecosystemmanagement remains primarily a philosophical con-cept for dealing with larger spatial scales; longer timeframes; and the requirement that management deci-sions must be socially acceptable, economically fea-sible, and ecologically sustainable. As the definitionand fundamental principles that make up the ecosys-tem management paradigm have not yet been resolvedand widely accepted, the challenge is to build theecosystem management philosophical concept into anexplicitly defined, operationally practical methodol-ogy (Wear et al., 1996; Thomas, 1997). Effectiveecosystem management processes are urgently neededto allow federal land managers to more effectivelyaccommodate the continuing rapid change in societalperspectives and goals (Bormann et al., 1993).

Ecosystem management represents a shift fromsimple to complex definitions of the ecosystems wemanage (Kohm and Franklin, 1997). Ecosystem man-agement will require the development of effective,multi-objective decision support systems to: (1) assistindividuals and groups in their decision making pro-cess; (2) support rather than replace the judgement ofthe decision makers; and (3) improve the quality,reproducibility, and explainability of the decisionprocess (Janssen, 1992; Larsen et al., 1997; Reynolds

H.M. Rauscher/Forest Ecology and Management 114 (1999) 173-197 175

et al., 1998). The complexity of environmentaldynamics over time and space; the overwhelmingamounts of data, information, and knowledge hi dif-ferent forms and qualities; and the multiple, oftenconflicting, management goals virtually guarantee thatfew individuals or groups of people can consistentlymake good decisions without powerful decision sup-port tools (Janssen, 1992).

This paper reviews ecosystem management deci-sion processes, the decision support systems availableto implement them, and discusses some issues thatmust be settled before ecosystem management canevolve from a philosophical concept to a practicaltool.

2. The nature of the ecosystem managementproblem

The lengthy and ongoing struggle to manage federalforests over the last 20 years has taught managersmany characteristics of the ecosystem managementproblem. First, societal goals, preferences, and valuesare numerous, ambiguous, and often in conflict Sec-ond, legal mandates are complex, unclear, and at timesself-contradictory. Third, policy direction is missing,ambiguous, or incomplete with a tendency to rapidlyshift in response to political pressure. Fourth, no well-defined and widely accepted decision making processexists. Decisions and the decision making process areusually based on trial-and-eiror methods and local,pragmatic inventions of necessity. Fifth, participatingdecision makers and stakeholders vary in the amountof time and effort they contribute to any one decisionwhile engaging only sporadically in the decisionmaking process. Sixth, no widely accepted methodis available for producing concensus among oftencontentious stakeholders. Individuals or small mino-rities have the power to block decisions at any timethrough judicial challenges resulting in managerialgridlock. Seventh, decisions must be made aboutactions and their consequences based on missingand uncertain data, and often inaccessible scientificknowledge about ecosystems. Also, the ecosystemmanagement problem is not as much about scienceas k is about politics (Rittel, 1972; Grumbine, 1994).The ecosystem management debate is a competitive,conflict-laden social process that determines how

power flows in resource management (Grumbine,1994; Chase, 1995).

2.1. The characteristics of wicked problems

Clearly ecosystem management is a 'wicked' orunstructured problem as defined by Rittel andWebber (1973) and introduced to forestry by Allenand Gould (1986). 'Wicked' is used here hi the senseof tricky, complex, and thorny. Wicked problemshave no definitively correct formulations. Stake-holders can define the problem on their own terms.Any one definition can only be more or less usefuldepending upon the definition of useful. Wickedproblems have no stopping rule to identify whenthey are 'solved.' Solutions are not true or false, butgood or bad and the only way to test the goodness orbadness of solutions is to execute them. Wickedproblems do not have an enumerable or an exhaus-tively describable set of potential solutions. Since theytend to be important with significant consequences,decision makers have no right to be wrong, making thedecision process intensely agonizing and usually frus-trating (Allen and Gould, 1986). Finally, we do nothave a theory that tells us how to identify a sociallybest state such as 'the greatest good for the greatestnumber' (Rittel and Webber, 1973). Problem-solvingprocesses for ecosystem management and otherwicked problems can be developed, but perhaps notby exclusive use of rationally based, operationsresearch or systems analysis methods (Rittel, 1972;Allen and Gould, 1986; Hashim, 1990). Such methodshave worked well for the 'tame' problems of scienceand technology where the impact of the human dimen-sion on the problem is eliminated or strictly con-trolled. Using tame problem-solving methods onwicked problems often results in failure. Barber andRodman (1990) provide a devastating^ powerfulillustration.

Given that ecosystem management, like humansurvival and welfare, is a wicked problem, how canwe tame it? We must use the same tools people havealways used for handling wicked problems - knowl-edge, organization, judicious simplification, andinspired leadership. The remainder of this sectionfocuses on examining a number of organizationalstructures for implementing ecosystem management.Three levels of organization are important hi the


Ecosystem Management Decision Environment

(Goals, Values, Constraints)

Ecosystem Management Organization andDecision Making Processes

Ecosystem Management Decision Support System

User DSS Software

(Social, Economic, Political, and Legal Context)

Fig. 1. Three levels of organization important in the ecosystem management decision making process.

ecosystem management decision process (Adelman,1992): the decision making environment, the decision-making organization and process(es), and the EM-DSS which includes human decision maker(s) andtheir software support tools (Fig. 1).

2.2. The decision-making environment

Creating a vision of the decision making environ-ment in which ecosystem management must functionis itself a wicked problem with no single best answer.However, thanks to Bonnicksen (1991), we do have auseful illustration of an ecosystem management deci-sion environment (Fig. 2). Ecosystem management iscomposed of two parts: an ecological subsystem and amanagement subsystem. Ecosystems are communitiesof organisms and their environment whose boundariesare defined by an observer to facilitate some humanpurpose such as research or management People needto be considered as part of the community of organ-isms that inhabit, use, or directly influence an eco-system (Behan, 1997). Thus people, in their role asusers, are part of the ecological subsystem, like trees,soil, and wildlife. Management is denned as makingdecisions about and controlling systems to achievedesired ends. People in the management role partici-pate hi ecosystem management in a very different role.They are the risk takers, the objective setters, the

Ecological AssessmentsEcosystem M^flg?1' tfsA. DSS

EcologicalSubsystem Resources

\Neutral Biophysical

Objects

Fig. 2. The ecosystem management decision environment.

judges of value, the substitutes, hi other words, thedecision makers.

Local management of ecosystems occurs within andis influenced by national and international social,economic and political systems. Similarly, local eco-systems operate within and are influenced by largerbiophysical systems such as eco-regions or biomes.An important distinction that Bonnicksen (1991)makes is that both the ecosystem and the managementsubsystems of ecosystem management are part of aninterlocking, nested hierarchy. FEMAT (1993) andKaufmann et al. (1994) support this important pointEcosystem management can and should occur at manyscales - global/international, biome/national, eco-

KM. Rauscher/Forest Ecology and Management 114 (1999) 173-197 177

region/multi-state, or forest landscape/national forest(GAO, 1994, p. 62). At the regional, national, andinternational scales, the ecosystem management deci-sion process should render the mosaic of environmen-tal issues manageable by: (1) labeling the issues; (2)defining the problems; and (3) identifying who iscausing the problems, who has the responsibility ofsolving the problems, who are the stakeholders asso-ciated with the problems, who will pay for finding andimplementing solutions (Hannigan, 1993). The deci-sion process at this macro-scale should also coordinatesolution efforts and supervise social and ecologicalsystem sustainability (Tonn et al., 1998). Decisionsupport systems that operate on the multi-state andnational scale have been developed and tested inEurope and can serve as illustrations of what is neededfor United States federal forest management (Van denBerg, 1996). There is as much work to be done inecosystem management at the larger scales as there isat the local national forest scale. At present, no oneagency, committee, or other organized body in theUnited States manages ecosystems at the scale sug-gested by Bonnicksen (1991) and Tonn et al. (1998). Itis not obvious that our society has addressed the needto manage ecosystems at the biome/national and eco-regional/multi-state scale to cope with the complexenvironmental problems we have created for ourselvesat that scale (Caldwell, 1996). It seems reasonable thatwe should try.

The specific values, goals, and constraints thatcharacterize public preferences and needs may beidentified through a group negotiation process invol-ving a variety of stakeholders and management deci-sion makers (Fig. 2). Management decision makers

• organize and lead the group negotiation process andensure that the resultant goals and desired futureconditions are socially acceptable, legal, economic-ally feasible, and ecologically sustainable. If ecosys-tem management decision support systems (EM-DSS)are available, all participants need to be able to rely onthem at a reasonable level of confidence for relevantinformation and analyses. This group negotiationprocess is probably the most difficult part of ecosys-tem management (Bonnann et al., 1993).

The ecological subsystem contains physical or con-ceptual objects such as trees, bkds, deer, rivers, smells,sights, and sounds. Each of these physical or concep-tual objects can be a resource if it has positive value in

the minds of people or a pest if it has negative value.Otherwise it is value-neutral (Behan, 1997). A criticalfeature of this view is that as goals and desired futureconditions change, objects can change from beingresources to value-neutral or even to being pests(Bonnicksen, 1991). For example, the white-taileddeer, once regarded as a sought after resource, isnow considered a pest in some eastern United Statesforest ecosystems. Most forest managers, administra-tors, and scientists are much more familiar with thestructure and function of the ecological subsystemthan with the management subsystem. This state ofaffairs is indicative of where we have put our attentionand energy in the past. One of the new messages ofecosystem management is that forest managers,administrators, and scientists need to rapidly redressthis imbalance.

While biophysical scientists at the national andinternational scale struggle to understand local, regio-nal, and global environmental systems, social andinstitutional scientists must struggle to understandhow to sustain societal systems that will protect theability of humans and nature to co-evolve (Bonnannet al., 1993; Tonn et al., 1998). Thus, defining andunderstanding the nature of sustainable societies andthe nature of sustainable ecosystems are equallyimportant Tonn and White (1996) described sustain-able societies as wise, participative, tolerant, protec-tive of human rights, spiritual, collaborative,achievement-oriented, supportive of stable commu-nities, able to make decisions under uncertainty, andable to learn over time. One of the defining character-istics of a society will be how effectively it manages tosustain both itself and its ecosystems. Even a cursoryreview of history reveals numerous extinct civiliza-tions that did not successfully sustain both society andecosystem (Toynbee, 1946).

The study of the management subsystem mustinclude understanding the dynamics of public prefer-ences, conflict management and resolution, and costevaluation and containment as it relates to ecosystemmanagement Defining and understanding stake-holders and their preferences is an important part ofecosystem management (Garland, 1997). Stakeholderand general public preferences are volatile and sensi-tive to manipulation through the control of informa-tion transmitted through public media (Montgomery,1993; Smith, 1997). Understanding the dynamics of

178 KM. Rauscher/Forest Ecology and Management 114 (1999) 173-197

social preferences and how they can be influencedover both the short- and long-term is a vital part of theecosystem management process. Ecosystem manage-ment processes and the institutions that use them mustbe able to detect and accommodate rapid, and some-times radical, changes hi public preferences (Kohmand Franklin, 1997).

Successful social conflict management is as impor-tant as understanding stakeholder preferencedynamics. Currently, the dominant means of settlingpublic land disputes have been either litigation orquasi-judicial administrative appeals. Such conten-tious methods of handling disputes expend muchgoodwill, energy, time, and money. These methodsproduce winners and losers, may leave fundamentaldifferences unresolved, and potentially please few ornone of the parties (Daniels et al., 1993). Decisionmakers need a fundamental understanding of thenature of environmental conflicts and disputes andhow to use conflict-positive dispute management tech-niques effectively (Daniels et al., 1993). Newapproaches to managing the social debate surroundingecosystem management, such as alternative disputeresolution (ADR) techniques (Floyd et al., 1996),should be evaluated, taught, and used. Adaptive man-agement techniques are as applicable to the manage-ment side of ecosystem management as they are to theecosystem side. They could be used to suggest a seriesof operational experiments that study actual publicparticipation and conflict management activities toquickly determine what works and what does not(Daniels et al., 1993; Shindler and Bruce, 1997).

The ability of federal land managers to avoid grid-lock is heavily dependent on stakeholder willingnessto negotiate and ultimately agree on the goals forecosystem management (Bormann et al., 1993).Unfortunately, people sometimes have preferencesbased on core values that are so strong and so con-flicting that no solution is acceptable (Smith, 1997).To avoid societal gridlock, we must design and imple-ment robust strategies that encourage voluntary con-flict resolution among contentious stakeholders andexplore other options leading toward a settlement ifvoluntary resolution is impossible. Such options mightinclude binding arbitration, an agreed upon delay inorder to improve our data and knowledge about theecosystem, or various other forms of conflict resolu-tion.

2.3. The cost of ecosystem management

Ecosystem management cost evaluation and con-tainment is a critical area for economists to study. As ageneral rule, increases hi problem complexity and thedegree of wickedness increase the cost of findingsatisfactory solutions (Klein and Methlie, 1990). Eco-system management should accommodate limits ontime, expertise, and money (Smith, 1997) becausesustainable forest management is impossible if thereare unsustainable social and economic costs (Craig,1996). Documentation of costs needs to be preparedand made public because few people know or appreci-ate the costs of efforts to solve wicked problems. Forexample, the USDA forest service has spent approxi-mately US$ 2 billion, equal to 16% annually of theentire national forest system budget, on planning sincethe National Forest Management Act was passed in1976 (Behan, 1990). The additional cost of imple-menting ecosystem management prescriptions andmonitoring and evaluating the results has not beenestimated. Are we willing or able to marshal thefunding to implement ecosystem management thatwill ensure that federal forest managers can complywith the law and satisfy public preferences? Theamount of money that could be spent on ecosystemmanagement nationally may be extremely large andidentifying clear benefits may be difficult (Oliveret al., 1993).

In the last century of federal forestiand manage-ment, timber harvesting has largely paid for multiple-use management activities. Many forecast that thelevel of timber harvesting under ecosystem manage-ment will greatly decline while the cost of ecosystemmanagement will greatly increase. Until managersevaluate the true cost and benefits, it will be difficultto determine whether the public is willing to pay forecosystem management programs. In any case, a newand rational means of capital resource allocation willbe required to fund the ecosystem management pro-cess adopted (Sample, 1990; Kennedy and Quigley,1993; Oliver et al., 1993). Refusing to fund ecosystemmanagement and opting for Ihe 'do nothing' alter-native is likely to result in unacceptable desired futureconditions. "Plant and animal species do not stopgrowing, dying, and burning; and floods, fires, andwindstorms do not stop when all management issuspended" (Oliver et al., 1993). Nature does not

HM. Rauscher/Forest Ecology and Management 114 (1999) 173-197 179

appear to care - either about threatened and endan-gered species or about humans. People care and peoplemust define goodness and badness. Nature will not doit for us. "Nature in the 21st century will be a naturethat we make; the question is the degree to which thismolding will be, intentional or unintentional, desirableor undesirable" (Botkin, 1990). Making the naturethat we want may be expensive. A good understandingof ecological economics will help society makerational choices.

2.4. Administering ecosystem management

Recently, regional ecological assessments havebeen used to describe the large-scale context forecosystem management and therefore can be consid-ered a decision support tool (Fig. 2). Regional assess-ments have been large, collaborative interagencyefforts, often with public stakeholder participation,that have taken 2-5 years and several millions ofdollars to finish. The objectives for integrated ecolo-gical assessments are to provide: (1) a description ofcurrent and historic composition, structure, and func-tion of ecosystems; (2) a description of the biotic(including human) and abiotic processes that contrib-uted to the development of the current ecosystemconditions; and (3) a description of probable futurescenarios that might exist under different types ofmanagement strategies (Lessard, 1997). The SouthernAppalachian Assessment (Southern Appalachian Manand the Biosphere, 1996) is an example although itdoes not address objective (3). Currently, preciselyhow these regional assessments fit into the ecosystem

. management process is unclear. One alternative would'• be to view regional assessments as ecosystem manage-ment at the eco-region/multi-state scale. In that capa-city, the current objectives for assessments focusentirely too much on the ecosystem component ofecosystem management SAMAB, for example,examines the social and economic activities of peoplewithin the southern Appalachian region, but only hitheir roles as ecosystem members and users. The roleof people as managers of ecosystems, including theirrole as managers of social, economic and politicalsystems hi the region, is ignored. To correct thisdeficiency, another list of objectives for regionalassessments might include the following: (1) identifya set of regional scale desired future conditions and

compare these to the current conditions; (2) identifyregional stakeholders, then: preferences and values,and how they compare to the general public in theregion; (3) identify the legal and political climatewithin which ecosystem management must function;(4) identify the regional costs and who will bear themas well as the regional gains and who will reap them;and (5) identify the major problems, who is respon-sible for solving them, and who has supervisoryresponsibility to monitor progress and assure that asatisfactory solution is eventually reached.

3. Ecosystem management processes

The decision making environment consists of thesocial, economic, political, and legal context hi whicha federal ecosystem management organization oper-ates. This decision making environment determinesthe goals, values, and constraints for the organization(Fig. 1). Organizational policy then translates themandates of the decision making environment intospecific decision making processes. A decision mak-ing process is a method or procedure that guidesmanagers through a series of tasks from problemidentification and analysis to alternative design andfinally alternative selection (Mintzberg et al., 1976).Ideally, decision support systems should not be devel-oped until the ecosystem management decision mak-ing processes they are to support have beenarticulated. In reality, both the ecosystem managementdecision processes and the software systems needed tosupport them are evolving simultaneously, each help-ing to refine the other.

'First generation' ecosystem management pro-cesses have evolved from two sources: (1) academiawhere several ecosystem management processes havebeen described at a general, conceptual level and theirmacro-level structures and functions have been iden-tified; and (2) federal forest managers at the field levelwhere numerous, local ad hoc processes have beendeveloped and tested under fire. The academic, highlevel descriptions of ecosystem management pro-cesses do not supply adequate details to guide thedevelopment of decision support systems and aretheoretical, lacking adequate field testing to determinehow they work in practice. The local, ad hoc ecosys-tem management processes are too numerous for

180 .HM. Bauscher/Forest Ecology and Management 114 (1999) 173-197

effective software-based decision support (approxi-mately 400 ranger districts in the U.S. national forestsystem each have then: own process) and few, if any,have been studied and described formally so thatsimilarities and differences can be identified. More-over, no particular ecosystem management decisionprocess(es) have been widely accepted and implemen-ted in federal forest management We should devote asmuch creative attention to devising good ecosystemmanagement decision processes as we do in assuringthe quality of the decisions themselves (Ticknor,1993). In this section the major elements of a genericecosystem management process are identified basedon a synthesis of the literature.

Adaptive management is a continuing process ofplanning, monitoring, evaluating, and adjusting man-agement methods (Bormann et al., 1993; FEMAT,1993; Lee, 1993). The usefulness of adaptive manage-ment as an ecosystem management process is beingfield-tested by the Forest Service in the Northwest andin other regions of the country (Shindler et al., 1996).Described at the most general level, adaptive manage-ment consists of four activities, PLAN-ACT-MONI-TOR-EVALUATE, linked to each other in a networkof relationships (Fig. 3). At each cycle, the results ofthe evaluation activity are fed back to the planning,activity so that adaptive learning can take place. With-out adding further detail to this definition, almost anymanagement activity could erroneously be labeledadaptive management In reality, adaptive manage-ment is a well-described, detailed, formally rigorous,and scientifically defensible management-by-experi-ment system (Walters and Rolling, 1990). Baskerville(1985) prescribes a nine-step process for implement-ing adaptive management correctly. Adaptive man-

Authorizationto Implement

ProblemIdentification

I

Alternative Alternative AutnorizationDevelopment Selection to Implement

Revised Goals

New Technology MONITOR

Fig. 3. The adaptive management process for ecosystem manage-ment

Feedback Loops

Fig. 4. A detailed view of the planning activity in the adaptivemanagement process for ecosystem management

agement requires that a series of steps be followed foreach of the four major activities described.

3.1. Plan

The Mintzberg et al. (1976) planning process can beviewed as a detailed description of the planning stageof the adaptive management process (Fig. 4). Janssen(1992) argued that the planning stage of any decisionprocess will generally need to be some variant of theMintzberg et al. (1976) method. The planning stage ofthe adaptive management process consists of foursteps: (1) problem identification including goal selec-tion, (2) alternative development (3) alternative selec-tion, and (4) authorization to implement the selectedalternative (Figs. 3 and 4). Each of these major stepscan be decomposed into one or more phases (Janssen,1992).

The problem identification step consists of twophases: (la) recognition - to identify opportunities,problems, and crises; the need for a decision launchesthe decision process and (Ib) diagnosis - to explorethe different aspects of the problem situation, identifygoals and objectives and decide how to approach theproblem. If the diagnosis phase, step Ib, is unneces-sary, it can be skipped (Fig. 4). The alternative devel-opment step consists of three phases: (2a) search - tofind previously designed and tested solutions to theentire problem or to any of its parts; (2b) design - todevelop new alternatives; and (2c) screen - to deter-mine whether the number and quality of the alterna-tives found, developed, or both, provide an adequaterange of choices for the selection step. The selectionstep consists of three phases: (3a) analysis and evalua-


tion - to evaluate and understand the consequencesover space and time of each of the proposed alter-natives and communicate these results clearly to thedecision makers; (3b) judgment and choice - whereone individual makes a choice; and (3c) bargainingand choice - where a group of decision makersnegotiates a choice. The authorization step consistsof two outcomes: (4a) authorization achieved -approval inside and outside the institutional hierarchyis obtained, marking the end of the planning processand the beginning of the implementation process; and(4b) authorization denied - evaluation of the cause fordenial and looping back to the appropriate part of thedecision process to make another attempt at achievingauthorization (Janssen, 1992).

A particular decision can take many pathwaysthrough Hg. 4 and iterative cycles are a normal partof how environmental decisions are actually made(Janssen, 1992). These cycles occur as the decisionparticipant's understanding of a complex problemevolves and when alternative solutions fail to meetadministrative, scientific, or political standards.Mintzberg et al. (1976) maintained that problemscan be classified into seven types and that the solution

: cycle for each type can be mapped on Hg. 4. Janssen(1992) illustrates this point by presenting and discuss-ing the solution cycles of 20 actual environmentalproblems in the Netherlands ranging from measures toreduce NH3 emissions, to clean-up of a polluted site,to protecting forests from acid rain. The effectivenessof competing EM-DSS may be evaluated by howmany of the above phases are supported, how wellthey are supported, and whether the complex iterative.cycling of real-world problems is supported (Janssen,1992).

3.2. Act

The planning stage of adaptive management,described above, results in decisions about goalsand constraints. The action stage determines how,where, and when to implement activities to achievethe goals and adhere to the constraints. Given a clearstatement of management goals and objectives, theimplementation stage creates testable adaptive man-agement hypotheses, explicitly describes the assump-tions supporting them, and generates an appropriateset of targeted actions (Everett et al., 1993). How each

hypothesis is tested must be carefully and clearlydocumented (Walters and Rolling, 1990; Lee, 1993;Kiinmins, 1995).

3.3. Monitor

Documentation in the action stage is stressedbecause the monitoring stage often occurs monthsto years later, and the individuals who implementedthe actions may not be involved in monitoring orsubsequent evaluation. Documentation may be theonly link between the two stages. The monitoredresults of experimental actions must also be recordedcarefully and in detail so that a complete, understand-able package exists for the evaluation stage.

This definition of the monitoring stage of adaptivemanagement has immediate consequences. What vari-ables are monitored and when, how, and where theyare monitored depend almost entirely on the hypoth-eses created in the action stage and on the type ofactions determined necessary to test those hypotheses.A unique goal - hypothesis - action sequence willprobably need to be designed for each specific man-agement unit. Each unit will probably have its ownunique monitoring requirements to distinguishbetween the adaptive experimental hypotheses pro-posed for that it Therefore, no general, broad-spec-trum monitoring program can or should be designed tosupport adaptive management. Adaptive managementmeans management-by-experiment Management-by-experiment requires hypotheses that must be imple-mented and tested. Monitoring can only occur after thehypotheses have been designed and theirtests devisedso that it is clear what needs monitoring (FEMAT,1993).

3.4. Evaluate

Finally, the documentation describing each adaptivemanagement experiment must be analyzed and theresults evaluated. Promising statistical methods havebeen identified (Carpenter, 1990), but using themrequires considerable expertise. At the end of theadaptive management cycle, a written report willcommunicate the results publicly to stakeholdersand managers and influence future cycles of theplanning activity of adaptive management (Everettet al., 1993). In fact, an analysis of all adaptive


experimental results should be compiled periodicallyand forwarded to the next higher planning level forcorrective change leading to new actions.

Adaptive management, when implemented asdefined by Walters and Rolling (1990), FEMAT(1993), and Lee (1993), is a complex and challengingprocess. The adaptive management process is not a'license to manipulate forest ecosystems haphazardlysimply to relieve immediate socio-political pressure(Everett et al., 1993). Adaptive management must beapplied correctly and rigorously as management-by-experiment if we are to achieve our stated goals."Managing to learn entails implementing an arrayof practices, then taking a scientific approach indescribing anticipated outcomes and comparing themto actual outcomes. These comparisons are part of thefoundation of knowledge of ecosystem management"(FEMAT, 1993). The whole point of adaptive manage-ment is to generate change in the way ecosystemmanagement is applied.

A number of institutional challenges must beaddressed before adaptive management can make itsexpected positive contribution in the ecosystem man-agement process (Lee, 1993). Adaptive managementrequires a greater level of statistical experimentaldesign and analysis expertise than other competingdecision processes. Kessler et al. (1992) suggested thatadaptive management can only be done through closecollaboration between forest managers and scientists."Finding creative ways of conducting powerful testswithout forcing staffs to do things they think are wrongor foolish is of central importance to the human part ofadaptive management" (Lee, 1993). Managers andinterested stakeholders must accept that adaptive man-agement means leaking small, controlled mistakes toavoid making big ones. Keeping adaptive manage-ment unbiased may be difficult. Research that hasconsequences is research with which managers maytry to tamper or keep from happening (Lee, 1993). Thecosts of monitoring the results and documenting theentire managerial experiment properly are unknown,but anticipated to be high (Smith, 1997). Adaptivemanagement, supplemented by the Mintzberg et al.(1976) planning process, is an attractive candidate foran ecosystem management process at several opera-tional scales. Despite much supportive rhetoric, theinstitutional and funding changes needed to imple-ment adaptive management as an ecosystem manage-

ment process for federal forestiand management havenot yet been widely accomplished.

The adaptive management concept appears to be themost well-developed candidate for an operationalecosystem management process. However, othersshould be investigated. Lindblom (1990), cited bySmith (1997), advocated a concept called 'probing'as a candidate for the ecosystem management decisionprocess. Probing is an informal process of observation,hypothesizing, and data comparison in which, peopleof all backgrounds can engage. Jensen and Everett(1993) pointed to a method called a 'land evaluationsystem' as a candidate for an ecosystem managementprocess. The land evaluation system (Zonnefeld,1988) has been used by the United Nations Foodand Agricultural Organization and by the InternationalSociety for Soil Sciences in forestry land-use plan-ning. Howitt (1978), cited by Allen and Gould (1986),offered a 'simple' approach to dealing with decisionprocesses for wicked problems that might be useful forecosystem management. Rittel (1972) advocated a'second generation systems approach' to wicked pro-blem solution based on the logic of arguments (Con-klin and Begeman, 1987; Hashim, 1990). Problemsand their consequences can be made understandable toindividuals and groups by asking and answering cru-cial questions while diagramming the process usingthe formal logic of argumentation. Vroom and Jago(1988), cited in Sample (1993), suggested their 'con-tingent decision process' may be used for problemslike ecosystem management. Any of the above-nameddecision making processes used to implement ecosys-tem management in the United States must satisfy therequirements of the 1969 National EnvironmentalPolicy Act (NEPA) and the 1976 National ForestManagement Act (NFMA).

Several formal, well-described candidates for anecosystem management decision process have beenintroduced here. In addition, numerous local, ad hocdecision processes have been developed and testedunder fire in every ranger district hi the U.S. NationalForest System. Few case studies (e.g. Steelman, 1996)have been published and not many evaluations (e.g.Shindler and Bruce, 1997) of the strengths and weak-nesses of these informal decision making processeshave been conducted. Surely a concerted effort tostudy the existing formal and informal ecosystemmanagement processes would result in some powerful

EM. Rauscher /Forest Ecology and Management 114 (1999) 173-197 183

candidates to implement ecosystem management Asthe adaptive management concept can also be used toimprove our management systems, it may not beoverly important which particular ecosystem manage-ment decision processes we choose. It is, however,critically important that we choose several and thenuse the adaptive management philosophy to test andimprove them in real-life situations (Kimmins, 1991).

4. Decision support systems (DSS) defined

DSS help managers make decisions in situationswhere human judgment is an important contributor tothe problem solving process, but where limitations inhuman information processing impede decision mak-ing (Rauscher, 1995). The goal of a DSS is to amplifythe power of the decision makers without usurpingtheir right to use human judgment and make choices.DSS attempt to bring together the intellectual flex-ibility and imagination of humans with the speed,accuracy, and tirelessness of the computer (Kleinand Methlie, 1990; Sage, 1991; Turban, 1993; Hol-sapple and Whinston, 1996).

A DSS contains a number of subsystems, each witha specific task Fig. 5. The first, and most important, isthe subsystem composed of the decision maker(s). Thedecision makers are consciously diagrammed as partof the DSS because without their guidance, there is noDSS. The group negotiation management subsystemhelps the decision maker(s) organize their ideas, for-mulate relationships surrounding issues and argu-

USER INTERFACE

Decision Makers

Fig. 5. The major components of a generic DSS.

ments, and refine their understanding of theproblem and their own value systems (Jessup andValacich, 1993; Holsapple and Whinston, 1996).Examples of group negotiation tools include: AR/GIS (Faber et al., 1997), the issue-based informationsystem (IBIS) (Conklin and Begeman, 1987; Hashim,1990), and various socio-ecological logic program-ming models (Thomson, 1993,1996). Group negotia-tion tools are used to construct issue-based argumentstructures using variants of belief networks to clarifythe values and preferences of group members in theattempt to reach group consensus. For example, IBISuses formal argument logic (the logic of questions andanswers) as a way to diagram and elucidate argumen-tative thinking (Hashim, 1990). By asking and answer-ing crucial questions, you can begin to betterunderstand the problem and its solution set Under-standing the meaning of terms, and through them ourthoughts, lies at the heart of collaborative manage-ment Greber and Johnson (1991) illustrated how themalleable nature of terms, in this case 'overcutting,'creates logically defensible differences of opinionwhich have nothing to do with a person's honestyor dishonesty in the argument DSS should specificallydeploy mechanisms by which biological realitiesguide and, if appropriate, constrain the desires ofthe stakeholders (Bennett, 1996). For example, com-promise is not acceptable for some issues. If theproductive capacity of an ecosystem is fixed whilekey stakeholders all want to extract a product from thatecosystem at a higher level, a compromise midwaybetween them will be unsustainable.

The next major subsystem, spatial and non-spatialdata management organizes the available descriptionsof the ecological and management components ofecosystem management Data must be available tosupport choices among alternative management sce-narios and to forecast consequences of managementactivities on the landscape. There is tension betweenthe increasing number of goals and desired futureconditions that decision makers and stakeholdersvalue and the high cost of obtaining data and under-standing relationships that support these choices.Monitoring disturbance activities, both natural andhuman originated, as well the disturbance-freedynamics of the forest ecosystems under managementis also extremely important if the FJM-DSS is toaccurately portray the decision choices and their


consequences. Barring blind luck, the quality of thedecision cannot be better than the quality of theknowledge behind it Poor data can lead to poordecisions. It is difficult to conceive of prudent eco-system management without an adequate biophysicaldescription of the property in question.

The next four subsystems - knowledge base, simu-lation model, help/hypertext, and data visualizationmanagement - deal with effectively managing knowl-edge in the many diverse forms in which it is stored,

. represented, or coded (see Rauscher et al., 1993 formore detail). Non-language-based knowledge is eitherprivately held in people's minds or publicly repre-sented as photographs, video, or graphic art Lan-guage-based knowledge is found in naturallanguage texts of various kinds, in mathematicalsimulation models, and in expert or knowledge-basedsystems. Data visualization software has been devel-oped that can manipulate photographic, video, andgraphic art representations of current and future eco-system conditions. Data visualization software is begin-ning to be incorporated into EM-DSS on a routinebasis to help decision makers see for themselves thelikely impact of their decisions on the landscape.

In the last 20 years, an impressive amount ofmathematical simulation software has been developedfor all aspects of natural resource management Schus-ter et al. (1993) conducted a comprehensive inventoryof simulation models available to support forest plan-ning and ecosystem management. They identified andbriefly described 250 software tools. Jorgensen et al.(1996) produced another compendium of ecologicalmodels that incorporate an impressive amount ofecosystem theory and data. The simulation modelmanagement subsystem of the EM-DSS is designedto provide a consistent framework into which models

• of many different origins and styles can be placed sothe decision makers can use them to analyze, forecast,and understand elements of the decision process.

Despite our most strenuous efforts to quantifyimportant ecological processes to support a theoryin simulation model form, by far the larger body ofwhat we know can only be expressed qualitatively,comparatively, and inexactly. Most often this qualita-tive knowledge has been organized over long years ofprofessional practice by human experts. Theoreticaland practical advances hi the field of artificial intelli-gence applications in the last 20 years now allow us to

capture some of this qualitative, experience-basedexpertise into computer programs called expert orknowledge-base systems (Schmoldt and Rauscher,1996). It is still not possible to capture the full rangeand flexibility of knowledge and reasoning ability ofhuman experts in knowledge-based software. We havelearned, however, how to capture and use that portionof expertise that the human expert considers routine.The knowledge management subsystem of the EM-DSS is designed to organize all available knowledge-based models in a uniform framework to support thedecision making process.

Finally, a large amount of text material exists thatincreases the decision maker's level of understandingabout the operation of the DSS itself, the meaning ofresults from the various modeling tools, and thescientific basis for the theories used. This text materialis best organized in hypertext software systems. Theoak regeneration hypertext (Rauscher et al., 1997a)and the hypermedia reference system to the FEMATreport (Reynolds et al., 1995) are recent examples ofthe use of hypertext to synthesize and organize scien-tific subject matter. Examples of the use of hypertextto teach and explain software usage can be found inthe help system of any modern commercial computerprogram.

The software subsystems of an EM-DSS describedso far help the decision maker(s) organize the decisionproblem, formulate alternatives, and analyze theirfuture consequences. The decision methods manage-ment subsystem (Fig. 5) provides tools and guidancefor choosing among the alternatives, for performingsensitivity analyses to identify the power of specificvariables to change the ranking of alternatives, and forrecording the decisions made and their rationale.There are many facets or dimensions that influencethe decision making process. The rational/technicaldimension, which concerns itself with the mathema-tical formulation of the methods of choice and theiruses, is the one most often encountered in the decisionscience literature (Klein and Methlie, 1990; Rauscher,1996). But there are others including the political/power dimension (French and Raven, 1959; O'Reilly,1983) and the value/ethical dimension (Brown, 1984;Klein and Methlie, 1990, p. 108;RueandByars, 1992,p. 61). The decision maker(s) might find themselves atany point along the political/power dimension definedby a dictatorship (one person decides) on the one


extreme and by anarchy (no one can decide) on theother. Intermediate positions are democracy (majoritydecides), republicanism (selected representativesdecide), and technocracy/aristocracy (experts or mem-bers of a ruling class decide). Currently threeapproaches seem to be in use at multiple societal,temporal and spatial scales: management by experts(technocracy), management by legal prescription(republicanism), and management by collaboration(democracy) (Bormann et al., 1993). No one approachpredominates. In fact, the sharing of power betweenthese three approaches creates tensions which helpmake ecosystem management a very wicked problem.

In the context of ecosystem management, the value/ethical dimension might be denned on the one extremeby the preservationist ethic (reduce consumption andlet nature take its course) and on the other by theexploitation ethic (maximum yield now and let futuregenerations take care of themselves). Various forms ofthe conservation ethic (use resources, but use themwisely) could be defined between these two extremes.The rational/technological dimension is defined by thenormative/rational methods on one hand and theexpert/intuitive methods on the other. Numerous inter-mediate methods have also been described and used(Janssen, 1992; Rauscher, 1996). The formal relation-ships between these dimensions affecting the decisionprocess have not been worked out Informally, it iseasy to observe decision making situations where thepolitical/power or value/ethical dimensions dominatethe rational/technical dimension. Choosing an appro-priate decision making method is itself a formidabletask (Silver, 1991; Turban, 1993) which influencesboth the design of alternatives and the final choice.

- Many EM-DSS do not offer a decision methods sub-system due to the complexity and sensitivity of thesubject matter. Unfortunately, providing no formalsupport in EM-DSS for choosing among alternativessimply places all the burden on the users and maymake them more vulnerable to challenges of theirprocess and choice mechanisms.

5. A comparison of existing ecosystemmanagement DSS

Mowrer et al. (1997) surveyed 24 of the leadingEM-DSS developed in the government, academic, andprivate sectors hi the United States. Their report

identified five general trends: (1) while at least oneEM-DSS fulfilled each criteria in the questionnaireused, no single system successfully addressed allimportant considerations; (2) ecological and manage-ment interactions across multiple scales were notcomprehensively addressed by any of the systemsevaluated; (3) the ability of the current generationEM-DSS to address social and economic issues lagsfar behind biophysical issues; (4) the ability to simul-taneously consider social, economic, and biophysicalissues is entirely missing from current systems; (5)group concensus-building support was missing fromall but one system - a system which was highlydependent upon trained facilitation personnel(Mowrer et al., 1997). In addition, systems that offeredexplicit support for choosing among alternatives pro-vided decision makers with only one choice metho-dology. The reviewers noted that little or nocoordination had occurred between the 24 develop-ment teams resulting in large, monolithic, stand-alonesystems, each with a substantially different concept ofthe ecosystem management process and how to sup-port it

Different EM-DSS appear to support different partsof the ecosystem management process. Table 1 lists33 EM-DSS, the 24 systems surveyed by Mowrer et al.(1997) plus nine DSS not included in that study.Nineteen of the 33 are labeled full service EM-DSSat their scale of operation because they attempt to becomprehensive EM-DSS, offering or planning to offersupport for a complete ecosystem management pro-cess. These EM-DSS can be further classified by thescale of support which is their primary focus: regionalassessments, forest planning, or project level planning.The remainder, labeled functional service modules,provide specialized support for one or a few phases ofthe entire ecosystem management process. Theseservice modules can be organized according to thetype of functional support they provide: group nego-tiations, vegetation dynamics, disturbance simulation,spatial visualization, and interoperable system archi- ;

lecture.

5.1. Full service EM-DSS

5.1.1. Regional assessmentsThe ecosystem management decision support

(EMDS) program is a software system specifically

186 HM. Rauscher/Forest Ecology and Management 114 (1999) 173-197

Table 1A representative sample of existing ecosystem management decision support software for forest conditions of the United States arranged byoperational scale and function

Full-service EM-DSS

Operational scale

Regional assessments

Forest level planning

Project level planning

Models

EMDSLUCAS*

RELMSPECTRUMWOODSTOCKARCRORESTSARATERRA VISIONEZ-IMPACTDECISION PLUS8

DEFINITE"

NEDINFORMSMAGISKLEMSTEAMSLMS

Functional service modules

Function

Group negotiations

Vegetation dynamics

Disturbance simulations

Spatial visualization

Interoperable systemArchitecture

Economic impact analysisActivity scheduling

Models

Ar/GISmisa

FVSLANDISCRBSUMSIMPLLE -

FIREBGCGYPSESUPEST

UTOOLS/UVIEWsvsa

SMARTFORESr

LOKICORBA

IMPLANSNAP

•References for models not described in Mowrer et aL (1997): EZ-IMPACT (Behan, 1994); DECISION PLUS (Sygenex, 1994); EBIS (Hashim,1990); DEFTNTTE (Janssen and van Hervijnen, 1992); SMARTFOREST (Odand, 1995); CORBA (Otte et al., 1996); SVS (McGaughey,1997); LMS (Oliver and McCarter, 1996); LUCAS (Berry et aL, 1996).

designed to support the development of ecologicalassessments, usually at the regional or watershedscales. It provides a general software environmentfor building knowledge bases that describe logicalrelations among ecosystem states and processes ofinterest in an assessment (Reynolds et al., 1997). Oncethese knowledge bases are constructed by users, thesystem provides tools for analyzing the logical strac-ture and the importance of missing information.EMDS provides a formal logic-based approach toassessment analysis that facilitates the integration ofnumerous diverse topics into a single set of analyses. Italso provides robust methods for handling incompleteinformation. A variety of maps, tables, and graphsprovides useful information about what data are miss-ing, the influence of missing data, and how data aredistributed in the landscape. EMDS also providessupport for exploring alternative future conditions.Finally, EMDS is general in application and can beused at the scale relevant to an assessment problem(Reynolds et al., 1997).

LUCAS is a multidisciplinary simulation frame-work for investigating the impact of land-use manage-ment policies (Berry et al., 1996). LUCAS has beenused to support regional assessments of land-usechange patterns as a function of social choices andregulatory approaches (Wear et al., 1996). LUCAScan be used to compare the effects of alternativeecosystem management strategies that could be imple-mented over any sized eco-region. These alternativescould be evaluated based on any number of socialchoice assumptions ascribed to private landowners(Wear et al., 1996). LUCAS could also be used toaddress the effects of land cover changes on naturalresource supplies and local incomes. The advantage ofEM-DSS operating at the eco-regional scale is thatregional decision making activities and their conse-quences can be forecast with reasonable credibility.

5.1,2. Forest level planningFor 17 years, from 1979 until 1996, a linear pro-

gramming, harvest scheduling model was turned into a

HM. Rauscher/Forest Ecology and Management 114 (1999) 173-197 187

forest-level planning tool, FORPLAN, and all nationalforest supervisors were required to use it as theprimary analysis tool for strategic forest planning.After years of increasingly fierce criticism that thenormative, rational, optimization approach to decisionanalysis implemented by FORPLAN and its succes-sor, SPECTRUM, was not adequate, the Forest Ser-vice finally removed its formal requirement to useFORPLAN/SPECTRUM (Stephens, 1996). The spe-cifics of the arguments critical of FORPLAN/SPEC-TRUM as an analysis tool for forest planning arebeyond the scope of this paper and can be readilyfound in the following publications: Hoekstra et al.(1987), Barber and Rodman (1990), Canham (1990),Howard (1991), Kennedy and Quigley (1993), Morri-son (1993), Shepard (1993), Behan (1994), Liu andDavis (1995), Behan (1997), and Smith (1997).

Forest-level planning, which corresponds to thestrategic planning process of each national forest,may be more successfully performed using soft, qua-litative decision analysis formalisms than the hard,quantitative methods employed in rational, linear ornon-linear optimization schemes. Many other decisionanalysis formalisms exist (see Rauscher, 1996; Smith,1997) along with the tools that make them useful andpractical (Table 1). Many of these techniques mayoffer greater support for dealing with power struggles,imprecise goals, fuzzy equity questions, rapidly chan-ging public preferences, and uneven information qual-ity and quantity (Allen and Gould, 1986). In particular,EZ-IMPACT (Behan, 1994, 1997) and DEFINITE(Janssen and van Hervijnen, 1992) are well-developedand tested forest planning analysis tools which usejudgement-based, "ordinal, and cardinal data to help.users characterize the system at hand and explorehidden interactions and emergent properties.

A forest plan should demonstrate a vision of desiredfuture conditions (Jensen and Everett, 1993). It shouldexamine current existing conditions and highlight thechanges needed to achieve the desired future condi-tions over the plan period (Grossarth and Nygren,1993). Finally, the forest plan should demonstrate thatrecommended alternatives actually lead towards thedesired future condition by tracking progress annuallyfor the life of the plan. The forest plan should be ableto send accomplishable goals and objectives to theproject implementation level and receive progressreports that identify the changes hi the forest condi-

tions that management has achieved. Ideally, all com-petitors in this class of EM-DSS should be objectivelyevaluated for their effectiveness hi supporting thesetasks, their ease of use in practice, and their ability tocommunicate their internal processes clearly and suc-cinctly to both decision makers and stakeholders. Suchan evaluation has not yet been conducted.

5.1.3. Project level implementation"Forest plans are programmatic hi that they estab-

lish goals, objectives, standards, and guidelines thatoften are general. Accordingly, the public and USDAforest service personnel have flexibility hi interpretinghow forest plan decisions apply, or can best beachieved, at a particular location, In addition, forestplans typically do not specify the precise timing,location, or other features of individual managementactions" (Morrison, 1993). EM-DSS at the projectlevel help identify and design site-specific actions thatwill promote the achievement of forest plan goals andobjectives. Several project level implementation EM-DSS have been developed hi the last few years(Table 1). Project level EM-DSS can be separatedinto those that use a goal-driven approach and thosethat use a data-driven approach to the decision supportproblem. NED (Rauscher et al., 1997b; Twery et al.,1997) is an example of a goal-driven EM-DSS wheregoals are selected by the user(s). These goals definethe desired future conditions, which define the futurestate of the forest Management actions should bechosen that move the current state of the forest closerto the desired future conditions. In contrast,INFORMS (Williams et al., 1995) is a data-drivensystem which begins with a list of actions and searchesthe existing conditions to find possible locations toimplement those management actions. Bothapproaches have their strengths and weaknesses.

Goal-driven systems, by definition, tend to ensurethat management actions move the ownership towardthe specified desired future conditions by committingto a particular ecosystem management decision pro-cess. This reduces the utility of the EM-DSS to that setof decision makers who wish to use the particulardecision process implemented. On the other hand, datadriven systems offer no guarantee that the results ofthe sum of the actions have any resemblance to thedesired future conditions as defined by the objectivesof the owner(s). Data-driven systems, however, do


allow competent and knowledgeable decision makersto choose the analyses they perform and how they putthem together to arrive at a decision.

5.2. Functional service modules

Full-service EM-DSS rely on specialized softwareservice modules to add a broad range of capabilities(Table 1). Tools to support group negotiation in thedecision process are both extremely important andgenerally unavailable and underutilized. AR/GIS(Faber et aL, 1997) is the most fully developed soft-ware available for this function and has been brieflydescribed in Section 4. IBIS, another group negotia-tion tool, is an issue-based information system thatimplements argumentation logic (the logic of ques-tions and answers) to help users formally state pro-blems, understand them, clearly communicate them,and explore alternative solutions (Conklin and Bege-man, 1987; Hashim, 1990). Vegetation dynamicssimulation models, both at the stand and at the land-scape scale, provide EM-DSS with the ability toforecast the consequences of proposed managementactions. Disturbance simulators simulate the effects ofcatastrophic events such as fire, insect defoliation,disease outbreaks, and wind damage. Models thatsimulate direct and indirect human disturbances onecosystems are not widely available. Models thatsimulate timber harvesting activities exist, but providelittle, if any, ecological impact analyses such as theeffect of extraction on soil compaction, on damage toremaining trees, or on the growth response of theremaining tree and understory vegetation. Models thatsimulate the Impact of foot traffic, mountain bikes, andhorse-back riding on high-use areas are largely miss-ing. Models that simulate climate change, nutrient

\ cycling processes, acid deposition impacts, and otherindirect responses to human disturbance are rarelypractical for extensive forest analyses. Stand andlandscape-level visualization tools have improveddramatically in the last few years. It is now possible,with relatively little effort, to link to and provide datafor three-dimensional stand level models such as SVS(McGaughey, 1997) as well as landscape level modelssuch as UVIEW (Ager, 1997) and SMARTFOREST(Orland, 1995).

No single model is likely to provide adequatesupport for ecosystem management (Grossarth and

Nygren, 1993; Mowrer et al., 1997). Familiarity withthe entire range of available decision analysis meth-odology and modeling tools that implement thesemethods is required. Retraining decision analystsand decision makers to use different tools for differentpurposes and to teach how various techniques andtools fit together to address management objectiveswill be a critical component to successful applicationof the ecosystem management paradigm (Grossarthand Nygren, 1993).

Even a cursory review revealed that none of theavailable EM-DSS has been found capable of addres-sing the full range of support required for ecosystemmanagement (Mowrer et al., 1997). The EM-DSSintroduced in this section hold great promise, but thispromise has not been fully achieved. A major reasonfor this situation is that system development has beenprimarily driven by technology, not by requirements.The requirements to guide EM-DSS development areunknown or poorly defined because the ecosystemmanagement decision processes have been inade-quately identified and described. The frequentlyobserved tendency to substitute technology for aninadequate or non-existent ecosystem managementdecision making process should be avoided becauseit is rarely satisfactory. Although formal evaluationprocedures are available (see Adelman,. 1992), few ofthe current EM-DSS have undergone an in-depth,critical evaluation of their suitability for ecosystemmanagement decision support. In the final analysis,EM-DSS software should be evaluated based on thequestion, "Does the EM-DSS improve the decision-maker's ability to make good decisions?"

6. Toward a large system architecture forEM-DSS

A key to effective decision support for ecosystemmanagement is interoperability of software systems(Potter et al., 1992; Bevins and Andrews, 1994; Fedra,1995; Mowbray and Zahavi, 1995; Otte et al., 1996).Open-architecture, interoperable systems provide asoftware interface standard which promotes commu-nication between modules and provides for the inte-gration of newly developed modules and refinementsas they occur over time. Interoperable, large systemarchitectures function like the conductor of an orches-


tra. The conductor is told by the user what composi-tion to play. The conductor looks up the musical scorefor that composition (developed by the DSS version ofthe composer) and then assembles the necessaryinstruments and organizes the flow of music. Notall instruments play in all compositions and all instru-ments have a special role to play at a particular timeand place in the composition. The conductor is thenultimately responsible for making sure that the musicis pleasant, the score is followed, and the user is happy.The conductor (software architecture) only conducts.It neither composes nor does it play an instrument Ourcurrent compositions (EM-DSS) are hardwired to playonly one composition like a simple music box. Theycannot play anything else unless extensively rework-ed; then they often only play the new composition.

The introduction and popular acceptance of theMicrosoft Windows3 and Apple Macintosh3 operatingsystems serve as premier examples of the advantagesof software interoperability. Monolithic software sys-tems, which are essentially islands of automation, arenow recognized as poor solutions to complex pro-blems (Otte et al., 1996). "As software technologycontinues to evolve, there is a growing trend towardsthe construction of systems from pre-existing compo-nents" (Mowbray and Zahavi, 1995). There is muchlegacy software available with potential for use insupport of ecosystem management (Schuster et al.,1993; Jorgensen et al., 1996). In addition, readilyavailable, high quality commercial software existsto support many needs of the ecosystem managementprocess. Finally, numerous, independently operatingdevelopment groups are producing software solutionsto ecosystem management problems that are poten-tially useful for a wide-ranging decision making audi-ence. The integration of these independent softwaresolutions in support of ecosystem management isessential to make rapid progress in the growth andevolution of effective EM-DSS.

Efforts at custom or localized integration are readilyapparent in the present generation of EM-DSS. TheUSDA forest service's forest vegetation simulator(FVS) is an integrated, multi-module software systemfor modeling forest vegetation dynamics (Teck et al.,

3The use of trade or firm names in this publication is for leaderinformation and does not imply endorsement by the U.S.Department of Agriculture of any product or service.

1996). It uses an event monitor to manage the user-defined stream of event activities that FVS can simu-late. Postprocessors, which are independently devel-oped computer modules, execute with FVS toreformulate results for specific purposes, includingthe design of input for the execution of subsequentmodules. A Windows-style interface is being devel-

'oped for FVS to facilitate designing complicatedsimulations and orchestrating the operations of linkedbut foreign computer modules. A different approach tointeroperability was used in NED. NED uses a cus-tom-developed Logic Server, based on the client-server paradigm, to create a communication and con-trol channel between the user interface and datamanager modules, written in C++, and the knowl-edge-base management module written in PROLOG(Rauscher et al., 1997b). The user interface module ofNED is the only module permitted to communicatewith the user, which enforces a uniform look and feel.All variable values are maintained by the data managermodule. Rather than send every known value of allvariables to the knowledge-base management moduleeach time it is called to perform a particular task, thedata manager responds to specific data requests fromthe knowledge-base module as they are needed. Suchintelligent communication capability makes the entiresystem more flexible and efficient The ecosystemmanagement decision support (EMDS) project pro-vides a final example of a different customized inte-gration approach. EMDS links together twocommercial software modules, the NetWeaver3

knowledge-base system (Knowledge Garden, WestPalm Beach, FL) and the Arc View geographic infor-mation system (Environmental System Research Insti-tute, Redlands, CA) to provide a flexible and powerfultool for environmental assessment analysis (Reynoldset al., 1997).

Custom integration approaches to software moduleinteroperability yield unique, point-to-point integra-tion solutions (Otte et al., 1996). Custom solutions aredifficult to extend into generic, general purpose inter-operable software architectures. Therefore, customintegration of software modules should be viewedas a temporary, stop-gap solution until more robustand generally applicable large system architecturescan be designed, tested, and adopted (Otte et al.,1996). Interoperable software architectures are notlarge monolithic systems; they only provide the com-


munication and control framework standards thatallow any number of independently developed soft-ware modules to function together.

The design, implementation, and maintenance ofinteroperable software architectures for EM-DSS arechallenging activities. System integrators face com-puter science problems with 'different hardware plat-forms, software languages, compiler versions, dataaccess mechanisms, module interfaces, and network-ing protocols' (Mowbray and Zahavi, 1995). In addi-tion, the ecosystem management arena contributeschallenges such as different data sources, ecosystemmanagement process visions, decision making meth-ods, and solution strategies. Future generations of EM-DSS must become more interoperable to provide thebest possible support for ecosystem management pro-cesses.

7. Implementation issues

Effective ecosystem management processes andDSS to implement those processes are urgently neededin the federal forest management sector. A number ofissues must be settled before ecosystem managementcan evolve from a philosophical concept to an effi-ciently functioning method. These issues can begrouped into four themes: scientific/technological,organizational/institutional, education/training, andsocial/political.

7.1. Scientific/technological issues

The scientific/technological issues are complex, butnot wicked (Allen and Gould, 1986). There is no doubtthat given a set of ecosystem management processes tosupport and adequate time and resources, effectiveEM-DSS can be developed. On the other hand, there isconsiderable doubt that explicit and widely acceptedprocesses for implementing ecosystem managementcan be crafted given the current institutional, educa-tional, social, and political climate.

Developing EM-DSS requires understanding of theecological as well as management subsystems thatmake up ecosystem management. One of the foremostrequirements toward that end is to establish clarity ofconcepts and definitions. In particular, it is necessaryto reinforce the fact that ecology is a science and as

such provides no value judgments. That principle hasbeen one of the cornerstones of the scientific enter-prise for over 400 years. Due to the highly chargedsocio-political climate in which today's environmentalconflicts are played out, some scientists have becomeunclear about fact - value distinctions and have chosento become advocates (Kimmins, 1993; Chase, 1995).The United States needs advocates who championimportant causes and it needs scientists to objectivelyevaluate courses of action and implement mandates(FEMAT, 1993). But to maintain scientific credibility,scientists must be expected to disclose publicly whenthey are playing which role (Dombeck, 1997). Thescientist who takes on the role of advocate "under themantle of objective science is not serving that processwhereby decisions are made that have profound con-sequences for the natural resources and on the peoplewhose livelihoods and lifestyles may be hi jeopardy"(FEMAT, 1993, p. H-80). A clear demarcationbetween these two roles must be maintained to ensurethat controversial decisions are based upon the bestknowledge available, that poorly defined concepts areidentified and unproved, and that mistaken facts areuncovered and corrected.

7.2. Organizational/institutional issues

Making ecosystem management work hi the USDAforest service and other federal agencies will requirestrong leadership, the willingness to accept organiza-tional change as normal, the support of the Presidentand Congress to change funding patterns, and a focuson results. Only strong leadership throughout an insti-tution can produce: (1) clear objectives and policydirection; (2) provide clear priorities and assignresponsibilities; (3) marshal adequate resources overa long enough time to meet commitments; and (4)assure that the desired results are achieved. It isaxiomatic in management science that wicked pro-blems, such as ecosystem management, should be theprimary concern of top management, not middle- orlower-level management (Donnelly et al., 1995). Thebasis .for this management principle is the fact thatmiddle- and lower- level management staff typicallydo not have the time, money, organizational power, orexpertise to deal effectively with wicked problems ontheir own. It is considered the responsibility of top-level management to ensure that unstructured pro-

KM. Rauscher/Forest Ecology and Management 114 (1999) 173-197 191

blems, such as ecosystem management, are increas-ingly more structured as they move down the manage-ment levels to the field. All administrative levelswithin the USDA forest service and its partner federalagencies will need to accept their own share ofresponsibilities to make ecosystem management alocal, regional, and national success (FEMAT, 1993;Kaufmann et al., 1994).

7.3. Education and training issues

Attempting to solve complex ecosystem manage-ment problems with inappropriate resources (lack ofadequate knowledge, untrained people, or inadequatetime) will is likely to produce poor decisions andquestionable ecosystem management results(Rauscher et al., 1993; Stock and Rauscher, 1996).Sound decisions are primarily influenced by the natureand complexity of the problem, accessibility to highquality and relevant knowledge, competent leadership,and well-trained decision makers who know how touse the knowledge wisely (Rauscher, 1996). Decisionmakers need sufficient training and education hi eco-logical decision analysis, decision support principles,and specific analytical technologies and tools. Main-tenance of a consistently and uniformly high level ofknowledge and skill at all levels in a large organizationduring turbulent times is difficult and potentiallyexpensive, yet critical. Continuing education pro-grams, implemented by many of the USDA forestservice's administrative regions, show great promise.For example, the Northern Region's program consistsof four modules: (1) basic ecological, evolutionaryconcepts; (2) ecosystem dynamics; (3) integratedecosystem inventory and analysis; and (4) ecosystemmanagement implementation (Bollenbacher et al.,1994). This course does not yet seem to adequatelyaddress decision analysis principles and practice, orevaluation and use of the available decision supporttools.

A democratic, collaborative group negotiation pro-cess for ecosystem management can only work wellwhere public control of policy preferences is informedcontrol (Kimmins, 1991). Do federal land manage-ment agencies, such as the USDA Forest Service, havea role in creating and maintaining an informed public?"The politically effective, emotive, but often factuallyincorrect rhetoric of both environmentalists and indus-

trialists must be challenged and put behind us since itdoes not provide an adequate basis from which todevelop effective policy for sustainable ecosystems"(Kimmins, 1991). One side of this issue urges that thepublic be proactively informed to increase the use offactually correct information to sway public prefer-ences. The other side of this issue argues that federalland management agencies must not be allowed toenter the public national environmental managementdebate. In the past, federal agencies have had clearmandates to educate, train, and otherwise inform theirperceived clients such as private forest landowners,local civic organizations, local environmental interestgroups, local hunting clubs, etc. However, by custom,federal ecosystem management agencies have notusually used influential local or national media hi aconcerted effort to inform the general public aboutenvironmental issues. When they have done so, as hithe case of the 'Smokey the Bear' campaign, they haveproven to be very effective. Since the ecosystemmanagement paradigm gives public stakeholders animportant role to play in the entire process, it isimportant to revisit this issue. Who is the public -vocal, minority special interest groups or the greatermajority of citizens? How can the public best berepresented in ecosystem management? How can theirinterests be identified? How can their questions best beaddressed? How can their factual misconceptions becorrected?

7.4. Social/political issues

A number of important social/political issues needlocal, regional, and national debate. These issues maynot be amenable to resolution, but their essentialelements should be understood by everyone becausethey help shape the deep-seated, core values on whichindividuals base their preferences.

Is sustainable ecosystem management even possi-ble? Ecosystem sustainability has been defined as theoverlap between what is biologically possible andwhat is socially desirable (Bormann et al., 1993;Maser, 1994). World population is expected to doublefrom its present 6 billion people to between 11 and 16billion people in the next 50 years (Marcin, 1993).Two hundred years ago, humans are estimated to haveconsumed about 1% of all energy captured by greenplants (Zeide, 1994). Today this figure is about 40%,


meaning that the proportion left over for all otherspecies has shrunk from 99 to 60%. This basic bio-physical fact makes species extinction and retrench-ment inevitable and the belief that we can save everyspecies a myth (Zeide, 1994). Despite numerousscientific conferences on global climate change, bio-diversity, and air pollution which point to humanoverpopulation as the ultimate cause of these environ-mental ills, resistance to accepting the consequencesof human population growth is common (Zeide,1994). It may be that given the current populationof the world (to say nothing of future increases) andpeoples' aspirations for 'the good life/ there exists nooverlap between what is biologically possible andwhat is socially desirable.

It seems unlikely that increasing environmentalquality and achieving sustainable ecosystem manage-ment can be accomplished without sacrifices (Zeide,1994). Our society has somehow come to expect thatwe can have all we want without paying a price."Lacking the resources or the will to address oursignificant (environmental) problems (in a substantiveway), we conduct a politic that is more and morecharacterized by images, rituals, and myths. Leaderssimply joust symbolically with problems. When theissues are politically too costly to be settled by thelegislative branch, symbolic legislation is passed thatseems to address the problem but, in fact, simplypasses the political hot potato to the bureaucracyand the courts. The bureaucracy is similarly stymiedas it casts about for public relations solutions to whatmay be politically unwinnable situations" (Shepard,1993; emphasis added). United States national politicshave increasingly turned to a reliance on fantasy(perception is reality) instead of visionary leadership(reality is reality) with absurd promises, consumptionof immediate benefits, avoidance of hard issues andchoices, and postponement of costs to future genera-tions (Shepard, 1993). Environmental costs are notadequately priced and tend to be passed on to publicswith the least power, thus creating conflicts betweenindividual, special interest group, and collective inter-ests (Janssen, 1992). Such a socio-political climateincapacitates the federal forest management decisionmaking process and until a resolution is found, eco-system management will probably remain merely asymbolic solution to our environmental managementproblems.

Powerful environmental interest groups emphasizethe benefits of doing nothing, assuming that naturewill know best But "nature in the twenty-first centurywill be a nature that we make; the question is thedegree to which this molding will be intentional orunintentional, desirable or undesirable" (Botkin,1990). Much of the answer to this question will dependupon whether we choose to manage human populationgrowth and economic development The species thatmake up ecosystems are continually changing andadapting to new stresses. "Nature, never having beenconstant, does not provide a simple answer as to whatis right, proper, and best for our environment There isno single condition that is best for all life." (Botkin,1995). In fact, the extinction of existing species andthe evolution of new species is a normal, naturalprocess. Nature does not care - either about threatenedand endangered species or about us humans. Peoplecare and it is up to people to define goodness orbadness. Failure to make a decision is a decisionand it is a different decision than to explicitly donothing, knowing the consequences. Consequenceswill happen without opportunities to evaluate themor mitigate them if they are undesirable. The statusquo will prevail and the range of our choices in thefuture is likely to shrink (FEMAT, 1993). In reality,there is no acceptable alternative to pro-active eco-system management.

Finally, concern over sustainable ecosystem man-agement is largely a luxury displayed by wealthysocieties where people no longer need be concernedabout food, shelter, or personal security (Kimmins,1991). The increasing prosperity and greater urbani-zation of the American public changed their principalconcerns from the supplying of economic wants toputting a greater emphasis on the non-consumptiveuses of federal forest ecosystems (Caldwell et al.,1994). Urbanization obscures the human reliance onconsumptive uses of natural resources (Fautin, 1995).Wood-framed single-family homes, toilet paper, wood-en furniture and cardboard boxes were all made fromtrees that were cut hi a forest somewhere. Many peopleseem to have lost the understanding that the two arefundamentally connected (Dekker-Robertson, 1997).Without a decrease in human consumption and/ornumbers, environmental degradation is inevitable(Zeide, 1994; Goodland, 1995) and resource exploita-tion will, in all likelihood, continue (Ludwig et al.,


1993). Deciding to reduce resource production byprotecting and preserving ecosystems without simul-taneously reducing consumption rates for thatresource means that other areas of the world will betaxed to make up the difference (Moir and Mowrer,1994). We have effectively exported our demand forthat resource to ecosystems somewhere on earth thatmay very well be less resilient than our own. The onlyother way to increase production without decreasingdemand is through scientific and technologicaladvances (Dekker-Robertson, 1997). Advances inresource production can provide more products andadvances in utilization can satisfy more demand perunit product One positive step we could take toaddress this issue is to organize an analysis of worldforest ecosystems that ranks their ability to produceconsumable resources sustainably. World demandshould be supplied primarily from those forest eco-systems best able to sustain the disturbance (Dekker-Robertson, 1997). We must all eventually realize andaccept that only by developing and maintaining ade-quate national wealth can we afford to preserve theenvironment (Marcin, 1993).

Many of the social/political issues address stronglyheld core values of the American public and thestakeholders in ecosystem management. The morenumerous and antagonistic the core values of stake-holders, the less likely satisfactory collaborative com-promises can be reached and the less likely ecosystemmanagement will be helpful in solving our environ-mental problems. In this situation, diversity is clearly adetriment, not a strength. In a democratic system, ameaningful national debate occurring primarily in the

. public media may be the most effective way to shape anational concensus about difficult environmentalvalues and preferences for federal land management

8. Summary

Ecosystem management has been adopted as thephilosophical paradigm guiding federal forest man-agement in the United States. The strategic goal ofecosystem management is, arguably, to find a sensiblemiddle ground between ensuring the necessary long-term protection of the environment while allowing anincreasing population to use its natural resources formaintaining and improving human life. Ecosystem

management has all the characteristics of 'wicked'problems. Given that ecosystem management, likehuman survival and welfare, is a wicked problem,how can we proceed to tame it? We need to use thesame tools that people have always used for handlingwicked problems - knowledge, organization, judi-cious simplification, and inspired leadership.

Adequately described and widely accepted ecosys-tem management processes do not yet exist, but aconcerted effort to study the many formal and infor-mal ecosystem management processes that do existwould result in some powerful candidates to supportthe practical implementation of ecosystem manage-ment.

The generic theory of DDS development and appli-cation is well developed. Numerous specific ecosys-tem management DDSs have been developed and areevolving in their capabilities. Given a set of ecosystemmanagement processes to support along with adequatetime and resources, effective EM-DSS can be devel-oped. However, major social and political issues pre-sent significant impediments to the efforts to makeecosystem management operational. A socio-politicalenvironment in which everyone wants to benefit andno one wants to pay incapacitates the federal forestmanagement decision making process. The very lawsthat were adopted to solve the problem, RPA/NFMA,have led to procedural paralysis at exponentially risingcosts (Behan, 1990). Developing a workable ecosys-tem management process and the decision makingtools to support it, is probably one of the most complexand urgent challenges facing us today.

It is becoming increasingly obvious that the federalforest management situation in the United States isunsatisfactory and that the ecosystem managementparadigm currently offers the best potential forimprovement The task is to end our paralysis andfind ways to operationalize the ecosystem manage-ment decision process. One concrete method to oper-ationalize ecosystem management is to design andbuild effective decision support tools. The theory andpractice of multiple-use forest management in theUnited States specifically for the sustainable produc-tion of timber, wildlife, water, and recreation productshas been researched and tested for over 100 years. Itwill very nicely take at least as much time and effort toresearch and test the theory and practice of ecosystemmanagement - a significantly more difficult problem.


Acknowledgements

I thank Wayne Swank for encouraging me to writethis paper and David Loftis for his willingness toengage in endless but informative discussions aboutecosystem management I also thank Paul Arndt,Klaus Barber, Erik Berg, Bob Giles, Dave Iverson,Gene Lessard, Shane Mahoney, Todd Mowrer, KeithReynolds, Terry Sharik, Boris Zeide, and two anon-ymous reviewers for their provocative review com-ments that have immeasurably improved the value ofthis paper.

References

Adelman, L., 1992. Evaluating Decision Support and ExpertSystems. Wiley, New York, 232 pp.

Ager, A., 1997. UTOOLS/UVIEW. In: Mowrer, H.T., Barber, K.,Campbell, I., Crookston, N., Dahms, C, Day, L, Laacke, J.,Merzenich, I., Mighton, S., Rauscher, M, Reynolds, EL,Thompson, J., TrencM, P., Twery, M., 1997. Decision SupportSystems for Ecosystem Management: An Evaluation ofExisting Systems. USDA For. Serv. Gen. Tech. Rep. RM-296,pp. 144-148.

Allen, GJ*L, Gould Ir, EM., 1986. Complexity, wickedness, andpublic forests. J. Forestry 84(4), 20-23.

Barber, KJL, Rodman, S.A., 1990. FORPLAN: The marvelous toy.J. Forestry 88(5), 26-30.

Baskerville, G., 1985. Adaptive management - Wood availabilityand habitat availability. The Forestry Chronicle 61(2), 171-175.

Behan, R.W., 1990. The RPA/NFMA: Solution to a nonexistentproblem. J. Forestry 91,20-25.

Behan, R.W., 1994. Multiresource management and planning withEZ-IMPACT. J. Forestry 92(2), 32-36.

Behan, R.W., 1997. Scarcity, simplicity, separatism, science - andsystems. In; Kohm, ILA., Franklin, J.F. (Eds.), Creating aForestry for the 21st Century. Island Press, Washington, DC,pp. 411-417. ,

Bennett, AJ., 1996. Sustainable land use: Interdependence betweenforestry and agriculture. Caring for the Forest Research in aChanging World, Congress Report, voL I, IUFRO XX WorldCongress, 6-12 August 1995 at Tampere, Finland, pp. 144-152.

Berry, M.W., Flamm, R.O., Hazen, B.C., Maclntyre, R-M, 1996.The land-use change and analysis (LUCAS) for evaluatinglandscape management decisions. IKKK Computational Scienceand Engineering 3, 24-35.

Bevins, C J>., Andrews, PL, 1994. The LoM software architecturefor fire and ecosystem modeling: A tinker toy approach. Proc.12th Conf. on Fire and Forest Meteorology, 26-28 October,1993 at Jekyll Island, GA, pp. 252-259.

Bollenbacher, EL., Samson, F., O'Hara, K., 1994. Continuingeducation in ecosystem management (CEEM), an interdisci-plinary approach. In: Foley, L.H. (compiler), Silviculture: From

the Cradle of Forestry to Ecosystem Management. Proc.National Silviculture Workshop, 1-4 November 1993 atHendersonvflle, NC, USDA Forest Service Gen. Tech. Rep.

. SE-88, pp. 224-228.Bonmcksen, TJM, 1991. Managing biosocial systems: A frame-

work to organize society-environment relationships. J. Forestry89(10), 10-15.

Bormann, B.T., Brooks, M.H., Ford, EJD., Kiester, AIL, Oliver,CD., Wtigand, J.F., 1993. A framework for sustainableecosystem management. USDA Forest Service Gen. Tech.Rep. PNW-331, 61 pp.

Botkffl, D.B., 1990. Discordant Harmonies. Oxford UniversityPress, New York, 241 pp.

Botkin, D.B., 1995. Our Natural History. GP. Putnam and Sons,New York, 325 pp.

Brown, T.C, 1984. The concept of value in resource allocation.Land Economics 60, 231-246.

Caldwell, L.K., Wilkinson, C.F., Shannon, M.A., 1994. Makingecosystem policy: Three decades of change. J. Forestry 92(4),7-10.

Caldwell, LJC, 1996. Science assumptions and misplaced certaintyin natural resources and environmental problem solving. In:Lemons, John (Ed.), Scientific Uncertainty and EnvironmentalProblem Solving, Blackwell Science, Cambridge, MA, pp.394-421.

Canham, H.O., 1990. Decision matrices and weighting summationvaluation in forest planning. Northern J. Applied Forestry 7(2),77-79.

Carpenter, S.R., 1990. Large-scale perturbations: Opportunities forinnovation. Ecology 71, 2038-2043.

Chase, Alston, 1995. In a Dark Wood: The Fight Over Forestsand the Rising Tyranny of Ecology. Houghtqn MifBin, Boston,534pp.

Conklin, J., Begeman, M.L., 1987. GffilS: A hypertext tool forteam design deliberation. Proceedings of Hypertext 87, 13-15November 1987 at Chapel Hffl, NC, pp. 104-115.

Craig, LJL,1996. Letter from the United States Senate, Committeeon Energy and Natural Resources, Washington, DC to TheHonerable Dan Glickman, Secretary of Agriculture. 20 June1996, 5 pp.

Daniels, S.E., Walker, G.B., Boeder, J.R., Means, I.E., 1993.Managing ecosystems and social conflict In: Jensen, ME.,Bourgeron, P.S. (Eds.), Eastside Forest Ecosystem HealthAssessment voL n Ecosystem Management Principles andApplications. USDA Forest Service, National Forest System,pp. 347-359.

Dekker-Robertson, Donna, 1997. Is the Northern Spotted Owlworth more than the Orangutan? Proc. National SilvicultureWorkshop, Communicating the role of silviculture in managingthe national forests, 19-22 May 1997, Warren, PA, North-eastern Forest Experiment Station Gen. Tech. Rep. NE-238, pp.19-28.

Dombeck, Mike, 1997. Integrating Science and Decision Making:Guidelines for Collaboration Among Managers and Research-ers in the Forest Service. USDA Forest Service, FS-608,11 pp.

Donnelly Jr., JJL, Gibson, JJL, Ivancevich, J.M., 1995. Funda-mentals of Management, 9th ed. Irwing, Chicago, pp. 127-128.


Everett, R., Oliver, C, Saveland, J., Hessburg, P., Diaz, N., twin,L., 1993. Adaptive ecosystem management In: Jensen, MJE.,Bourgeron, P.S. (Eds.), Eastside Forest Ecosystem HealthAssessment vol. n Ecosystem Management Principles andApplications. USDA Forest Service, National Forest System,pp. 361-375.

Faber, E.G., Wallace, W.W., Croteau, K., Thomas, VI., Small,L JL, 1997. Active response CIS: An architecture for interactiveresource modeling. Proc. CIS World 97, Integrating spatialinformation technologies for tomorrow, 17-20 February 1997,at Vancouver, BC, pp. 296-301.

Fautin, Daphne G., 1995. Sustainability - more than a buzz-word?Annual Review of Ecology and Systematics 26, pp. 5-9.

Fedra, K., 1995. Decision support for natural resources manage-ment Models GIS, and expert systems. AI Applications 9(3),3-19.

FEMAT, 1993. Forest ecosystem management An ecological,economic, and social assessment Report of the ForestEcosystem Management Assessment Team, US GovernmentPrinting Office, chap. I-K plus Appendices.

Floyd, D.W., Germain, RJL, ter Horst, K., 1996. A model forassessing negotiations and mediation in forest resourceconflicts. J. Forestry 94(5), 29-33.

French, J.R.P., Raven, B., 1959. The bases of social power. In:Cartwright, D. (Ed.), Studies hi Social Power. Institute forSocial Research, University of Michigan, Ann Arbor, pp. 150-167.

GAO, 1994. Ecosystem management Additional actions needed toadequately test a promising approach. A Report to Congres-sional Requesters, GAO/RCED-94-111, 87 pp.

Garland, JJ., 1997. The players in public policy. J. Forestry 95(1),13-15.

Goodland, Robert, 1995. The concept of environmental snstain-ability. Annual Rev. Ecol. Systematics 26, 1-24.

Greber, BJ., Johnson, KN., 1991. What's all this debate aboutovercutting? J. Forestry 89(11), 25-30.

Grossarth, O., Nygren, T., 1993. Implementing ecosystem manage-ment through the forest planning process. In: Jensen, ME.,Bourgeron, P.S. (Eds.), Eastside Forest Ecosystem HealthAssessment voL n Ecosystem Management Principles andApplications. USDA Forest Service, National Forest System,pp. 291-299.

Grumbine, RJi., 1994. What is ecosystem management? Con-servation Biology 8(1), 27-38.

Hannigan, J., 1995. Environmental Sociology: A Social Construc-tionist Perspective. Routledge, London.

Hashim, S JL, 1990. Exploring hypertext programming. WindcrestBooks, Blue Ridge Summit, PA, 365 pp.

Hoekstra, T.W., Dyer, A.A., LeMaster, D.C. (Eds.), 1987.FORPLAN: An Evaluation of a Forest Planning TooL USDAForest Service, Gen. Tech. Rep. RM-140, 235 pp.

Holsapple, Clyde W., Whinston, Andrew B., 1996. Decisionsupport systems: A knowledge-based approach. West Publish-ing, Minneapolis/St Paul, 713 pp and Appendices.

Howard, AJ7., 1991. A critical look at multiple criteria decisionmaking techniques with reference to forestry applications.Canadian J. Forest Res, 21, 1649-1659.

Howitt, W.S., 1978. The value of the simple approach. Interfaces 8,64-70.

Janssen, R., 1992. Multiobjective Decision Support for Environ-mental Management Kluwer Academic Publishers, Dordrecht,The Netherlands, 232 pp.

Janssen, R., van Hervijnen, M., 1992. DEFINITE: Decisions on afinite set of alternatives. Institute for Environmental Studies,Free University Amsterdam, The Netherlands, Kluwer Aca-demic Publishers, software on 2 disks.

Jensen, M.E., Everett, R., 1993. An overview of ecosystemmanagement principles. In: Jensen, ME., Bourgeron, P.S.(Eds.), Eastside Forest Ecosystem Health Assessment: vol. nEcosystem Management Principles and Applications. USDAForest Service, pp. 9-18.

Jessup, L.M., Valacich, J.S. (Eds.), 1993. Group Support Systems:New Perspectives. MacMuTan, New York, 365 pp.

Jorgensen, S.E., Halling-Sorensen, B., Nielsen, S.N., 1996. Hand-book of Environmental and Ecological Modeling. LewisPublishers, Boca Raton, FL, 453 pp.

Kaufrnann, MIL, Graham, R.T., Boyce Jr., DA., Moir, WJL,Perry, L., Reynolds, R.T., Bassett, Ri., Hehlhop, P., Edminster,CJJ., Block, WM., Com, P.S., 1994. An ecological basis forecosystem management USDA Forest Service Rocky Moun-tain Forest and Range Experiment Station, Gen. Tech. Rep.RM-246, p. 22.

Kennedy, JJ., Quigley, T.M., 1993. Evolution of Forest Serviceorganizational culture and adaptation issues in embracingecosystem management In: Jensen, ME., Bourgeron, P.S.(Eds.), Eastside Forest Ecosystem Health Assessment voL nEcosystem Management Principles and Applications. USDAForest Service, pp. 19-29.

Kessler, WB., Salwasser, H., Cartwright Jr., C.W., Chaplan, J.A.,1992. New perspectives for sustainable natural resourcesmanagement EcoL Applications 2(3), 221-225.

Kimmins, J.P., 1991. The future of the forested landscapes ofCanada. The Forestry Chronicle 67,14-18.

Kimmins, J.P., 1993. Ecology, environmentalism, and greenreligion. The Forestry Chronicle 69, 285-289.

Kimmihg, jj>, 1995. Sustainable development hi Canadian forestryin the face of changing paradigms. Forestry Chronicle 71(1),33-40.

Klein, M., Methlie, Li., 1990. Expert Systems: A DecisionSupport Approach. Addison-Wesley, Reading, MA, 539 pp.

Kohm, KLA., Franklin, JJ7., 1997. Introduction. In: Kohm, KA.,Franklin, JJ7. (Eds.), Creating a Forestry for the 21st Century.Island Press, Washington, DC, pp. 1-5.

Larsen, DJL, Shifley, SJL, Thompson HI EJR, FJL, Brookshire,BJ.., Dey, D.C., Kurzejeski, E.W., England, K., 1997. 10Guidelines for ecosystem researchers: Lessons from Missouri.J. Forestry 95(4), 4-9.

Lee, K.N., 1993. Compass and Gyroscope: Integrating Science andPolitics for the Environment Island Press, Washington, DC,243pp.

Lessard, G., 1997. A national framework for integrated eco-logical assessments. USDA Forest Service, EcosystemManagement, P.O. Box 96090, Washington, DC 20090-6090,p. 56, hi press.


Lindblom, C.E., 1990. Inquiry and change: The troubled attempt tounderstand and shape society. Yale University Press, NewHaven, CT, 334 pp.

Liu, G., Davis, L.S., 1995. Interactive resolution of multi-objectiveforest planning problems with shadow price and parametricanalysis. Forest Science 41(3), 452-469.

Ludwig, D., ffilbom, R., Walters, C., 1993. Uncertainty, resourceexploitation, and conservation: Lessons from history. Science260, pp. 17 and 36.

Martin, T.C., 1993. Demographic change: Implications for forestmanagement J. Forestry 91(11), 39-45.

Maser, Chris, 1994. Sustainable Forestry: Philosophy, Science, andEconomics. St Lucie Press, Delray Beach, FL, 373 pp.

McGaughey, RJ., 1997. Visualizing forest stand dynamics usingthe stand visualization system. Proc. 1997 ACSM/ASPRSAnnual Convention and Exposition, 7-10 April, 1997. Amer-ican Society for Photogrammetry and Remote Sensing, Seattle,WA, voL 4, pp. 248-257.

Mintzberg, H., Raisinghani, D., Theoret, A., 1976. The structure ofunstructured decision processes. Administrative Science Quar-terly 21, 246-275.

Moir, WH., Mowrer, H.T., 1994. Unsustainability: The shadow ofour future. In: Covington, W.W., DeBano, L.F. (Eds.),Sustainable Ecological Systems: Implementing an EcologicalApproach to Management 12-14 July 1993 at Flagstaff,Arizona, USDA Forest Service Gen. Tech. Rep. RM-247, pp.143-149.

Montgomery, CA., 1993. Socioeconomic risk assessment and itsrelation to ecosystem management In: Jensen, MJL, Bourger-on, P.S. (Eds.), Eastside Forest Ecosystem Health AssessmentvoL n Ecosystem Management Principles and Applications.USDA Forest Service, pp. 325-335.

Morrison, J., 1993. Integrating ecosystem management and theforest planning process. In; Jensen, ME., Bourgeron, P.S.(Eds.), Eastside Forest Ecosystem Health Assessment: voL nEcosystem Management: Principles and Applications. USDAForest Service, pp. 281-290.

Mowbray, TJ., Zahavi, R., 1995. The Essential CORBA: SystemsIntegration using Distributed Objects. Wiley, New York, 367 pp.

Mowrer, H.T., Barber, K., Campbell, J., Crookston, N., Dahms, C.,Day, J., Laacke, J., Merzenich, J., Mighton, S., Rauscher, M.,Reynolds, K., Thompson, J., Trenchi, P., Twery, M., 1997.Decision support'systems for ecosystem management Anevaluation of existing systems. USDA Forest Service, Inter-regional Ecosystem Management Coordination Group, Deci-sion Support System Task Team, Rocky Mountain Forest andRange Experiment Station, RM-GTR-296, 154 pp.

Oliver, CD., Knapp, WJL, Everett, R^ 1993. A system forimplementing ecosystem management In: Jensen, M.E.,Bourgeron, P.S. (Eds.), Eastside Forest Ecosystem HealthAssessment vol. n Ecosystem Management Principles andApplications. USDA Forest Service, pp. 377-388.

Oliver, CD., McCarter, JJJ., 1996. Developments in decisionsupport for landscape management In: Heit, M, Parker, H.D.,Shortreid, A. (Eds.), GIS Applications in Natural ResourceManagement 2. Proc. 9th American Symposium on GeographicInformation Systems, Vancouver, BC, pp. 501-509.

O'Reilly, CA, 1983. The use of information in organizationaldecision making: A model and some propositions. In:Cnmmings, Li., Shaw, BJvL (Eds.), Research in Organiza-tional Behavior, voL 5, pp. 103-139.

Orland, B., 1995. SMARTFOREST: A 3-D interactive forestvisualization and analysis system. In: Power, J.M., Strome, M.,Daniel, T.C. (Eds.), Proc. Decision Support-2001 Conference.12-16 September 1994 Toronto, Ontario, pp. 181-190.

Otte, R, Patrick, P., Roy, M., 1996. Understanding CORBA: Thecommon object request broker architecture. Prentice-Hall, NJ,267 pp.

Overbay, J.C., 1992. Ecosystem management Proc. nationalworkshop: Taking an Ecological Approach to Management,27-30 April 1992 at Salt Lake City, UT. USDA Forest Ser-vice, Watershed and Air Management Branch, WO-WSA-3,Washington, DC, pp. 3-15.

Potter, WD., Byrd, T.A., Miller, JA., Kochut, KJ., 1992.Extending decision support systems: The integration of data,knowledge, and model management Annals of OperationsResearch 38, 501-527.

Rauscher, ILM., Alban, DJL, Johnson, D.W., 1993. Managing theglobal climate change scientific knowledge base. AI Applica-tions 7(4), 17-̂ 10.

Rauscher, RM., 1995. Natural resource decision support: Theoryand practice. AI Applications 9(3), 1-2.

Rauscher, HM, 1996. Decision making methods for ecosystemmanagement decision support Caring for the forest Researchin a changing world, Congress Report, voL n, IUFRO XXWorld Congress, 6-12 August 1995 Tampere, Finland, pp. 264-272. ;

Rauscher, ILM., Loftis, D.L., McGee, C.E., Worth, C.V., 1997a.Oak Regeneration: A knowledge synthesis. The Compiler15(1): pp. 51-52, insert, 3 disks, 3.8 Megabytes; 748 chunks;742 links (electronic).

Rauscher, H.M., Kollasch, RJ"., Thomasma, SA, Nute, D.E.,Chen, N., Twery, MJ., Bennett, DJ., Cleveland, H., 1997b.NED-1: A goal-driven ecosystem management decision supportsystem: Technical description. Proc. GIS World 97, Integratingspatial information technologies for tomorrow. 17-20 February1997, Vancouver, BC, pp. 324-332.

Reynolds, Keith M., Rauscher, H. Michael, Worth, Christopher V.,1995. A hypermedia reference system to the forest ecosystemmanagement assessment team report and some relatedpublications. USDA Forest Service Pacific Northwest ResearchStation, Gen. Tech. Rep. PNW-GTR-357,13 pp.

Reynolds, K., Saunders, M., Miller, B., Murray, S., Slade, J., 1997.An application framework for decision support in environ-mental assessment Proc. GIS World 97, Integrating spatialinformation technologies for tomorrow. 17-20 February 1997,Vancouver, BC, pp. 333-337.

Reynolds, K., Bjork, J., Riemann Hershey, R., Schmoldt, D., Payne,J., King, S., Moeur, M., DeCola, L., Twery, M., Cunningham,P., Lessard, G., 1998. Decision support systems for ecosystemmanagement In: Szaro, R., Sexton, W., Johnson, N., Malk, A.(Eds.), Ecosystem management A reference guide to con-cepts and implementation. Island Press, Washington, DC, inpress.


Rittel, H., 1972. On the planning crisis: Systems analysis of thefirst and second generations. Bedriftokonomen 8, 390-396.

Rittel, H., Webber, MM., 1973. Dilemmas in a general theory ofplanning. Policy Sciences 4,155-169.

Rue, L.W., Byars, LI., 1992. Management Skills and Application.Richard C. Irwin, Homewood, IL, 421 pp.

Southern Appalachian Man and the Biosphere (SAMAB), 1996.The southern Appalachian assessment summary report Reports1-5, USDAForestService,SouthemRegion, Atlanta, GA, 332 pp.

Sage, A.P., 1991, Decision Support Systems Engineering. Wiley,New York, 344 pp.

Sample, V.A., 1990. The Impact of the Federal Budge Process onNational Forest Planning. Greenwood Press, New York, 250 pp.

Sample, V.A., 1993. A framework for public participation innatural resource decision making. J. Forestry 91(7), 22-27.

Schmoldt, D.L., Rauscher, ttM, 1996. Building Knowledge-basedSystems for Natural Resource Management Chapman andHall, New York, 386 pp.

Schuster, E.G., Leefers, L.A., Thompson, I.E., 1993. A guide tocomputer-based analytical tools for implementing nationalforest plans. USDA Forest Service, Gen. Tech. Rep. INT-296,269 pp.

Shands, W.E., 1994. National forests and the human legacy: Somehistory. In: Foley, LJL (compiler), Silviculture: From the cradleof forestry to ecosystem management Proc. National Silvicul-ture Workshop, 1-4 November 1993 Hendersonville, NC,USDA Forest Service Gen. Tech. Rep. SE-88, pp. 3-11.

Shepard, W.B., 1993. Ecosystem management in the ForestService: Political implications, impediments, and imperatives.In: Jensen, ME., Bourgeron, P.S. (Eds.), Eastside ForestEcosystem Health Assessment voL H Ecosystem Management:Principles and Applications. USDA Forest Service, pp. 31-37.

Shindler, B., Steel, B., List P., 1996. Public judgments of adaptivemanagement J. Forestry 94(6), 4-12.

Shindler, Bruce, Neburka, Julie, 1997. Public participation in forestplanning: 8 attributes of success. J. Forestry 95,17-19.

Silver, M.S., 1991. Systems that Support Decision makers:Description and Analysis. Wiley, New York, 256 pp.

Smith, GJL, 1997. Making decisions in a complex dynamic world.Kohm, ELA-, Franklin, J.F. (Eds.), Creating a forestry for the21st century. Island Press, Washington, DC, pp. 419-435.

Steelman, T.A., 1996. Public participation in national forestmanagement A case study of the Monongahela NationalForest West "Virginia. Proc. TEEE 1996 International Sympo-sium on Technology and Society. Technical Expertise andPublic Decisions, 21-22 June 1996 Princeton University,Princeton, NJ, pp. 226-230.

Stephens, JA., 1996. A new era in planning analysis tools.Analysis notes, USDA Forest Service, WO/Ecosystem Manage-ment Analysis Center, Ft Collins, CO, voL 6,2, pp. 1-2.

Stock, MW, Rauscher, RM, 1996. Artificial intelligence anddecision support in natural resource management New ZealandJ. Forestry Sci. 26,145-157.

Sygenex, 1994. Criterium Decision Plus: The Complete DecisionFormulation, Analysis, and Presentation for Windows. User'sGuide. Sygenex, 15446 Bel-Red Road, Redmond, WA 98052,327pp.

Teck, R., Moeur, M, Eav, B., 1996. Forecasting ecosystems withthe forest vegetation simulator. J. Forestry 94(12), 7-10.

Thomas, J.W., 1995. The forest service program for forest andrangeland resources: A long-term strategic plan. Draft 1995RPA Program, USDA Forest Service, Washington, DC.

Thomas, J.W., 1997. Foreword. In: Kohm, K-A., Franklin, JJF.(Eds.), Creating a Forestry for the 21st Century. Island Press,Washington, DC, pp. ix-xii.

Thomson, AJ., 1993. Paradigm green: AI approaches to evaluatingthe economic consequences of changing environmental view-points. AI Applications 7(4), 61-68.

Thomson, AJ., 1996. Asimov's physchohistory: Vision of thefuture or present reality?. AI Applications 10(3), 1-8.

Ticknor, WD., 1993. Sustainable forestry: Redefining the role offorest management In: Aplet, G.H., Johnson, N., Olson, J.T.,Sample, VIA. (Eds.), Defining Sustainable Forestry. IslandPress, Washington, DC, pp. 260-269.

Tonn, B., White, D., 1996. Sustainable societies in the informationage. The American Sociologist 27(1), 102-121.

Tonn, B., English, M, Travis, C., 1998. Frameworks for under-standing and improving environmental decision making.ORNL/NCEDR-7, Oak Ridge National Laboratory, Oak Ridge,TN,45pp.

Toynbee, Arnold J., 1946. A study of history. Oxford UniversityPress, New York, 742 pp.

Turban, E., 1993. Decision Support and Expert Systems: Manage-ment Support Systems. MacMfflan Publishing, New York, 833pp.

Twery, MJ., Bennett, D J., Kollasch, RJ>, Thomasma, S A., Stout,SI., Palmer, J.R, Hoffman, RA., DeCalesta, D.S., Hornbeck,J., Rauscher, KM., Steinman, J., Gustafson, E., Miller, G.,Cleveland, H., Grove, M, McGuinness, B., Chen, N., Mute,DJEL, 1997. NED-1: An integrated decision support system forecosystem management Proc. Resource Technology 97 Meet-ing, 7-11 April 1997, Seattle, WA, pp. 331-343.

Van den Berg, J.C.J.M., 1996. Ecological Economics andSustainable Development Theory, Methods, and Applications.Edward Elgar, Cheltenham, UK, 312 pp.

Vroom, VJL, Jago, A.G., 1988. The new leadership: Managingparticipation in organizations. Prentice Hall, Englewood Cliffs,NJ, 389pp.

Walters, CJ., Rolling, C.S., 1990. Large-scale managementexperiments and learning by doing. Ecology 71(6), 2060-2068.

Wear, D.N., Turner, MG., Flamm, R.O., 1996. Ecosystemmanagement with multiple owners: Landscape dynamics in asouthern Appalachian watershed. Ecological Applications 6(4),1173-1188.

Williams, S3., Roschke, DJ., Holtfrerich, D.R., 1995. Designingconfigurable decision-support software: Lessons learned. AIApplications 9(3), 103-114.

Zeide, B., 1994. An unsustainable technology: Sustainabledevelopment In: ECOTECH '94, Internet Address: et-tor@-seam-sunetse, paper ET-004.

Zonnefeld, LS., 1988. Basic principles of land evaluation usingvegetation and other land attributes. In: Kuecbler, A.W.,Zonnefeld, LS. (Eds.), Vegetation Mapping. Kluwer AcademicPublishers, Dordrecht The Netherlands, pp. 142-157.

Forest Ecology

and Management

Submission of manuscripts.Manuscripts should be submitted in triplicate. Authors from the Americas, Australia, New Zealand and the Pacific are requestedto send their manuscripts to Forest Ecology and Management, Dr. Richard F. Fisher, Department of Forest Science, TexasA&M University, College Station, TX77843-2135, USA; all other authors are requested to send their manuscripts to the EditorialOffice, Forest Ecology and Management, P.O. Box 181,1000 AD Amsterdam, The Netherlands.

Authors In Japan please note: Upon request Elsevier Science Japan will provide authors with a list of people who can checkand improve the English of their paper (before submission). Please contract our Tokyo office: Elsevier Science Japan, 1-9-15Higashi-Azabu, Minato-ku, Tokyo 106-0044, Japan; tel. (+81)-3-5561-5033; fax (+81)-3-5561-5047.

Electronic manuscripts: Sectronic manuscripts have the advantage that there is no need for the rekeying of text thereby avoid-ing the possibility of introducing errors and resulting in reliable and fast delivery of proofs.For the initial submission of manuscripts for consideration, hardcopies are sufficient For the processing of accepted papers,electronic versions are preferred. After final acceptance, your disk plus two, final and exactly matching printed versions shouldbe submitted together. Double density (DD) or high density (HD) diskettes (3.5 or 5.25 inch) are acceptable. It is important thatthe file saved is in the native format of the wordprocessor program used. Label the disk with the name of the computer andwordprocessing package used, your name, and the name of the file on the disk. Further information may be obtained fromthe Publisher.

Enquiries concerning manuscripts and proofs: questions arising after acceptance of the manuscript, especially those relatingto proofs, should be directed to Elsevier Science Ireland Ltd., Bay 15K, Shannon Industrial Estate, Shannon, Co. Clare, Ireland,tel. (+353-61) 471944; fax(+353-61) 472144.

Advertising information: Advertising orders and enquiries can be sent to: USA, Canada and South America: MrTino de Carlo,The Advertising Department Elsevier Science Inc., 655 Avenue of the Americas, New York, NY 10010-5107, USA; phone: (+1)(212) 633 3815; fax: (1) (212) 633 3820; e-mail: [email protected] Japan: The Advertising Department Elsevier ScienceK.K., 9-15 Higashi-Azabu 1-chome, Minato-ku, Tokyo 106, Japan; phone: (+81) (3) 5561 5033; fax: (+81) (3) 5561 5047. Europeand ROW: Rachel Gresle-Farthing.The Advertising Department Elsevier Science Ltd., The Boulevard, Longford Lane, Kidling-ton, Oxford OX51GB, UK; phone: (+44) (1865) 843565; fax: (+44) (1865) 843976; e-mail: [email protected]

Orders, claims, and product enquiries: please contact the Customer Support Department at the Regional Sales Office nearestyou:New York: Elsevier Science, PO Box 945, New York, NY 10159-0945, USA; Tel.: (+1) (212) 633 3730 [toll free number for NorthAmerican customers: 1-888-4ES-INFO (437-4636)]; Fax: (+1) (212) 633 3680; E-mail: [email protected]: Elsevier Science, PO Box 211,1000 AE Amsterdam,The Netherlands;Tel.: (+31) 20 4853757; Fax: (+31) 20 4853432;E-mail: [email protected]: Elsevier Science, 9-15, Higashi-Azabu 1-chome, Minato-ku, Tokyo 106-0044, Japan; Tel.: (+81) (3) 55615033; Fax: (+81) (3)5561 5047; E-mail: [email protected]: Elsevier Science, No.1 Temasek Avenue, #17-01 Millenia Tower, Singapore 039192; Tel.: (+65) 434 3727; Fax: (+65)337 2230; E-mail: [email protected] de Janeiro: Elsevier Science, Rua Sete de Setembro 111/16 Andar, 20050-002 Centra, Rio de Janeiro - RJ, Brazil; Tel.: (+55)(21) 509 5340; Fax: (+55) (21) 507 1991; E-mail: [email protected] [Note (Latin America): for orders, claims and helpdesk information, please contact the Regional Sales Office in New York as listed above]

- US mailing Info: Forest Ecology and Management (0378-1121) is published monthly by Elsevier Science B.V. (Molenwerf 1, Post-bus 211,1000 AE Amsterdam). Annual subscription price in the USA is US$ 2307.00 (valid in North, Central and South America),including air speed delivery. Second class postage rate is paid at Jamaica, NY 11431.USA POSTMASTERS: Send address changes to forest Ecology and Management Publications Expediting, Inc., 200 MeachamAvenue, Bmbrrt, NY 11003.AIRFREIGHT AND MAILING in the USA by Publications Expediting Inc., 200 Meacham Avenue, Elmont, NY 11003.

Forest Ecology and Management has no page charges

For a full and complete Guide for Authors, please refer toForest Ecology and Management, Vol. 113, No.1, pp. 105-108.The instructions can also be found on the World Wide Web:

access under http://www.elsevier.nl or http://www.elsevier.com

© The paper used in this publication meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).

Printed in The Netherlands

LANDSCAPE AND

URBAN PLANNING

An International Journal of Landscape Ecology,

Landscape Planning and Landscape design

, Editor-in-Chief:J.E. Rodiek, College of Architecture, Texas A&M University, College Station, TX 77843-3137, USA

AIMS AND SCOPEA journal concerned withconceptual, scientific anddesign approaches to landuse. By emphasizingecological understanding anda multi-disciplinary approachto analysis and planning anddesign, it attempts to drawattention to the interrelatednature of problems posed bynature and human use ofland. In addition, papersdealing with ecologicalprocesses and interactionswithin urban areas, andbetween these areas and thesurrounding natural systemswhich support them, will beconsidered. Papers in whichspecific problems areexamined are welcome.Topics might include but arenot limited to landscapeecology, landscape planningand landscape design.Landscape ecology examineshpw heterogeneouscombinations of ecosystemsare structured, how theyfunction and how theychange. Landscape planningexamines the various wayshumans structure their landuse changes. Landscapedesign involves the physicalstrategies and forms by whichland use change is actuallydirected. Landscape andUrban Planning is based on

the premise that researchlinked to practice willultimately improve the humanmade landscape.

Editorial Advisory Board:LD. Bishop, PanwSle, Vic., Australia,E.G. Bolen, WSmington, JVC, USA,ID. Brims, Schomdorf, Germany,JJ3. Byrom, Edinburgh, UK,T.C. Daniel, Tucson, AZ, USA,RJH DeGraaf, Amherst, MA, USA,J.G. Fabos. Amherst, MA, USA,S. Gonzalez Alonso, Madrid, Spam,M. Hough, Etobicoke, ON, Canada,P. Jacobs, Montreal, PQ, Canada,D.S. Jones, Melbourne, Vic., Australia,H. Lavery, Milton, Oft/., Australia,WJW. Harsh. Hnt,M/, USA,D-Llfitehen.Oa/fes,7X,US4.D.Q. Morrison, Athens, GA, USA,JJ. Nassauer. Si Paul.MN. USA,M. Nellscher, Guejph, ON, Canada,DJ). Paterson, Vancouver, BC, Canada,A. Ramos, Madrid, Spain,P. Shepard. Claremont, CA, USA,QJR. Skage, Alnarp, Sweden,R.C. Smardon, Syracuse, NY, USA,C. Sorte, Alnarp, Sweden,F. Steams, Rhinelander, Wl, USA,

ELSEVIER

SCIENCE PUBLISHERS

R.C. Szaro, Washington, DC, USA,J.W. Thomas, La Grande, OP, USA,P.J.Trowbridge, Ithaca, NY, USA,T.HJ>. Turner. London, UK,M.J. Vroom, Wageningen,The Netherlands,W.V. Wendler, College Station, TX,USA,B.-E. Yang, Seoul, Korea,EMTube, Tucson, AZ, USA

ABSTRACTED/INDEXED INApplied Ecology Abstracts,Biological Abstracts,Current Contents B & S,Environmental PeriodicalsBibliography, Geobase,Geographical Abstracts,LandSearch.

1994 SUBSCRIPTION DATA

Volumes 28-30 (In 9 issues)Subscription price:

Dfl. 1080.00 (US $584.00)Inch Postage

ISSN 0169-2046

Ebevier Science PubDshers,P.O. Box 211, 1000 AE Amsterdam,The NetherlandsFax:(020)5803-203Customers In tfw USA and Canada:Ebevier Science PublishersP.O. Box 945, Madison Square Station,New York, NY101 604757, USAFax:(212)633-3680

ricequaed^Kin to America aty. Non VXTteglDmdcuitoimrainllwEinpemCooiiniirilrtlxwUtddlhtipptopiuBVAT rate applicable in Ihehcouitiy to lh«pic«.

reprinted from forest ecology and management forest ...coweeta.uga.edu/publications/63.pdf ·...

Documents