adaptive management in habitat conservation plans

Essay

20

Conservation Biology, Pages 20–29Volume 16, No. 1, February 2002

Adaptive Management in Habitat Conservation Plans

GEORGE F. WILHERE

Wildlife Program, Washington Department of Fish and Wildlife, 600 Capitol Way North, Olympia, WA 98501, U.S.A., email [email protected]

Abstract:

Habitat conservation plans (HCPs) allow incidental take of threatened or endangered species inexchange for conservation measures that minimize and mitigate such taking. Habitat conservation plans en-tail a compromise between regulatory certainty and scientific uncertainty. This compromise is controversialbecause many HCPs are thought to inadequately address scientific uncertainty. Adaptive management is thesystematic acquisition and application of reliable information to improve natural resource managementover time. Ideally, under adaptive management, conservation strategies are implemented as a deliberate ex-periment. This approach can establish cause-and-effect relationships and point the way toward optimal strat-egies. Adaptive management has been promoted as essential to management under uncertainty, but fewHCPs incorporate genuine adaptive management. Habitat conservation plans will continue to lack adaptivemanagement until certain conditions are met, such as acknowledgment that an HCP is a management hy-pothesis, landowner interest in improving biological outcomes, and sufficient financial resources. Economicincentives would encourage adaptive management in HCPs. Habitat conservation plan permittees might re-ceive direct payments or tax deductions for reliable information that benefits a species. “Mitigation credits”could be awarded for information produced through adaptive management. In effect, habitat would be ex-changed for information that benefits a species. Successful use of mitigation credits would depend on cor-rectly valuing information and enforcement of the U.S. Endangered Species Act. Under a “precautionary pol-luter pays principle,” an HCP permittee would put up an environmental assurance bond. Portions of thebond are returned with interest as adaptive management demonstrates that environmental damages are un-likely to occur. Funds spent on adaptive management are funds unavailable for habitat protection, so limitedfinancial resources may force a compromise between protecting habitat and acquiring knowledge.

Manejo Adaptivo en Planes de Conservación de Hábitat

Resumen:

Los planes de conservación de hábitat (PCH) permiten la captura incidental de especies amenaza-das o en peligro a cambio de medidas de conservación que minimicen y mitiguen tales capturas. Los planes deconservación de hábitat conllevan un compromiso entre certidumbre en los reglamentos e incertidumbrecientífica. El compromiso es controversial porque se piensa que muchos PCH enfrentan la incertidumbrecientífica inadecuadamente. El manejo adaptivo es la obtención sistemática y aplicación de información con-fiable para mejorar el manejo de recursos en el tiempo. Idealmente, bajo el manejo adaptivo, las estrategiasde conservación se instrumentan como un experimento deliberado. Este enfoque puede establecer relacionesde causa – efecto y señalar el camino para estrategias óptimas. El manejo adaptivo se ha considerado esen-cial para el manejo bajo incertidumbre, sin embargo, pocos PCH incorporan el manejo adaptivo genuino.Los planes de conservación de hábitat seguirán careciendo de manejo adaptivo hasta que se cumplan ciertascondiciones, tales como reconocer que un PCH es una hipótesis de manejo, el interés de los propietarios detierras en mejorar los resultados biológicos y suficientes recursos financieros. Los incentivos económicos fo-mentarían el manejo adaptivo en los PCH. Los permisionarios de los PCH podrían recibir pagos directos o de-ducciones de impuestos por información confiable que beneficie a una especie. Se podrían otorgar “créditosde mitigación” por información producida a través del manejo adaptivo. En efecto, se canjearía hábitat porinformación benéfica para una especie. El exitoso uso de los créditos de mitigación depende de la correctaevaluación de la información y aplicación del Acta de Especies en Peligro de E.U.A. Bajo una aproximaciónde que “el contaminador precautorio paga el principal” de los PCH, un permisionario podría fijar un bonode seguro ambiental. Porciones del bono se devuelven con intereses a medida que el manejo adaptivo de-

Paper submitted August 10, 2000; revised manuscript accepted March 14, 2001.

Conservation BiologyVolume 16, No. 1, February 2002

Wilhere Adaptive Management in HCPs

21

Introduction

Adaptive management is widely recognized as an intelli-gent, if not essential, approach to the management ofnatural resources under uncertainty ( Walters & Holling1990; Irwin & Wigley 1993; Parma et al. 1998). As origi-nally conceived, adaptive management can be defined asthe systematic acquisition and application of reliable in-formation to improve management over time. Althoughdeveloped in the 1970s ( Holling 1978), adaptive manage-ment is still a poorly understood concept that has been re-peatedly misapplied or misappropriated (Halbert 1993;Walters 1997 ). Many natural resource managers are notaware of the scientific principles underlying genuine adap-tive management. The meaning of adaptive managementand its proper application have been particularly ambig-uous with respect to habitat conservation plans (HCPs).

Habitat conservation plans show promise as an effec-tive means for resolving conflicts between the U.S. En-dangered Species Act ( ESA) and the legitimate use of nat-ural resources ( Bean & Wilcove 1997; Noss et al. 1997 ).An HCP is the basis for a contract between an “applicant”(typically a private landowner) and the U.S. Fish and Wild-life Service or the National Marine Fisheries Service (theservices), who are responsible for protecting federallylisted threatened and endangered species. The contractallows a permittee (formerly the applicant) to incidentallytake listed species in exchange for conservation measuresthat minimize and mitigate such taking. Habitat conserva-tion plans attempt to reconcile the economic costs of habi-tat protection with the risks of species extinction. Ourknowledge of species and habitats is usually poor; conse-quently, crafting such compromise is fraught with scien-tific uncertainty. To deal with uncertainty and its attendantrisks, adaptive management is often invoked. To date, how-ever, the “adaptive management” described in most HCPsdoes little to reduce risk. A study of HCPs indicates thatonly about 5% of HCPs have a “monitoring plan sufficientto evaluate the [HCP’s] success” (Kareiva et al. 1999).

Many of the problems associated with adaptive man-agement in HCPs can be attributed to (1) misunderstand-ing of the adaptive management concept, (2) misunder-standing of Section 10(a) of the ESA and regulationspromulgated pursuant to the act, and (3) a lack of will, orincentives, to implement HCPs based on adaptive manage-ment. My goal here is to (1) contribute to a common under-standing of the adaptive management concept; (2) reviewthe current relationship between adaptive management

and HCPs; and (3) discuss the future of adaptive manage-ment in HCPs, including incentives to encourage it.

Adaptive Management

Adaptive management is a misunderstood concept, whichis not surprising. The numerous parties involved in re-solving contentious natural resource management issues—scientists, managers, landowners, lawyers, concerned citi-zens, political activists—possess different professional back-grounds and perspectives. For instance, some managersbelieve that good management is by definition “adap-tive.” But to them,

adaptive

simply means a willingnessto change. They have not been informed, or have not un-derstood, that adaptive management is a thoroughly newparadigm for managing natural resources. Other stake-holders equate adaptive management with “flexible man-agement” ( Halbert 1993) or see it as an opportunity tocontest policies they consider objectionable. All theseviews ignore a key aspect of adaptive management, thesystematic acquisition of reliable information.

Understanding of adaptive management will be enhancedby first understanding the alternatives: deferred action andtrial and error (Walters & Hilborn 1978; Walters & Holling1990). Under deferred action, an ecosystem is not man-aged until after it is understood. Only minimal disturbanceis allowed while basic research is conducted to determinekey processes and relationships. Deferred action is an eco-logically cautious approach, but it has an economic costdue to the discounting of future revenues. The longer man-agement is deferred, the larger the loss in net present value( Herfindahl & Kneese 1974). Consequently, deferred ac-tion is unattractive to most private land managers. There isalso a sound scientific argument against deferred action.Behavior of an undisturbed ecosystem can be dramaticallydifferent than that of a managed ecosystem, so knowl-edge acquired while deferring action may not be validfor the managed ecosystem ( Walters & Hilborn 1978;Parma et al. 1998).

Trial and error has been and continues to be the domi-nant paradigm in natural resource management. Trialand error typically emphasizes the “trial,” which entailsresource utilization and produces revenue, but neglectserror detection, which entails costly monitoring. Trial-and-error approaches are also referred to as “learning bydoing” (Haney & Power 1996) or “evolutionary” ( Walters& Holling 1990); both monikers are misleading. Managers

muestra que es poco probable que ocurran daños ambientales. Los fondos utilizados en manejo adaptivoson fondos no disponibles para la protección del hábitat. Por lo tanto, los recursos financieros limitados

pueden forzar un compromiso entre la protección del hábitat y la adquisición de conocimiento.

22

Adaptive Management in HCPs Wilhere


undoubtedly learn by doing, but particular types of learn-ing do more harm than good. Casual observations, anec-dotal reports, and unreplicated case studies lack statisti-cally valid experimental design and are likely to yieldunreliable information. Managers relying on these typesof learning may fail to detect errors—damage to the en-vironment—and risk perpetuating harmful policy. Al-though biological evolution might be a trial-and-errorprocess, each individual in a population is a trial, sothere are many trials running simultaneously. Under trialand error, managers typically implement a single policyand assume it is satisfactory until proved otherwise.

Trial-and-error approaches lead to small, incrementalchanges over time ( Halbert 1993). Major changes, whenthey occur, are often driven by political pressure stem-ming from social concerns. Social concerns could be in-cited by a species threatened with extinction or the col-lapse of an overexploited fish stock. Given thesecommon scenarios, trial and error may be more accu-rately described as “reactive learning” ( Hilborn 1992) or“crisis management” ( Halbert 1993). Trial and error un-dervalues information, so data collection is poorlyfunded. For this reason, trial and error appears to be rela-tively inexpensive, and it will be if all goes as expected.But natural resource management is full of surprises. If allcosts are considered—the costs of undetected environ-mental damage, management inefficiencies, interruptedoperations, lawsuits—then trial and error may be rela-tively expensive over the long run. The heated contro-versy over management of national forests in the PacificNorthwest ( Thomas et al. 1993; Marcot & Thomas 1997 )demonstrates the myriad shortcomings of trial and error.

In contrast, under adaptive management, major driversof change are generated internally. Regular adjustmentsto policies are acknowledged as necessary and desirable.Information is highly valued, so data collection is wellfunded. The costs of research and monitoring make adap-tive management seem relatively expensive. But if allcosts are considered, then adaptive management may berelatively inexpensive over the long run. In theory, invest-ments in reliable information should yield excellent re-turns in the sustainable use of natural resources (Smith &Walters 1981; Hilborn et al. 1995; Walters & Green 1997).

Adaptive management comes in two varieties, passiveand active ( Walters & Hilborn 1978). Passive adaptivemanagement has been confused with trial-and-error ap-proaches ( Irwin & Wigley 1993; Parma et al. 1998). Asoriginally conceived, passive adaptive management is ascientifically rigorous process of formulating predictivemodels, making policy decisions based on those models,and revising the models as monitoring data become avail-able ( Walters & Hilborn 1978; Hilborn 1992; Fig. 1). Themodel is used to predict ecosystem responses to man-agement activities. In theory, activities can be intention-ally designed to perturb the ecosystem in a way that en-hances parameter estimation (Isermann 1981). Monitoring

and evaluation systems are in place before managementcommences, but monitoring is done without controls,replication, or randomization—the three essential aspectsof statistically valid experimental design ( Hurlbert 1984).Consequently, passive adaptive management has a flaw: itcannot establish cause-and-effect relationships betweenmanagement activities and changes in ecosystem condi-tions. Without controls, replication, and randomization,managers often cannot determine whether the observedresponses were caused by the management activity ofinterest, by some other activity, or by some “natural” pro-cess. The advantages to passive adaptive managementare that it is relatively simple and cheap.

Under active adaptive management, management ac-tivities are conducted as a deliberate experiment ( Walters& Hilborn 1978; Fig. 2). Alternative policies are viewed astreatments and are implemented through statistically validexperimental design. Monitoring is the data-collectionstep of the experiment. Active adaptive management canestablish cause-and-effect relationships between manage-ment activities and changes in ecological conditions. Ac-tive adaptive management leads to a better understandingof how and why natural systems respond to management.Active adaptive management has another advantage overthe passive approach: responses to a range of treatmentscan point the way toward an optimal policy. The disadvan-tages of active adaptive management are that it is morecomplex and more expensive.

Walters (1997 ) said, “The essential idea of adaptivemanagement is to recognize explicitly that managementpolicies can be applied as experimental treatments.” Acrucial implication of Walter’s statement is that monitor-ing activities must be integrated with management activ-ities. Under adaptive management, monitoring is not thelast chapter of a conservation plan; rather, monitoringand conservation plans are developed concurrently toform a single adaptive-management plan. The evolutionof the adaptive-management concept has been awayfrom passive approaches and toward active approaches( Forest Ecosystem Management Assessment Team 1993;Irwin & Wigley 1993; Lee 1993). I will henceforth usethe term adaptive management to refer to active adap-tive or experimental approaches to management.

Habitat Conservation Plans

The ESA of 1973 did not allow compromise. The U.S.Congress legislated an absolute prohibition against thetaking of endangered wildlife ( Bean 1983). This seem-ingly idealistic approach to the conservation of endan-gered species created several problems. First, govern-ment officials were reluctant to enforce a law thatprohibited and potentially punished an extremely widerange of otherwise lawful activities, such as forestry andreal-estate development. Second, fear of land-use restric-



23

tions incited some landowners to deliberately destroyhabitats of endangered species ( Hood 1998). Third,landowner apprehension interfered with scientists’ abil-ity to monitor and understand endangered species; somelandowners refrained from reporting occurrences of en-dangered species on their property. Finally, the ESA un-fairly penalized landowners who happened to own thelast available habitat of a species after all surroundinghabitat had been destroyed (Hood 1998). According toBean and Wilcove (1997 ), the absolute prohibitions ofthe act did little to conserve habitats of federally listedspecies on nonfederal lands.

In 1982, Section 10 of the ESA was amended to allowcompromise. The amendment “establishes a procedurewhereby those persons whose actions may affect endan-gered or threatened species may receive permits for theincidental taking of such species, provided the actionwill not jeopardize the continued existence of the spe-cies” ( U.S. Congress, House of Representatives 1982).Section 10(a) authorizes such incidental take subject tothe approval of an HCP. The amendment was, in effect,Congressional approval for a compromise between theeconomic costs of habitat protection on private landsand the likelihood of species extinction.

Under Section 10(a), issuance of an incidental takepermit requires that (1) the taking is incidental to other-wise lawful activities; (2) the taking is, to the maximumextent practicable, minimized and mitigated; (3) ade-quate funding for the plan is ensured, (4) the taking will

not appreciably reduce the likelihood of the survival andrecovery of a species in the wild; and (5) such other mea-sures be implemented that the services may require asbeing necessary or appropriate for purposes of the plan.The first criterion is easily satisfied. The last criteriongives the services authority to require monitoring and/oradaptive management. The relationship between the sec-ond and fourth criteria warrants discussion. The secondcriterion is an economic test establishing the maximumconservation (i.e., mitigation and minimization) requiredof an applicant. In effect, it says that applicants are re-quired to bear the greatest cost they can practicably af-ford. The fourth criterion is a biological test establishingthe minimum conservation required by the applicant.The logic of these two criteria is clear: satisfying the bio-logical criterion is necessary but not sufficient; a validHCP also must demonstrate that it satisfies the economiccriterion. At present, the services do not require an ap-plicant to disclose financial analyses of their HCPs ( U.S.Fish and Wildlife Service & National Marine FisheriesService 1996), so the validity of many HCPs could be le-gitimately questioned.

Only 12 HCPs were approved between 1983 and 1992(Hood 1998). For many reasons, HCPs were not as at-tractive to private landowners as Congress may havehoped. In 1994 the Secretary of the Interior issued the“no-surprises” policy, which gave private landownerslong-term regulatory certainty, an attractive incentive. Itsaid that the services would never require additional

Figure 1. Block diagram of passive adaptive management. As originally conceived, passive adaptive management was a scientifically rigorous process based on a quantitative model. Monitoring was used for system identification and parameter estimation. Modified from Walters and Hilborn (1978).

24



commitments of land or financial compensation beyondthat agreed to under the terms of the HCP. Even if un-foreseen circumstances arose that could jeopardize thespecies’ survival, the permittee would not have to pro-vide additional habitat conservation. In effect, the no-surprises policy precluded future modifications to anHCP and led to a proliferation of HCPs. By 1997, over212 HCPs had been approved and at least 200 morewere being developed ( Kaiser 1997 ). About 78% ofHCPs cover less than 200 ha each, but at least two HCPseach cover more than 400,000 ha ( U.S. Fish and WildlifeService & National Marine Fisheries Service 2000). Themajority have dealt with commercial forest managementand real-estate development ( Hood 1998).

As written, the no-surprises policy did not allow com-promise. The permittee was given absolute regulatorycertainty, with no provisions for the myriad scientificuncertainties associated with complex ecosystems. Thiselicited considerable criticism from scientists (G. Meffeet al. 1996. Endangered species legislation shouldn’tflout scientific reality, a letter from 169 scientists to Sen-ator John Chafee; D. Murphy et al. 1997. A statement onproposed private lands initiatives and reauthorization ofthe Endangered Species Act from the meeting of scien-tists at Stanford University). They considered no-sur-prises to be a flawed policy and a troubling obstacle toadaptive management. The services were not ignorantof the need for adaptive management. Their own hand-book for HCPs ( U.S. Fish and Wildlife Service & National

Marine Fisheries Service 1996) said that “Where signifi-cant uncertainty exists, it can be addressed through theincorporation and implementation of adaptive manage-ment measures into HCPs,” and that “... there should bea clear understanding and agreement between the ser-vices and the permittee as to the mitigation range of ad-justment which might be required as a result of any adap-tive management provisions.” The services’ handbookand the no-surprises policy were inconsistent, however,and the heavily promoted policy instilled in HCP appli-cants expectations of absolute certainty.

The policy was codified in 1998 ( U.S. Fish and Wild-life Service & National Marine Fisheries Service 1998),but the new regulations now contain a mechanism foraddressing uncertainty. Compromise between the regu-latory certainty sought by the applicant and the scien-tific uncertainty associated with complex ecosystemscan be negotiated. Compromise is based on a distinctionbetween “changed circumstances” and “unforeseen cir-cumstances.” Changed circumstances are those that“can reasonably be anticipated” and “can be plannedfor.” Fire in areas prone to wildfire is an example. Incontrast, unforeseen circumstances are those that“could not reasonably have been anticipated.” The assur-ances given to permittees depend on the nature of fu-ture circumstances. If a circumstance changes in a wayprovided for in the HCP, then the permittee will imple-ment the specified conservation measures. This part ofthe new regulations is an improvement over the original

Figure 2. Block diagram of active adaptive management, which is natural resource management conducted as an experiment. Cause and effect are established through an active approach. Ideally, treatments ( T1 and T2) and the control (C) should be replicated and assigned randomly. Natural disturbances affect T1, T2, and C.



25

no-surprises policy because HCPs can change in responseto future events. But if there is no provision in the HCPfor that particular changed circumstance, then no addi-tional conservation measures are required from the per-mittee. If an unforeseen circumstance occurs, changes tothe HCP cannot involve additional land or financial com-pensation or additional restrictions without the consentof the permittee. This is the same deal granted under theno-surprises policy.

About a year after the new regulations were announced,the services presented their “5-point policy initiative,”which is an addendum to the services’ HCP handbook( U.S. Fish and Wildlife Service & National Marine Fisher-ies Service 2000). Two points provide additional guid-ance regarding expectations for adaptive managementand monitoring, the most important of which is a discus-sion of experimental approaches for evaluating alterna-tive conservation strategies.

How will the new federal regulations and five-pointpolicy initiative affect the use of adaptive managementin HCPs? The services clearly expect that HCPs willbe modified in response to changing circumstances, butthey also accept limits on those changes. The extent ofthe limitations is unclear and will depend on how thecategories of “changed circumstances” and “unforeseencircumstances” are addressed during negotiations. Antic-ipating, planning for, and monitoring “changed circum-stances” is the mechanism through which adaptive man-agement can be incorporated into HCPs. An HCP basedon adaptive management should produce three prod-ucts: habitat for covered species, financial profit for thepermittee, and reliable information.

Negotiating Adaptive Management

The new regulations for HCPs open the door to adaptivemanagement. But adaptive management, like many as-pects of an HCP, will be negotiated. Although negoti-ated solutions to habitat conservation are considered un-seemly by some scientists, this approach is gainingacceptance ( Noss et al. 1997; Richter & Redford 1999).The outcome of negotiations—the level of commitmentto adaptive management—will be affected by the appli-cant’s situation and values. Lee (1993) describes institu-tional conditions favoring adaptive management. Someof Lee’s conditions suggest that HCPs and adaptive man-agement should be compatible; others reveal why HCPsrarely incorporate adaptive management but point theway toward corrective policies. According to Lee (1993),the following conditions must be met before decision-makers will adopt genuine adaptive management:

(1) There must be a mandate to take action in the faceof uncertainty. This is true of all HCPs. The principal par-ties, the HCP applicant and the services, have a strongmotivation to act. The applicant desires economic profit

from ownership of a natural resource such as timber orundeveloped land but is frustrated by the ESA. The ser-vices perceive HCPs as a means to acquire both ecologi-cal benefits for listed species and political support forthe ESA ( Moser 2000).

(2) Preservation of pristine environments must nolonger be an option. This is almost always true for HCPs:the habitats involved have already been affected by hu-man activities. In some rare cases, preservation is an op-tion if sufficient funds are available to purchase land or aconservation easement. Stronger Congressional supportfor and stricter enforcement of the ESA are also options,but these seem politically unrealistic at this time.

(3) Human intervention must not be able to producedesired outcomes predictably. Our understanding ofmost species is too poor to predict outcomes, so thiscondition is likely to be met. If it is false, however, thenthere is no need for adaptive management.

(4) There must be sufficient stability to measure long-term outcomes. Lee is referring to institutional stability,and such stability is a fundamental goal of HCPs. An HCPis formalized through a binding contract, the implemen-tation agreement that typically lasts many decades oreven a century (e.g., Washington Department of NaturalResources 1997 ). The services claim to enforce the con-tract through compliance monitoring of the permittee( U.S. Fish and Wildlife Service & National Marine Fisher-ies Service 1998). Although implementation agreementsusually include provisions allowing a permittee to termi-nate the contract, such provisions often require mitiga-tion for incidental take that has occurred under theHCP. This acts as an effective disincentive against termi-nation. Adaptive management may not be feasible forshort-term HCPs.

(5) It must be possible to formulate hypotheses. MostHCPs are nothing more than a set of management hy-potheses. A plausible hypothesis is built upon existingtheory and fact patterns. Likewise, development of anHCP requires at least a rudimentary understanding of themanaged ecosystem. Adaptive management should bedesigned to experimentally test the implicit hypothesesof an HCP. This condition should be met for a legitimateHCP. Again, an HCP must “not appreciably reduce thelikelihood of the survival and recovery of the species inthe wild.” This test would be difficult to administer with-out basic information about the effects of habitat man-agement on the listed species. If the fifth condition is notmet, then implementation of an HCP should be deferred.

(6) Theory, models, and field methods must be avail-able to estimate and infer ecosystem-scale behavior. Arudimentary understanding of the managed ecosystem isneeded to develop the monitoring and evaluation com-ponents of adaptive management. The most importantthings to monitor and how to monitor them must besomewhat apparent. An ability to identify indicator vari-ables and design efficient sampling schemes will enable

26



more effective monitoring. Evaluation should be basedon specific biological objectives. If adequate understand-ing is lacking, then an “adaptive monitoring design”(Ringold et al. 1996) could be considered.

( 7 ) Decision-makers must be aware that they are ex-perimenting. And, I would add, they must be aware ofthe potential hazards associated with those uncertain-ties. By “experimenting” Lee means that actions haveuncertain outcomes. With uncertainty comes risk, riskthat an HCP will jeopardize the survival of a species. Sci-entists rarely have the opportunity to convey uncertain-ties to decision-makers, and decision-makers often fail tofully appreciate the potential hazards stemming from un-certainty. This condition might be satisfied through a re-quirement to explicitly express the uncertainties associ-ated with an HCP, and the mechanism to do so exists inSection 10(a). Again, an HCP must “not appreciably re-duce the likelihood of the survival and recovery of thespecies in the wild.” This requirement implies a risk as-sessment, but many if not most HCPs do not present riskassessments for the survival or recovery of a species(Smallwood et al. 1999; Smallwood 2000). This condi-tion also entails a public relations problem for the appli-cant. Citizen groups may be unwilling to accept anyrisks associated with adaptive management ( Walters &Holling 1990; Lee 1993). Hence, admitting uncertainty, akey step toward adaptive management, may be per-ceived by the applicant as creating vulnerability. In thecontentious atmosphere surrounding some HCPs, appli-cants negotiating in good faith may be unwilling to admituncertainty when third parties can use it to their advantage.

(8) Organizational culture must encourage learningfrom experience. And, I would add, the culture mustvalue reliable information. By “learning from experi-ence,” Lee means active adaptive management. Al-though nearly all natural resource managers professtheir faith in learning from experience, few institutecredible monitoring programs for doing so. Many man-agers obtain most of their information from casual obser-vations and unreplicated case studies. Both sources arebound to be biased or misleading. An HCP applicantmust believe that an important measure of success is ac-quisition of reliable information. Learning from experi-ence is critical to success, but Lee (1993) cautions thatadaptive management that conflicts with economic ob-jectives will rarely be considered.

(9) Resources must be sufficient to measure ecosys-tem-scale behavior. At present, this condition is gener-ally not met by HCPs, but stricter adherence to Section10(a) would make it so. An HCP must “ensure that ade-quate funding for the plan will be provided.” “Adequatefunding” usually includes funding for monitoring, andthat monitoring could be done in conjunction withadaptive management. Monitoring that provides reliableinformation for adaptive management is likely to be ex-pensive, which discourages adaptive management.

(10) Decision-makers must care about improving out-comes over biological time scales. Lee’s tenth conditionaddresses two psychological factors, motive and pa-tience. A private landowner’s main motive for an HCP iseconomic; the applicant desires a positive economicoutcome. An HCP applicant may or may not care aboutbiological outcomes, but an applicant will be persuadedto care if economic opportunities are linked to habitatconservation. Patience is the cardinal virtue of adaptivemanagement; research and monitoring may take years oreven decades to produce results. Therefore, the appli-cant must be persuaded through incentives or forcedthrough regulations to be patient.

Conditions 1–4 are generally met by HCPs. Conditions5 and 6 should be met for an HCP to be legitimate andscientifically based. The biggest impediments to basingHCPs on adaptive management are conditions 7–10. IfLee (1993) is correct, then as long as these conditionsremain unsatisfied HCPs will lack adaptive management.The problems posed by condition 7 can be diminished if(1) decision-makers and the public become educatedabout uncertainty in natural resource management, and(2) scientists insist that scientific uncertainties are con-sidered thoughtfully and addressed adequately. Withoutstricter regulations or economic incentives, conditions8–10 will remain unsatisfied and HCP applicants will re-sist the cost of adaptive management.

Incentives for Adaptive Management in HCPs

To encourage adaptive management in HCPs, they shouldbe negotiated such that reducing uncertainty results intangible benefits to the HCP applicant. I propose two ap-proaches for doing this. The first relies on various formsof compensation for reliable information. The second de-pends on an entirely new way of structuring HCPs.

As explained above, Section 10(a) of the ESA containsan economic criterion that requires the applicants tobear the maximum cost that they can practicably afford.This maximum cost, whatever it may be, establishes thetotal financial resources available for the HCP. When as-sessing this maximum cost, an applicant undoubtedly in-cludes the costs of research and monitoring. Therefore,funds spent on research and monitoring are funds un-available for habitat protection. On the other hand,funds spent on habitat protection are funds unavailablefor adaptive management.

The lack of adaptive management in HCPs may be a ra-tional outcome of economically based decisions madeby the applicant. Reliable information needed for adap-tive management comes at a cost to applicants, but thefull economic value of that information is not returnedto them. The services, other government agencies, andother HCP applicants will use the information at no cost.In the market of scientific information, a permittee’sHCP generates a positive externality. That is, adaptive



27

management releases information as a byproduct, andothers benefit from it at no cost. In theory, an economicexternality leads to an inefficient allocation of resources( Herfindahl & Kneese 1974; Pearce & Turner 1990); inthe case of HCPs, it leads to insufficient resources di-rected toward the acquisition of reliable information.

These issues raise the following questions: How canmore and better adaptive management be encouraged inHCPs? How should the total cost of an HCP be split be-tween conservation measures and adaptive manage-ment? What is the value of information that helps con-serve an endangered species?

More and better adaptive management would be en-couraged by compensating permittees for the reliable in-formation they produce. Several types of compensationare possible: direct payments, tax deductions, and mitiga-tion credits. Each would award fair compensation for in-formation that benefits the conservation of a species. Di-rect payments could be accomplished through grants fromfederal agencies and would cover the permittee’s costs forresearch and monitoring. Tax incentives for adaptive man-agement would require modifications of the current taxcode. Either payments or tax deductions would reduce thepermittee’s costs for adaptive management, and, accord-ing to the “maximum extent practicable” criterion, the sav-ings must be redirected toward conservation measures.

Mitigation credits allow a permittee to do less habitatconservation in exchange for reliable information thatbenefits conservation efforts across a region. In effect,mitigation credits are an accounting mechanism for thefinancial resources allocated to adaptive managementand not allocated to habitat protection. The creditscould be awarded as information is produced or as stepsin its production are completed. Mitigation credits forinformation would lead to better adaptive managementif, and only if, the services (1) enforce the “maximumextent practicable” criterion, (2) are informed about theapplicant’s costs and expected profits from the HCP, (3)correctly value the information to be obtained, and (4)negotiate a fair exchange of mitigation credits for reli-able information. The first condition is already requiredunder the ESA, but—as discussed earlier—enforcementhas been lax. The second condition is needed to enforcethe first, but the services do not currently require finan-cial disclosure by the applicant. Correctly valuing infor-mation, the third condition, poses an enormous chal-lenge, especially when research or monitoring may takedecades to yield useful results. Analytical methods forvaluing information in natural resource managementhave been studied ( Walters & Green 1997 ), but muchwork remains to be done.

Generally, the services do not consider research to bea form of mitigation ( U.S. Fish and Wildlife Service & Na-tional Marine Fisheries Service 1996). Presumably, theservices have the same attitude toward the costs of adap-tive management. Given that the third condition cannot

be met at this time, the services should maintain this atti-tude. But the services must face the fact that funds spenton adaptive management are funds unavailable for habi-tat protection, and they should explore the potentiallong-term net benefits of exchanging habitat protectionfor reliable information. The information obtained fromone HCP can be applied to future HCPs and other conser-vation programs. The amount of research and monitoringneeded by future HCPs could be reduced and the effi-ciency of conservation measures could be increased. Asrequired by Section 10(a), the financial savings would beredirected toward more habitat conservation.

The second way to encourage adaptive managementin HCPs relies on the “precautionary polluter pays prin-ciple” or “4P” (Costanza & Cornwell 1992). The 4P is apractical formulation of the precautionary principle.There is no canonical precautionary principle, but mostexpressions of the principle make two demands: (1) hu-mans must be extremely risk-averse when using naturalresources, and (2) the burden of proving minimal riskmust be shifted to those who exploit natural resources.The 4P approach has been promoted as a rational meansfor dealing with scientific uncertainty (O’Riordan & Jor-dan 1995; Gullett 1997), but implementation of a precau-tionary principle remains controversial (Bodansky 1991;Dover & Handmer 1995; Gray & Brewers 1996; Morrill1996; Santillo et al. 1998). Much of the controversy sur-rounds the subjective nature of risk. The 4P approachwould use economic instruments, in the form of an envi-ronmental assurance bond, to control the behavior of anHCP applicant and to objectify risk. The size of the bondwould equal the costs of potential worst-case damages.A portion of the bond would be returned, with interest,after the HCP applicant demonstrates that worst-casedamages are less than had been expected originally. Ulti-mately, the entire bond could be returned with interestwhen it is demonstrated that an HCP will not result inany unacceptable damages. Such a demonstration woulddepend on a satisfactory reduction in uncertainty, whichmight be obtained only through adaptive management.

The 4P approach reverses a common scenario in HCPs.Some HCPs require additional mitigation from the permit-tee if new information demonstrates a need for it (e.g.,Washington Department of Natural Resources 1997 ).Clearly, this creates a disincentive against research andmonitoring by the permittee because such an investmentmay result in additional costs for future mitigation. In con-trast, the 4P approach affixes a positive economic valueto information. A reduction of uncertainty enhances eco-nomic opportunity for the permittee.

The 4P approach forces HCP permittees to act cau-tiously. If they make a mistake due to a poor understand-ing of the ecosystem, then they lose their bond. Withbetter understanding, the services return a portion ofthe bond and may assent to further use of a natural re-source. Better understanding can also lead to innovative

28



management practices that protect species and theirhabitats but still allow economically practicable re-source use. In this respect, the 4P approach is differentfrom other bonds recommended for HCPs. Hood (1998)recommends a performance bond to cover the cost ofadditional mitigation that may be necessary over theterm of an HCP. As described, this bond offers no incen-tive for adaptive management because it is not returnedto the permittee as a reward for better information or in-novative management practices. As with Hood’s (1998)bond, the 4P approach might resolve misgivings mem-bers of the public feel toward HCPs. The public shouldbe more willing to accept the risks of an HCP when abond is posted to pay for damages.

The main problem with the 4P approach is that it re-quires an amendment to the ESA. Neither Section 10(a)nor its legislative history say anything about bonds. Butin an era when other environmental problems, such asair and water pollution, are addressed through eco-nomic solutions such as pollution taxes and tradable pol-lution-discharge permits, the 4P approach is within therealm of possibility. Another problem arises from thecost of precaution. As discussed earlier, deferring re-source use imposes an economic cost on the permittee,who could pass this cost on to society by reducing habi-tat conservation in the future. In conjunction with themaximum-extent-practicable criterion of Section 10(a),the 4P approach trades precaution in the short-term forhabitat over the long-term. Because the bond generatesmonetary interest, however, habitat lost due to the cost ofprecaution could be minimal (Costanza & Cornwell 1992).

The human activities, time span, and land area cov-ered by HCPs vary widely, so the proposed economic in-centives may not be applicable to all HCPs. The informa-tion obtained from HCPs covering a small area (e.g.,

�

100 ha) or a short time span (e.g.,

�

10 years) may havelittle value. On the other hand, information obtainedfrom large HCPs in which ecosystem-scale behavior canbe studied should have considerable value. The 4P ap-proach may not be feasible for HCPs developed by smallbusinesses, because the cost of an assurance bond equalto worst-case damages would be too much to bear (Cos-tanza & Cornwell 1992). Therefore, the 4P approachmight be most appropriate for large commercial forests orlarge real-estate developments, which Costanza and Corn-well (1992) claim can be regulated by a 4P approach. Ap-propriate incentives will be different for different humanactivities. Mitigation credits may be inappropriate for HCPscausing irreversible habitat loss (e.g., real-estate develop-ment), but may be appropriate for forest management.

Conclusion

When the Secretary of the Interior issued the no-sur-prises policy, he made a mistake similar to that of theoriginal ESA: there was no mechanism for compromise.

If applicants met specified conditions, then they weregranted “general assurances” that amounted to absoluteregulatory certainty. Fortunately, recent regulations cod-ifying the no-surprises policy increase the potential foradaptive management in HCPs. Unfortunately, the pros-pects for genuine adaptive management remain unprom-ising. The quality of adaptive management in HCPs willbe determined through negotiations, and without eco-nomic incentives applicants are unlikely to adopt costlyexperimental approaches for managing natural resources.

Economic incentives for adaptive management are notideal because they too involve compromise. Direct pay-ments to private landowners may divert funds awayfrom public conservation agencies, and tax incentivesmay reduce government budgets for environmental pro-grams. Awarding mitigation credits entails weighing thevalue of conventional mitigation (i.e., habitat protectionor restoration) versus the value of reliable information.Delays imposed by the 4P approach may lead to con-flicts between the cost of precaution versus the benefitsof more tangible conservation measures.

Most conservation dilemmas are enmeshed in consid-erable scientific uncertainty. Although we will nevereliminate uncertainty, we must strive to reduce it. Asconservation biologists, we need information that in-creases our understanding of managed ecosystems. Weshould also appreciate the substantial costs of acquiringsuch information, especially when dealing with naturalprocesses that operate over large areas and long timespans. Most conservation dilemmas are resolved throughpolitical compromise, and adaptive management may beno different. When financial resources are limited, wemay have to reconcile the competing goals of habitatprotection and reliable information. If we value knowl-edge about managed ecosystems, then we should deter-mine its fair price and find ways to pay for it.

Acknowledgments

Many ideas for this paper were originally developedfor a presentation at the National Center for EcologicalAnalysis and Synthesis (NCEAS). I thank D. Brosnan andS. Courtney for the opportunity to discuss adaptive man-agement in habitat conservation plans at their NCEAS–sponsored forum. J. Buchanan, B. Glass, P. Swedeen, andtwo anonymous reviewers provided insightful commentsand criticism. Editing by E. Main and W. Brown greatlyimproved this paper. Parts of this work were supportedby the Washington Department of Natural Resources andthe Washington Department of Fish and Wildlife.

Literature Cited

Bean, M. J. 1983. The evolution of national wildlife law. Praeger,New York.

Bean, M. J., and D. S. Wilcove. 1997. The private-land problem. Con-servation Biology

11:

1–2.



29

Bodansky, D. 1991. Scientific uncertainty and the precautionary princi-ple. Environment

33:

4–5, 43–44.Costanza, R., and L. Cornwell. 1992. The 4P approach to dealing with

scientific uncertainty. Environment

34:

12–20, 42.Dover, S. R., and J. W. Handmer. 1995. Ignorance, the precautionary

principle, and sustainability. Ambio

24:

92–97.Forest Ecosystem Management Assessment Team. 1993. Forest ecosys-

tem management: an ecological, economic, and social assessment.Report. U.S. Forest Service, U.S. Department of Interior, U.S. De-partment of Commerce, and U.S. Environmental ProtectionAgency, Washington, D.C.

Gray, J. S., and J. M. Brewers. 1996. Towards a scientific definition ofthe precautionary principle. Marine Pollution Bulletin

32:

768–771.Gullett, W. 1997. Environmental protection and the “precautionary

principle”: a response to scientific uncertainty in environmentalmanagement. Environmental Planning and Law Journal

52:

52–69.Halbert, C. L. 1993. How adaptive is adaptive management? Imple-

menting adaptive management in Washington State and British Co-lumbia. Reviews in Fisheries Science

1:

261–283.Haney, A., and R. L. Power. 1996. Adaptive management for sound

ecosystem management. Environmental Management

20:

879–886.Herfindahl, O. C., and A. V. Kneese. 1974. Economic theory of natural

resources. Merrill Publishing, Columbus, Ohio.Hilborn, R. 1992. Can fisheries agencies learn from experience? Fisher-

ies

17(4):

6–14.Hilborn, R., C. J. Walters, and D. Ludwig. 1995. Sustainable exploita-

tion of renewable resources. Annual Review of Ecology and Sys-tematics

26:

45–67.Holling, C. S., editor. 1978. Adaptive environmental assessment and

management. Wiley, New York.Hood, L. C. 1998. Frayed safety nets: conservation planning under the

Endangered Species Act. Defenders of Wildlife, Washington, D.C.Hurlbert, S. H. 1984. Pseudoreplication and the design of ecological

field experiments. Ecological Monographs

54:

187–211.Irwin, L. L., and T. B. Wigley. 1993. Toward an experimental basis for

protecting forest wildlife. Ecological Applications

3:

213–217.Isermann, R., editor. 1981. System identification. Pergamon Press,

New York.Kaiser, J. 1997. When a habitat is not a home. Science

276:

1636–1638.Kareiva, P., et al. 1999. Using science in habitat conservation plans.

American Institute of Biological Sciences, Washington, D.C.Lee, K. N. 1993. Compass and gyroscope: integrating science and poli-

tics for the environment. Island Press, Washington, D.C.Marcot, B. G., and J. W. Thomas. 1997. Of spotted owls, old growth,

and new policies: a history since the interagency scientific commit-tee report. General technical report PNW-GTR-408. U.S. Forest Ser-vice, Pacific Northwest Research Station, Portland, Oregon.

Morrill, W. I. 1996. The precautionary principle, weakening the linkbetween long-term research and management. Transactions of theNorth American Wildlife and Natural Resources Conference

61:

483–490.Moser, D. E. 2000. Habitat conservation plans under the U.S. Endan-

gered Species Act: the legal perspective. Environmental Manage-ment

26:

S7–S13.Noss, R. F., M. A. O’Connell, and D. D. Murphy. 1997. The science of

conservation planning. Island Press, Washington, D.C.O’Riordan, T., and A. Jordan. 1995. The precautionary principle in

contemporary environmental politics. Environmental Values

4:

191–212.Parma, A. M., and the NCEAS Working Group on Population Manage-

ment. 1998. What can adaptive management do for our fish, for-ests, food, and biodiversity? Integrative Biology

1:

16–26.Pearce, D. W., and R. K. Turner. 1990. Economics of natural resources

and the environment. Johns Hopkins University Press, Baltimore,Maryland.

Ringold, P. L., J. Alegria, R. L. Czaplewski, B. S. Muldur, T. Tolle, and K.Burnett. 1996. Adaptive monitoring design for ecosystem manage-ment. Ecological Applications

6:

745–747.Richter, B. D., and K. H. Redford. 1999. The art (and science) of bro-

kering deals between conservation and use. Conservation Biology

13:

1235–1237.Santillo, D., R. L. Stringer, P. A. Johnston, and J. Tickner. 1998. The

precautionary principle: protecting against failures in scientificmethods and risk assessment. Marine Pollution Bulletin

36:

939–950.Smallwood, K. S. 2000. A crosswalk from the Endangered Species Act

to the HCP handbook and real HCPs. Environmental Management

26:

S23–S35.Smallwood, K. S., J. Beyea, and M. L. Morrison. 1999. Using the best

scientific information for endangered species conservation. Envi-ronmental Management

24:

421–435.Smith, A. D. M., and C. J. Walters. 1981. Adaptive management of stock-

recruitment systems. Canadian Journal of Fisheries and Aquatic Sci-ences

38:

690–703.Thomas, J. W., et al. 1993. Viability assessments and management con-

siderations for species associated with late-successional and old-growth forest of the Pacific Northwest. U.S. Forest Service, Port-land, Oregon.

U.S. Congress, House of Representatives. 1982. Endangered SpeciesAct amendments of 1982. 2nd session, House Report 835.

U.S. Fish and Wildlife Service and National Marine Fisheries Service.1996. Habitat conservation planning handbook. U.S. Fish and Wild-life Service and National Marine Fisheries Service, Washington, D.C.

U.S. Fish and Wildlife Service and National Marine Fisheries Service.1998. Habitat conservation plan assurances (“no surprises”) rule.Federal Register

63:

8859–8873.U.S. Fish and Wildlife Service and National Marine Fisheries Service.

2000. Availability of a final addendum to the handbook for habitatconservation planning and incidental take permitting process. Fed-eral Register

65:

35241–35257.Walters, C. 1997. Adaptive policy design: thinking at large spatial

scales. Pages 386–394 in J. A. Bissonette, editor. Wildlife and land-scape ecology. Springer-Verlag, New York.

Walters, C.J., and R. Green. 1997. Valuation of experimental manage-ment options for ecological systems. Journal of Wildlife Manage-ment

61:

987–1006.Walters, C. J., and R. Hilborn. 1978. Ecological optimization and adap-

tive management. Annual Review of Ecology and Systematics

9:

157–188.Walters, C. J., and C. S. Holling. 1990. Large-scale management experi-

ments and learning by doing. Ecology

71:

2060–2068.Washington Department of Natural Resources. 1997. Final habitat con-

servation plan. Washington Department of Natural Resources,Olympia.

adaptive management in habitat conservation plans

Documents